## Sunday, March 27, 2011

### Speedster Redux and the Dynomometer Puzzle

In this week's episode, we take the Speedster Redux, with it's 57 CALB 180AH cells, brand new Netgain Warp9 improved motor, and Soliton1 controller to task. We do this by putting it on a dynomometer and actually measuring both power IN to the system from the batteries and power OUT of the system on the rollers. We found this exercise illuminating and suspect you will as well.

The bottom line is that we can put about 111kw or 145 HP into the system and we seem to get about 125 HP out the other end.

This would appear to compare very well with the 76 HP we had reported on previous dynomometer tests. And we really should compare Apples to Apples. But we didnt'

And here's why. Our earlier tests have been primarily of interest to me to learn how much current, and how much voltage, comprising how much power, was REQUIRED to go a certain speed and RPM on the road. As such, we would load the dynomometer with the approximate weight of the vehicle, and then go to speeds in ten mile increments in each gear.

At that point, we measured RPM, HP out, Torquie out, and voltage and current IN. Of course, we're lookihg for power consumption at these various speeds and RPMs with an eye on our battery capacity and potential range. We all know that you get MORE miles at 35mph in town than you do at 80 mph on the freeway. Those dynomometer tests, lacking a wind/airpressure component, were illuminating.

This time, we're really interested in something else entirely. We face the daunting task of the Winston Battery Elescalade project. This is gonig to require us to move about 8200 lbs of stuff down the road. We have a couple of strategies to do that.

First, we're going to go to a larger diameter motor AND we are going to use TWO of them. Why does motor diameter matter? It is not precisely that it is capable of more current or voltage. It's not capable of more voltage and the differences in current capacity are nominal.

The power of an electric motor is applied at the air gap. This is a very small gap between the armature, or rotor, and the stator - the field windings. At that point, two magnetic fields collide and being of opposite polarity, they exhibit an irrestably repulsive force.

But we take the output of the motor at the shaft, which is in the dead center of the motor. The amount of "torque" we get is then to a very persuasive degree a function of how much leverage we have between the point of most intense magnetic flux interaction, ie the air gap, and the center of the shaft. The SAME force caused by the SAME voltage and current, will appear as a much higher torque value if it has a longer arm. And for this reason, the 11 inch motor is more powerful than the 9 inch motor, and similiarly a 13 inch motor would be more powerful yet.

At the same time, the longer the arm, the faster the surface of the rotor must travel for any given RPM. And so the higher the centrifugal forces that seek to tear the motor asunder and turn your drive train into a claymore mine. Primarily, with DC series motors this is a function of the ability to hold the commutator together. We observer a nominal limit of 6500 rpm, while AC motors, which do not have this commutator, are often rated to 9000 or even 11000 rpm. They don't really make any power at those lofty RPMs at all, but they don't explode either.

So while the 13 inch motor is impressive in torque, we start to get into some RPM limitations that are not to our advantage.

In the case of the Elescalade, we went to a PAIR of 11 inch motors, instead of a single 13 inch motor.

THIS allowed us to do a couple of things. For one, by using TWO controllers, one for each motor, we can DOUBLE the amount of power our controllers can handle. In this case, we selected the Soliton1 for a couple of reasons. It has an "idle" feature that is quite well thought out and let's us use an automatic transmission. And it purports to handle 300kw - 1000 amps at 300v.

The 11 inch motor is pretty much limited to 192v. Traditionally, Netgain motors have been limited to 170v to prevent arcing of the commutator. Helwig insists that their new Redtop brushes are arcless at up to 192v. And so that is what Redux, really a dress rehearsal in many ways for Elescalade, is powered with.

The Soliton 1 won't of course do 300 kw at 192v. But with two of them delivering 1000 amps and assuming we sag to 150v applied while producing 2000 amps out of our 400Ah cells, we would have a 300kw drive train. 150 x 1000 x 2 = 300,000.

We are also plowing some new (for us of course) ground with the automatic transmission. We then become interested in the concept of efficiency. Efficiency is for these purposes a comparison of the amount of power IN to the drive train compared to the amount that shows up at the wheels. If we put 100 HP of electrical power IN to our drive train and get 85HP to show up on the dynomometer, we would be 85% efficient. There are GOING to be some loses in an automatic transmission. If we could chart our efficiencies using the Soliton, a 9 inch Netgain and a manual transmission, we could expect the efficiencies of the 11 inch and Soliton to be quite similar on an automatic transmission. And so the difference in efficiency between the Elescalade and the Redux would be MOSTLY isolated to the use of hte automatic transmission. That difference would be intensely interesting to me.

The best laid plans of mice and men go aft aglay....

So on THIS series of dynomometer tests, we are interested in something quite different. First we load the dynomometer to 5500 lbs. Then instead of going to specific speeds, we simply tromp the pedal to the metal. Between the increased rate, and the constant acceleration, limited in rate only by what the drive train will do, we can get maximum power at various speeds and rpms. Not how much is REQUIRED to maintain that speed and rpm, because we are not maintaining it, in fact we are accelerating THROUGH it.

This poses some problems in capturing current and voltage. That we attacked with a video camera. We can go to any point on this acceleration curve on the dynomometer data and retrieve torque, horsepower, and RPM for any given mph speed. We pick the usual 10/20/30/40/50/60/70/80/90/100.

Then with a video of the meters on the car, we can go back and get voltage and current off the meters for the resulting RPM values.

THAT gives us our voltage, current, kw power, and indeed HP IN to the drive train. By comparing HP in to HP out, we get efficiency.

Here was the problem. We did this FULL RPM range acceleration in ALL FOUR gears. The most current we ever measured was 755 amperes at 3900 rpm in fourth gear.

IF this holds true on the Elescalade, we're looking at a total power input of not 300kw, but more like 225kw. And I'm trying to move 8200 lbs with a drive train slightly more powerful than the Tesla Roadster.

Oh, it will drive fine. But it won't do any 7 second 0-60 or impress anyone. Nothing to apologize for, but 12 seconds more like.

I did confer with EVnetics about the shortfall and we haven't come up with much yet. Mr. Jenkins insists that they MOTOR CURRENT 1000AMPS inside the controller is what they are talking about when they claim 1000 amps and that this is an industry standard widely observed with ALL DC controllers and some magic magic explanation about the difference between motor amps and current amps that I'm not smart enough to make out. With the pedal floored and the car accelerating uphill, they should all be the same.

I posed this question to Otmar of Cafe Electric regarding the Zilla 1K. His replay was no, at full power it does 1200 amps motor current for the 1000 amps battery current but that this difference disappears when the ripple goes away at 100%. So no industry "standard" that I can tell.

We also have measured in all cases MORE current than spec on our AC controllers, the Curtis, the TIMS600 and the Rhinehart Motion Systems.

There remains Seb's argument that we are not really loading the motor. I don't know how much more we can load the motor. It is doing all it can under constant acceleration and at 5500 lbs. If it could do more, we should accelerate faster - that's all. His contention is that if we bring the motor to a stall we'll see the 1000 amps.

No thanks. No stalled motor with maximum current. I am starting to understand their absolutely mystifying score of 7:0 vs the Netgain Motors. They've melted 7 motors without killing a Soliton. Impressive, but I don't want to blow up my motor. I want it to make 1000 amps at 150v while driving the car.

We do cover the Soliton settings in this video. We publish the data. Make me smart. Make me go. Make me more power. I'll be your buddy. I don't know how to make it go harder than this. If it would do more, it should accelerate faster - not demand to be stalled.

Ergo our 9.00 second mile. We're really putting about 125 HP down on the road. If we could do 1000 amps, that figure would be more like 165HP and we probably would be down around 7 seconds, if not 6.

I view it as an open question and a mystery at this point. I still love the Soliton1, but can't make it do those power levels we were hoping for. If you can, chime in.

BTW, as several of our viewers have noted, apparently Otmar and Cafe Electric have come to terms with a production facility and the Zilla will be back in production - of all people, Manzanita Micro. I'm not very popular with them either. Rumor has it ANOTHER fire with the Rudman Regulator - all very secret of course so I can't mention it.

Jack Rickard

1. Jack,

Doesn't the current depend on the voltage difference and the resistance of the circuit? Could it be that you are not seeing 1000A because of the physics of the situation and not a limitation of the controller design? Could you actually get 1000A of current if you hooked the batteries directly to the motor, bypassing the controller? I'm not talking about the motors being at 0 rpm.

2. I have two suggestions

Slew rate
Instrument lag

One think in particular does not look right: I would expect torque and current to stay in step. To have approximately flat torque curves with rapidly rising current looks wrong. So I suspect that real currents are higher than what the ammeter needle is saying, but still not all that it should be because the slew rate limit is holding it back

3. "One thinK in particular" = "One thinG in particular"

4. Gizmo:

It does not precisely depend on the "resistance" of the circuit. No. A funny thing happens on the way to Ohm's law when it comes to motors. The appropriate term I suppose is REACTANCE. There is a back force against the current caused by the interaction of the fields in the motor. Those coils have very tiny "resistance" values but the inductive reactance, or resistance to changes in current flow, are sizable - a result of the amount of work being performed by the motor, and the counter electromotive force created by motor rotation.

The confusion is quite understandable. And I really do not have a truly excellent explanation of electric motor physics to offer in any simplified form.

Jack

5. Well, we can and will certainly change the slew rate. But that should simply be a function of how FAST we change current in reaction to changes in the pedal, and we have the pedal full on. We should be able to reach full current then in 200 ms.

Jack Rickard

6. re slew rate: if you were to replot the current curves with time on the horizontal axis instead of RPM and if the curves were roughly together up to the knee regardless of rpm, I'd suggest you might have a smoking gun.

But I do wonder more about the mismatch between torque and current curves

7. Jack, seeing as I have some experience building a controller I think I know what's going on. would you like me to tell you? I ask because when I volunteer intelligent feedback you spit in my face and that gets old so maybe if you ask me you will appreciate it more.

8. also, I take it you put the current measuring on the battery side, not the motor cables?

9. Looking at the jump in current for 4th gear maybe you needed 5th to hit 1000amps? Which would seem to fit with the not enough load theory.

10. Jack,

I work at a utility company and in the summer when we have a highly reactive loads from all of the air conditioner motors, we put in capacitor banks to bring the current and voltage back into phase. I actually worked at a plant close to a major load that adjusted "VARS" on the system. I am not an electrical expert but would it be possible to have a capacitor bank that would be added to the circuit to help bring the current and voltage back into phase?

11. Jack, admittedly I'm a bit lagging behind understanding the problem. I'd appreciate some clarification for the simple minded.

I would expect the Amps gradient (of any electric motor) in the charts to climb much faster to Amax and then fall off slowly in relation to rpm. But maybe the charts show whats typical for the Netgain?

I also wonder why you reach different Amax levels depending on gear, the higher the gear, the higher Amax.
Probbably related to that, the torque is different in all gears, which I would expect to be more or less constant at the same dyno setting accelerating the 5500lbs faster in lower gears.

So do I understand your complaint correctly that the spec sheet of the motor actually promises 1kA at 150V at such and such torque, and the controller somehow chokes the motor?

I thought current drawn is a funktion of torque demand of and voltage provided to the motor.
So the simpleton in me thinks its just a problem of too little a load on the motor?

12. Ok, this is going to be a long comment. First; the Soliton 1 is limited to 1000 Amps MOTOR current for a reason. This is typically how buck-converters are built (no matter if they're a motor controller for an EV, a PSU in a computer or whatever) because it's the output current that will kill the transistor if the limit's exceeded.

The pulse wave can vary between 0-100%, which could also be described as the variable D (for duration) that can have a value between 0-1. Since a controller is a buck converter it means that D will control the proportion between in- and out voltage. So:

Umotor = Ubattery * D

The controller doesn't control the resulting current, the motor does! The motor current will be:

Imotor = (Umotor - backEMF) / Ri

Where Ri is the internal resistance in the motor. When the motor is at zero RPM back-EMF will be zero so the only thing that limits the current is your Ri. If you'd get full pulse width (D = 1) it means that motor Voltage will be full pack voltage and since the internal resistance is just a few tens of milliOhms it means your motor current will sky rocket to some thousand amps and the controller will blow up.

Therefore throttle doesn't control the pulse width and the Soliton won't give you 100% pulse width unless it's safe. Instead what the Soliton does is control the motor current so your throttle controls the torque of the motor and it does that by adjusting the current 100 times/second to match the desired current (throttle) with the motor current.

Now, since the controller literally convert power to power (ie Pbattery=Pmotor + some losses) it means that:

Pbattery=Ubattery*Ibattery=Umotor*Imotor=Pmotor!

But since Umotor=Ubattery*D this can be rewritten to:

Ubattery*Ibattery=Ubattery*D*Imotor

Or simpler:

Ibattery=Imotor*D

And since D can never be greater than 1 it means that Ibattery is always less or equal to Imotor. This means that Ibattery always has to be <=1000 Amps (with your settings) and if you don't reach 1000 Amps on the battery side it probably means that your voltage over the motor never reaches the full pack voltage.

So when you hit full throttle at 0 RPM you get a motor Voltage of approximately 30 Volt and this means that even with 1000 Amps motor current you'll likely see something like 150 Amps from the pack despite the motor current spikes up to 1000 Amps. If the controller were to give you 1000 Amps on the battery side, it'd mean that the motor current would hit 6.4kA (ok, less due to sag) and that would make the silicon in the Soliton burn up. Literally. We know from experience.

I have the fullest respect for Mr Otmar and I can't verify or falsify his claim that the Z1k can peak at 1200 motor Amps, but I know from real life data collection that the Z2k does, indeed, keep the motor current at a maximum of 2000 Amps. You can see it yourself for example on this graph from Waylands site:

http://www.plasmaboyracing.com/graphics/graph10.jpg

Between 3-6 seconds of the running time it's very visible that motor current hovers around 2000 Amps but it takes quite a while for battery current to peak 2000 Amps and pretty immediately afterwards the motor- and battery current falls off completely synchronized. What you also can see between 3-6 seconds is that motor voltage, battery current and RPM rise pretty proportionally.

It's next to impossible for me to speculate why you don't get more than 225kW in your current setup without you providing me with actual logs from the Soliton, but we know that the Soliton 1 provides 1000 true Amps. On the MOTOR side! The only time you will see 1000 Amps on the battery side is when you run out of Voltage, and that means you're starting to run out of torque because of back-EMF!

Provide me with logs from the Soliton and I'll help you analyze the situation.

13. "but I know from real life data collection that the Z2k does, indeed, keep the motor current at a maximum of 2000 Amps."

Unless you do a "Plasma Boy cabling change" John talks about here: http://www.plasmaboyracing.com/blog/?p=239

Be sure to read both parts of the blog. He tells how they arrive at the ~3000A figure from a Z2K and it doesn't blow up. John Wayland is an entertaining writer.

Jack: thank you for the response. You basically said what I was trying to but I didn't use the proper terms.

Martin, thank you for that post. It clarified my basic understanding of the situation.

14. ok... So.. next week Mr. Brain will take the Spyder to the Dyno and swap the controller with the one with 650A... and so on...
So, why don't you put Mr. Matt to swap the Soliton1 with one of your Zilla controller that I am sure that you have at least one 1k and one 2k on your junk pile...
That would kill all those doubts about which does which doesn't.

15. We don't need to kill any doubts and don't appear to have many. I've published what happened, and if the answer is go do it again and send me something else, we're going to pass.

We'll do the Spyder to compare the Curtis 7501 output to 7601 performance. Somehow, I don't think it will be the motors fault there either.

Jack Rickard

16. We do have a couple of Zillas on hand actually. I haven't done much with them because you all haven't been able to get them. If they do indeed get back into production, we'll undoubtedly do a project with one of them. I don't see much purpose in titillating our viewers with tales of our adventures using equipment they can't get.

The point was raised that there is a valid industry concensus on how controllers are advertised and that the Soliton1 1000A claim was to comply with that "industry standard".

After consulting with Ryan Bohm of Netgain Drives and Otmar of Cafe Electric, it is clear that this is nonsense.

The AC motors routinely exceed their current claims. Somehow, the concept was posited that DC series controllers are somehow different and by convention listed motor amps as somehow the definitive parameter.

Apparently this is bullshit.

If anyone has an viable suggestion for a change in operation to increase this to the advertised levels, I would love to hear it. An endless series of "yes but" requests that we connect things differently and rerun the test is actually of zero interest.

Jack Rickard

17. what limits have you set for battery current and low voltage?

the graphs look alot like you are measuring battery current:
battery current is roughly propotional to power
motor current is roughly proportional to torque.

18. All of the limits are discussed and actually scrolled up the screen in the video. If you haven't watched it, the value of your input might be a little limited.

Your definitions make no sense to me - roughly.

Power is a function of both battery current and voltage - rather precisely in the case of DC.

Torque is the result of power applied to an electric motor. Actually, that is precisely what it is, torque to produce rotation. Rotation over time is measured as RPM. And by "convention" to account for both torque and time we have horsepower.

The graphs look a lot like graphs.

Jack

19. What did you have programmed into the Soliton for a minimum pack voltage? If you hit that it will effectively limit current.

The 4th gear graph looks like what I expect to see. Torque is flat as the amps rise, up to peak power. Then everything except rpm starts going down. I suspect things where happening a bit to fast to accurately capture the peaks in 1st and 2nd gear, complete with a bit of lag.

20. Hello, I'm Sebastien from Evnetics.

To be completely clear the Soliton1 is a 1000A controller both on the motor and battery pack side.

Why Jack's car is only seeing 750A on the battery side is likely because:

1. As many of you pointed out, not enough load. In Jack's graphs the net power increases as the gears increase, so it would stand to reason that a fifth gear (or additional load in existing gears) would yield increased horse power, ie battery Amps.

2. The controller is reaching a user limit. From the setup data Jack has provided, that seems unlikely, although we really haven't gotten a logfile from the controller itself.

3. The controller is thermally de-rating. Again, doesn't sound likely, but we do not have a supporting logfile to show either way.

4. The controller's internal calibration is out of whack. Jack's controller is an older version of our current sensing strategy, which is less accurate (in either direction) than the current generation controller. So that's a possibility.

5. The battery pack can't put out anymore than 750A. This is unlikely to me as the sag is still under control at this amperage.

In any event, I would love to fly in and get to the bottom of this. I see our controllers perform to their full rating envelop every day, and I see no reason why Jack's, or anybody else's shouldn't do the same. Please let me know if this would work for you guys.

Best regards,
Sebastien Bourgeois
seb@evnetics.com

21. I apologize of this gets posted twice!

Could it be that when you have the pedal fully pressed (engaged) you are not actually sending a signal that indicates you want full power to the controller?

22. Anonymous said:

"battery current is roughly propotional to power
motor current is roughly proportional to torque"

That seemed odd to me too when I first read it, but actually it makes sense with one major caveat. The power output from the batteries is current x pack voltage. Neglecting voltage sag (hence "roughly") pack voltage is a constant, so power is proportional to current.

On the motor side, neglecting windage and other losses ("roughly" again), the torque output of a given DC series motor is dependent on current, regardless of the combination of applied voltage, back emf, resistance and reactance that led to that current. Torque is directly related to the strength of the magnetic fields in the motor, and the strength of the magnetic fields is proportional to the current. However in a series motor doubling the current roughly doubles the strength of both rotor and stator fields so torque is ~proportional to the square of motor current.

That also comes back to a point I attempted to make earlier: something is wrong with the dyno graphs. I do not believe a graph that shows a huge variation in current with little change in torque.

23. Jack, could you connect an oscilloscope to the output of the controller and take a look at the voltage waveform? If so, wouldn't that answer the question on what the controller is actually doing and if the lower current is due to the controller or the motor / load?

24. @Gizmo: "Unless you do a "Plasma Boy cabling change"..."

Yep. It's of course possible to do dirty tricks with the Soliton as well if you're prepared to void the warrant. But by obvious reasons it's not something I'd recommend. ;)

@Jack Rickard: "After consulting with Ryan Bohm of Netgain Drives and Otmar of Cafe Electric, it is clear that this is nonsense."

I quote from the OwnerÊ¼s Manual for the Zilla Motor Controller Package with Hairball 2 Interface", page 2:

"The Zilla line of controllers includes six models with current ratings of 1000 and 2000 amps..." And in a table right below that sentence it's clearly stated that it's "Maximum Motor Amps" that's 1000/2000 Amps.

On page 3 one of the features listed is "2000 motor Amps available with proper cooling for Z2K, 1000 Amps for Z1K." There is no mention of battery amps more than how to limit it to protect the pack. The manuals for Netgain WarpDrive and Curtis 1209B/1221B/1221C/1231C don't specify which amps they're talking about which unfortunately opens for misunderstanding and confusion.

From a technical point of view it doesn't make sense to specify battery current instead of motor current as the limit for a DC-controller since it has very little with the actual technical limitations of the hardware. So I'd be very interested to hear what Ryan Bohm and Otmar told you.

@Jack Rickard (again): "If anyone has an viable suggestion for a change in operation to increase this to the advertised levels, I would love to hear it."

Yes. Provide me with logged data from the Soliton when it's working hard. I don't care if you make it sweat on a dyno or just slam the throttle on an empty road, but I need raw data from the actual controller to analyze.

The dyno graphs shows what you get, but they don't give any hints why.

25. Hi Jack I would like to know why too. I have the same thing happening on my car the motor see's 500 to 550A but the best l get out of my batteries is about 450A on an extremely well cooled curtis controller while driving up a big hill. The motor current hovers around 500-550A and the battery curent starts @150A and increases to 450A as speed increases then drops off similar fashion to your graphs. Its not a battery thing, tried different makes and capacities over the last decade something with the controller same thing happens on mates car with Altrax controller. Even tried 70mmsq cable no difference.
Thanks Leigh
Ps:- keep up the good work experience counts theories change to accomadate reality.

26. No changing of theories to accommodate reality here, Leigh.

Jack seems to understand that voltage x current = power, and that a PWM motor controller will only ask for what it needs to satisfy the power demanded by the motor.

Jack also seems to understand that RPM is proportional to voltage for a given load and that torque is proportional to current for a given RPM (and torque x RPM = power, just like current x voltage = power).

What I find extraordinarily puzzling is that Jack seems to not be able to connect these two things together. That, for example, it takes 100V to maintain 2000 RPM with 1000A of torque loading on a WarP-9 motor. To increase RPM to 3000 while maintaining current at 1000A will require 50% more voltage, or 150V. That's more or less what we've measured ourselves.

What this simply means is that the dyno needed to be set to a higher torque load so that it would need 150kW of power from the controller BEFORE the battery pack runs out of voltage.

If Jack (Brian, actually) had actually bothered to run the data logger program they would have instantly known whether the throttle was asking for 1000A (if not, maybe because of mis-calibration or a bind in the cable?), whether motor current was making it to 1000A, whether duty cycle was making it to 100%, etc.

Hell, just moving that Snap-On current clamp over to the motor side would have quickly verified if motor current was making it to 1000A.

Finally, it is trivial for anyone to verify our performance claims - all you need is a DC clamp meter - so why in the world would we lie about them? This just doesn't make any sense at all, but it didn't stop Jack from calling me a liar anyway. Thanks, Jack. Much appreciated.

27. Jeffrey:

You say "...torque is proportional to current for a given RPM...".

I think with a series motor this should be

"...torque is proportional to current SQUARED REGARDLESS of RPM..."

28. John Hardy - that is true up until the field enters saturation, at which point torque increases linearly with current. Saturation in a WarP-9 motor seems to occur around 250A, IIRC, so at 1000A it is deep into saturation.

29. Jeffrey:

First let's get it on here buddy. This is NOT DIYElectric. If you want to be all butt hurt about someone "calling you a liar" and get into a flame war, trot your ass right on back to DIYelectric and the other kiddies and have a party. If you want to talk about reality, we're very interested both in your controller and its performance and there may indeed be a solution here. But you're going to have to grow up and play with adult people if you want to be a part of that conversation.

You, and your company, have made a product claim that I cannot verify on a car. If you can make it do so, I'd love to hear about it. If you want to personalize it into he said she said and defending your deep honor as a truth teller, I just don't have time for it and I really don't want the blog cluttered up with that crap. Right now, the claims for performance on the SOliton 1 do not appear to be accurate. That could be for a variety of reasons.

1. We may have it installed incorrectly.
2. We may have it configured incorrectly.
3. It might be defective.
4. It might just not do that level of power.

That can derive from a number of things, but in this case I rather suspect that it is this splitting of hairs over your switching currents in the device itself, rather than viewing it as a component of a car. And if so, it is certainly my intention to make that QUITE clear to our viewers.

It doesn't even make it a bad controller. But in designing/planning your car, a cognizance of what it will actually do is quite important. We're heading for a pretty serious shortfall in power in the Elescalade actually times two, by way of example.

I have shared this data both before and after. Instead of answers or a suggested change in configuration, I THINK what I'm hearing is that I need to LEARN to tell your story the same way you tell it. I need to "view it differently".

Pass.

As to loading, you're correct. We don't understand what you all keep saying. The "load" on the dyno was 5500 lbs. That is substantial but it isn't hte point. It should actually work the same way at 2500 lbs.

If we floor the pedal, the power produced will be exhibited by the acceleration. If it would produce more power, we would accelerate faster. If we loaded it less, we should accelerate faster yet.

YOU calculated we should be doing 0-60 in six seconds. We do it in 9.00 seconds. This is akin to saying it would produce more power if we had a heavier car? That makes no sense. We would be forever stuck at 9.00 seconds in that event.

NOW STEP BACK a minute and think about this. Let's view the controller and the motor as black boxes.

In the case of the motor, rated at 52 hp for one hour is it capable of using 150kw of power to produce torque for 30 seconds?

Demonstrably and historically so.

NOw to the controller. Can it apply 150kw of power to the motor?

Demonstrably not. We get 111kw.

30. We DID measure current before and after the controller with no notable differences in what was read.

Now, can you LOG something to the screen that indicates otherwise? No doubt. Clearly you think it performs to these claims and I have no doubt your "log" will support that. I have no interest in that as it is kind of circular. And as you full well know, we already sent you logs of our 0-60 run. If I understand your response, it was "shift faster to go faster".

Now let's view it as a larger black box consisting of a motor and your controller. Can I make 150kw from the battery pack? Demonstrably so.

Can the black box put 85% of that on the ground - measured as NOT.

I think the batteries will do it, I think the motor will do it, and I think the transmission will do it. I even think the dynomometer will do it. In fact, I think ALL of those components will do quite in excess of that.

The controller appears to be limiting factor. We are measuring 111/112kw with a 191.5v pack. IS your story that it can do more?

And if so, how do I get it to do more?

If your story is that Im' not interpreting this the way you do, I cannot help or hurt any of that.

Jack Rickard

31. Sebastien:

Come ahead. Our car and shop are yours.

Jack

32. Jack,

Thank you for the opportunity.

I'll make some travel arrangements, and keep you posted.

Seb

33. I think Mr. Persson is exactly correct in his analysis. He could add that motor torque is directly proportional to motor current and back emf is directly proportional to motor speed. These constants can be derived from the data sheets provided on the Warp web site. It would help if you would plot motor current and voltage.
Robby

34. I thought I would point out that while the maximum motor current in the Zilla is 1000 amps for the Z1k and 2000 amps for the Z2k the battery current limit is indeed lower. From the next page of the manual:

• Maximum Battery Current at 200V: 1900 Amps for Z2K, 950 Amps for Z1K
• Maximum Battery Current at 300V: 1770 Amps for Z2K, 885 Amps for Z1K
• Maximum Battery Current at 400V: 1600 Amps for Z2K, 800 Amps for Z1K

35. Hello Jack,

I will open with some butt kissing.

I love the show. I really do. I wish you guys all the success in world. At the very least I hope something like the Discovery or Speed Chanel picks up your show in place of some of the silliness that they serve us.

Unfortunately there is no real controversy here. I say it is unfortunate because controversy and drama is what makes for good TV.

Take for example those very popular guys on The Discovery Channel who "build" choppers. All they really do is a lot of chrome plating and fancy paint jobs, but even my mother (the furthest thing from a motorcycle enthusiast) loves the show because the father and son never stop yelling at each other.

Anyway I like your strong opinions. Don't get me wrong I am not suggesting that they are fabricated for the sake of your show. I am saying that I hope you find the right public controversy/fight to pick to take your show and the industry to a wider audience.

I can tell you that the Soliton1 does exactly what the spec sheet says it does. I know because I have bench tested and calibrated every single one. Well expect for when I make a mistake which happens from time to time. However, I don't think I made a calibration mistake in your controllers case.

I am tending to agree with Sebastien here. I think it is a question of loading the motor/gearing/back EMF. Whatever you want to call it.

Jack Said:
"There remains Seb's argument that we are not really loading the motor. I don't know how much more we can load the motor. It is doing all it can under constant acceleration and at 5500 lbs. If it could do more, we should accelerate faster - that's all. His contention is that if we bring the motor to a stall we'll see the 1000 amps."

Is this not why gearboxes exist? To exploit the optimal RPM of whatever they're coupled to, no? Seb is simply suggesting that it is quite possible your data was taken past the optimal RPM of the system. You do not need to completely stall the motor to see 1000A, just slow it down a bit.

Allow me to try to explain how my simple mind understands back EMF to work (which is probably widely inaccurate in which case Jeff will call me a dumb mechanical engineer later):

I think that at the motor speed your testing, the pressure in your battery (pack voltage) matches the pressure produced by the spinning motor (back EMF). So in order to force the flow of the maximum amount of electrons in there (1000A @ 150V) you need to reduce the pressure produced by the spinning motor (loading it/gearing it down).

In any event I assure you your controller is not the limiting factor (unless it's malfunctioning), and Seb will be up there in a few days to help find out what is. He is not the best at advertising his work, but in the past 3 years he has built some really great high power conversions. In the process he has turned into to quite the EV Whisperer.

Again, I love the show. Especially the episode where you gush over our products. Please continue to keep us honest when you think we are full of it. The industry needs it if it's ever going to grow up.

--
Christian Ruoff
Director of Manufacturing Operations
Evnetics

CRuoff@evnetics.com
727.895.8989

36. On the low side I thought it odd the Solitons were reading numbers high on the bench. I did feel there was a measure of de-rated output and uprating of quoted numbers so the Soliton was playing safe and wouldn't end up a smoking ruin.

Jacks 1kA ammeter (10,000rpm) on the car showed an easy top reading when the car was caned. Will you guys be temporarily fitting the other Zeva2 between the Soliton and the motor to read both simultaneously?

Another gentlemen suggested an oscilloscope to read output wave form. Good idea! I'd get the old laptop out and listen to its audio with oscope (Linux) or equiv. One can then firstly record and study in bench time.

This can be sorted. The Soliton is a very neat bit of kit and also an advertiser on here so I do hope everyone is happy at the end of the day.

Lets hope its down to the carpet behind the pedal and slightly over careful engineering. ;)

Andrew

37. Jeffrey - good point about saturation. That does indeed limit torque, but I'd be surprised if it cuts in anywhere near as low as 250 Amps. The published figures for the Warp 9 go up to 330+ with no sign of it. I'd also expect that when it does occur it would be a glass ceiling.

38. I forsee Jack handing Seb his sword for next weeks show :) In which case everybody wins, Soliton puts out and the Speedster gets faster.

39. Hi John... I think you might be right about the WarP-9 motor, and that saturation doesn't occur until 500A or so. That is why I tacked on that "IIRC" bnefore ;)

Jeff Major, however, seems to think saturation might be occurring a lot earlier than the NetGain curves otherwise suggest (ie - that they are extrapolated). We have a dyno with two warp-9's on it so it wouldn't be too much work to measure when torque stops increasing exponentially with current.

However, saturation in a motor is "soft" because of the air gap between the stator (field) and rotor (armature). This will make the changeover from a quadratic function to a linear one very gradual. I suspect measurements at 250A and 500A, then 400A and 800A will give a reasonable idea of the 2nd derivative of torque vs. amps.

40. Jack, are you sure those batteries can deliver 1000 Amps? 1000 A / 180 A = 5.6 C, isn't that well above spec? I have read these large prismatic lithiums don't really like high currents. A great test would be to short a battery and measure the current to see if it is even possible to draw 1000 A from them. I'm actually impressed you could draw 755 A / 180 A = 4.2 C.

41. Jack,

I think I have had a revelation. It seems to me that what is really required for EVs, with Warp9s, is a CVT that can keep the motor RPM at its max power point for acceleration, and then actually raise the RPMs to the Max efficiency point when cruising. Not practical for the speedster. I understand.

For now you could always try bigger wheels?:)

Just a thought.

DGS

42. What's missing from your graphs is motor RPM.
That would be really useful in determining if back EMF is the problem.

43. Err... motor RPM is at the bottom of each graph except torque and horsepower. The only RPM we have is motor RPM.

David Kois claims 2400 amps out of a 100AH Thundersky to full discharge.

If you all missed it, we did 30 second discharge pulses on 100AH CALB cells for a full discharge curve at 400Amperes. The batteries can do it.

There is a strenuous effort to portray in the forums that I don't understand the difference between motor amps and battery amps. They apparently draw some smug comfort from this but it's never been the center of the controversy.

I do find this concept of rating the controllers are motor amperes very annoying and tremendously misleading.

I am interested in power into the drive train and power out of it mostly to achieve accleration. I don't think we're getting it. And I've said from the beginning that it's probably something I've got configured wrong. But no one goes there. Haven't had a single suggestion on configuration.

Sure, if we stall the motor the current will go to the limit and perhaps we would see 1000 AMPS or not. I don't want to do that to the motor. I'd rather it accelerate the car. I would expect more than 112kw and 125 HP from the "1000 amp controller". We're not getting it and apparently because we are going to fast???

Jack Rickard

44. The motor is going too fast, not the car. A higher gear ratio might get you there.

45. "I would expect more than 112kw and 125 HP from the "1000 amp controller". We're not getting it and apparently because we are going to fast??? "

Thats funny!!! I thought I could look at the data and I'd come up with some ideas, but every idea I've come up with is contradicted by the data. This is a real puzzle. Let me look at it some more. Can't wait to hear the diagnosis from Sebastien.

The first response from Evnetics gave me some pause regarding my future purchase of a controller for my next project, however it looks like the two following responses were more in line with what I'd expect from a professionally run company. I hope they continue to be supportive of their customer and help to find the problem with the setup.

46. Make that taller.

47. I think the first place to check is the controller's duty cycle. Look at pulse width at max load with o'scope as suggested above.

48. Jack, I don't see that 400 Amps from a 100 Ahr battery proves a 180 Ahr battery can do 1000 Amps. If things scaled it would be 400 / 100 * 180 = 720 Amps. It's probably just coincidence but that's very close to the 722 Amps max you measured. BTW the CALB specs say 3*180 = 540 Amps is the max continuous current, that would be very impressive if they could do twice spec for maximum current.

That must be a typo, 2400 Amps for full capacity from a 100 Ahr Thundersky... I don't believe it. That's 8 times the max spec continuous current. If it's not a typo, has someone not selling batteries been able to replicate that performance?

Jack I trust you more than anyone selling stuff. How about a test on just one cell showing it can truly provide 1000 Amps? If the batteries aren't up to it, then the controller is not at fault.

49. OMG, I work late doing OT for 2 nights and come back to this!! a very heated debate!!

Mean while Chris and I are busy testing the new secrete squiral EVTV website for the finalist system!

Play nice now! I hope Seb can sort things out and everyone can make up.......

50. Here is an idea. Perhaps the Soliton1 thinks its a JR! If you have the latest software for the Soliton it might have the JR code which might be selected by an internal jumper or SW config parameter. I think there is just one set of code for both units. Also, they have already said that you do not have the latest hardware, and it could be that whatever method the are using to select between the two was not set correctly in your unit.

Something to look into.

David Seabury

51. Well I wouldn't jump out the window over this one guys. I not only didn't say it wasn't a good controller, I rather specifically said I still like this controller. We DID go from 76 HP to 125/130 HP. True, we also went from 120v to 192v, but the Soliton1 does the increased voltage just fine. And in fact, one piece of the puzzle is that the controller isnt' even getting warm. It's true we ran our quite capable liquid cooling full on the whole time, bypassing the thermostatic switch. But their fin fans only came on at the very end of the 4th gear test. With an IR gun, we're seeing a trivial 98F on the case.

As to taller gears, we're kind of already there guys. This VW transaxle has a 3.44:1 R&P, a kind of hard to find configuration and a 0.93 overdrive 4th gear - it really comes into play at about 80 miles per hour as you can see from the graphs.

This is where I fall out of my chair on your "loading" theories. With a rolling start of 300 or so RPM, we are flooring it in 4th gear against 5500lbs of load. The peak IS in 4th gear at about 3900 rpm.

IF the 1000 amp claim is at stall - zero RPM, obviously we arent' there long enough to see it, and I have no interest in it. We're not goinog to drive the car stalled.

It's a CAR. We're driving it. And this all started with Jeff Jenkiins claim that we should be able to do 0-60 in 6.0 with his controller. We do 9.0. But we never see the levels claimed for the controller in current either.

Further, I was ABUNDANTLY clear right from the introduction that the MOST LIKELY cause was that I had something misconfigured. Except no one has pointed me toward the misconfiguration. Instead, i get a whole lot of what I don't understand about what 1000 amps IS is, in truly Bill Clintonesque style.

I think the restriction on power IS the controller. We didn't test at 4C for a spurt. We did an entire DISCHARGE CURVE at 4C. And we need 12 seconds here. David Kois claims 2400 amps from a 100A cell to full discharge and Ihave no reason to disbelieve that.

I suppose we could throw a Z2K on. If we still were at 125HP, then I guess it's not the controller at all. Hmmmm....

Jack Rickard

52. I don't think your instrumentation is up to par with with what you are trying to measure, Jack. Even quality true rms meters don't display voltage and current accurately when PWM is involved, not to mention the resulting mayhem when you toss in reactive components. Your efficiency numbers seem too good, especially those 97% numbers in 3rd gear. I think your electrical side meetering is under-reading. Perhaps a good digital scope with the ability to mathematically analyze the data (trace) is in order. -Klaus

53. Perhaps so. The sudden efficiency peak is of course troubling, and very unlikely.

It's what we have to work with.

Jack

54. I'm confident that you are pushing 1000 amps to your motor. When you hit the 4k RPM and at 150 or so volts of sag you are hitting the back-EMF wall where the 1000amps cuts down and your battery current and torque also drop. You are at full PWM at the point where the power drops beyond those speeds. Before that you are at 1000 amps of limit. If you want to see 1000 amps from your battery end you need a higher voltage or a series DC motor of the appropriate diameter to move your back-BMF point.

You are measuring battery amperage both with the EVision and the other meter, seems the EVision has a little lag to it but the numbers are both measuring the same thing. You aren't clamped to the wiring going out to your motor where the 1000 amps is.

Look back to your video where you did your 10 second runs near the airport, you were showing the 1000 motor amps on your own instrumentation. Your motor couldn't suck in the amperage once the motor voltage met the battery voltage and the current to the motor dropped out before the total power would hit a point where battery current would hit 1000 amps.

Ohms law, it's this simple.

Everything I said above simplified:
1. Your motor voltage is lower than battery voltage at low speeds to due to the 1000 amp motor limit.
2. 1000 amps at a lower voltage will be less current from the batteries that are at a higher voltage(volts*amps=power, voltage is different in the two places).
3. By the time your motor voltage matches battery voltage there is too much back-EMF to have any more than the 700 or so amps you are able to push to the motor.

Your torque is flat and the amperage to the motor appears solid at whatever amperage is pushed to it, when the torque starts its drop dramatically there is your back-EMF.
Let me know if you are still confused.

Gerald Williams

55. Jack,

GREAT JOB, THANK YOU +++++

-Youri
1-ev.com

I tend to agree. From the graphs above, 4th gear current at 900 rpm is said to be ~159 Amps and the torque is said to be about 163 ft lbs. The Warp 9 data sheet shows 154 Amps generating just 20 ft-lbs.

Given the modest 0 - 60 times, there may be other issues (controller set up being a possibility as Jack has said); but for sure there is something amiss with these data readings.

57. I think Mr. Williams nailed it. Power = power = power (toss in efficiencies, of coarse). Motor voltage is the time average of the PWM duty cycle times sagged battery voltage. When duty cycle is < 100%, current out is greater than current in. I'd blame Ohm for that, not Evnetics.

Don't blame the controller for self preservation when the impedance of the motor/load/speed is low.

If there were a chart like http://www.go-ev.com/images/003_09_01_WarP_9_Graph.jpg, only at constant current of 1000A instead of constant voltage of 72V, I'd bet a nickel that Jack's data mapped to it up to his sagged battery voltage.

Great show!

Mike

58. Hey Jack!

It just occurred to me that, in the absence of wind resistance and with perfect gearbox efficiency, testing the various gears at "5500 lbs." is the same as testing the same gear at various loads. If you scale the data by motor RPM instead of MPH, do you get peaks in 1st, 2nd, and 3rd gears that are "cropped" because an unfortunate selection of speeds to plot didn't coincide with the curve peaks?

Mike

59. Jack said: "I think the restriction on power IS the controller. We didn't test at 4C for a spurt. We did an entire DISCHARGE CURVE at 4C. And we need 12 seconds here."

A test at 1.3 times above spec doesn't prove it'll do 1.9 times above spec. You could well be right, but the 400 A test doesn't prove it.

Jack said: "D... claims 2400 amps from a 100A cell to full discharge and I have no reason to disbelieve that."

Jack if you believe that I have a BMS to sell you. 8^) Seriously, you must admit that is a pretty extreme claim to be able to run a battery at 8 times its max continuous current spec for over 2 minutes. Let's do some back-of-the-envelope on this one:

From Thundersky's curves, it looks like the internal resistance is 0.8 milliOhms.

Power lost to resistive heating
= I^2R
= 2400 A * 2400 A * 0.0008 Ohms = 4.6 kW
That's like 3 to 5 microwaves going at once!

Full capacity at 2400 A, or 24C, implies a 60 / 24 = 2.5 minute = 150 second run time

Energy = 4.6 kW * 150 seconds = 690 kJ

Ethyl Acetate (EA) is a component of the electrolyte and has a heat of vaporization of 398 kJ / kg

That means that's enough energy to boil 690 kJ / 398 kJ / kg = 1.7 kg of EA. That's 1.5 liters of the stuff! The battery's dimensions are 2.4 liters of volume.

Enough energy to boil 1.7 kg of EA is alot for a 3.5 kg battery!

I see no way you could run a 100 Ahr Thundersky at 2400A for at 24C, or 2.5 minutes. There is a single claim for this number -- is there anyone else that claims they can repeat this performance? There must be some mistake in the measurement, a typo on the capacity, or something. BTW I'd love to be proved wrong!

60. Jack –

I wonder if the controller is acting as a chopper circuit and not utilizing half of what would be a “simulated” AC waveform. If that is true (that it is not utilizing true AC), then you would only be dumping half of the potential current into the motor that would exist if it was a true AC wave form (powered on both halves of the wave). A way to test this would be to power the controller at really low current and rectify the output in both directions on the output through a diode rectifier and see if you are getting power off of both directions (both halves of the wave). Another thing I was wondering (if you had controllers to spare and you could afford to blow one up) is how much current that controller would allow if you ran it into a hard short rather than a motor. It would need to be a measurable, slowly increasing hard short with decreasing resistance down to zero to see what the real amperage limit is on that controller. What kind of resistance measurement do you get if you throw an ohm-meter on the controller with the “throttle” wide open? Fun to think about. Take care.

- David Hathaway

61. Do you have any spare cells like the ones in the new speedster? presumably it wouldn't take that long to do an experiment where you try and get 1000 amps out of the cell for say 20ish seconds. It would make a fun little segment for the show and potentially silence some doubters ...or not, depending on what happens.

-Nick F :-)

62. No banana Gerald Williams.

Yes, we DID show 1000 amps. We also had the WRONG pulse count to the tachometer as it turns out. We were showing DOUBLE the amperage we were getting as it turns out.

Jack Rickard

63. Mike:

Yes and no. There is only one peak. It always occurs at about 3900 RPM apparently regardless of gearing. On several of the charts you will see the 3900 strike between the 10 mph points graphed. The power IS higher there. Nothing striking.

I think next time we are going to include the peak power point in the graphs regardless of speed. That will lead to a whole lot of OTHER confusions as to why we did THAT. But what you noticed is quite real.

On the graph, it looks like a little flat top bridging the two MPH points. It's actually peaked.

Jack

64. David Hathaway.

It IS a "chopper" or square wave PWM circuit. You get "half" if you are AT half (50% duty cycle). We are trying to apply all the power the controller will apply in a full on acceleration.

The leading posit is we aren't getting a full 100% throughput THEN because it isn't fully "loaded". I'm struggling with this. Does this mean the ONLy way to get full power from this controller is to put it in a car that weighs twice as much?

The question of course is how you size a controller, motor and battery to a car. And how do you get maximum performance from the car. IF the controller is the limiting factor, how do you get it to NOT limit.

And so round and round we go. The batteries can easily do 10C and actually more like 20C for this time period. You can believe it, doubt it, or whatever. I'm not doing ANOTHER battery show for you on this.

The motor appears to do twice what we're doing - anecdotally.

And so we are focused on the controller. I'm having to fight off one bullshit explanation after another to get there. Gradually they are falling away.

ALL subject to measurement mistakes, and configuration mistakes along the way. Methodology is always a problem.

Jack Rickard

65. I said this in an earlier post and no one commented.

Has it been determined that when the accelerator is fully depressed you are actually sending the proper signal that the controller will interpret as give me full power? Has a meter reading been taken recently, not when the accelerator was installed? Is there any configuration in the Soliton that affects how the accelerator signal is interpreted?

In any scientific study you must test and verify each step of the way. Please tell me that this has actually been done before going down the road with all these possible theories!

66. Ok. I'll comment. Granted.

Actually its a good point. I'll check it again tomorrow.

That said:

The Soliton features a unique and I think very usable calibration feature. You simply select this using a web browser. It disables the controller. With the accelerator off, you simply click on a button to calibrate the accelerator minimum. Then press the acclerator fully and click the other button to set the accelerator maximum.

At that point, you can also set two variables. One gives you a little dead zone above minimum. The other sets the mapping where you can set teh percentage of power applied during the first 50% of travel.

I think we have a 5% dead zone and map 55% of power in the first half o fthe pedal. Doing this from memory.

Jack Rickard

67. Jack, thanks for checking Into this. For grins, when doing the dyno test, change the parameter from 55% to 100% for the power in the first half of the pedal. This would eliminate any anomilies that could be encountered at top half of pedal travel. Just a thought.

68. Easy way to draw more current would be to do field weakening. Dennis Berube and http://www.poormansev.com have experimented with field weakening. Just make up some long segments of 1/0 cable, put them all in series and across the field terminals. Take out a segment at a time. You could risk motor damage if you weaken too far and work the motor too hard. You know your total current, you could put a clamp on Ammeter on the field weakening cable and see how much current is bypassing the field.

Seems like you'd want to weaken it until it just hit 1000 battery amps at the one point where it just drops out of current limit -- provided you don't hit the arcing limit first!

The other alternative is to get a higher voltage motor and controller.

It very well could be the controller meets spec and it is your setup that can't deliver 1000 Amps.

69. everyone has got an opinion haven't they......lots and lots of online engineers typing themselves smart.......

Inertia dynos are crap at giving true results, the only way to test is with a continuous load dyno like a water brake or eddy current brake, and actually measure the torque and RPM.

Inertia dynos are only good at giving you a acceleration result, not a sustained power output result, horse power and Torque is only an interpolation.

If you had the ramp rate on the controller set to low (amps/second) the rolling road won't have an opportunity to load the car enough. The slower you accelerate the
drum the less power you need, vise versa, if you restrict your acceleration rate you cant accelerate the drum at a fast enough rate to load the motor.

Can you get 1000amps while driving the car on the road??? If yes, the dyno sucks, If no, the acceleration rate is probably set to modest......

70. http://www.land-and-sea.com/dynamometer/dynamometer-comparison.htm

71. Simon: I didn't know that about dynos. If "...horse power and Torque is only an interpolation..." that explains the absurd numbers in the graphs.

If (Simon again) "...you had the ramp rate on the controller set too low (amps/second) the rolling road won't have an opportunity to load the car enough..."

Jack - back to the second comment in this impressive string: bad data + slew rate. I know you said slew was set to 0.2 milli seconds per amp, but is it possible that you are not actually getting that either as a result of finger trouble or a software glitch?

72. Oops "Hardy" not "Hatrdy"

73. Jack for a smart bloke...
Its obvious
The Flux capacitor is faulty
Man its allways the Flux Capacitor

74. To Simon.
If I'm not mistaken. If you look at the video you will see a huge banner on the wall stating it is a hydro brake dyno.

Why blame the dyno. Where's the 1000amps?
======================================
There are overunity/youtubers out there who farm the back EMF off DC motors to power other things. Hey, we have an opportunity here folks. Regen when driving harder. Just rectify and condition that back EMF between controller pulses. >:-))
======================================
Nice to see Jack has sorted the 10krpm meter to show true amps. Brian said he used around 1A/mile before the test. It would be a hoot to find Redux is actually using 500mA per mile.
Which I doubt.
======================================
Linux/Audio/Xoscope. I have had lot's of fun with that. Lets see what the waveforms being kicked out by the controller under load are like. The answer might even save Seb a ticket. Even though I'd love to see him, happy on your show.

75. Grrr! Trust an Anonymous to find the right answer.

Of course! That damned flux capacitor. It only kicks in at 88mph.

If only Brian thumped the dashboard. The batteries would of given out 1.21 Gigawatts then he could of come back with a Mr. Fusion strapped to the back of your Redux mkIV plus a couple of couple of 'TS AAA' cells to have a look at..

76. I have changed the slew rate from 5000A/sec to 9,000A/sec. I had originally heard you could go as high as 25000 a/sec but it appears that is no longer the case.

In any event, I do not believe slew rate is the problem. This should define how quickly we build to maximum current. As the maximum is 1000Amps, at 5000A/sec this should take 200ms. It takes 10-12 seconds to accelerate under this load.

Jack RIckard

77. Hi Guys
regarding what I said previously I will reword it, Theory is proved or disproved by experimental outcomes, then the theory is rewritten to explain the experimental outcome ie:- reality.

Now guys if we look at the graphs we will notice in the 4th gear more so than the rest, the motor peaks out at around the 160 ftlbs torque give or take a few, this if you connected an amp meter across the motor leads you will most likely find the controller is limiting the current to the motor at about the 1000 amp point give or take a few amps. Also it is very flat from about 900rpm up to the 3900rpm point from that point on there is not enough volts passed through the controller to continue to supply the 1000amps. Torque is dependent on amps and revs is dependent on the voltage the motor gets. The question is why didn't the motor see full voltage and hence get 1000 amps from the battery? It got lost somewhere? I asked the similar question with the experience I have had with 2 different makes of controllers?
Come on Guys what's the answer?
My intuition is that the controller has a software glitch or a hardware issue and is limiting the pulse width to the motor. If the motor gets battery voltage it will increase in revs until it matches the 1000amp most likely around the 5000rpm and all will be well.
regards Leigh

78. Jack,
What is the Tranny on Speedster Redux?

Thank you.
-Youri
1-ev.com

79. Apologies to the physicists and theoreticians, but how about some empirical data to form our equations around? If someone with access to a Warp9 motor, dyno, and Soliton1 controller could take steady state torque and motor voltage (duty cycle * sagged battery voltage) readings at motor current controlled to 1000A and various RPM (500, 1000, 1500, 2000, ...), this would give us better insight into what to expect during a current-limited acceleration. It would also sort out all the nonlinearities between current and mechanical power (field saturation, etc.), resolve instrumentation lag (and maybe accuracy; "99.7% efficiency"??), define the motor reactance at 1000A, and give Sebastion a reference for comparing to Jack's situation. It would open up Jack's "black box". The "inertia" of the drivetrain should be sufficiently small relative to the vehicle inertia that steady state readings are pretty darn close to acceleration realilty, and easier to set up. But then, what's the fun if the light is shed and all the armchair engineering quiets down?

I'd guess if Evnetics doesn't already have this data, that they could pull it off a calilbration bench more easily that anyone else could get it.

BTW Jack, don't miss the bigger issue here. 99.7% efficiency! That's a huge leap toward perpetual motion. I'd have never guessed that the Porsche 356 replica was the missing link. With just a litle more work ... ;-)

Mike

80. We took 6C out of our SkyEnergy / CALB 130Ah cells with no problems or total collapse of voltage for estimated 10 secs. (We never needed more or a longer peroiod of time).
So Jack should be able to get at least 1080A out of his 180Ah cells.

81. Youri:
VW SWING AXLE TRANSMISSION
1961-1966 Short Axle (26-5/8”) transmission with short splines.

A few notes on this transmission:
Gearing
1st 2.64
2nd 1.93
3rd 1.14
4th 0.93
Ring & Pinion 3.44

Super Diff - The reinforced differential housing carries four spider gears instead of two which are mounted by an additional shaft. The end gears are kept in place by two snap rings per side rather than the usual one per side.

Heavy-Duty Aluminum Side Cover – This gear box cover reduces the spreading of the ring and bevel gear, which often in VW transmissions under high power loads leads to fracture of these parts.

Welded 3rd and 4th Gear Hubs - The original two piece construction of the synchromesh gear hubs are not designed for high performance use. Welded synchros are able to handle more power.

Steel Shift Fork – A much stronger steel shift fork. Hardened Woodruff Keys - Are used to secure the individual gears in the main
shafts.

Redline MTL Manual Transmission/ Manual Transaxle Lubricant is used in the transaxle to reduce friction and increase efficiency. This makes the gear noise a little louder, but decreases rolling resistance.

82. Mike:

The 99.7% efficiency is clearly some sort of measurement anomaly. However, I kind of am bound to report what we saw, not what we expected to see.

I have no doubt that EVnetics has done all you propose. I don't think they are faking any data or not making data, however much Jeff wants to posture and preen that that is what this is about.

If you read between the lines, the point we are working toward here IS black box. If you can't make it do it on your car during normal acceleration, it is useful as a comparison of WHAT????? Or useful in your componnent selections HOW????/

If we just put the car on a dynomometer, with NORMAL loading, we should be able to achieve the maximum utility from the batteries, the motor, and the controller in a normal acceleration and if there is any to spare, it should allow us more speed.

But it doesn't. We are looking to explore how, why, and what that leaves you with. The theoretical explanations only become interesting as they show up in this process to explain observed behavior.

We have more than adequate battery capacity, more than adequate motor capacity, and the Soliton1 controller, purported to do 300kw and 1000 amps. We're not getting it in any normal driving/acceleration motif I can mount.

If we can't, it's unlikely our viewers can either, and in fact, our "calculations" on the Elescalade are pretty much out the window.

I think this is a very valuable exercise for our viewers who are planning conversions or are experiencing this with the conversions they've done. If a few of the old guard are all offended by all that, or that anyone would QUESTION their claims, why would that concern me at all. They will talk BAD about me in the forums????

They will all get into a little group and vote me off the island, take a holy oath to never watch EVTV again (and then talk about what?)

We're plowing good ground here. There's not a right or wrong in it for me, or really for any of the vendors, although I have no confidence they will see it that way.

There's kind of a division of labor around here. Brian gets to worry about vendors and advertisers, and I get to worry about viewers who are spending their ducats to make electric cars go in the night. Fortunately for the viewers, guess who is the big stick around here?

Jack Rickard

83. For the commenteers (as in musketeers) who take issue with the lack of peaks on the graphs. Look again. The peaks don't occur because no reading was taken where they happened.

If double the data points were taken, one would see all the peaks.

Jack, you should of drawn them in to avoid confusing the natives.

Now go get liquered up. Play with high voltages with your sillyscope and go for a drive.

84. Another enjoyable show, especially the white board insights. I also like how you pursue the truth even though it might ruffle some feathers. The Speedster on the dyno reminded me a bit of the Lexus LFA "sound" commercial. I wonder why the manufacturer doesn't just send you another Soliton? Also, how about dissecting the Soliton as a white board lecture? Could the failure of a single part allow for what you are experiencing--it functions perfectly until the critical 1000 amp piece is activated?

85. He has got at least 3 pb. 2 on the bench and one in the car.

-Nick F

86. Quite so Andyj. But we should have put them in. I actually had the data. I just liked the symmetry of 10/20/30/40/50. We'll put em in next time somehow.

You can almost draw it in yourself it is fairly obvious.

I don't know that anything is WRONG with the Soliton1.

We DID do some Spyder 550 stuff today and learned quite a bit. But the one thing we DID learn, again, two visits in a row now. This concept of "loading" the motor to achieve maximum current - appears to be total 100% bullshit. We actually get a decrease in current. I did not believe it, so I went with them today. It's real. The car just can't keep up and the rpm and current drop quite in order. The "stall" thing just isn't happening on a dynomometer at RPM.

Jack Rickard

87. If loading it up does not raise the current then we back to the Soliton not producing its max output. I did notice that one of the recent changes to the software (ver 1.3)was a change in the MAX current for JR to 600 Amps. Could they have made a mistake and done it for both 1 and JR? I know it is just pure speculation, but possible.

Also, some have suggested putting a scope on the motor to see if you are getting 100% duty cycle on the PWM signal. My guess is that there will be so much noise on the signal, from the motor, that you really won't be able to tell much, but it might be worth a try. I tried this recently with a 12V 60Amp PWM controller on a robot with a DC motor and the signal was basically just noise on the scope.

88. Since we aren't measuring current to the motor and only measuring battery side current, how can anyone possibly be saying that its not providing its rated current? Jack said his meter wasn't reading right before at the airport because it was showing double as it wasn't measuring the pulses correctly. ...if that was measuring motor current, what is it showing now that its corrected? If it is showing motor current. The two numbers WILL be different, if they are the same or close to the same, they are both measuring battery current.

Measure the right thing and come back because at this point I can't believe how this conversation could possibly be going in so many directions when its so obvious. ...without a single motor current measurement to boot! Get the right data, post that data, then we can actually talk about something useful. Until then, you have nothing without this critical piece of information.

Gerald Williams

89. Gerald:

You have so MANY things so EMPHATICALLY confused in one little paragraph combined with so much condescending shit in it to season it, that I just don't quite know how to approach the meal.

We got the right data, we posted that data, and most assuredly you are never going to be able to talk about anything useful. I MIGHT. You wont. You've got complaint, ire, arrogance, affrontery, and so much else all balled up into one blog comment I don't know how to unwind it.

In the first place, battery current and motor current can absolutely be the same. You have simply read some shit and extended it to nonsense. They don't HAVE to be the same and in fact will often be different. At maximum power, they should be the same.

We HAVE compared motor current and battery current, and for the purposes of what we seek, the maximum power point, they appear to be the same. It IS true that what you see graphed is the battery current.

The graphed measuremenst were taken with three devices. And ONE of them has anything to do with your EARLIER wildly condescending claim that we had used IT to measure 1000 amps at the airport. As it turns out, it WAS misconfigured and we were reading 500 amps at the airport. That has been corrected, and is part of the reason we might want to compare three different devices to see if they all come out the same.

Further, not that it matters for this discussion, but the EVision does not read more than 500 amperes accurately.

It is going in so many directions because you have such strongly held beliefs about things you have no comprehension of apparently. The reason it IS going in all directions is that we are searching in all directions for possible outliers, answers, and ideas for what could explain measured and observed data.

I don't fully understand why the current would go up at peak RPM under increased load. So it is very difficult for me to then explain why it actually goes DOWN instead of the UP everyone assures me it will. It even seems a stronger DOWN than the drop in RPM would suggest.

Further, the Spyder test today was an ignomious failure because of a throttle calibration issue with the controller, and the whole thing will have to be repeated anyway.

Tomorrow, we return to the Dyno with Speedster Redux, and Sebastion Bourgois attending. Cant wait to see what that turns up.

Bottom line is that the little charts and theories you are used to on the forums are all very neat because none of them are real. In real testing, leads fall off, magnetic pickups quit picking up, Xantrex's throw alarms to put the throttle in limp mode, meters run out of batteries, cameras get knocked over, etc. etc.

If you graph it by miles per hour, you skip rpms and if you graph it by RPM you skilp miles per hour. And yes, we have a 98% efficient motor, at about 2 rpms out of all of them.

I love this stuff. But it isn't for the faint of heart, and if you want to armchair, relax and posit your theories as a suggestion, rather than a "command" to "do better or else."

Jack

90. Gizmo (and Jack):

Motors are basically big inductors, so they oppose a change in current. But with a constant current, they just look like a wire. This is what you get with a constant DC voltage - it does actually obey Ohm's law. But...the back EMF opposes the applied voltage so that under zero load, the BEMF cancels the applied voltage and you get zero current. If there is a load, then the motor spins slower than the rated RPM at applied voltage and you get some current (but once it is constant DC, then it follows V = IR, where V = V_applied - BEMF).

On the point about getting max current, I am almost sure that you need to load the motor more. Having 100% throttle will put you at 100% duty cycle, but that doesn't mean 100% current. When you drew the pulse wave, you labeled the y-axis current. This is incorrect. The y-axis is voltage, and the current will be whatever the motor is "asking for".

Just to make sure, I tested this with my Alltrax 7245 in my electric motorcycle. Sure enough, at 100% throttle, I get 100% duty cycle (as measured motor side with an oscilloscope), but maximum current was 80A. This was with the wheel up on a lift. If 100% duty cycle meant 100% current, I should have seen 450A.

I posted a video here. You can see the scope on the left, and the output from the Alltrax on the right. It's hard to read the numbers, but I have the output log and the most amps I pulled even at 100% throttle was 80A.

Under hard accelerations on the street, I pull over 400A. It's the load.

-Noah Podolefsky

91. Oops - video is here:

-Noah

92. My suggestion. With whatever method you choose, hall effect or shunt, measure the current output, to the MOTOR, at 2000RPM, throttle floored. What do you get? Should be the same or very close at 3000 RPM.

Let us know what you get. Controllers are rated at maximum motor current, so we should have a reading for that and your graphs and video haven't posted it yet in either the video or here on the blog. YOU may have compared it yourself but none of this information is here and you haven't suggested what it is.

None of my comments ever suggested load was the issue, I don't think it is, you seem to be mixing up my two comments so far with what others have said.

Gerald Williams

93. Jack there is a way you could get 1000 motor amps the whole time and never reach 1000 battery amps. You statement that you should always hit the motor current = battery current sweet spot is true in an instantaneous, no delay world.

The system has lag, lump it all into a parameter tau. Lag can be from the inductance of the motor, delay in the measure-command-response system loop, etc.

Suppose you are accelerating the motor fast compared to this tau. Suppose you are at 1000 motor Amps and 3900 rpm and 722 battery Amps. By the time you get through your next tau time, the motor has already overshot the motor = battery amps sweet spot, and is well into the back EMF zone.

This would explain why loading the motor down more would make it more likely you hit the motor = battery current sweet spot. There would be more time for the system to equalize. You'd approach 100% PWM duty cycle slowly enough to get motor = battery current.

There is a way to check this. Measure how slowly the current decays down, and compare this to the time your back EMF drops current. If the decay time is comparable to the time it takes current to decay 1000 - 722 = 278 Amps, then you have your answer. A more fun test would be to drive test it. Find a hill. The hill plus wind resistance would slow things down, you could see if you got higher battery current.

The fact that 4th gear had so much more current than the lower gears is circumstantial evidence for this conjecture.

Anyway, thanks for the great shows and the actual measurements you do and share.

94. Hey Jack,

Does a series wound DC motor really work that way, where you can set both voltage AND current arbitrarily under any load and speed condition? Wouldn't that capability be neccessary to achieve what you describe as "full performance" from the controller at every condition?

Say at a given condition, the motor has an impedance of 0.25 Ohms. You'd need 250 volts to make 1000 Amps. At another condition, say 0.125 Ohms, you'd need 125 volts to make 1000 Amps. In niether condition can you put precisely 160 volts AND 1000 Amps into the motor simultaneously. You could put in precisely 160 kW of electrical power, but the motor's impedance will dictate the combination of voltage and current.

I'm guessing that there is a condition where the motor impedance is 0.16 Ohms, and power from the controller is very close to 160 volts AND 1000 Amps, limited by either, niether, or both the sagged battery voltage and controller current limit. Every other condition (where impedance is not 0.16 Ohms) will not allow 160 kW because of a voltage OR current limitation.

This is my grand thoery. Shoot holes in it, attack my character, whatever, l'm just trying to understand how this works.

Mike Kaindl

2. That means the controller or the batteries are limiting the current. Otherwise the current would increase with load as Noah Podolefsky
has said.

3. Since the batteries should be able to supply that much current that means the controller is limiting the current.

4. The only remaining question is why. Is it a configuration issue, a spec issue, or defective unit.

I dont think there is much more to say on this issue until more data is available.

David Seabury

96. So, if it turns out that motor reactance, or resistance, or impedance, or whatever, really does vary with load and speed, in a repeatable and predictable way, wouldn't it be interesting to see dyno data that characterizes it vs. torque and RPM? Then we could predict the max power that could be applied at any time, develop decent simulations for vehicle 0 to 60 times, validate vehicle dyno tests, understand this beast, ...

Mike Kaindl

97. One last comment. The max voltage drop on the pack under test seams to be 192v-154v or 38v. If the internal resistance of the pack is ~.0008 Ohms * 56 cells or 0.0448 Ohms, then the current draw on the pack at that point is 38/.0448 or 848 Amps! So then power from the pack is actualy 154*848 = 130,592 Watts at that point. Input power would be ~175 HP and output assuming 85% eff. would be ~148 HP. For what it's worth.

David Seabury

98. One last comment. The max voltage drop on the pack seems to be 192v-154v or 38v. The total internal pack resitance should be ~.0008 Ohms * 56cells or 0.0448 Ohms. That means the current at that point is 38v/.0448 or 848Amps. The input power is then 130.592 KW or ~175 HP. Assuming 85% eff then the output HP is ~148 HP.

For what it's worth. Just another way to look at the data.

David Seabury

99. Mike Kaindl:

I like the thought process, and indeed I am attracted to the load theory. It is popular, and it has symmetry.

It also has a problem. As the power out is a function of the voltage and current applied in, if this is limited by the load, we have a huge problem here.

That would mean that with a larger motor, larger current capacity, and larger controller, NOTHING would change and we would use exactly the same amount of power, and achieve the same acceleration.

What I'm pointing out is that your theory paints us in a corner from which there is no escape. It implies we cannot produce more power to the road without increasing voltage from the pack. Yet either a larger motor or controller will indeed increase our acceleration.

In any event, we increased the load through braking, twice, and observably had less current. It was kind of hard to pick that out from the less rpm, but if I understand your hypothesis, less RPM would also lead to more current through decreased CEMF. We see the opposite. Less, not more.

Jack Rickard

100. I am thinking your next show will be one of the most anticipated shows yet...

101. Don't PWM DC controllers need to limit their duty cycle to something significantly less than 100%? I thought the physics of IGBT switching dictate this by their finite on-off performance at high frequencies. Perhaps inductive reactance effects of switching 1000A on a square wave also contribute. At any rate, and whatever the cause, it appears the Solitron1 operates to a maximum of approx 75% duty cycle switching a 1000A input current, judging by Brian's max amperage dyno run. This is the simplest explanation of Jacks' observed results. It also seems to jive with the manufacturers explanation which was briefly touched upon in the show.

So, why are we complicating it more than that with all the above theories? It looks like the design max duty cycle for this particular product is around 75% plain and simple. Am I missing something?

102. In theory Jack should have a 200KW heating element called a Netgain WarP9 almost as soon as the throttle is floored. Instead the current rises almost linearly from nothing to 4Krpm. Get your heads around it folks.

Yes, there is a max VA per C.
(36V down@722A) but why is it not the 'limited to' ..and Jeff's witnessed this; 1000A?

Seems the only recourse for more power as things stand is more voltage.

Methinks the Soliton is doing its job too well :)

Apart from lacking an inbuilt DC-DC converter.
Using the IGBT's to control recharging the batteries from the mains.
Holding downhill speed when off pedal. Using regen. For hilly, country UK roads where bloody speed camera's abound.

Then it will be the cornerstone of my shopping list. >:-p)

103. RG,
Its the switching that makes it hot.

That is why there is the 8Khz performance (efficient) mode and the 14Khz quiet mode for the Soliton.

104. I'm not going to comment the Speedster since Seb's in place to help JR, but I want to sort out some of the more odd speculations about the Soliton.

@David Seabury: "Here is an idea. Perhaps the Soliton1 thinks its a JR! If you have the latest software for the Soliton it might have the JR code which might be selected by an internal jumper or SW config parameter."

If not completely impossible it's extremely unlikely. However, if a Soliton got misconfigured it would also show up in the web interface and Jack wouldn't for example be able to set max motor current to more than 600 Amps. In that case it'd be pretty obvious why he doesn't see the expected current...

@GoFigure: "I'd guess if Evnetics doesn't already have this data, that they could pull it off a calilbration bench more easily that anyone else could get it."

Jeffrey did and published the data here:

Enjoy.

@Anonymous: "I did notice that one of the recent changes to the software (ver 1.3)was a change in the MAX current for JR to 600 Amps. Could they have made a mistake and done it for both 1 and JR? I know it is just pure speculation, but possible."

Yes, you're right. It's pure speculation.

Big changes like that are thoroughly tested and it would be such an obvious mistake we'd have caught it pretty much instantly. All controllers are also tested that they make 1kA on the dyno before they're shipped.

@RG: "Don't PWM DC controllers need to limit their duty cycle to something significantly less than 100%?"

No. They have to limit the duty cycle to limit the current and then duty cycle has to be slowly increased to maintain the desired current as the vehicle accelerates and back-EMF increases.

"It looks like the design max duty cycle for this particular product is around 75% plain and simple. Am I missing something?"

Politely stated: Yes.

The only reason for a DC-controller to limit the upper duty cycle at a level like ~75% is if you have a pack voltage like 250 Volt and don't want to risk to blow up your motor. The Soliton has no problems reaching 100% duty cycle if pack voltage, back-EMF etc allows it.

105. Jack,

When you loaded the system with additional braking did you also see an additional drop in pack voltage? At some point the voltage will drop to the point where you cannot get any more power out of the cells. If there was minimal drop then the controller is the limiting factor.

I think I squared R may be killing us and any atemp to increase current just ends up as more heat in the battery, controller, and motor. Higher voltage would help.

106. Just do what I did in my video. Put a scope on the output and look at the PWM signal. Measure battery side and motor side current. That will end the whole debate. If you see 1000A motor side and a duty cycle less than 100%, that's your answer. The controller is performing at spec, but you don't see it battery side. The average voltage motor side is lower than battery side (due to the PWM) and so with lower current battery side you still have power in = power out.

107. We are all stabbing in the dark here. Including me. I feel I've already "contributed" too much.

There is one possibly embarrassing thing. That involves Solitons latest software update.
It has a new feature. A setting for maximum motor KW to stop it from melting.

It's embarrassing for me to write something as obvious as this. Jack is no silly sausage... is he?
If I'm uber wrong. Feel free to kick me.

108. Hey Jack!

Jack said, "That would mean that with a larger motor, larger current capacity, and larger controller, NOTHING would change and we would use exactly the same amount of power, and achieve the same acceleration."

Fortunately for you, I disagree.

1. I'd expect the "resistance" load of a Warp11 at a given torque/speed to be different from that of a Warp9, if only because the "lever arm" is longer.

2. With more controller output current capacity, more power can go into the motor. The tested accelerations are current limited to 1000A (+/- the controller's output current measurement accuracy at this current), although it is on the controller output, which I know is an unfortunate twist for you. Doubling the controller output rating to 2000A would double the power into the motor during current limiting, if your batteries, etc. could take it.

3. More battery voltage does NOT give you more power WHILE THE CONTROLER IS CURRENT LIMITING. It will just act to shorten the PWM duty cycle during current limiting. The current limiting controller will control the duty cycle to create the voltage that results in the the 1000A (limit) to the motor. More battery voltage will, however, give you more top speed when the controller is NOT current limiting.

So, what I'd expect, looking at torque vs. motor RPM:

Increasing controller output current rating should cause a corresponding increase in torque at speeds left of (below) the "peak" speed.

Increasing battery voltage should move the peak to a corresponding higher speed (to the right), but not increase the torque to the left of the peak.

Going to a bigger motor should be like concurrently increasing controller current capacity and decreasing battery voltage. By how much? If we had dyno data ... (see my above posts).

My theory is completely consistant with "bigger motor and/or more current produces higher acceleration".

Disclaimer: By "corresponding", I don't mean to imply "linear" (like double the torque for double the current). I just mean to expect a monotonic increasing, probably non-linear, relationship between torque and controller output current rating.

Can't wait to see what you and Sebastian find out today. Got to say I'm looking forward to a big "Ah-ha" moment in the near future. :-)

Mike Kaindl

109. Mike,
Occams razor.

That's a big explanation for a battery pack that will give 1000A and a motor that will accept 1000A.

110. 1000amp motor test.........

111. (1000amp motor test.........)

Where's the car?

112. Are you making it with tesla stock?

I had a feeling two days ago we were at a buy point. With forecasts of 5 dollar a gallon gas this summer...sounds like the blood bath will be in the oil burners corner!

http://green.autoblog.com/2011/03/31/tesla-shares-skyrocket-on-morgan-stanleys-overweight-rating/

Keep up the good work.
JMS

113. My my my my my. Such a tremendous number of very strongly held opinions. Well, well, well, well....

I'm having a very happy end to a marvelous day thank you. And truly delighted to meet the EV Whisperer, Mr. Sebastien Bourgouis. We had a lovely day, and thanks to his very rapid analysis, we were drinking whiskey and partying lock Rock Stars by two in the afternoon.

Hate to give away this week's show, but I can't read any more of this battery amp/motor amp theory nonsense that of course never was and never could have been the issue. The motor loading issue was interesting, but as I noted, rather paints us into a corner of never getting well even with a nuclear power plant driving a 12000 lb motor.

Apparently, if the calibration of the hall effect sensor INSDIE the Soliton is off, it is quite possible for the Soliton to be entirely convinced it IS putting out 1000 amps, while the rest of the real world, including the fat kid, is kind of convinced it's more like 722. In that event, the device has a cunning strategy for self preservation, it simply limits itself to 1000 amps, even if they are not real.

If you replace that unit with a DIFFERENT Soliton1 from the magic bag, that does NOT have this calibration issue, guess what you get on the very first test drive?

A TOTALLY SMOKED Stage 2 clutch and pressure plate assembly. I mean NASTY SMELLING and pretty much destroyed.

Throught the mircacle of time lapse photography, we should have a new CB Performance kevlar double sided ceramic trick double throw me down stage 4 unit with 4000 lb pressure plate by AM tomorrow.

0-60 times to follow.

Quite a relief. And quite an impressive show on the first run.

Jack Rickard

114. To EVnetics Guys.......

Personally, If I was in the business of making a professional product like a Soliton 1, I would not be commenting in any forum, whether it be this forum or DIYelectric........there should be one contact person only, not four people, especially when they all say different things, direct communication only, phone or email.

Bleating in Forums at prespective customers is not professional conduct.

NO PUBLIC FORUMS..........

115. Your right!! Does Tesla, Ford, GM, MicroSoft, Apple make excuses for there products in an open forum?? NO!! They quietly come up with a new firmware update and say nothing.....

116. Jack,

Wonderful news. Good to hear it is good enough to destroy your clutch pressure plate too. Interesting about the calibration issue too. Dispels the myth about the cells not providing the required power and the myth of gotta load it up.

Glad to hear you can get your clutch and pressure plate replaced so soon. Impressive that you need to go with that high of performance parts. Time for the track I presume.

Good news for Evnetics too. Guess the boys need to get on the same page.

Hats off to the EVTV team and Seb. Party hardy.

This news will put a run on the Evnetics team to get more controllers out to the community. I may now be game for one myself.

117. @Jack Rickard: Very good news, indeed. Hope both the car and the whiskey now are up to your standards. ;)

@Anonymous & @Simon: Of course we can always get better at communication, but I hope we never get THAT professional...

118. "Could the failure of a single part allow for what you are experiencing--it functions perfectly until the critical 1000 amp piece is activated?"

So, is the "hall effect sensor" a single part? ;)

119. So that first Sol1 was not putting out 1000 Motor amps. I'm not gonna start to "if you had done this and if you had done that..." But just one thing: that clamp-on Current meter on the motor side of the controller would have shown this.

Anyway really good to se that you start breaking parts now, that shows the real power of the system...

Superb Blog, it has been really interesting to follow this week.
Looking forward to the show this week.

Regards
/Per

120. So that first Sol1 was not putting out 1000 Motor amps. I'm not gonna start to "if you had done this and if you had done that..." But just one thing: that clamp-on Current meter on the motor side of the controller would have shown this.

I think Jack already new this
Quote
We HAVE compared motor current and battery current, and for the purposes of what we seek, the maximum power point, they appear to be the same. It IS true that what you see graphed is the battery current.

Do you really think he just shoots his mouth off without reason or rhyme
He actually is rocket scientist for friggs sake.
Im sure he has the horsepower up top to calculate exactly whats coming out that controller even if he had not measured it.

121. @Anonymous bah, you beat me too it!

122. So hall effect sensors measure FLUX
and the output is summed by a CAPACITOR
Hate to tell you I told you so:)

123. Occams Razor.
Always so simple. The cells can give 1000A and the motor can't help but accept 1000A.

The controller was not consuming it in heat so there was no point in measuring current in front and behind the controller. The power was linear until the back EMF took over. Every gear peaked at 125Bhp so all this loading talk, internal resistances(no BMS, lol) etc. etc. was garbage.
============================

I hope Seb did not show Jack where the Hall effect sensor is. Jack (knowing the IGBT's are double rated on current and voltage) might squirrel back at night when everyone is asleep and go for that blaze of glory... More, more! I want more power!!!! >:-))

I can just imagine it now. The red glow of the Soliton666 illuminating Jacks features from below with his wicked grimace and it's unusual radiation causing horns to grow through his hair while bolts of tesla lightening are arcing across them. His arms becoming a twisted mass of wires connecting to everything, doing unthinkable things to the rest of the car and those poor CALB's, slaves to the great master. Unable to escape; to serve or die.

124. You do realize that this whole slew of drama is EXACTLY the reason why so many businesses do not want to deal with end users like (especially like) Jack Rickard and prefer to do OEM deals exclusively?

Not providing the datalogs with the reason they'll be used by Evnetics to cover their ass was below the belt, in my opinion.
Jack Rickard's been ranting about supply chains and product support on other products but if you want product support you've got to work with the supplier not against them.
Evnetics is not in the business of selling finely crafted two-digit-kWh electrocution devices for 10k a pop, unlike some other vendor out there.

Don't get me wrong; one of the reasons I like EVTV is because it features a grumpy senior citizen that is ranting and raging about his electronic petunias getting wrecked by "those damned kids".
But this drama is jumping the shark; there was no reason it had to spin out of control like it did.

125. @Fisher and Anonymous.

Jack Did not explicitly show that he had measured Amps on the motor side because he means that they should be "about the same" as the Battery amps. Well yes at full DutyCycle yes. And that turned up to be 722Amps when read on the batteryside.
If we had seen a measurement of the motorside amps showing about 720amps all through the acceleration we could have said straight away that the controller was in fact not putting those 1000 whopping amps out to the motor.

I have the outmost respect for Jack and his computing capacities. BUT he was not at the dyno doing the tests and setting things up.
Had he been, I'm pretty sure other conclusion would have been done.

Just my opinion.
I'm glad they found the fault and i'm eagerly awaiting this weeks episode.......

Regards
/Per

126. Quote
You do realize that this whole slew of drama is EXACTLY the reason why so many businesses do not want to deal with end users like (especially like) Jack Rickard and prefer to do OEM deals exclusively?

What so they dont get people asking them real questions?

Trust me having worked on both sides of the fence real questions are asked and if real answers are not given then the threat of big \$\$\$ implications follow very fast

Just so sweat of you

127. I have found all these comments hugely entertaining. I have been checking back regularly. It's a happy ending to the story. Jack gets a faster car and knows he can use the controller for the Escalade, Evnetics endure a slightly difficult week, but then get a large big deal made about how their controller does do 1000 amps on a show lots of people watch, and anyone with the controller with the same problem will now know what the problem is.

I'm looking forward to this weeks show. :-)

128. I trust Jack Rickard's repeated remarks that being an OEM supplier is not all roses and sunshine.
But I believe there is a reason Jack is running into unobtanium so often.

Having an engineering team on standby to deploy to a client for any unexpected component issue is OEM level service.
Since Evnetics' entire market is watching EVTV they have no choice on the matter.

Look I'm not saying Jack should hold his punches but did he really have to spout vitriol about someone's person maturity level?
Or demand real answers without providing the real information that goes with asking the real question?

And yes the aftershow blog is usually very interesting to keep an eye on!
Definitely going to be an interesting show.

129. @Khalen
Well the problem would not have been found even if the logs where sent. The logs just records what the controller think it sees. The fault/unadjusted controller thaught that 722amp was infact 1000amps and would surely had logged it as 1000amps also.......

Regards
/Per

130. Hey!

At 2:05 yesterday, I posted:

"The tested accelerations are current limited to 1000A (+/- the controller's output current measurement accuracy at this current)."

I must have felt the vibes and smelt the whiskey.

Can't wait to see the runs at 1000A if only to compare them to the 722A runs.

Got to say, this whole episode of confusions could have been avoided if measurement points were identified correctly, and Jack hadn't posted comments like "battery current should be about the same as motor current". Must have missed that day in middle school where they talk about the Scientific Method. It's one thing to be a couple guys messing with cars out in the garage, and quite another to be publishing data for the enlightenment of a community.

I only say that because I sense that Jack has the maturity to accept constructive criticism.

Hope you include the smoke and motor whining in the show.

How's the calibration on the Escalade Solitons?

Mike

131. Precisely so. We HAD sent them logs from the 0-60 test a week or so ago. We also sent them the entire xcel spreadsheet with the graphs, and on the show scrolled the entire configuration up the screen while I chanted it.

Had you watched the video , you would know the third gear dash closeups were motor side.

There was NEVER any sentiment to withold logs from EVnetics - a ridiculous concept. I didn't see any purpose since they already had a log and the LOG showed 1000 amps. But we couldn't measure 1000 amps. Indeed, just to clear all the bases we were going to remeasure and log the whole thing. As you'll see in this weeks' show, I've invested in some further equipment, including a Fluke meter that has a wireless receiver display (about \$800). Seb loved it by the way.

And so the next step in a long gruesome tale was to be to dutifully, like a moron, measure battery amps, and motor amps, while doing the log thing - knowing nothing was going to come out any different at all.

And indeed it wouldn't have.

Seb's troubleshooting is superbly analytical and indeed he mentioned in an e-mail as one of his four possibilities just what he found within minutes of arrival.

Here's the problem with several of the after theories, everything WORKED from the beginning. THere was no single component failure IN the Soliton. The hall effect sensor works perfectly.

To measure current, the hall effect sensor simply puts out a voltage proportional to the current. The ones I like, not necessarily the EVnetics ones, use a single 5 v supply and put out a 2.5v signal at 0 amps. This signal is then varied proportional to current flow in EITHER direction. I'm almost sure EVnetics would be unidirectional but in any event otherwise work the same.

This signal is then fed into an A/D converter to produce a digital number. In the case of my Blue Sentry remote sensor units, this varies between 0 and 65535 with 32767 NOMINALLY establishing 0 amps. The problem is, this 0 point can drift with temperature. And you have to derive a scale for the measurement based on the size of the hall effect sensor.

The easiest way is to just wire it all up and calibrate it. Which is what I understand that they do.

The EVnetics was working. The hall effect was working. But the calibration was off. At about 700 amps, the EVnetics is measuring 1000 amps. Familiar number. That's the current limit on the EVnetics. The software duly notes the 1000 amps and limits any further output to just that 1000 amps. We read that as 722 amps.

And that IS what made this all a little confusing. I HAVE THREE other SOliton1's laying around here and it takes less than an hour to swap one.
I should have. Mea Culpa.

Why didn't I? Well 722 amps is pretty impressive actually. That was what mislead me. EVERYTHIGN WORKED. ANd had we sent logs, they would have shown 1000 amps - everything working.

132. If every thing scales down,

9.0 seconds * 722A/1000A

= 6.50 seconds (0 to 60 mph)

Jeffery, brace yourself. You may still suffer wrath for the last half second! Hope that flux capacitor kicks in.

133. And I have to point out in my defense, it was working WELL. We were getting 125 horsepower. We were at a 14.5 second quarter mile. As I've said a million times, I"m not precisely a NEDRA kind of guy. THe performance of the car was quite acceptable for me, but it fell a little short of claims.

Had it put out 100 amps or 200 amps or something like that, I would have certainly swapped it with another one to confirm. But it was just good enough that I thought it was doing the best it could.

Oddly, it was. It was just a software calibration issue. And kind of a unique one.

We're too late for a dyno test this week. I'm hoping Brain can get a clutch in at least for a quick 0-60 this morning.. I assure you we WILL revisit the dynomometer this next week as soon as we can get an appointment. Because this thing is now a MONSTER and probably DANGEROUSLY overpowered. It didn't smoke the clutch on the dyno, it smoked it on the FIRST test drive of about 5 minutes. And it smelled to HIGH HEAVEN. We're not talking about a little slippage. We could smell it coming in the door.

This augurs VERY well for the Elescalade. Two Jim Husted turned Netgain Warp 11's with TWO Soliton1's and 400AH cells? We will have to be VERY careful.

And Duane Ball's Porsche Carrera GTS 904? No more weight really than the Speedster Redux - maybe 100 lbs, and a NEW model Netgain Warp 11 with Soliton1? Get out of Dodge City. That car will be a man killer. NOT POSSIBLE for an ICE version of Chuck Beck's 904 to every appear NEXT to it for more than a few milliseconds.

Did I mention our total sag was to 147volts? How about 197 HP in a 356 Speedster?

Jack Rickard

134. Great to hear it is all worked out with the controller, things were getting heated on this blog post.

Great to know that the dual Warp11 setup will provide plenty of power available in the Escalade.

I can't wait to watch this coming video and see this 356 fly! Did the clutch fail right away or was it after a few hard romps and slippage?

135. GO Figure. And he may not. Acceleration is kind of an integral function. I don't think your little division trick will quite do it. Perhaps. Perhaps 5.0 seconds. We'll see.

Jack RIckard

136. I posted too late, I see you answered the question about the clutch.

137. *******You do realize that this whole slew of drama is EXACTLY the reason why so many businesses do not want to deal with end users like (especially like) Jack Rickard and prefer to do OEM deals exclusively?*******

We just do the same thing but in the public view. Drama still continues within the large corporations. Had this issue not been worked out in a big corporation and they put out all of the controllers with a faulty calibrated hall then the company would be facing a very large recall. Trust me, the the same drama takes place but in a different arena.

I prefer the public.

Now EVnetics will need to be prepared to build a crap load of these things. I am going to be on the list. Guaranteed.

138. Mike:

You are a puddle of negativity. You could find the cat shit at a Church picnic.

There was nothing unscientific about any of it. I'm very pleased with all the measurements. And now what I'm hearing is a bunch of the backbench hecklers and naysayers busted by reality. Their wild theories of how Jack screwed it up didn't pan out. And so in the face of an EXCELLENT outcome, you grouse on anyway?

Here's the bottom line. This community has TOO MUCH SCIENTIFIC METHOD> In fact, here's how that method goes. One of you types some almost accurate misundersting of physics into the screen. The next person takes that, adds a little bit of what "makes sense" to them and retypes it into the screen. And the next person adds a little of "their understanding of the laws of physics" and types that into the screen. 100 iterations of this only take 4 and a half hours on the Internet. But not satisfied with all that it goes on for a year and a half.

And it does it with HUNDREDS of people. Not a one of them ever goes and puts a meter on it, because they're all lazy as shit. It's easier to type yourself smart.

Meanwhile, the constant influx of new enthusiasts gets online and reads all this crap. It becomes its' own religion - complete with hubris and immense anger at anyone who doesn't follow the party line. But very confusing and making it quite difficult for our newer members to join the party , because nothing in their garage quite works that way. They must be doing something wrong....

We come along and measure it, and it doesn't match the by now horribly confused misundestanding of something tenuous from the beginning. And so the online body politic goes BERSERK ranting and raving about what SHOULD have been done by the guy with the meter to uphold their crackpot copper foil helmeted theories.

Some of these things require hours to setup, and as I said, cameras get knocked over, leads come loose, tie down straps break, batteries run out on the meters, etc. etc. But if you make ANY mistake and release the information publicly, you just might as well paint a bullseye on the back of your shirt.

Armchair engineers, pontificating from their lofty perch, espousing theories and criticizing methodologies in minute detail, never quite leaving the chair to do any measurements of course, but offering sage advice in all directions.

Meanwhile Seb wants all our Solitons back. They've suspected some calibration issues and have been working on that procedure. Some more work now. That's how it REALLY gets done guys. And I guess I'm going to bet when YOU order one, it will be properly calibrated to avoid such incidents.

That IS the process. More complicated on the AV8-B Harrier, but essentially the same. You build. You test. You rebuild. You retest. And the realityville version is a little messy. But that's how it progresses.

I would guess a lot of people DO test and measure. But I can well undestand why they don't bother to contribute the results. They generate a lot of heat.

We'll take the heat. It's fun to learn. At the point where I'm not learning any more, we'll be moving on to the Next Thing....

In the meantime, we're kind of taking it as a mission to actually test things, AND share them. For those with an ear to hear and an interest in the topic.

Jack Rickard

139. Khalen. I think your point view is without value.
Seb's attitude and ability has more than any other on the Soliton team has given me great confidence in the product and I'm certain his triumph being paraded on EVTV will (I hope) bring Evnetics great rewards.

We have seen ebay sellers giving nasty rhetoric to irate customers. According to the seller it was never their fault. Would YOU deal with them?

140. Hands up everyone who blamed anything but the Soliton.

You get an F
Next time listen to teacher.

The AV8B! Never touched one of those. Apart from a little job fixing a widget... and that came in a bag.

141. Khalen:

What show have YOU been watching. Spout vitriol? Demand answers? Attack someone's personal maturity level?

I haven't spouted ANY vitriol, I have a very positive attitude about all of it. I have READ a lot of DEMANDS and DIRECTIONS aimed at me, but iI haven't "demanded" anything. I published what we found and absolutely and without any possibility of confusion on your part announced that I didn't know the answer, considered it an open question, and invited anyone and everyone to explain it to me.

In fact, I think I posited that the MOST LIKELY problem was something I had "configured" wrong as that had been my most common discovery in such instances. How do you resolve this, which can be no more on record than a video, with what you have presented as an accusation?

I did get kind of an attack from Jeff and I admonished him that we were going to investigate this in an adult manner and if he simply wanted to spout juvenile crap to go back where it was most appropriate - DIYelectric where he's accustomed to holding court. My attempt at elevating the level of discourse to some civil discussion of a fairly simple data set of importance certainly to me but potentially to a number of our viewers facing component selection.

And my understanding is that he did, and tried to enlist a small army to come here and do the attack for him - certainly an adult approach. Apparently most of them simply declined to do so. I assume pending further data or outcome.

And that outcome is I think tremendously positive news. It is an easy fix for EVnetics really. We have a car so strong we have to buy a new clutch. And the Soliton1 will soon be published with a complete dynomometer data set on an actual car by a third party not known for pandering.

Brain and Matt have a new toy so fast they can probably hurt themselves with it. George is basking on his new model Warp9 doing 192volts flawlessly. It actually doesn't even get warm.

Our viewers have a front row seat to a great little drive train combination that will work on a variety of cars and in the grand scheme of things is really quite reasonably priced.

But I've got a couple of armchair professors here that went out on a limb with a bunch of theories that turned out to be kind of publicly busted, and they're all butt hurt about it. Unhappy apparently at the happy ending?

What am I supposed to do with that?

Jack Rickard

142. Thank you Martin. Your software is still the best. The car IS up to my expectations and really quite beyond. As to the Whiskey, you'll have to confer with Seb on that for a third party evaluation. My own opinion would be quite biased and not very worthwhile.

Ok. In my opinion, I make GREAT whiskey....

143. ============================================
Personally, If I was in the business of making a professional product like a Soliton 1, I would not be commenting in any forum, whether it be this forum or DIYelectric........there should be one contact person only, not four people, especially when they all say different things, direct communication only, phone or email.
============================================

Here too, I find this a very strange comment. You mean you advocate that companies should be remote, unresponsive, and very closely guard their secrets and manage their message. And you think this is preferable to being open, available, and responsive to existing customers and potential buyers.??/

That is certainly a new and interesting point of view.

I find it a bit naive and kind of in the same vein as the para-reality of the forums themselves. Engineering is not the precise, analytical, white walled development lab you apparently picture - ANYWHERE that I' have personally observed. Product development is a messy business and I advocate open and honest as posture simply because it dispels distrust.

I was on a dozen very large DOD programs. I would say almost every one started out under a cloud and several were the subject of congressional investigations. Ten years later and sometimes 20 years later they were being touted as the most successful weapons program ever developed. The difference was of course the 10 years of grudging, step by step, painful reevaluations and improvements/upgrades.

I would love to write a software program that worked perfectly the first time, or even on first release. I would LOVE to design a piece of hardware that worked perfectly out of the box. And after I woke up, I would LOVE a big hearty breakfast and a cup of coffee....

It is very educational to observe the emergence of a new product, the problems it has, and the fixes deployed.

Contrast this to our experience with the MES-DEA 200-250 motor and TIMS600 controller. No public debacle there. And no support there either...

Jack Rickard

144. Jack,

In the original blog for this episode, you made the comment:

"With the pedal floored and the car accelerating uphill, they (motor and battery current) should all be the same."

Just a little understanding of how a current limited PWM controller works flies in the face of this statement, and caused a number of us to want to help you. Would you prefer that we just let it pass? Do I need to give credentials before I'm allowed to point this out?

Mike

145. PmDawn (Per) - Understood. False negatives, or false positives depending on point of view.

AndyJ - Not sure how you got the impression I was ripping on EVnetics. I'm expressing my support for them. I've seen, heard and read nothing but positive things about their product. They even post on those forums that Jack Rickard reviles and I found those posts very interesting to read and they are essentially a goodwill commercial for their crew, product and business. Even their posts NOT about their product involve guiding people away from dangerous designs by Armchair Engineers before someone builds it and sets their house on fire. To top it off, the support that Jack got from EVnetics was above and beyond from what you could expect. My argument is that you cannot do that every time a customer has an issue with your product; and that's why I mention that if you want more businesses to do single unit sales and support it requires customers to participate in the support by means of diagnosis; like measuring motor current or some sort of troubleshooting beforehand. I felt Jack was leaning exclusively on his show's reach to elicit premium support from EVNetics instead of ringing them up and discussing the issue in person.

Jack - The reason I kept an eye on this particular blog entry was exactly because of your most recent show. After watching that and reading your posts here in this thread I felt you were sliding towards a Jim Husted style or even a Metric Mind style confrontation and I strongly felt that was unjustified for the reasons I mentioned to AndyJ. Call it silly, but that's my motivation for posting.

The "demands" you refer to was my retort to an Anonymous who responded to me regarding his experience in dealing with or working in an OEM supplier.

Your response to Jeffrey's post was indeed what I referred to as "vitriol". While my opinion on this is likely ineffectual I don't think it's helping you achieve your goals in weening people of BMS'es or similar beliefs, it'll likely do the contrary. Your many hours of excellent documentation and argumentation on EVTV for these things should be enough.
What I DID do wrong there was assuming Jeffrey was in one form or another associated with EVnetics. My bad.

I am not part of any army. Well, except the one that watches EVTV.

146. Mike,
It was, duh!

The controller acts effectively as a series resister.

So which ever side you measure the current, (cell side or motor side) its the same.

147. As an observer of all of this quarrelling, and I mean an observer because I don't know spit about any of this stuff you guys are talking about. From the start my odds were on the Soliton fellas making it right - and they did.

To Jack's credit he was consistent in saying he didn't know the answer and was open to suggestions. Perhaps a little strong, but that's his style.

Now, let's make peace and get on with it. I've got lots to learn

randy

148. Andyj: Good April 1 joke!

149. Now it seems that we have a SECOND Jack Rickard posting on my blog? You kids ARE exasperating.....

Jack Rickard

150. Especially when these cheeky fraudulent posters go missing. ;)

151. Jack
Turn the acceleration rate back down to 700A/sec.
the clutch will love you for it.
Doc Hiller

152. I can't wait until this weeks show! Jack I hope you treat us to an early release (like today at 9pm) :)

153. Battery current or motor current, which specification would I prefer ?

Well, from a designer's point of view I specify a motor depending on the load it will be subjected too and the speed it will need to rotate. You can always get a bigger motor if you want one, so I specify what I need in terms of power.
I know the voltage of the motor and due to the power conversion I can make a rough guess at the current the motor requires.
That is how I specify a controller when I shop for one, so that is how I want them to be specified for me. It also helps make sure I don't exceed the motor current.
Of course, if the controller manufacturer lies about their efficiency I might find out that the voltage I expect to have to feed to the controller isn't enough.
Battery amps tells me nothing compared to motor amps (at the design stage at least), because I expect to see losses in the controller and more often than not the losses aren't always as good(?) as the makers say - If I specified a drive by it's battery current based on just an efficiency figure I know for sure I would get it wrong.

154. Jack you da man! This is truly one of the best April Fools jokes of all time! Conveniently forgetting to measure motor amps, continuing to not measure, getting hundreds of stirred up posts, and then coming clean on the first of April!

155. Please, white board the Soliton design, flaws, etc. I hope Seb. B. gets some air time. I like that you are your own 'warts and all' 'show me' and industrious American gem. I respectfully think of you as a varying mixture of Prof. Julius Sumner Miller, Nero Wolfe, W.C. Fields, John Gault, and some other notable characters I can't put my finger on at this time. Keep doing great things, and don't get hurt in an super-powered Speedster.

156. Hey Jack,

After all the rantings and ravings I think you should reward the person closest to the correct guess as to what was wrong :)! Maybe a case of Stag beer or a bottle of your home brew!

If so, I would like to claim that reward! My guess was posts #21 and #65 (I think if I can count). These were the ones where I wanted you to make sure that your Hall effects pedal was putting out the correct value. If I read correctly as to what was truly wrong it was the sensor that the pedal talked to that was out of calibration. Close?

157. No NO it was me me with the flux capacitor!
and forget the stag beer
I want some of that fancy wine
Hey is it true its called stag beer because it contains Viagra!?

158. The world needs 2 Jack Rickards!

159. Hi Guys,
now we know, we can all sleep well except for the pontificating arm chair experts. Luckily the Porsche has good aerodynamics otherwise it would literally fly away with so much power.
Well done Jack.
regards Leigh

160. Yes Jack please do a whiteboard.
Some of these rarer posters who are still trying to prove Jeff correct when you and Seb proved otherwise are doing my head in.

I'd love you to give a minute to show the difference between battery current and apparent motor current.

The filth reflected back off the motor are making some peeps bark up the wrong tree.

161. GoFigure:

Ok, you were right. Or rather it was the OTHER way. 6.98 seconds in yesterdays video.

Jack

162. Hey Jack!

If you could allow another guess, everything in the 0 to 60 time fairly scaled to amps except gear shifting. Yeah, it's integrated over time, but now from 0 to 6.98 seconds instead of 0 to 9 seconds.

But hey, in the sixes! Another qeuestion just occured to me. In the redux upgrade, did you beef up any fusible links and/or circuit breakers? If so, what are you using for 1000A?

Mike Kaindl

163. A 6.98 second kitcar *drool*

Must see this!!

164. Great stuff: glad you found the main problem (and the only one that matters in in the car). The dyno figures are still duff, but what the heck...

Clutches and gearboxes. The GT40 community over here generally use Renault UN1 transaxles with some Porsche and Audi.

I have no idea how well a Beetle box will handle big torques or whether you can find a civilised clutch that will fit in the housing to handle it (mine is an 11" plate in a bell housing the size of a young jacuzzi)

If you can find a clutch, watch out for the box.

165. All is well, that ends well.

Although, I much have rather not have this happen in the first place, really. I can nevertheless assure you that this "reality check", will invariably help shape procedures for us.

I'd like to thank Jack for his generous hospitality, company and quite frankly the best whiskey I've ever tasted. This has truly been an unforgettable visit for me.

We're looking forward to our pilgrimage to EVTV in September 2011, it'll be quite the event, I suspect;)

All the best,
Sebastien Bourgeois
Evnetics

when is the Friday show coming out?

167. ================================
Turn the acceleration rate back down to 700A/sec.
the clutch will love you for it.
Doc Hiller
=================================

I am sorry Doc. We've never turned anything down at EVTV and we don't really know how to do that. I think it's COUNTER clockwise, but that is kind of a 1970's thing and we just don't know how to do it with this digital stuff.

The Stage IV clutch worked well.

Jack Rickard

168. Beautiful spring afternoon. Got the video edited by noon and it is rendering now. Loaded the wife and took Redux off to the River Ridge winery for lunch. Got two bottles of Chardonney in her and back home late afternoon.

Please to report that the Soliton1 worked its magic and I did indeed get lucky on arrival home.

It works every time.....

Jack Rickard

169. "I can't wait until this weeks show! Jack I hope you treat us to an early release (like today at 9pm) :)"

You know, I can't either. But it's rendering now and it will likely be another 8 hours or so the little indicator shows. That's 12 processors on solid state hard drives. H.264 video compression in 1280x720 HD just takes a lot of computer - usually about 16 hours AFTER I've got it edited and start the process. That's on the most top of the line 12processor MacPro made. It just chugs for hours....

In a couple of years, that will take 20 minutes. I sure wish we had Moore's Law for batteries...

Jack Rickard

170. ======================================
In the original blog for this episode, you made the comment:

"With the pedal floored and the car accelerating uphill, they (motor and battery current) should all be the same."

Just a little understanding of how a current limited PWM controller works flies in the face of this statement, and caused a number of us to want to help you. Would you prefer that we just let it pass? Do I need to give credentials before I'm allowed to point this out?

Mike
======================================

Mike:

I just can't do this. I can't go there. You ARE the problem and it is precisely the credentials that get you wrapped 30 times around an axle. nobody can unwind that. You just kind of have to twist with it.

THEY ARE THE SAME. And now they measure the same - within a handful of amperes. Yes, theoretically there is a little bit in there for you to argue for months on how many angels dance on the head of a pin.

We have about 1002 motor amps and about 998 battery amps measured RIGHT NOW at full acceleration. My meters aren't any better than that.

Your entire POINT is on the top of your head. It doesn't matter, and it never DID matter to this discussion, and was NO part of the problem.

Now if you submit a curriculum vitae going back to 1916, what part of all that is going to change? I absolutely REFUSED to get drug off into the bushes on this one, and ALL the old guard wanted to go there. I knew it wasn't the problem from the get go, and it was one of the best moves I made here.

If then you want to conclude that I don't understand what your are talking about IT'S OK WITH ME. I'm not applying for a position with your organization. I DO happen to know what you're talking about, and knew full well it simply was not pertinent to this problem. If you want to use this very direct situation to pontificate on waveform analysis TAKE IT TO DIYELECTRIC WHERE THEY CARE. You will be thought INCREDIBLY brililiant by a very small group of very confused people there.

For any reasonable purpose, at the point of operation we were concerned with THEY ARE THE SAME. And now, lo and behold they ARE the same as measured, close enough for government work.

The problem was the Soliton was reporting 1000 amps while it was REALLY making 700 - measured on the motor or the battery. What you didn't know was we already HAD measured it both and indeed, they are about 16 inches apart in the car. We don't even bother to determine which is which when we are measuring. THAT's how little it matters.

Hijacking a problem discussion off into the bushes to make WHAT point. This is already confused enough.

Jack RIckard

171. ======================================
Yes Jack please do a whiteboard.
Some of these rarer posters who are still trying to prove Jeff correct when you and Seb proved otherwise are doing my head in.

I'd love you to give a minute to show the difference between battery current and apparent motor current.
======================================

Well some further damage headwise then. Seb and I didn't prove anyone right nor did we prove Jeff wrong.

I went to Jeff with some data and a problem and he responded in a kind of condescending manner I take badly. I tried to get it back on track and he went there a second time in the usual forum fashion in all respects. I responded that if that was the notion, take it to EVDL and DIYELECTRIC where that juvenile level of discourse is of course ALWAYS appropriate. We need an adult on this problem.

Unfortunately, he did just that and tried to enlist support there among a lot of people who other than taking a moment to note that they indeed were wise and knowledgeable and Jack Rickard was not, continued to graze placidly and look out over the fields of grass. He rather mischaracterized our entire conversation to make his case to them and I do fault him eggregiously for that.

BUt he was not "wrong" about the problem in any sense because he never really addressed it. He prefered to feel superior and note at every opportunity that I really didn't know what I was doing.

As I myself note that in every episode, it kind of didn't have the effect he hoped. And I really didn't have time for it this week. So I sent him to DIYELECTRIC until he can learn manners and associate with big people.

But it had nothing to do with the problem. We weren't getting the amps and weren't getting the power. Seb solved it entirely by himself. And it took him not even 10 minutes. He jabbed a clamp on a cable and had Matt put on the accelerator and brake simultaneously, (PLEASE don't do this- ever) with the vehicle in gear. The log showed 1000 amps and the meter showed 680 - just as I'd been telling them. Out came one Soliton and in went another. Batta boom batta bing.

From there, we did what BIG people do, drank whiskey and talked about our daughters. THeyr'e great at 2 1/2 and miserable at 14 1/2 years old. But the WHISKEY is EXACTLY THE OPPOSITE.

Jack RIckard

172. Update. The stage 4 Kenedy clutch and 3000lb pressure plate continue to work well. YES we DID then blow a fuse.

We also had another anomaly. Occasionally the Soliton just faults out and will not complete the startup sequence. I've played with the pedal, and got it to come up that way. I think it is sensitive to accelerator input on startup, and we probably forgot to put a deadzone or an insufficient one when we calibrated the throttle on the Soliton1.

Or we have an intermittent. I'll know more tomorrow.

Jack RIckard

173. But hey, in the sixes! Another qeuestion just occured to me. In the redux upgrade, did you beef up any fusible links and/or circuit breakers? If so, what are you using for 1000A?

Mike Kaindl

Good point. You might have mentioned it BEFORE we blew the 400 amp fuse.

Jack

174. I should probably say I'm sorry Seb had to get on a plane and come all this way to fix our car for us, particularly since we have THREE other Soliton's we could have thought to swap out at any time.

But I can't. We had a hoot spending the day with him, and he' made GREAT video - he's really very good on camera and I think you'll find this week's show one of our better ones.

Is this what I have to do to get industry guests onto the set?

I'm learning this video thing....

Jack Rickard

175. Jack,
I have question that has been bothering me. In general batteries are modeled as a pure voltage source in series with a resister (the internal resistance). So when you draw 1000 Amps from the pack it would seem that the pack should be dissipating 1000Amps * the voltage drop (sag) in the pack. That would be 40KW if you had a 40 volt sag. Is this truly the case? If so it seems like a tremendous amount of heat would be generated in the battery boxes (like 40 small space heaters). If not, then do the cells just lower their voltage without actually dissipating that many watts internally? Sorry if this sounds like a stupid question, but you have spent a lot of time characterizing these cells and I thought you might have some insight.

Thanks,

David Seabury

176. I do, I can't really, and I will. Basically you are going to get me into a lot of trouble here with an endless series of critiques again from the peanut gallery who have been taught some formal things and cannot get around them.

But I will.

They do not. I have never liked that characterization and eschew all discussions of internal resistance for this reason. If we take our 191v and pull 1000 amps from it, we drop in this case to 147 volts. We can account for this drop with what is actually technically termed an EQUIVALENT SERIES RESISTANCE. The main thing I want you to note is that if it was a REAL series resistance, we would not need to refer to it as EQUIVALENT.

If we divide the 44 volts by the 1000 amps, we can derive a result of 0.044 ohms or 4.4 milliohms which is sometimes misreferred to as the "internal resistance" of the cells (this is forum speak). The cells indeed are OPEN as NO ELECTRON FLOW OF ANY KIND is supported within the cell. And indeed if you measured with an ohmeter, you would simply confuse the meter, but it would always read OPEN if it were not for the voltage of the cell screwing up the measurement. And so this 4.4 milliohms is properly an EQUIVALENT SERIES RESISTANCE. and satisfies tis portion of the ohms law equation to account for the voltage drop, which is quite real.

So what causes the voltage drop? Let's see. First do I know? And second, can I explain it?.

There is a solid electrolyte interphase barrier at the surface of the active material on both the anode and cathode side. Once this is penetrated, the ions have to migrate or "tunnel" into the kind of hairball screwy maze that is a very complex crystalline structure, until it finds a location where it can share the exact number of electrons necessary from adjoining atoms to satisfy the desire for the outer vallence band of the Li+ cation AND the olivine crystalline structure of the material. This is termed intercalation. This takes time and there is a term for it which I can't recall but is essentially diffusion delay.

When you put a very low ohmic load such as a motor coil across the terminals of the cell, the "charge" on the cell lowers as electron flow migrates from the negative terminal through the motor to the cell positive terminal. In order for this flow to be maintained, additional Li+ cations must intercalate into the material, adding to the charge potential, which produces additional currrent. This diffusion delay will result in a lower apparent charge.

This is all VERY different from dissipating power as current passes through a resistor. THere is no resistor. IN FACT there is no current flow. Inside the batteries, charge is carried ionically, - by the migration of Lithium Ions. The "shuttle" back and forth between anode and cathode as a result of charging and discharging. ALL the current flow is external to the battery, and ALL the power dissipation is external to the battery as well.

Internally, there is no electrical connection, no current flow, and so no resistance.

This does NOT mean there is no heat generated in the cell. The intercalation and deintercalation cause a warping of the crystalline structure, and indeed there IS electron flow from the current collector plates through the terminals to the external circuit. So they will heat on discharge and on charge, but not due to internal resistance or equivalent series resistance.

177. In reading this, I can't make sense of it either. I guess the two points I would leave you with is the concept of internal series resistance is entirely misunderstood int he forums and I'd like you to think of and refer to it by its proper term EQUIVALENT SERIES RESISTANCE.

Second, there is NO electron current flow from anode to cathode or cathode to anode INSIDE the battery. There is a charge shuttle mechanism with Lithium ions as the carrier. But NO current flow.

Let's see, is there a more fundamental analogy. A magic box with a chemical reaction inside of it. This chemical reaction, like a slow fire or a rapid rust, either oxidiizing or a reduction, CAUSES EXTERNAL CURRENT FLOW. But it doesn't have any.

SImilarly an Oil Fire beneath a boiler can CAUSE STEAM, but it never gets wet, it isn't steam, and in fact steam would put hte fire out. Of course, if you take away the fire, you have no steam. So we can develop a formula for heat into the fire, and steam out of th boiler, and even define the energy input as steam but it is EQUIVALENT STEAM when it is in the fire. It can never be real steam.

Similarly we can view a battery as a current source, and indeed it is. Ironically, there is NO current flow within the cell itself.

If you'll recall from a recent show, I said I've come to view voltage as entirely mythological. It's the ONLY thing we can measure, but it never IS anything. A battery is NOT a voltage source, it is a CURRENT source. We can measure it's POTENTIAL for making current at the terminals and in fact we call that VOLTAGE. BUt it really isn't anything. THe current IS a thing itself - the movement of electrons. Like water in a tank, we can measure PRESSURE, but you can't drink pressure. It isn't anything. It's a SYMPTOM of water. And you can old watera nd you can measure water flow, but pressure is just a SYMPTOM of water - a byproduct. It could also be a symptom ofr air and indeed it would be a great way to measure HOW MUCH air, but it is NOT air. It's just pressure. You can't fill a tire with PRESSURE you can only fill it with AIR.

It is quite possible to receve a formal education in all this, and become very good at the formulas and relationships that we use to manipulate electricity to do our bidding. And in fact, some of these people become the very best at that manipulation.

Then there are 75 guys on the planet that really understand parts of it. I've never met one who understood all of it. And unfortunately, I'm not one of the 75.

But they used to let me hang out with them..... I was very good at making coffee you see and told really quite entertaining jokes....

Jack Rickard

178. Andy J said:

"Hands up everyone who blamed anything but the Soliton.

You get an F

Next time listen to teacher."

Wait...who's the teacher?

In hindsight, yeah, it's kind of obvious - but it always is after the fact. There were lots of ideas that deserved an A, they just happened to be wrong. But they were good ideas - they might have been right.

Jack blamed the Soliton, but for the wrong reason. That's OK, his ideas were good, they just didn't happen to be exactly right (and he did hedge, which is usually good scientific practice).

Several people (including me) said put a scope on the output. This may or may not have turned up any more clues, but it was one thing that hadn't been tried (as far as I know). This would have directly tested a hypothesis (that the controller was intentionally limiting the current).

Actually, everyone gets an A for effort. Extra credit if you were nice. :)

179. Andyj:

I didn't go there. And you are actually correct, you can put a scope on the output, if you REALLY want to.

Have you ever done this? With 190v and 1000 amps or even 722 amps?

It's a pretty noisy environment, and you have to be pretty good with a scope to get anything but nonsense. On a Dyno?

Jack Rickard

180. Very enlightening - here is what I learned:
- Jack correctly determined that his controller was not up to snuff, but could not perform the precise tests necessary to prove it.
- A Soliton controller had a bug so did not produce the current it should. The manufacturer found the problem, fixed it, and are digging further to see if it affects anyone else. Full marks.
- Jack does not know how a PWM motor controller works (battery current is not equal to motor current) and is not willing to admit it.

181. One final note on diffusion.

You may or may not notice, but many of the manufacturers of cells offer two variants, a high power cell or a high energy cell.

A high energy cell provides more energy (AH) in the cell than the high power cell.

The high power cell can do more amperes for short duratoions. This is why some specific cells can do 20C, 40C, 60C, or in very rare events 100C.

What they are doing and talking about is a trade off in battery design.

If you make the active material that is smeared on these aluminum cathodes (primarily) VERY VERY thin, the intercalation of lithium ions into the crystaline structure can be made easier, or sped up quite a bit - SIMPLY by making it thin. This allows very HIGH current levels with relatively LOW levels of voltage sag.

When you do this, the ratio of active material to current collector naturally goes down, and so your total available energy from the cell measured in AH ALSO goes down. A high POWER cell.

The reverse trade off is of course to put MORE active material on the current collector. This causes the capacity of the cell to go UP with more active material, but of course the diffusion factor goes up as well, limiting how much current can be delivered at any one instant. A high ENERGY cell.

Jack Rickard

182. ===============================
Very enlightening - here is what I learned:
- Jack correctly determined that his controller was not up to snuff, but could not perform the precise tests necessary to prove it.
- A Soliton controller had a bug so did not produce the current it should. The manufacturer found the problem, fixed it, and are digging further to see if it affects anyone else. Full marks.
- Jack does not know how a PWM motor controller works (battery current is not equal to motor current) and is not willing to admit it======================================

There's a slow one in every class. Would one of you brighter kids take Clyde here back in the corner and go over his multiplication tables with him? Any volunteers?

Jack

183. Thanks Jack!
That clears up a lot alot. It is a relief to know you are not actually dissipating that much power in the pack itself.

Regards,
David Seabury

184. =========================================
Jack blamed the Soliton, but for the wrong reason. That's OK, his ideas were good, they just didn't happen to be exactly right (and he did hedge, which is usually good scientific practice).
=============================================

WHERE do you guys GET this shit. Does it grow on trees? The creativity is mind boggling. But it's mind bogglingly STUPID.

Jack blamed the Soliton? For the wrong reason?

What freakin part of IT WON"T DO 1000 AMPS did I get wrong? It DIDN'T do a thousand amps. The guy who MAKES it looked at it for 10 seconds and said IT DOESN"T DO 1000 AMPS. Everyone on the planet now, lacking perhaps YOU knows it didn't do a 1000 amps. And he replaced it with one that DID do 1000 amps. And now I've GOT one that does 1000 amps.

I never OFFERED a solution other than an observation that it doesn't do 1000 amps. I RESPONDED trying to eliminate a number of crackpot theories, most of which I new to BE crackpot theories.

So you got the long bouncy limb you had built yourself with a bunch of bullshit theories sawed off from underneath your dumb ass RIGHT when you were getting warmed up in public and your all butt hurt over it.

How does that make any part of anything I've said WRONG in your ILLUSTRIOUS and EXHALTED and obviously very VALUABLE freakin opinion?

The good part of this, or the bad part of this, depending on who you are and how you are looking at it, is that there might be a BUNCH of people out there very happily doing 700 amps with Soliton controllers. Since they don't measure anything, and certainly not on a dynomometer publishing charts for half the world to see, they are relying on dumb theories from cranks like YOU to assure themselves that they are indeed getting 1000 amps- they're just "special" amps that only you can understand.

185. I can just picture them driving down the road, fat dumb and happy thinking their Soliton 1 works GREAT at 3/4 power.

And frankly, it DID drive quite well. We just weren't getting the advertised power. We are now, and it IS a big difference. In factd, I would characterize it as an ENORMOUS difference. But with precisely one data point to compare, we could have driven that car for the rest of my life and the life of the car and never known there was a problem if we hadn't measured it.

We brought this to light. The manufacturer is now thoroughly aware of it, and I'm persuaded quite motivated to fix it so it doesn't become an issue in the future. I think it will be yeoman's service to those who have Soliton1 controllers. I think it will be good for EVnetics. I think it will be good for the future of the product. I think it will be good for Jeff. It's kind of good all around.

But you've managed to find a negative in it. We noted a problem in that the controller was not putting out a 1000 amps. We never offered a theory, we just kept measuring and trying out things EVnetics and others offered, never with any success. We never GOT to wrong. We asked a question, and we measured to find a result.

Sebastien solved the problem. You are correct in that I did NOT. I DID observe that the Soliton thought it WAS making 1000 amps as shown in the logs. And I noted quite publicly that I COULD NOT CONFIRM THAT with available test instruments. In fact, I consistently got a lower reading.

I heard dozens of theories quite specifically why my measurements were wrong, and in fact could never BE right. None of them were. They were spot on from the first one to the last one.

Seb flew up and confirmed EVERY BIT OF THAT. And he had the solution in his bag - a different Soliton. Fair enough.

IN THE FACE OF ALL THAT, the hysterical, womanish, homo, poseurs among you who all critiqued endlessly but almost ALWAYS attacking the messenger and the measurements as impossible, now continue to TYPE FURIOUSLY hope against hope to alter REALITY to fit your typing? You want to go back and make it NOT so by rewriting how it happened? Without EVER being ANY PART OF IT or even in the same State at the time? But you're going to summarize and in fact rewrite it for us and explain what it really MEANS??????

Jack

186. "Womanish" and "homo" as insults? I can only hope that was the "SECOND Jack Rickard" you referred to before.

187. Hey Jack, with all your typing this evening have you gotten the impression they are not willing to learn? I suppose that means you can't type someone else smart either.

188. Maybe people need to register and sign in to have there two cents worth, it would stop the cowards that remain anonymous filling the blog with crap.

189. Oh yeah. Where have you been. The REAL Jack Rickard left the blog a couple of days ago in disgust. I'm all that's left.

He told me "good luck" with you girl scouts right after my first post.

I hear tell he's looking for two more for some other work he is behind on.

Jack

190. Very interesting show and blog again. I find Evnetics a impressive bunch of people. OK, Jack and his team too. Have to say that too of course. But that evnetics team is really a bunch of talented nerds. Very open, honist and skilled. Man, I wish they started on an AC inverter.

Regards, Jan

191. Hi Jack, your a better men than I to put up with all this verbal Diarrhoea. l would have just left them to there madness. But thank God for you Jack while they are pontificating their theories most of us are getting on in life and finding the real truth by actual experience which has this fine knack of waking us up with all sorts of unpredictable outcomes.
Welcome to the real world guys. Get out of our lives and out of your armchairs and get a life and do something worthwhile instead of criticising others its a wonderful world out there if you DARE TO TRY.

Enjoy yourselves
Thanks Jack
Regards Leigh

192. Never was a truer word said

Get out of your armchairs and get a life and do something worthwhile instead of criticising others its a wonderful world out there if you DARE TO TRY.

193. Jeepers creepers Noah, give it a rest will ya? The problem was with the Soliton but Jack always gave it the benefit of the doubt saying it could have been something he was doing wrong. Missed that part, did you?

Also, brilliant ideas that don't work can't get an "A". A "nice try" star is all you get.

You know who gets an A here? Seb. He had the business-sense to look beyond the VIP customer complaints and rather than posting a jab he got on a plane and fixed the problem. He gets an A. And when they're splitting the company stock, the Board members need to remember that it was Seb showing that level of commitment and foresight that brought in all that business.

JR

194. "The problem was the Soliton was reporting 1000 amps while it was REALLY making 700 - measured on the motor or the battery. What you didn't know was we already HAD measured it both and indeed, they are about 16 inches apart in the car."

Just think of all the theorizing you could have avoided if you had mentioned that fact.

195. Jack said:

"Have you ever done this? With 190v and 1000 amps or even 722 amps?

It's a pretty noisy environment, and you have to be pretty good with a scope to get anything but nonsense. On a Dyno?"

Admittedly, no...but I've done it with 80V and 450A, and I had a perfectly clean PWM signal.

Anyway, obviously 1/2 of you folks are more interested in bashing and name calling than having a rational conversation. Those people are why internet forums and comment sections have such a bad reputation.

Too bad. I'm going to head back to my usual spots where people treat each other like adults.

cheers!
Noah

196. "the hysterical, womanish, homo, poseurs among you"

You tell 'em Jack! The world of EVs is for heterosexual males only - women and gays need to keep using gasoline!

I trust the sarcasm in my response your sexist attitude is self-evident.

197. Oh No!
Having come back on here after my last look at Jacks last video and seeing a flurry of new posts all because someone was upset for getting a few "F"'s as suggested by me.

I like to stir the pot just to make minds tick. Ego's were severly hurt. I'm sorry its got too far.

Attacking Jack over the Soliton1 in my opinion is seriously outside the bounds of sanity.

I'm personally accepting an 'F' for thinking Jack might not of set max KW for the motor. A new feature on the latest software. It's on the video, he did! I'm happy to be proven wrong because next time I'm stronger for it.

There were times some made simple statements in order to correct others but were not heeded or considered. Why not?

Examples:
(sic)
'The pack will give 1KA and the motor cannot help but accept 1KA. What else is in the way?'

'A PWM controller is like a series switch. It does not dump the current. So motor side and cell side currents are the same apart from the controllers losses.'

All those people who did not listen or blamed the dyno, internal resistances, CEMF, no BMS, blah blah. Then listen to me. I have title deeds to all of Libya's oil if you wish to make a wise investment.

Er, Jack, you are absolutely correct about using a scope on the o/p of a controller. Its as said before, a filthy environment. But it's do-able ;)

This week has upped my faith with Jacks findings and Seb's product beyond measure. Anybody else wishing to run a business, you should remember. The customer is always right..... Until proven guilty in a court of law.

198. this thread is so awesome. it makes every asshat I have to deal with in my daily life seem like a reasonable human being. thanks jack, you rock.

199. I'll keep this short. Andy J said:

"Er, Jack, you are absolutely correct about using a scope on the o/p of a controller. Its as said before, a filthy environment. But it's do-able ;)"

Indeed it is doable. Looks pretty clean to me: