Monday, March 7, 2011

Not Everything Works the First Time

Interesting show this week. Mostly a rundown on our additions to the EVTV test bench, which is coming along quite nicely.

We've added a control panel for the generator, with a contactor, current shunt, control switch and voltage/current displays. This allows us to turn on the generator or leave it off, depending on what kind of load we want to put on it. And we can measure the current back into the battery pack.

I'm once again disappointed to report that we get less power OUT of our perpetual motion machine than we put in. But it gives us a bit of a load we can cut in. I'd kind of like to add a PWM circuit or controller to this eventually so we can dial in just how much of a load. But the 150 v battery pack doesn't present much of one actually. We have to turn the generator up to a couple of thousand RPM to really get 80 amps or so out of it.

We also added a control panel to the input side, quite a bit more extensive. It has a Speedhut tachometer with the RECHARGECAR magnetic pickup to display RPM. We used a handheld laser RPM counter to check it and it probably reads about 50 rpm low - very acceptable. We're using this at 4 pulses per turn and it is working well with the tach.

We've also added a vernier 5K potentiometer to act as our throttle control. And we use a precharge resistor/contactor/shunt box originally used on our first Speedster. The control panel has a switch and a light to turn this power on. And a 1000 Amp 50mv meter that displays the current through the matching 1000 amp 50mv shunt in the box.

This gives us a little extra safety. We can of course turn the ignition voltage off to the controllers from the panel, but we can also cut off the power through the contactor. Turns out to be a handy feature as you'll see in the video.

We wired up the two Soliton1 Controllers to the two Netgain Warp 11 motors. Therein lies a tale but also one of the reasons for the test bench during the Escalade conversion.

The EVnetics Soliton1 is a relatively new controller and has some very interesting features. Most importantly to this application an IDLE function that maintains a low level RPM control on the motor with the throttle off. We know of no other controller with this feature.

Normally, with Siamese motors, a series/parallel two speed electronic shifting system is used. This requires a lot of contactors, and at the powers we are applying, that is a bit of a problem.

The advantage is of course that you can apply the full current from the controller through both motors at low speeds. This allows you to get maximum torque from both motors since in series they both get ALL the current put out by the controllers through their series armature/field windings.

The disadvantage is that it drops the voltage applied across each motor in two. So a 192v pack putting out 1000 amps would put a little over 95 volts to each motor at 1000 amps or 95kw each for 190kw. But as the motors turned up in speed, they generate back EMF (electro-motive force). Think of them as also acting as generators in the reverse direction. These motors normally would start to drop in torque at 3600-3900 rpm at 192 volts but with half the voltage, the torque dropoff from BEMF would also decrease - maybe 2500 rpm or so.

And so once your vehicle is up to speed, you would shift into parallel mode. In parallel mode, the current output of the controller is applied to each motor separately, or in parallel. The advantage here is that each motor receives full voltage and so can move the torque drop off back up to 3600 rpm. The disadvantage, of course, is that each motor only sees 500 amperes maximum. This really is NOT a terrible disadvantage, because by the time you are going down the road at 2500 rpm, your need for power is quickly diminishing in normal driving, and 500 amps is generally a lot, particularly from two motors on the same shaft. It's still 192kw.

I don't like the system. THe problem is:

Failure items
Shift points
Operational complexity

The wiring is simply more complex with several contactors required. The system is of course switched with the car underway and so the contactors have to break some current. The contactors can do that - up to a point. Generally we use contactors ( a misnomer really for a high current capacity relay) to apply power or remove it but in normal operation they are not carrying ANY current at the time. You close it when you start the car. You generally open it when you shut it off. And you are parked both times and drawing near zero current.

In a pinch, you can use the contactor relay to break current in an emergency and shut down the system. The Kilovacs we use, can break up to a couple of thousand amps - about once. And they DO fail. Ergo the mechanical disconnect switch backup. There have been real incidents of contactor relay FAILURE to break current in a high current situation.

It certainly decreases their life expectancy when they are opened with significant current and this is spelled out quite graphically on the data sheet of the contactor. The mechanical life is a million cycles. At 200 amp current you get 12 cycles. At 2000 amps you get 1 cycle, maybe.....

So they become failure items.

Then too, you have to have some means of initiating all this shifting. And indeed, THEN you have to actually do the shifting. This of course COULD be automated. But now we have a controller and series of relays to control our contactor relays and so complexity builds again.

How about we just use two controllers.

We put one controller on one motor and the other controller on the other motor. Now each motor can have the full 192v AND the full 1000 amps all the time.

Of course, we have the cost of the controllers. And then they have to have pretty similar outputs. You would think they would have to have exactly matching outputs or the two motors would fight. I didn't think this was the case. Both are applying torque in the same direction on the same shaft. Even if one was applying half torque while the other was applying full torque, you should get the sum of the two torques. There is no war going on here unless one is commanding it backwards.

But I had never heard of anyone doing it.

Basically, it ought to work. The outputs of the controllers really can't feed each other, they are on entirely separate sets of windings. The only common point would be the batteries and the control inputs. But having never tested it, it was kind of a theory, not a knowledge. As I stress over and over, your EV does not care IN THE SLIGHTEST what you THINK about it's operation. Your theories might entertain you, but the car just does not give a shit. It will follow the actual lawas of physics as it interprets them, which is generally a might differently than how YOU interpret them.

Ergo the test bench.

And we did learn quite a bit rather quickly. One is that the output of the magnetic pickup is probably two light for two Solitons's and a tachometer. Our Soliton's were giving some erratic and very erroneous RPM readings - generally 100-150 rpm high, but also not very stable. I might be able to dress this up with some resistance value across the output. We'll have to play with that.

We did cal the two controllers to fairly precisely measured voltage outputs from the 5K pot. We tied the 5v and signal grounds together, and then the 5v signal as well. And calibrated each controller separately for 1.00v min and 4.00v max. That gets us off on the right foot. The motors turned very smoothly and had no apparent problem working out the torque sharing at any rpm. Even noise potentially fed back to the input just wasn't a problem. The controllers obviously have some capacitors on the input to smooth things a bit and we might be able to augment that (future Top Secret video). But it doesn't seem to be a problem at all.

The other target of opportunity was the idle. Obviously idling we don't NEED two motors and two controllers. To turn the transmission pump, the steering/brake pump, and an air conditioning compressor should only require 2 or 3 horsepower. So EITHER motor could be used. So we set up ONE controller with idle and the other without.

The idle function in the Soliton1 is pretty cunning. It uses a PID algorithm to seek the target RPM and provides whatever current is necessary to get there - up to a limit you can actually set separately. I like the design.

This simple concept, maintaining RPM through the controller, is actually a black art and heinously problematical. It looks easy, but any correction tends to overshoot, and cause another error input, which causes another correction, which of course overshoots again. The cycling can hit all sorts of resonances and self enhancing oscillations. Generally lumped under the term hysterisis.

This from Wikipedia:

"A proportional–integral–derivative controller (PID controller) is a generic control loop feedback mechanism (controller) widely used in industrial control systems – a PID is the most commonly used feedback controller. A PID controller calculates an "error" value as the difference between a measured process variable and a desired setpoint. The controller attempts to minimize the error by adjusting the process control inputs.
The PID controller calculation (algorithm) involves three separate constant parameters, and is accordingly sometimes called three-term control: the proportional, the integral and derivative values, denoted P, I, and D. Heuristically, these values can be interpreted in terms of time: P depends on the present error, I on the accumulation of past errors, and D is a prediction of future errors, based on current rate of change. The weighted sum of these three actions is used to adjust the process via a control element such as the position of a control valve or the power supply of a heating element.
In the absence of knowledge of the underlying process, a PID controller is the best controller. By tuning the three parameters in the PID controller algorithm, the controller can provide control action designed for specific process requirements. The response of the controller can be described in terms of the responsiveness of the controller to an error, the degree to which the controller overshoots the setpoint and the degree of system oscillation. Note that the use of the PID algorithm for control does not guarantee optimal control of the system or system stability.
Some applications may require using only one or two actions to provide the appropriate system control. This is achieved by setting the other parameters to zero. A PID controller will be called a PI, PD, P or I controller in the absence of the respective control actions. PI controllers are fairly common, since derivative action is sensitive to measurement noise, whereas the absence of an integral term may prevent the system from reaching its target value due to the control action. "

The Soliton1 allows you to individually specify the proportional, integral, and derivative values. I wouldn't have a clue if you e-mailed them to me. I took the defaults. It works pretty well. By cutting in and out the transmission and the generator, we could vary the load. And while the Soliton can't accurately measure RPM, it did a good job of maintaining it.

All of this scratches my ongoing itch for subtle ironies, which I mostly use to entertain myself. In this one case, I'll share. A Soliton is a standing wave, first observed in a canal of water. The soliton phenomenon was first described by John Scott Russell (1808–1882) who observed a solitary wave in the Union Canal in Scotland. He reproduced the phenomenon in a wave tank and named it the "Wave of Translation".

I rather associate this with the wavy pattern in the Soliton heat sink. I would have named Soliton Jr. the Compacton instead, but there I go.

In any event, this PID idle control was why we selected the Soliton1 for the contest and the Escalade, NOT the purported 1000 amp output. And it appears to work very well. I think it's a unique feature they almost added as an afterthought, but promises to differentiate their product from most others. Simply holding a throttle input with the A/C kicking in and out, and the transmission cycling and who knows what else would not really make this work in any satisfactory fashion. It would have required an entirely additional circuit just to control the controller had we wanted to do an automatic transmission without it.

The one fly in all of this is that to START the idle, you first have to blip the throttle past your target RPM. I don't like this and I do not think it is necessary. It might be salutory to have a separate input to start it. In this way, we could use the START signal, separately from the ignition signal, to start the idle. But if the Soliton powered up on an ignition 12v input and established idle after a brief delay, there really isn't an issue here. Automatic transmission vehicles really only let you start them in Park or Idle anyway. They did not need to take this "safety issue" on themselves.

THe problem is that it makes operation of our car nonstandard. In an ICE engine vehicle with automatic transmission, you turn the key and the engine starts and idles. Period. In our Escalade, we'll have to turn the ignition key and then give it some throttle to "start it." How do I explain this useless feature to my daughter. She'll turn the key. Nothing will happen. And she'll get out of the car and ask me why it is broken.

I think we should build our cars where they operate as expected. These standard operational issues were worked out over the past 100 years without any input from me, and I don't think they need to be reworked by the crew at EVnetics.

The controllers and motors worked quite well on our test bench. Unfortunately, the transmission somewhat less so. We had it completely full of very good transmission fluid. We had no external heat exchanger but no intention of operating under any serious load, for any appreciable length of time, or at anything over about 2500 rpm. But Matt noticed early on some heating of the shell, which I measured at a peak of 140F. This is hardly warm by transmission temperature standards, but we were on the other hand hardly turning it. At one point, I pulled 80 amps out of the generator - maybe 14 horsepower - through a transmission purportedly capable of handling 800 horsepower.

But it appears to have failed anyway. At about 2000 rpm while filming, it suddenly threw on its own load and started a rather noisy vibration from within. We quickly shut off the system. Restarted it at VERY low RPM's and was immediately able to isolate the problem to the TCI transmission. We'lll be contacting them to see if they have any thoughts on the topic this morning.

You'll no doubt enjoy the onscreen panic that ensues. Kind of a KeyStone Cops meets the transmission shop.

Jack Rickard


  1. Another good show, and I have to say the cut you make after the transmission gums up had me laughing for five minutes. (Not at you, of course, I've never had a transmission fail, and certainly not one that costs 7-10k.)

    I appreciate the good humor in spite of the setback. I do have to say that I was initially surprised at how much power you were losing in the transmission and now I wonder if it was a sign. Who knows? Anyway, let's hope a new one is more efficient and reliable.

  2. Jack , regards the current displayed on the laptop versus the ammeter. Were you reading motor amps on the screen? That may explain the difference.


  3. Jack, it seems from a quick look around the net that not only does the Soliton version 1.3 software include PID loop control for idel, it apparently also has a 'start' input to get the idle control loop running (apparently it is a requirement to meet certain standards). Yoiu might want to look into that.
    Also, if it is not fitted to your model, then a push to make switch which bypasses an open throttle with a fixed resitsance will give the controller enough of a signal to get the motor running, that might require just a simple 'push to start' button on the dash.
    As for tuning the PID control, okay, I won't say a thing...


  4. I greatly enjoyed the new episode. When things go wrong in engineering land maybe it's a problem (although you learn a load of stuff), but when things go wrong in TV land ...well that's brilliant. That's what everyone wants to watch. There are whole TV formats dedicated to people screwing stuff up and many millions of people watch them. ...Anyway it seemed like a fairly minor hiccup to me in the general scheme of things. I hope you work out what went wrong.

    Jack you should get your viewers to do your cataloguing for you. Crowd source it. You have 900 or so viewers so devoted that they are willing to write you an essay. You have their email addresses. write back and tell them they have the chance to win a place at this conference your doing and in return all they have to do is help you catalogue your videos. provide them with a website and a standardised way to do it and you'll have the job done in no time.

    -Nick F :-)

  5. Great show but I would really recommend the operator be in a separate room in case it all comes apart. Something else I noticed was that the dual 11's had no form of air cooling. (Weren't the fans taken out?) Also you were just in first gear at 2000 RPM's on the input shaft, whats the gear ratio in 1st? It probably about 4 or 5:1 so that would be spinning the genny at 500 RPM or less. I would not think the genny would be very efficient at those speeds.

    Thanks, Jim

  6. In terms of programmable chargers, you might consider Piktronik KOP series, these are pretty robust, usually used in marine application, excellent detailed setting and smooth software, EU made, i.e. certified.

    Unfortunately for your application, these are at the max ~48V@20A, so about one 1kW power, you would have to divide into batt. subpacks, which is preferable anyway. Perhaps they could take a custom project, e.g. combine three units into single one?
    software info:

    PS in some of the recent episodes you mentioned, that you play Tesla stock as reaching top at $352 some future day, that was a joke or viable strategy? Thanks

  7. Jack, not sure about the yearly insolation at your place, but there is also Aurora Wind inverter, which is design for feeding from <7kW perm motor/generator. Basically, you drop a floating contraption into the river with submerged rotor positioned against the stream of the water and you are golden: electricity 24/7/365 - unlike solar and its not shiny days. There are some pilot setups like this in the UK underway now, you can coppycat their basic hydro shape and suspension.

  8. What Fun!

    It is just a million times cooler to blow up a tranny and not have to call the flatbed. I've actually never had it so good. The best I've ever lucked out was being able to limp home 20 miles IN REVERSE. That was in farm country, in broad daylight, and 25 years ago. It was OK. To watch you two do it in the garage on TV is even funner. Jack, the look on your face...

    It seems to me you lunched the torque converter. I have no idea whether you were supposed to be controlling the lockup electronically, and I couldn't see how Matt was shifting the box (whether with the standalone control box or manually.) Its going to be very interesting to see what TCI finds in the post-mortem on their "triple tested quality" unit that lasted 10 minutes in service.

    If I were you, I'd send Mr. Hauber with the gearbox to film the teardown. Either you made a mistake and broke it, or you've gotten yet another hugely expensive, custom made flagship product that coughs and dies right out of the gate. I wouldn't leave that investigation undocumented. You really need to know either way, and not just that "you wrecked it," yes? If you were at, say 200 Amps, that's about 40 horsepower, and a very difficult failure to understand at that power level, unless you blew the lockup TC because it was uncontrolled...

    It will also be really interesting to see what TCI says about how you set up your bench run, too. Did you use the standalone control box? If not, can the transmission (and especially the torque converter) be safely operated without it?
    I'm not sold on the fluid loop-back line, either. A transmission makes more heat in a lower gear than a higher one, and that builds up fast- and a danger level can arise far quicker internally than it takes to warm up the whole case. Still, I'm with Matt's nose that it probably isn't a burned transmission, but in the future you need to be taking your trans temperature readings off of that loopback line, not the outer casting shell. It seemed like you were trying to do that, but the instrument doesn't have the resolution at the distance you were from the line to really get it.

    I've been concerned about a lockup TC mated to an electric motor all along, and I've said so. It will be very interesting indeed to see this unit come apart.

    Don't get me wrong, guys, I'm not being critical, just incisive. The transmission was always the hardest part of this conversion, and guess what- it really is. How cool is it that you confirmed that in the first five minutes of testing, before you actually put anything together?

    On the bright side, that segment was easily your most entertaining moment.




  9. With regard to the bench test failure, there are basically three options (or their combo) what could have gone wrong there:
    -manuf. defect in the tranny
    -badly alligned dual warps as you mentioned one of bought as used and it took many delays to have them ship the dual configuration, perhaps they encoutered similar problems
    -buildup of harmonics introduced into the bench setup, you should have mounted all three components on single beefy steel rail in the first place (look at similar dynos at big manufs. research facilities):
    ============================= single steel rail
    ----------------------------- table bench
    ps you should get datasheet for the siemens generator, as you can easily read from their efficiency map curves, what are the proper energy losses at given RPMs

  10. Very pleased to announce there was NOTHING wrong with the transmission after all.

    Spent the weekend editing and uploading the video of course. We went into the shop Monday morning. Young Hauber disconnected the tranny from the generator and noticed that the shaft of the GENERATOR was extremely difficult to turn even though it was unloaded.

    Some quick troubleshooting indicated that the 300A 1200V 3-phase rectifier module had blown. So much for that. We reinstalled our original 150 amp module, and everything works VERY smoothly now.

    This was an inexpensive Chinese module we had ordered on eBay from Hong Kong. I've had mixed results there and this one was very mixed. Apparently their 300A module blew at about 80 amps.

    Yesterday, we did some further testing. Matt got the software working with the transmission control module. We were getting right at 60% efficiency end to end - 100A into the motors - 60 amps out of the rectifier. Very smooth rotation even in 6th gear.

    Jack Rickard

  11. I hope it's a cheap fix for the tranny.

    Speaking of PID loops, I had a crazy thought of putting a current sensor on both Soliton1's and slaving the throttle input of one to the current of the other, which is connected to your 5K vernier, such that the currents were always equal. However, what you did is simpler and seems to work.

    I noticed that the rpm's were erratic at the beginning under minimal load. Weird. Any idea why? -Klaus

  12. Great to hear you have tracked down and identify the culprit. Hah, so that was another unsuspected area, since your video indicated most likely a tranny problem.. Btw. any idea why the show from Dec. 10 -
    "Clubman it's alive" is not uploaded on the ytube channel? Cheers

  13. Hello Jack and crew:

    The video this week was absolutely terrific.
    My first thought was that something had happened to the generator to cause the stall, but Brian seemed to think that, from the noise in the transmission, it was the cause. Just goes to show you that jumping to conclusions, remotely, can lead to a red face.
    I tend to agree that the tables will probably work safely until some means of controlling the load torque has been developed. You certainly are not, currently, applying huge, sudden forces anywhere.
    One question: Do you thing that you would have moved as quickly, had there been no 'Weed Whacker' comments?
    Anyway, another fine video. Thank you.
    Joe Jackman,
    Vancouver Canada.

  14. Jack,

    It was the reluctor ring that fell on the floor. Watched slow mo. Thought you might like to see this moment frozen in time. When it went thermal nuclear on you guys. :)

    Nice moving guys shutting it down. Good to hear it was neither the trans or generator.

  15. Another splendid show. Glad to hear that the problem was less messy and expensive than it might have been. I'm looking forward to seeing the bench running at higher speeds and loads, but please don't get hurt, will you?

  16. Ahhh! I did think Matt had not aligned the generator so well. It did seem a bit lumpy when it was turning over after the wotsit failed.

    That vibration must have disappeared now its fixed.

    Unlike that John Hardy guy, I'm NOT looking forward to the bench running. I'm happy for it to stay where it is. >:-P

  17. Awesome!!! scared the crap out me as it did you!

    Glad to hear everthing "expensive" is OK


    Simon and the guys.......

  18. Great show. Thanks for showing the failure, too.

    You could have a torque converter or commutator flinging shrapnel like a grenade under several scenarios: If a controller failed as a short circuit, or the torque converter suddenly unlocked at high rpm, or the tranny slipped into neutral, or the generator loading suddenly quit, or the u-join broke, pot box short, etc.

    You should put a scattershield over the bellhousing

    You should move your control board out of the planes of the torque converter or the commutators. It really should go at one end or the other.

    You could do a dead man's switch where things spin only while a button is pressed. You could make it wireless so you could wander about.

    You should put a cover over the u-joint.

    The exposed high voltage connections are begging for a belt buckle or tool to short them out.

    It wasn't clear, but is the cable consistently red for positive and black for negative?

    The generator and tranny and motors should be attached to the same strong steel beams.

  19. Is a scattershield related to a scattersuggestion?

    If we had all THAT on this test bench, I fear we'd never get it to roll down the road at all....

    Jack Rickard

  20. Enjoyed the .pdf Owner's Manual for Rickard Eletric. The language and style is a bit like out of 1980's Microsoft FlightSimulator handbooks, which is a good thing. It would
    serve as very good primer for any EV novices, cudos for you. However, there is no information about procedures for re-balancing the pack, should the owner even by mistake overdischarge, or your method of distributed subpacks around the vehicle trhow them out of balance after few dozen thousand miles driven. Also high temp environment charging pre-cautionary warning is also not mentioned, e.g. car sitting in a summer concrete jungle, let it cool and charge overnight. What is also missing, perhaps some kind of warning that the vehicle platform has got zero passive safety features, even not a rollbar of average kitcar, so crashing the speedster is a fatal endevour even in moderate speeds. It's great joyrider for the backstreets, but people should be warned in terms of application, e.g. for regular highway commuting it's a bit unsafe, especially on todays US roads full with 2t SUVs. Pls. keep it as constructive criticism not offense against you. PS the joke about Toyota's pedal issues in the text is a bit misplaced there, given we know today it was mostly GM paid propaganda campaign, and rather unsuccessfull at that, since asians (Japanese,S. Koreans) continue dominate the market no matter what falls from the sky.

  21. Here is another nice conversion:


  22. Yes CocoEV. I agree. In fact, all owners manuals for all objects should be MOSTLY warnings with SAFETY foremost. You see, if even ONE child is will be well worth having user manuals for all devices that have so many pages of warnings that you can't FIND the operating instructions in them.

    I don't know WHAT I was thinking. I forgot to mention that it's a replica of a 1957 Porsche and doesn't even HAVE a roll bar - you could be KILLED in an automobile crash you know. That should have been good for five plages of indecipherable shit all by itself....

    I'll rewrite the whole thing...... We can go perfect bound once it's at about 150 pages.....

    Jack Rickard

  23. Jack, perhaps you have to read my previous post again. I namely stressed the very good value (clarity of the text and involving narrative style) of that manual for general EV operation. I was very pleased with that.

    However, you can't obfuscate the fact, that somehow bottom balancing doesn't require an informed and actively involved user, it's not a plugin and forget appliance. This simple pack is fluctuating in capacity because of environmental conditions (namely temp), the method of wiring (subpacks placed all over the vehicle), calendar life, and whatnot in terms of other contributing factors. Simply, if you have had overstressed the point in the literature that people shouldn't go under ~40-60% DoD under any conditions/time of a year I'd not have a said word on that front.

    On the point of vehicle platform, however, you are very well aware about the basic point I'm trying to make here and you failed to address that - "special editions" former brazilian speedster replica based on the vehicular diy standards of 50-60's is not comparable to say a FFR Cobra replica (and their other products), which is regularly (and safely) being crashed on race tracks(not mentioning streets), similarly for other kitcars and experimentals with racing pedigree or general safety in mind. Btw. I gues there is somewhat marginally safer Porsche Spyder model S from Special Editions/Chamonix with rollbars.

    Obviously, you are right that due diligence should be placed also on the buyer/customer not only the producer. But as the high performance ev conversion components trickle down to the masses, I guess it's not very good idea mounting anything bigger than say ~60-80hp into these sub 1t resin/few tubes joyriders. I'm afraid that's the way you are actively promoting in recent shows for these light donors, on another level put 500kW into that Escalade/Canyonero as you wish that's another story - because the vehicle is made for the safety situation on the streets of today. That's why on the track days, you won't be even allowed enter with such a spyder/speedster machine, because of this very reason it may easily endanger not only the pilot but others as well.

  24. There are probably 100 ways you can tackle the initiation of the idle on the Soliton. One possible solution would be to just use the start circuit that exists in the Escalade. I'm assuming that the signal that gets sent to the starter solenoid will not function unless the gear selector is in Park. If this is true then you can just use this signal to go into a small circuit that applies a partial throttle to the controller, kind of like the "limp mode" circuit. The operator turns the key to ON and then waits for the lights to come on, on the dash, and then turn the key to the start position and that will provide a partial throttle to start the motor. Since this circuit will only work in Park, it should be fairly safe. It would be much easier to take the starter solenoid circuit to the Soliton inputs and use one of the inputs to initiate, or easier yet to just have it software enabled at bootup, but this is an option that could work with the current configuration of the Soliton.


  25. Jack,
    Great show. Perhaps one of the more valuable episodes.

    Not that it has anything to do with anything, but I wonder why you would not/could not set your bench up such that you could use a very short driveshaft between the tranny and generator. It would suggest a little bit of flexibility in terms of mating up the generator to any trans configuration regardless of misalignment, as well as mitigate vibration issues or unwanted axial loads to the gen shaft.

    But, maybe it's not an issue at all, although using one would relieve (to an extent)major readjustments to the generator adjustment sled for different tests.

    Was there a reason y'all may not have set up that way?


  26. This episode was a treat, a dream make true documentary. I enjoyed it fully, and congratulate you upon the coverage. Only half a minute was edited out with the nuclear blast ;)
    Epilogue makes me even happier.
    The bottom line is no one got hurt, neither the damage is considerable, nor would be costly its repair. Are we having fun?! Thank you very much, I believe we do.
    As a matter of fact, I am no longer sure what do I like better, the show itself or the intermezzo. Some of the comments on the blog are marveling me. EVTV is engineeringly thrilling. Congratulations Mr. Rickard

  27. @colin - yes, I was wondering why not a short driveshaft too. Direct coupling as has been done here requires a very high level of accuracy whereas a short shaft would make things easily changeable with no realignment issues or requirement to even move the generator at all.

    Another thing i dislike is that the gearbox is cantilevered off the adapter plate with no support at the tail end. This can transfer radial loads ( up/down, side/side) to the input shaft of the gen and overload the UJ bearings. In fact, for correct operation UJ should actually have a small angular offset otherwise the needle bearings in the cups never actually "roll" and they end up indenting the bearing race. Not that this is an issue with a test rig as it is a short term setup.

  28. Regarding the idle start, is it really that big a deal that the motor doesn't idle until you first press the pedal? You turn on the vehicle, put it in gear, then press the A-pedal. The motor spins and the vehicle starts to move and the idle is engaged. Nothing to learn, and I think your chopper pilot daughter, and anyone else can easily deal with it.

  29. We'll take a closer look at the drive shaft issue. We talked about a short slip shaft and then Matt ordered what he ordered from his father.

    It works well enough, and we were trying to keep it simple. I thought less joints the better. But we'll take another look.

    As to the vibrations, not too bad yet, but some. We DID add a fairly light bracket supporting the aft end of the transmission with a bit of "spring" to it. We'll test that tomorrow.

    The idle issue will be fixed. They've already run into this and fixed it. They just didn't include anything about it in the 1.3 documentation.

    They've added a START ipnut item that we can wire to a start button or as you say, to the START output on the Elescalade extant. You press that until it gets up to idle speed. So apparently this problem has come up and been solved earlier.

    We're also going to try some different pullup resistor arrangements they've suggested to improve the tach input problem to the Solitons.

    jack Rickard

  30. CocoDev:

    I did respond. I'm not remiss. i refused to address because it is too embarassingly stupid to deal with.

    We do NOT bottom balance and do not advocate bottom balancing for the 3235th time.We occasionally use this as a remedial measure for packs that have been BMS abused, or in the event of an expansion of the number of cells.

    NONE of that does what you THINK it does. There is no drift. We don't EVER need to balance. Not to temperature. Not to age. It is simply a myth and everything you think you know about these cells is declared herein and forevermore to be wrong. I do NOT want vehicle owners bottom balancing or attempting to do so. If they will just NOT SCREW THESE CELLS UP BY INCESSANTLY PLAYING WITH THEM AS YOU SUGGEST, tehy can charge them, drive the car, and charge them some more, for about 29 years before there will be a problem. And the problem THEN will be a seriously sad looking interior carpet and upholstery problem in all likelihood.

    As to the safety of the basic car - it is what it is. I understand your problem here is that they won't let it IN on RACEDAY???? And I should address that in the manual????? Or what? Someone might get hurt driving a car????

    Having you been drinking with the girls again.....?

    Jack Rickard

  31. Just in case anyone else was wondering "what manual" I found the link here...

  32. Jack, the claim that there is no drift in your batt. pack after few thousand miles is rather optimistic (given you constantly tweak it). Make a release candidate vehicle, make it run a few dozen thousand in variating climate in one go and no tweaks along the way, that will be a confirmation (so far it's an experimentation), that there is no user input needed (bottom balancing).

    As mentioned before, I'd not have commented should your properly address it in the manual, i.e. inform the user that even regular ~80% DoD is bordering on problems in the long run, so get prepared discharhing to only lower bracket. That's perhaps just psychological barrier you don't have to admit it's better to stay well bellow the advertised maximum range, make it less than 100km.

    So far, it's not working, e.g. the guy following your example with bmw z3, fully acknowledges he had to laboriously balance the pack from time to time, especially when nearing full capacity discharges. There is another similar project but w. AWD vehicle where the owner plans only 40-50% DoD, that will be interesting to watch.

    Simply, it's impossible to be draining upto ~80% (or more) of the capacity in regular fashion with a system of haphazzardly placed several subpacks around the vehicle, each of different size to begin with, and not to endup with unbalanced pack. That's why no OEM will ever use such system, and if they had to compromise for the lack of real estate (new EV dedicated platform not ready yet), think about the upcomming gen1 Ford EV hatchback, their subpacks on different places at least share the same environmental overall packaging, be it re-circulation of air/dedicated batt. pack HVAC, water heating circuit etc. I know you mentioned planing similar setup for the Escalade/Canyonero project, so that's your double standards in plain sight - an acknowledgment it is needed, while supposedly not necessary for smaller-summer cars.

    I'm sorry to be watching you here obfuscating the safety point in such blatant fashion, the special editions-chamonix is not "a car" it's a joyrider category kitcar replica with no passive safety what so ever. Your reply was a cheap classic strawman tactics, I said there are plenty of racing approved and safe kitcars, which are for that very reason also popularly used in street application. There is a clear distiction, a car is your BMW Mini and your Escalade-Canyonero.

  33. Thanks for deleting my rebuttal post, sadly I did not save a copy. Very interesting you can't face up to logical and argument based discussion.

  34. First, the latest manual is at

    Second, no one deleted your rebuttal post. I have no need nor necessity to "face up" to logical and argument based discussions CocoEV. That said, I've been on online forums since about 1983, predating virtually everyone in the game. And I have NO problem taking the heat, or having an argument online.

    But just as your points are weak, so is your understanding of the online millieu. Google has a limit on content length, and if you go over, it just kind of shitcans it without really giving you an opportunity to correct. We've also had some problems with Google Blogger's SPAM filter, which I am unable to turn off.

    I have a guy kind of/sort of working on migrating this blog to a service called WordPress - actually one of the original blog software programs. THere are some issues retaining our photos etc that are attached to this blog. We have a number of viewers who refer to the back postings for various components etc. So I can't just wave a magic wand here. But I've had offers of some help from some very knowledgeable people. Hope to be free of the Google limitations soon.

    So I'm sorry you're having trouble with this blog software. I do too. I've lost a number of really quite good posts that I spent an hour on and they just went up in smoke.

    But your paranoia about some human actually deleting your precious points of logic is simply errant. No one has. No one does.

  35. CocoDev:

    I found your comment in Google's SPAM folder. SOME commenters appear to get sent to the sandbox, most of us have no trouble with it. I suspect it is based on activity QUITE off our site. Someone has simply reported your OTHER errant e-mails and comments somewhere else to Google as SPAM. It happens to certain individuals in the online millieu. Not to take it personally.

    You pose another in a long series of "YES THEY DO NO THEY DON"T" posts regarding cell drift with these batteries. This is left over from the Pb cell experience and the guys with the MOST Pb cell experience simply cannot let it go. They are SURE if we somehow use and abuse these cells long enough, they will drift.

    I don't know what to tell you. At this point we have six or seven cars, several years with them, and the ability to measure to the nth degree. I'm at this point not guessing, and rapidly waning in interest in the debate. If you want to believe in all that, they're your batteries. But I'm not writing MY user's manual based on your myth and superstition belief system. And you can post everywhere and always to the contrary and it won't change how these batteries behave.

    Part of the problem in that you THINK they move around is that you keep moving them around. All this balancing and shunting and BMSing is what I call "wearing the spots off the cells". You actually think you are observing something. In reality you are CAUSING it. Leave em alone.

    This too is not a guess. I HAVE had some problems with these cells. Almost ALL of them derive from my well meaning attempts to measure to confirm or deny these myths. But in doing so, I imposed parastic loads on some cells, but not others, LEADING to an imbalance. They do not imbalance themselves, but YOU can imbalance them quite easily. They don't balance themselves either. This is what you are seeing.

    Beyond that, there's nothing here for me to do to convince you. I am a EE and have worked with literally hundreds of people who were, but of course 85% of them mostly stood around criticizing what the ones working were doing. It's no vaccination against incompetence.

    Your vague and anonymous appeals to some "authority" fall on deaf ears. Around here, we experiment and we demonstrate and it is what it is. Why do you think that you knowing somebody, who had a friend, who knew somebody, that disagrees with me, would matter to me or any of our viewers? Opinions it seems, are quite like assholes. Everybody has one. I don't even value mine very highly, so where would that put my cognizance of yours. SHOW ME THE DATA.....and I'll be happy to look at it and interpret it for you if you don't know how.

    Jack Rickard

  36. If that deleted post was a software glitch not a strike of censorship then I offer an apology on that, I posted here last time year ago and it was ok. What is lost is lost and for the sake of time and energies I can't recreate the whole multi-layered piece in the same quality here again. But vague it was not. For one thing I said produce a "release candidate" vehicle which will be tested for dozens of thousand miles in various climate conditions in accelerated fashion (even hire pool of dudes to drive it 24/7), without tweaking it, i.e. replacement or babysitting batteries not allowed. Simply, you have accumulated very interesting knowledge over the years, but you don't have a relevant dataset yet for the one single same pack config beyond few thousand mi of range. In that fashion you can only recreate true appliance approach and "no dumb user input" testing conditions. The recommended 80% DoD is still too much, e.g. the Z3 BMW conversion which followed your example already needs from time to time laborious individual cell service, after accidentaly discharging too low (80-90%). I know you added features like limp mode, but is it enough? Also, there is a guy with big AWD setup, which for these very reasons plans only 40-50% shallow discharges with his ~35kWh non BMS decreased charging voltage strategy for LiFePO4 pack. Disclaimer: I'm not a BMS dealer, groupie or advocate. Btw. I'm not anonymous, and even if it was a case that's sidestepping the arguments here again.

  37. Oh Jack,
    Regarding the incompetence vaccine: early research indicated the need for added stabilizing compounds to prevent decay of potency but the added chemicals caused a psychosis, often misdiagnosed as autism, which sometimes presented signs of continual rocking while chanting, mantra-like, word strings, making little sense. No anti-psychotic drug had any effect. Research was discontinued….. Sorry, I don't have a tongue-in-cheek emoticon.
    Now you're being 'interpreted', just the way Prof. Whitacre was, even though his frustration with BMS proponents was clear, from his almost flippant statement near the end, something like, "Well, then perhaps you'd better use a BMS."
    To me, that sounded like, "They're your cells, and if you feel safer with BMS, then use it."
    I've heard you say, several times, "Install the pack and charge it. Don't mess with it!" Yes, you have shown bottom balancing of cells which were mismatched for partial pack use, (Vantage Van) or the Duh pack, where you had used 10 cells to power instrumentation. Have you had to re-balance the Vantage Van's Pack Since? I bet not.
    In the 1960's I had a '61 Lincoln Continental convertible, and oh m'gosh, it had NO roll bars and no ABS. As a First Nation's friend of mine said, "There's no problem, if you drive like a sober white man!"
    Please accept my condolences for your frustration. I never thought that it would be so hard to get concepts through to your audience.
    Keep up the good work though, and maybe just have done with the BMS argument. Everyone SHOULD know your position by now, and should you be proven wrong, your record shows that it will be exposed in a video! I have NO doubt.
    This is SCIENCE, not PSEUDOscience.
    Joe Jackman,
    Vancouver, Canada.

  38. I also said, there is a reason (multiple) why none of the OEMs of any significance during the early/mid 1990s or now with predominantely Li-ion based chemistries attempted to build haphazzardly distributed subpacks placed around the vehicle, with different sized modules on top of that. Because they now it's detrimental, for instance the fortcomming first gen Ford EV hatch, which had to compromise for lack of dedicated EV platform on that battery space issue for now (gen1), has got subpacks but they are enclosed in one overall sealed body w. HVAC and/or water temp management. You mentioned recently, that's the way you plan to go with Escalade-Canyonero, so that means you are well aware of these issues, and it's no only the all year weather performance. I'm pleased with that and looking ward to it, because that's largely mising feature on the diy scene at the mooment. On the question of suitable vehicle platforms I only warned about the trend continuously spicing up the performance envelope for the "Special editions" - Chamonix Speedster & Spyder is not good, the vehicle is very unsafe for say >60hp drivetrain, in the same price category it can't be even distantly compared with FFR production (and others) which is crash proofed both on track and in the street version for almost two decades. I guess it's worth make the disctinction among the true cars as your BMW Mini and Escalade-Canyonero and the Porsche replicas, which are just rudimentary zero safety joyriders where extra pre-caution is necessary.

  39. What a mess it should be posted in reverse order, even the LiveJournal login is buggy. If that deleted post was a software glitch not a strike of censorship then I offer an apology on that, I posted here last time year ago and it was ok. Thanks for posting it up. It seems though you still didn't tackle the main arguments. But vague it was not. For one thing I said produce a "release candidate" vehicle which will be tested for dozens of thousand miles in various climate conditions in accelerated fashion (even hire pool of dudes to drive it 24/7), without tweaking it, i.e. replacement or babysitting batteries not allowed. Simply, you have accumulated very interesting knowledge over the years, but you don't have a relevant dataset yet for the one single same pack config beyond few thousand mi of range. In that fashion you can only recreate true appliance approach and "no dumb user input" testing conditions. The recommended 80% DoD is still too much, e.g. the Z3 BMW conversion (people who you personally know - not my buddies) which followed your example already needs from time to time laborious individual cell service, after accidentaly discharging too low (80-90%). I know you added features like limp mode, but is it enough? Also, there is a guy with very well documented big AWD setup, which for these very reasons plans only 40-50% shallow discharges with his ~35kWh non BMS decreased charging voltage strategy for LiFePO4 pack. Disclaimer: I'm not a BMS dealer, groupie or advocate. Btw. I'm not anonymous, and even if it was a case that's sidestepping the arguments here again.

  40. CocoDev:

    I can't do it pard. It's just too repetiive. Go back and watch some of the videos and do the best you can for Christ sake. The OEM LiIon cells have as much to do with htese LifePo4 prismatics as apricots do with apes. They are entirely different chemistries, have different cycle lives, and live in a different world. Beyond that, the OEM's aren't even DOING what you apparently think they are. It's just too much to recant over and over. I'm sympathetic. Ishould be able to do this better and with more patience. But I can't. You don't even have the BMW Z3 right. I just can't face it all.

    If you guys would show up with ONE point to talk about, I might be able to do SOMETHING with it. With such a broad bucketfull of misconstrued confused crap, I just can't respond to it. One thing is tangled with another thing and then that is misunderstood with ANOTHER thing for a conclusion of WHAT????? I can do nothing with it.

    If your point is that cars are dangerous, I could not agree more. If you saw how I drive you would understand. I am horrified at the prospect of ignomiously dying in a car wreck, the stupidest and generally easiest to avoid way of parting from this earth that there is.

    Beyond that, there is nothing I can do about it. And adding safety notices in my users manual is just absurd. Of course they are dangerous. So are wet floors. What can I tell you. Get a life. And be careful.

    Similarly all the concern over my test bench. Yes, it will probably kill me. It's ok. I'm 55. My kids are all grown. My father, two uncles and an aunt died of ALzheimers. It's ok. It's all good. I don't mind. You're right. We're not going to do it that way, but you are of course right. Better safe than sorry. Safety first. A stitch in time saves nine. Don't let the bed bugs bite. A penny saved.... ok I can't think of any more platitudinous maxims and definitely words to live by. I know I should be touched by your concern. I am. Truly.


    Jack Rickard

  41. CocoDev:

    It appears reversed because I had to go dig it out of the SPAM folder which Google keeps sending you to. AN dyou sound remarkably like MikeyPEE under another login. He had the same difficulty. And the same story line.

    I can't do anything about your login, online identity, or Google Blogger. And I can't manually keep digging you out of the pile. Sorry. We're going to migrate to Word Press, but I'm a little busy at the moment.

    I don't like this blogging software either. It looks good on the surface, but it's skin deep. Underneath, it's very poor.


  42. Thanks Joe. Very pleased you survived the Lincoln. You should have seen me in my Falcon or the 54 Dodge Coronet. It was mayhem.

    Matt drives the Vantage Van DAILY all over Cape Girardeua. We not only don't have a BMS< but not even a proper charger. It works fine. 103 miles was the max run. We're having some difficulty in the rain. They mounted the controller and contactor UNDER the chassis in the weather, and when it gets wet it quites with an ERROR 39. It dries out, and you can drive it off.

    The batteries made it through extreme cold in the winter and Matt's batt rack is all exposed underneath at that. I've promised to buy him an onboard charger but havent' gotten a round tuit.


  43. Jack, What a great trip your taking us on. However, it seems to me that your video operator needs a pair of JR-548321 rubber gloves so as not to possibly electrocute himself. I appreciate your work. mike

  44. Hello Jack,
    I remember seeing your show introducing the Vantage van, but do not recall any further info on the actual make-over. Did you keep the same sepex motor? If so does the regen braking work ok?


  45. We did a couple of followups. We replaced all the AGM with HiPower LiFePo4 cells. Matt rebuilt the battery box to accommodate them and bottom balanced the mix set we put in. It drove 103 miles on a single charge. He drives it daily now.

    It was an AC induction motor with Curtis 6501 controller. They work fine, but we're having issues with the controller in the rain. It's mounted underneath the frame rather low on the car.

    Jack Rickard

  46. Ah, my mix up. It was the G-van with the sepex.


  47. I'm having controller issues with the G-VAN. I have THREE of the controllers laying here. We have brand new wheels and tires on it. And I don't have time to fool with it at the moment.

    Make me an offer. Comes with 3000 lbs of lead acid AGM's.

    Jack RIckard

  48. Was just curious regarding the sepex option. These are not as common as series motors. Don't really know why as on paper they do offer more control possibilities.
    Regarding making you an offer, alas I hail from the other side of the pond, Malta actually, hence my obvious ignorance of US van models.

    My planned conversion is a Citroen C4 with auto tranny. So I am really looking forward to your bench testing and your approach re interfacing with the tranny ecu.

    Great show.


  49. SEPEX is not really a motor. Almost all DC series can be seperately excited. It simply means the current through the field windings and the current through the armature are separately controlled.

    The CONTROLLER is a bit of a different animal. The motor not so much so.

    Jack Rickard

  50. Jack that much I understand but there must be some subtle differences in the design/construction, like wire gauge choice as people like Kostov list them seperately with different power ratings. SEPEX being rated slightly lower.


  51. And bdw, the company I work for has a contract with a French OEM to manufacture the busbars for their EV'S. From what we can see from the schematics they are using modules of around 32V. The values vary depending on the document. They are quite secretive of their battery technology, but on each module busbar there are outputs for voltage measurement and mounting for temperature sender. The cells in the module are connected in series with no tracks for BMS. Not even individual cell measurement.
    I am quite sure that this is not news to you but thought you would like to know from someone seeing it first hand.

    And I especially like your choice of wines.



  52. Fascinating. It is of course news to me. As you said, it's a secret.

    I love secrets.

    Jack Rickard

  53. I think at about 45 minutes or so n to the March 4th video you state that voltage is equal to current divided by resistance, Ohm's law.
    Shouldn't that be volts = current X resistance ?