Friday, September 2, 2011

Testing, testing....one... two ... .three


The online milieu offers enormous opportunities for learning and information about topics of interest. In the early dreams of a global internet, this ability to pool the expertise of many individuals around very narrow topics of interest was one of the greatest boons foreseen. Unfortunately, in many ways it has turned into a cesspool.

There are a number of people out there successfully converting electric cars. I am disheartened to observe that the majority of them view anything they learn along the way to be a "proprietary secret" they can somehow leverage into untold riches from their little shop in what they see as a viciously competitive world.

Well, if you see it that way, it is that way - for you.

That mostlly leaves the online forums as a source of info about electric vehicles where roving bands of clueless people copy/paste what the last guy said into the next conversation - with a few typos and often some expert opinion on what it really means. This leads to comically large bodies of total disinformation misleading anyone who reads it into believing it is real. The pool of knowledge becomes a cesspool.

We attempt to counter that in two ways. First, we actually DO build electric cars, and we share all of it with our viewers. The good, and too often the embarrassing bad. To get the most from it, you have to watch over time unfortunately. The neat device and grand theory as to why that's just the tits in this week's episode may be followed in six months and a few thousand miles by an examination of the charred pieces and a rebuild using something else. Real world. I should apologize for it.?

The second prong of our attack on disinformation is to actually DO testing, not just talk about theories we've heard. And better yet, to demonstrate how to DEVISE tests you can do to learn first hand yourself. We encourage others to also perform our tests and report their results = either confirming or contradicting our findings.

Part of this goes again to the 12 blind men around an elephant. Corresponding about IDENTICAL cars, for example with Eric Kriss who basically built Speedster Duh, can be quite frustrating. Different results from two identical cars? Or maybe not so identical. And maybe not exactly the same test conditions... you see the problem.

Part of it is contrast and comparison. Most people don't HAVE two vehicles to compare. Indeed, often you have NO other frame of reference at all. You build your first electric car, and often take your first electric drive in it. Does it feel like it's supposed to? How IS it supposed to feel?

Our first car was the red speedster of course. It ran flawlessly. Oh, I guess I was a little disappointed in the performance. But it certainly kept up with traffic and was pleasant to drive.

Six WEEKS later, I was playing with the Kelly Controller software and noticed the minimum battery voltage setting. I see that voltage a lot so I must be at minimum voltage a lot. I wonder why?

I changed the setting to a lower value. Instead of 100v for my 108v pack, I lowered it to 90. Reset the car and hopped in for a drive. Backing out of the driveway in REVERSE I very nearly shot through the fence around the property in an almost uncontrolled acceleration. I immediately stopped the car and checked everything again. I reset the voltage to 100v and the car behaved quite normally. I changed it back to 90v and VERY carefully applied the accelerator. The car shot ahead like a rocket (comparatively speaking).

When the controller notes the battery voltage approaching the "minimum voltage" it cuts back current to the motor to prevent the pack from going below that voltage. Of course it does this several hundred times per second. As a result, when I applied throttle, it put out power to the motor until the voltage dropped to 100v. At that point it limited current to precisely the value necessary to hold 100v. If the voltage crept up, it increased current. If it began to sag below 100v, it would decrease current. And it did this so well, it felt quite smooth. The car operated flawlessly, at about 1/4 power.

The batteries DO sag in voltage when you put them under load. So by decreasing this "minimum voltage" all of a sudden I could get full power out of the device.

While this is almost comical, it illustrates the problem. No frame of reference. I might have gone two years before discovering this. Never really gotten excited about electric cars, which felt pretty tame up to that point. And so never shot our first episode - no EVTV. For want of a nail....

So I'm very sympathetic to the guy trying to build an electric car with no "go-by." No frame of reference or model to build too. It is ALL a discovery process.

We did this comparison between the Speedster and the Spyder largely as an exercise to demonstrate some of the components of range and specifically the impact of aerodynamics on range. Whey you get one range at 45 mph and a very different one at 75 mph.

But real world testing, as opposed to staged scripted test demonstrations, always have outliers and anomalies. In this case a huge one. A very marked difference between two cars with ostensibly identical drive trains.






The problem was that the aerodynamically slightly pudgy looking 356, which was also 200 lbs heavier, got DRAMATICALLY more range than the sleeker and lighter Spyder 550. And everything we tried made the disparity more pronounced.

We spent $3100 on running gear upgrades using really kind of extreme aluminum rotors and calipers, low rolling resistance tires, lightweight wheels, and indeed shaved some 60 lbs from the car, dropping the curb weight from 1905 to 1844 lbs.

Of cousre, we also learned that the Michelin Energy Saver A/S low rolling resistance tires on the Speedster weren't set to 42 lbs pressure as we thought, but a more squishy 33-34 lbs. When we aired the tires up, the Speedster got dramatically better, as good as 1.000 AH per mile at 40 mph steady speed. That's a 180 mile range.

Meanwhile, all that work on the Spyder produced improvements, but improvements so meager they are almost in the noise level of the test.

ENter by far our most popular test technique. The Soap Box Derby. We've received so many comments on this test that we are thinking of incorporating it into our race day at the Electric Vehicle Conversion Convention (EVCCON 2011) I'm searching for some kind of incline or ramp we can use. In addition to the electric AUTOCROSS RACE and the electric DRAG RACE we have scheduled for Friday afternoon, I would like to have a SOAP BOX DERBY where we can see who's car rolls the furthest at 2 mph.

The Speedster will pretty much roll down the street pretty nearly to the house in neutral. The Spyder almost goes into reverse and backs up the hill. It barely moves at all.

We received a lot of input from viewers on this. My favorites suggestion came from John Hardy of the UK. This involves some residual pressure valves in the brake lines. In cars where the master cylinder is below the calipers, the brake fluid of course has a tendency to drain back down into the master cylinder and you have to pump the brakes to get the fluid back up into the brake calipers. On drum brakes, the wheel cylinder holds much more fluid than disk brakes. And these cars were all originally equipped with drum brakes on the rear, disk brakes typically as an upgrade.

So to keep the fluid from flowing back, they use a little check valve that maintains just a bit of pressure on the line to keep the fluid from flowing down by gravity. On drum brakes, this residual pressure valve typically provides 10 ps, while for the lest voluminous disk brake calipers, a more modest 2 psi. John's suggestion was that they had upgraded the rear brakes to disk but kept the 10 psi residual pressure valve from the drum brakes. This causing the rear pads to drag.

I loved it. Our master cylinder IS below the calipers and we don't have to pump the brakes so we must have that.

We went through the car and can't find one of these residual pressure valves anywhere, front or back - or of any value. But it was a GREAT theory.

We had the car aligned this week. I haven't' retested yet but I would be astounded if that made that much difference. Perhaps. But we did reroll the Soap Box Derby and no joy.

Several viewers have commented on the negative camber on the Spyder. I think it is slightly more visible on the Spyder. But all these cars have negative camber - the book is 2 to 5 degrees. The Spyder is more on the 2 degree side actually.

About all that is left that is obvious is the difference in the Ring and Pinion gears - essentially the "differential" on this car. The Spyder has a 3.88:1 and the Speedster has a 3.44 to 1. Eric Kriss spoke to a racing transmission expert who claims that could indeed do it. So it's our main culprit at this point.

It is unlikely that I would swap out the transmission to find out. The car runs nicely and has a 100 mile range now. But we will certainly stick to the 3.44:1 for future builds.

You always get a result from testing. But in the real world, often it can be weeks or months before you really know what it all means. Aerodymics still has the predicted impact on range. But why these two cars are so very different is so far beyond me.

We have a new Canon AHX1 camera we're trying out. It has built in XLR microphone inputs but we apparently did not adjust our audio very well because it is truly awful in this episode. My apologies. Testing, testing. ...one...two....three.

Jack Rickard



37 comments:

  1. Regarding audio: when I was watching it using the headphones I noticed that your microphones are connected to oposite channels: I can clearly hear you, Jack, in my left ear when you are on the right side of the screen, and Brian sounds in my right ear.
    I put my earphones in opposite way and it cured situation.

    ReplyDelete
  2. I think that you are right on the differential causing some of your lower RR.

    From:http://www.klenger.net/dodge/general-reference/Mileage-Manifesto.html

    GEARING

    The main effect of gearing is upon engine rpm, so I will cover that when we get to the engine. But there are other elements to consider. First, is differential lube. A higher viscosity lube increases the work necessary to turn the differential gears.
    Second is the effect that ratio has upon gear design. Namely, the more torque multiplication a gearset must do, the less efficient it is at transferring energy. A 3.54 axle ratio has less of a difference in tooth count between ring and pinion than does a 4.10 ratio gearset. As you go higher numerically, the efficiency gets less as the teeth must be machined in numbers and at angles that make them less efficient conduits of energy.

    ReplyDelete
  3. Jack:
    Go back to the dyno guy and test the cars at set MPHs using 1:1 gearing, at the same time monitor the current draw and equate that to the friction in the drive lines; aero is taken out of consideration, as is the rolling resistance of the front wheels and friction in the tranny is minimized by the 1;1 gearing.

    ReplyDelete
  4. How do you tell if the rear wheels are out of alignment?

    ReplyDelete
  5. Maybe some others have a point with the tyres adjustment. They normally set a cars tyres to specific angles for good handling so Hike up the back end so the rear shafts are straight and have all toe-in and out removed from the equation. That will shut everybody up.

    There is going to be no end to this conjecture. All I know is there is 19% more drag for 12.5% more shaft speed in the gearbox.

    So its doing our little heads in.

    Enjoy the car. It's gorgeous. Better than mine. When you put the power on and your car pulled away silently and oh so smoothly... Priceless.

    Andrew, (m1aws)

    ReplyDelete
  6. When gears mesh, the contact point (really a line as long as the gears are thick) between a driving tooth and a driven tooth slides from the root to the tip of the driving tooth, and from the tip to the root of the driven tooth. Friction opposes this sliding, and is the source of the inefficiency of the gears. The friction (and so the power loss) is proportional to the normal force at the point (line) of contact, which is, on average, proportional to the torque that is being transmitted through gears.

    When the car is coasting, there should be no torque transmitted through the differential gears (unless there is drag on the system upstream of the differential), and so no significant energy loss due to inefficiency of the gears. Zero times anything, like gear efficiency, is still zero. I think the difference in rolling resistance between the speedster and the spyder has to do with the transmission/differential, but NOT due the different efficiencies of the respective gear ratios.

    The car at the top of the hill has potential energy proportional to the weight of the car and the height of the hill. The difference in distance of the soap box derby coast (SBDC) relates to how much of that potential energy becomes kinetic energy before it turns to heat. The faster you generate heat, the less distance you travel. If all of the energy of the SBDC was lost only to viscous gear lube drag on the ring gear of the differential, and viscous drag is proportional to rotaional speed times the pitch diameter of the ring gear, the speedster should go 13% farther than the spyder (3.88 is 13% larger than 3.44). If other sources of loss (rolling resistance of tires) are significant, the 13% becomes some smaller number, and possibly very small. If the difference in SBDC distance is greater than, say 6%, I dont't think you can write it all off to viscous gear lube losses. I think something else is rubbing, possibly in the transmission.

    How much farther did the speedster coast than the spyder?


    Mike Kaindl

    ReplyDelete
  7. Pardon my ignorance of VW transaxles, but can you simply pull out the driveshafts so that the wheels are free spinning and take the transmission drag out of the equation?

    ReplyDelete
  8. You can if the transaxle is the IRS style but with the swing axle style you can no do this. The weight issue was addressed and even with an extra person the coast was no better. So weight is not the issue. Front alignment is not a factor either. It's easier to turn the 3.44 R&P than a 3.88 R&P. The shaft of the pinion also holds a large pack of gears. After looking at what is still moving within the transmissions I'd have to agree that it is likely the R&P. If they were both IRS it would be real easy to rule out the transmission.

    ReplyDelete
  9. The Speedstere was probably 250 feet further down the road than the Spyder. And that's with the Spyder rolling about 600 feet.

    We did a four point alignment with zero toe-in on Thursday. On Friday, we gained about ONE CAR LENGTH on the Spyder.

    Jack Rickard

    ReplyDelete
  10. "we gained about ONE CAR LENGTH on the Spyder."

    Bummer. I was hoping for more. Still it seams like it is more than just that R&P gear set. But I sure can be wrong on that. :(

    I was thinking that maybe the internals are just tighter in one than the other. One shops way of building a transmission vs another shops way of building a transmission. Parts are different and tolerances are different and one could just be tighter. Different brands of gears too. All combined just result in more drag. little here and a little there add up to a lot.

    ReplyDelete
  11. I will however use the 3.44 R&P in my VW Bus along with a super diff and heavy duty side plates welded synkros and heavy duty gear sets. It won't be hauling much.

    ReplyDelete
  12. If you have the vehicle off the ground and rotate the tires by hand, how much resistance do you feel? Brake drag can be heard and felt. There is enough of a difference between the two cars (range and rolling distance) that the difference in rotational resistance could probably be felt by hand. Jack, you definitely bring up an interesting issue, regarding a point of reference. I wonder how many individuals have converted a vehicle and have not been satisfied due to some little idiosyncrasy. Or were satisfied with less than what was possible from the vehicle. Well I'm off to the races at Infineon Raceway.

    ReplyDelete
  13. Jack,
    It looks like it is time to get out the torque wrench and see how much is required at all eight wheels with the cars on jack stands. You might be able to use the wheel nuts if they are the same bolt pattern. If not, then building a jig which attachs to the nuts and has a center to the wheel torque point. You would hopefully just need a inch-pound torque wrench for this test.

    ReplyDelete
  14. The 3.44 ring and pinion is an aftermarket item while the 3.88 is almost always good used VW set. There may be some efficiency difference there. The Spyder has the ring gear flipped for mid engine operation (unless you chose to have the motor turn the opposite direction), I have no idea if that effects efficiency.

    You could lift the rear tires and run both at the same rpm through the gears no load while measuring current to see if the transaxle could be the cause of the poor coasting performance. You could likely tell by driving both around the same course for 25 miles and comparing gear oil temperatures too.

    I watch this with interest because I run a VW kit car with the swingaxle transaxle. I have even lower gearing that was carried over from its gas days, 4.375 R&P with a 0.93 4th. That is partly offset by the large diameter of the P215/70R15 Goodyear FuelMax LLR rear tires (3060 rpm at 60 mph.) I'm not as efficient as your cars but am looking for places to make improvements while retaining the beach buggy style.

    ReplyDelete
  15. The Speedster looks to have a considerably larger frontal area than the Spyder (links below). Since the Spyder was a race car, it might have been designed to display best performance at higher speeds, while the sports car Speedster might have been optimized for mid-range speeds. I'm suggesting the graphed efficiency lines might narrow, converge, or even cross at 80, 90, 100 mph. Negative lift forces and or perhaps some sort of drag crisis could allow for unexpected results. I would like to see the soap box derby conducted from two or more >slope hills with the weights of the two cars initially being the same [weight added to Spyder] and the tires inflated to the same pressures with equally weighted cars. Spyder tires inflated to 42 under an 1800 lb car could have a different rolling properties than similar tires on the 9% heavier Speedster. It would also be interesting to see how the Spyder does in the rolling competition with the larger dia. tires. According to High-Performance Vehicles http://www.estg.ipleiria.pt/files/283161_cap%208%20-%20altaperformanc_43a9b269d732b.pdf the 1954 Porsche 550 Spyder has a cD of 0.45 and a frontal area of 1.03 m2. Additionally, "The closed Porsche 356A of 1950 came with an incredibly low drag coefficient of cD = 0.28 and a frontal area of A = 1.68 m2, though with a very high lift coefficient of cL = 0.28. This, however, was quite tolerable since top speed was relatively low." Side by side photo http://www.constructorscarclub.org.nz/Articles/porschecomp.html

    ReplyDelete
  16. Sounds like alignment was not the problem.

    I wonder if one could take an regular, plug-in, electric drill and plug it in through a watt meter. Then use the electric drill to spin each of the wheels, via some home made gizmo to couple the drill chuck to the wheel, to get an idea of how much power each one takes to spin. When doing the rear wheels, hold the opposite wheel still, with the tranny in neutral, to get some idea of gear-box plus axle power figures. Do this for both sides and get some power readings for the drill alone.

    I don't know if this would work but it seems that it might stand a good chance. -Klaus (Murphy is my copilot)

    ReplyDelete
  17. Go Figure:
    I'm sorry but must correct you. Gears do not rub, they roll across each other. That is why they have a carefully calculated and shaped "bulge" on the flanks of the teeth. Believe it or not, even the bevel gears do this. They have the spiral also ground into them so the tooth contact is constant This way you don't hear a constant whine or feel vibration from it when under load.

    http://www.cs.cmu.edu/~rapidproto/mechanisms/chpt7.html#HDR108

    ReplyDelete
  18. EVfun said...
    "...The Spyder has the ring gear flipped for mid engine operation (unless you chose to have the motor turn the opposite direction...".

    Interesting.. You have really pricked my ears and there might be more in it than we know. Is the motor running in reverse on this car? A couple of drive cables had to be swapped. Just wondering about the thrust line and thrust bearings and all that.

    Maybe I'm wrong.
    Andrew, (m1aws)

    ReplyDelete
  19. The inch pound torque wrench idea that many of us have suggested seems like a simple and accurate comparison. Does Duh have significantly more miles on it? Maybe all bearings and such are simply more broken in and running more smoothly?

    ReplyDelete
  20. We gained a car length by a four point alignment. The guy at Ranch did say that there was quite a bit of variation from transmission to transmission just on parts fit and how you put them together. And the R&P will definitely be different.

    We're kind of down to the transmission at this point.

    The Spyder has the entire TRANSMISSION flipped for mid engine operation. If you think about it, clockwise is still clockwise and forward is still forward.

    Jack Rickard

    ReplyDelete
  21. Watching the latest episode I must continue our LEAF sales debate. LEAF sales are higher than many other models, conventional and hybrid, from Nissan and other OEM's, are all those vehicles going to be failures as well? By your logic any vehicle with similar or worse sales numbers than the LEAF simply are not selling enough to continue.

    Regarding the Costco charger removal, do you think the fact that until this year there were almost no EV's that could have used the chargers has anything to do with them not being used, and that the few EV's that could have used them had different plugs?

    Regarding the CA law about unplugged vehicles, I understood it differently. I don't think it's illegal to unplug a vehicle, it's illegal for your vehicle to be in an EV charging space and not be plugged in. So if your car is finished charging and someone else comes along and unplugs you to charge their vehicle, YOU get the ticket for being unplugged. Either way seems extreme and unnecessary.
    While I don't use public charging, (there are none within range), I constantly see Tesla and LEAF owners talking about using them or wishing they were available.

    ReplyDelete
  22. This comment has been removed by the author.

    ReplyDelete
  23. Can the gearbox be bolted upside down onto the differential?

    ReplyDelete
  24. VW transmissions are not flipped upside down. They are turned around. By turning the transmission around you need to flip the ring to the other side so you then get your forward gears back in proper order. Better than having 4 reverse gears and one forward.

    No the gear box can't be installed upside down.

    The VW Bus with Reduction Gear Boxes also have the flipped ring gear. That is because the reduction gears change the direction of the gearing so the ring needs flipped from the others.

    Pete :)

    ReplyDelete
  25. I do not understand, if you turn the transmission around, rotate the differential and flip the ring gear does the motor still rotate clockwise?

    ReplyDelete
  26. What difference does it make? It's an AC induction motor. To reverse it you swap any two leads. It cannot tell the difference. It has no forwards and backwards.

    Jack Rickard

    ReplyDelete
  27. Mark,

    Yes if your using DC motors. The motor is not being touched. With AC! As Jack said. Makes no difference.

    ReplyDelete
  28. Gotchya now. Thanks fella's.

    Flip bevel ring to other side = runs backwards. Was wondering incorrectly if the pinion thrust bearing was correctly in use.

    Maybe you have a perfectly tightly assembled gearbox on top of the 12.5% extra gear speed for 18%(?) losses. I'm not going to guess on too tight shimming for the bevel gear. There's no vibrations or whines anybody spoke of.

    Let's get back to this in another 40,000 miles then let us know how its improving. ;)

    ReplyDelete
  29. Yes Jack, sorry I was thinking DC for some reason. So the Spyder motor is running counter to the "Speedster Duh" motor and that, you state, makes no difference.

    I do understand to reverse a 3 phase motor you swap any two of the three leads.

    ReplyDelete
  30. Just because the motor doesn't care if it spins one way or the other doesn't mean the transmission doesn't care. The gearbox is a spash oiled system, and I find it hard to believe that it works just as well and efficiently in either direction. I have only owned a VW for 2 weeks, so I haven't looked in one of these gearboxes, but its very likely that the internals are made to reduce drag, windage and pumping losses of the rotating gears and shafts beating into the lube.

    This, plus the 3.44 v. 3.88 R&P, plus variance stack-up in assembly, could all be conspiring to a very significant drag difference.

    There's a pretty straightforward way to test this: just take Duh and the Spyder back to the dyno facility. Pop for a pair of lug pattern hats for your new rear hubs on the Spyder that matches the wheels on Duh and you can then use the same tires to test both cars. That eliminates everything but the rear wheel bearings and brakes from the equation, and if you were obsessive you could manually reach in with a pry bar and retract the brake pads before your runs to remove that, too.

    Long dyno runs in each gear with carefully measured energy consumption would give you some very good data about the gearboxes- or maybe the motors, or something...

    My thinking is that the transaxle just doesn't spin backwards from its design direction very efficiently. Monitoring the temperature of the gear lube on a long run would be a triangulating data point there.

    More generally, you guys are bumping into the hard wall of mechanical efficiency with these cars- and kind of finding out the hard way how very good they really are:

    First, you thought you'd get 150-160lbs off the suspension with lightweight parts. It was 59lbs, and more than 1/3 of that was removing Dwayne's non-stock wheel adapters and coilovers. Shaving weight off of a Porsche or VW just isn't so easy, and you had to go with marginal street brakes to really get anywhere.

    The other things just aren't worth very much, either. Yes, they add up, but they're all expensive, time consuming, and there just is no low-hanging efficiency fruit on these cars. I'm really interested to watch this process now that I have a VW-based kit car to convert, and I'm grateful for the data points on all these things, but you guys are really scratching the pavement with your fingernails looking for tiny improvements in range. Don't get me wrong, somebody needs to do this, but man, what a hard road...

    Looking forward, I think you'll bump into this situation over and over. I'm dubious about any big weight savings with an aluminum chassis. Same for the carbon fiber body. Maybe the Special Editions cars are heavy on the mat and shot out of a chopper gun, so there's a lot of pork to be shed just changing the manufacturing process, but CF isn't a magic bullet on weight savings over S2 glass, especially on an unstressed panel.

    The plain fact is these VW/Porsche gliders are very, very good and efficient cars right out of their 60 year old boxes. Yes, you can tweak here and there at considerable expense, but the great news is that if you do NOTHING but clean, tune and lube the chassis, you've got 85%-95% of the efficiency you'd have spending $10k more on it.

    Kind of amazing, really...

    TomA

    ReplyDelete
  31. I agree that you can quickly reach a point of diminishing returns. But I think the electric concept does change the economics. Sixty pounds is 60 pounds and in this case it was unsprung weight. The difference in handling is marked.

    These tests showed nothing of range gains from weight savings. It was all about aerodynamics and rolling resistance.

    I am fairly confident we have isolated the problem to the transmission. What part of the transmission begins to become a bit disinteresting, unless we are going into transmission design. We'll probably just stick with the Long 3.44 in future builds, though we may talk to them about any ideas they have to make them turn more easily.

    Similarly I think an aluminum chassis would be a big play, simply because of weight. Again, it won't change the rolling resistance. But on acceleration, it really does count.

    As to carbon fiber, you may be right. At some point, the pounds simply become too expensive. But what we're hearing is that it may not be that much different. And there are some epoxy/glass combinations that are lighter and stronger and reduce the fume problems of polyester resins in manufacture substantially.

    Indeed, we are not going to change the transmission to get increased range on the Spyder. It's a pretty nice car as she rolls, but we find the difference to Speedster Duh remarkable.

    The point, Tom, is that moving forward with electric vehicles, things like weight, aerodynamics, and rolling resistance are important in ways that they simply are not with ICE powered cars. The economics really quite are different. And so our thinking should be different.

    In the case of the Speedster, I have some counter requirements. I like creature comforts. I would like a roadster version with rollup windows and a little more upholstery in the seats. That ADDS weight. And so it is nice to find ways to compensate. The heavy 3 inch DOM steel chassis makes good copy for the kit car guys, but how strong does a Speedster chassis have to be? It looks to me like you could build a Ford F-150 on the amount of steel they have under there. We can stiffen the body with battery boxes.

    And so it goes...

    Jack Rickard

    ReplyDelete
  32. Mark/Tom,

    Nothing changed with the motor. It is still running as it was in the beginning. Both motors are running the same direction in relation to how its mounted. The transmission was just turned around to face the other direction and the ring was just flopped to the other side. Nothing has changed internally either except that the ring was flopped. The pinion still rides on the ring exactly the same. The only change is that the ring is now on the other side of the transmission but now that the transmission is facing the other direction you will find that the ring is still on the same side of the car.

    I mentioned that parts are parts and one persons torque arm is not the same as another persons. So one tranny may be internally tighter than the other or looser. I suspect a few things that are adding up. The ring and pinion only because of the ratio. Put in a 3:22 R&P and I bet it turns easier yet. But if tolerances are tighter it will take more to move them and when things heat up that increases.

    ReplyDelete
  33. Tom,

    "There's a pretty straightforward way to test this: just take Duh and the Spyder back to the dyno facility. Pop for a pair of lug pattern hats for your new rear hubs on the Spyder that matches the wheels on Duh and you can then use the same tires to test both cars. That eliminates everything but the rear wheel bearings and brakes from the equation, and if you were obsessive you could manually reach in with a pry bar and retract the brake pads before your runs to remove that, too."

    Wouldn't it just be easier to run the soap box derby test backward to see if it makes any difference?

    ReplyDelete
  34. No. This isn't about coasting the cars.

    A coast-down doesn't say much about the transmission, especially if its in neutral, and in any case you can't isolate the transmission drag in the soap box test.

    If you put the cars on the dyno with the same wheels and tires, then you can actually see which drivetrain takes more energy to spin the dyno drums a given distance. You can also do that in different gears and at different speeds for additional comparison.

    Since you can (rightly or wrongly) assume the motors, controllers, brakes and wheel bearings have about the same drag, or account for the differences somehow, what you're left with is how efficiently the transaxles are delivering motion to the dyno drums for the energy you're putting into them from the electric motors.

    That's the best test. As Jack points out, though, its a moot issue because he isn't changing the transmission.

    ReplyDelete
  35. Jack,
    I find it interesting that you seem to have an aversion to using gauges that measure mechanical movement. You are quick to point out that the electrics is easier as you can make measurements with your many meters, but that the mechanics is more of a mystery to you.

    Might I make a constructive suggestion that you invest in a few mechanical measurement tools that will allow you to more easily pin point problems and verify correct alignment of components before spending a lot of money replacing parts.

    At the very least, you should have torque wrenches and dial indicators. Brake pressure gauges might also be helpful at times.

    With the torque wrenches you can measure break-away torque, rotational torque and of course, correctly torque nuts/bolts to the manufactures specifications. With dial indicators you can measure run-out on brake disks and flywheel/couplings.

    These tools can be purchase for modest prices and will take a lot of the guess work out of assemble and trouble shooting and should save you time and money in the long run.

    Thanks for the great shows! I don’t know how you find time for all that you are doing.

    Karl

    ReplyDelete
  36. Tom,
    Even in Neutral the bevel gears and the driven shafts are still engaged. This is where the best guess is blaming the drag.

    Karl,
    Jacks already stated everything spins freely and we have already seen all the equipment you stated as the video's rolled by through time.

    ReplyDelete
  37. I came across this interesting way of estimating Crr.

    http://ecomodder.com/forum/showthread.php/calculating-rolling-resistance-62.html

    ReplyDelete