Monday, September 5, 2011

Be Careful What you Ask For

Six months ago we received a request for a quick list of parts to make a B&B Manufacturing Turnkey Minus 66 Cobra kit car into an electric drive vehicle. We responded that it wasn't quite that easy. The components required kind of had to be worked out based on what you wanted the car to do and it required a bit of thought. We were more correct in this than we realized. And therein lies a tale.

I have enjoyed and endured a variety of experiences in my brief life. Racing automobiles never was one of them. Vehicles of my youth revolved around farm and construction equipment and we worked on them of necessity. They were not a lot of fun frankly. So I never got into the passion for automobiles and specifically not the race scene.

The 66 Cobra was produced specifically for racing. And it is one of the more popular and enduring replicas, usually available in kit form. An entire culture has grown up around this mythic beast. But I was learning Z80 assembly language with Rodney Zaks at the time all that was going on.

But a clean sheet design using someone else's idea of what an electric car should do might be a challenge. Er, duh, yeah.

The topic has actually come up again with the same car. The answer is the same only more so.

This week we pretty much completed assembly of the battery pack for this vehicle. I think it illustrative of how one basic design decision/component choice drives three others, and those in turn three others until you are in a quagmire of components all deriving from the initial suppositions. Be careful what you ask for.

The basics go like this. There are three factors in an electric car conversion:

1. Range

2. Performance

3. Cost.

You can optimize for two. By way of example, the little spreadsheet I keep on component costs for the Cobra currently shows $42,519.40. No labor. A bit of outside fab. But by far and away just components. And it doesn't include the original cost of the "Turnkey - Minus" at all.

The original concept involved a car that would do a true 120 mile range, 120 miles an hour, and a 0-60 time in six to eight seconds.

To do 120 miles reliably, we have to have a 100% discharged range somewhere in the neighborhood of 140-150 miles. That takes a LOT of battery to do in a conversion - which adds WEIGHT. Weight is the enemy of performance. To launch a 3000 lb vehicle to 60 mph in six to eight seconds requires a lot of POWER. In this case, I calculate about 250 kw of power.

To do this, we chose to go with a Netgain Warp 11HV motor. Why? Series DC motors are just very good at producing torque, which is what we need to launch this beast. But they are power limited by the voltage you can run them at before brush arcing occurs. The 11HV features Interpoles - windings between the field windings that are aligned so as to neutralize the induced voltages in the windings NOT connected to the brushes so when they DO make brush contact they do not arc. This allows higher voltages, and thus higher power. THe normal Warp 11 is traditionally limited to 170volts. This HV version let's us go to a higher voltage, and so a smaller current for the same power.

But not much smaller. To get 250kw, at 231volts, we have to have over a thousand amps. Of course, when you take 1000 amps out of a 231 pack, it really is NOT 231 volts any more. It's more like 175 volts. And so we're back to needing more current.

We are going a bit out on a limb opting for the reallly still beta Netgain Controls Warp-Drive Industrial version. But it purports to do 260volts (actually higher) and 1400 amps (we hope). So even if our pack sags substantially, we can still be up around 250 kw total electrical power applied to the motor.

Of course, we don't want to burn up the motor. Our best effort at avoiding that is the addition of a cooling air blower. Not all blowers are created equal. But we can get 450 cubic feet per minute from a converted Garrett Turbocharger. Unfortunately, it is $350.

We also have to cool the controller - rather substantially. So big glycol cooling system.

But the biggest wrestling match is the battery cells. We have to have a lot of them in series to get to 231volts (69). That points toward a smaller cell. But what about current? We've done 1000 amps from these basic types of cells using 180Ah cells. Could they do 1400 amps? Perhaps. For a few minutes. But 180 AH cells are kind of bulky. And as it turns out, a lot of our spaces are pretty constrained in this car.

So we opted to strap two 90Ah cells together. That gives us a very different dimension and granularity of positioning to the cells. But it also means 138 cells total.

And that drives us to a LOT of battery boxes - seven that have to be fabricated and assembled. We go to Southeast Fabricating to do the basic foldup/weldup. But then we have to add angle aluminum, attachments, lids, paint, terminals, etc. It's a good bit of work with seven boxes. Two underneath only hold seven cells each. Another two hold eight cells. Etc. It turned into a wrestling match to fabricate, install, strap up, and wire all these cells more or less securely.




Unfortunately, a LOT of these cells then pose accessibility issues. We have a two tier box in back. You have to completely disassemble, in the right order, the rear top box to access any of the cells underneath. The two rib boxes underneath are a little easier. We made little doors on the outside of the car you can remove with five bolts to expose the terminals of the cells. But many in front are likewise rather a procedure to access.

So we bottom balanced those 138 cells rather carefully,, many in the lab before installation. We tried to pair higher capacity cells with lower capacity cells to diminish the variation in capacity across the pack.

And it wound up being a LARGE pack for a car this size. It weighs in at 973 lbs for the cells, not counting boxes and straps, and provides 41,607 wH of energy storage.

But that gets us several things. First, 300 wH per mile for a 3000 pound car is about the right 10:1 ratio we see in all our cars. If you've followed the Spyder/Speedster fiasco you will now be much smarter that range is not always range and indeed it is a very ethereal thing. But we've found the 10:1 rule pretty good in practice.

The 300 wH and the 41,607 wH work out to 140 miles at max 100% discharge. 120 miles would then be 36000 wH or 86% DOD. That's a little higher than we like but we're making tradeoffs here and doing our best. It will safely deliver 120 mile range in a pretty good variety of conditions. We're hoping he charges more often than that.

So you see the game we are playing.

Second, we get our higher voltage for the Netgain Warp 11HV. And finally we get our 1400 amps, at least for a few moments. We're hoping for a full 250 kw on the dynamometer.

Speedster Redux got us 147 kw in 2385 lbs for 62 watts per pound and does a zero to sixty in 6.5 seconds. At 3000 lbs, that will require 186 kw. But frankly we are hoping for some upside at 250kw, if we can pull it off. And there's a lot of slips between here and 1400 amps.

Of course, to handle that amperage we have to have a kind of massive high current relay. We took two fairly ordinary ones and coupled them together. In doing so, we hooked up quite a bit of copper plate to help dissipate the heat.

As to cable, the choice at these currents is between 2/0 and 4/0. The 4/0 would be better, but it is heavier. MOre to the point, it is harder to work with. With seven battery boxes, that's a challenge. And since we don't have any very long runs connecting all those boxes, we opted for the 2/0.

To keep the noise/EMI down, we have gone with Champlaign Cable Company shielded cable designed specifically for electric vehicles. Unfortunately, it's also designed to fetch $8.57 in ducats for each linear foot of the cable.

To fuse all that current, we use a Ferraz Shawmutt A30QS800-4. This part number indicates a 300vdc rating and a 800 ampere rating. It's not precisely a slow blow, but it will take 1400 amps for a few seconds.

As to a high current relay, we faced a bit of a problem. We used two of the Tyco Kilovac EV200 relays. The proper relay for this is probably their EV500 series, which will do 1600 amps for 10 seconds.

But the EV500, called "Bubba" is typically available for $1200 to $1400. Occasionally they come up on eBay at $350-$500 each. But the EV200, rated at 500 amps, are common as dirt at about $70 each. So we used two of those in parallel.

The current capability of the EV200 is kind of dependent on the size terminal lug connected to it. The terminal lug actually acts as a heat sink. So to do the advertised 500 amps, you need a pretty largish lug and cable. While we're ok there, more is better. So to parallel them, we actually used a pretty hefty piece of copper bar stock. This should help dissipate heat and allow our pair to do 1400 amps briefly. We'll see. At $140 instead of $1400, it's worth a try.


On the other end, we spent a couple of grand on lighter wheels and low rolling resistance tires, which did save us 10 lbs each on the front and 16 lbs each on the rear for 52 lbs. We also spent $1120 on an aluminum third member for the differential. Again, the weight savings was substantial

The result is our car is 32% battery by weight. But the components to do it would total over $42,000.

All of this from a couple of basic assumptions regarding what kind of car we'd wind up with. Had we said 11 seconds 0-60, and 80 safe miles with 100 miles to 100% DOD, we would have a much lower weight and cost and a much easier time fitting battery cells.

My experience with these cells comes out the same way every time. They want to make a car with 100 max 80 safe miles. Yes, you can get them to do more, as we have in Redux and now Cobra. But it grows exponentially more difficult and more expensive.

The question you have to ask, is what do your REALLY need and want in an electric vehicle. Everything in design is a tradeoff. There are no little easy answers. I can't give you a "kit list" until I've done the design and build, tested it, revised it probably several times, and so forth.

Will a go fast Cobra with a 120 mile range scratch the itch? Let's hope so. But be careful what you ask for.

Jack Rickard

60 comments:

  1. Jack I guess some of your builds push the edges of the envelope - Mini Cooper for sophistication, Escalade for mass, Cobra for a nutty combination of range and acceleration. So what are you going to do when those are out of the door? A hypermiler (min watt hours per mile)

    ReplyDelete
  2. Greetings Jack - another superb blog.

    Today's blog has this long time follower of EVTV (and proud member of the unauthorized Heavyweight Division of the EVTV Fan Club) worrying that your Factor #3 - Cost - alone seems likely to make large/heavyweight vehicle conversions (including anything remotely similar in weight to the much anticipated Elescalade) out of reach for most of us EVTV fans with a mortgage.

    Just taking your $42K Cobra components costs number alone, but then of course adjusting downward for more modest performance/range (e.g., using your “11 seconds 0-60, 80 safe miles w/100% DOD”) -- then adjusting (significantly) upward for much more weight for the Elescalade or other beloved large/heavy set vehicles -- makes one worry components cost-wise whether large/heavyweight vehicle conversions are simply completely outside the budgetary reach of most EVTV fans not inside the “Pass the Grey Poupon” financial sector.

    Admittedly, guest-imating adjustments to the $42K Cobra components number may not be a completely useful predictor for those of us contemplating a large/heavy (but less-performance/less-range than Cobra) conversion -- given the obvious differences - apples vs. oranges - between the Cobra/Porsche Universe vs Elescalade Universe. And sure, until one puts pencil to paper, item for item, just roughly adjusting that $42K downward for less performance/less duration, then adjusting upward for 2X or 3X the weight - isn't meant to yield anything quite specific

    - but it does just give one pause if large/heavy vehicle conversions remains a desire for some of us.

    - makes one wonder if for large/heavy vehicles we're looking at the need to be prepared for component costs alone (no labor) to FAR exceed that $42K Cobra number -- perhaps far exceed even a $50-$60K range?

    Do you have any relatively current Elescalade component budget numbers -- that might ease such large/heavyweight conversion budget anxiety>

    -- and a chaser of any Rickardian words of hope wouldn’t hurt either -- to buck up the Heavyweight Division of EVTV fans . . . that it still might be reasonably financially possible given adjustments downward from Cobra performance/duration – and yet still yielding a desirable performance/range – & without robbing a bank – for us to venture forth in large/heavyweight electric car conversions?

    Or is the $50-$60K components alone range OR EVEN MORE? -- what we really might have to contemplate?

    ReplyDelete
  3. @jasg2

    The first thing to remember is this scale of EV conversions is not meant for the budget conscious. These projects start with kit cars that run upwards of $25-30K just for the roller. The EV components can double the price of the car, and that's not even counting labor.

    However, this also leads to the second point to remember. Even this high end build is ultimately less expensive than the Tesla Roadster.

    As Jack has said many times before, this is the early adopter phase. The components are expensive because they're primarily hand-made.

    Even using a VW Beetle, for which the parts to convert it are the most common and nearly plug and play, you're still spending $1-2k for the car, but $15-20k for the EV parts.

    If cost is anywhere on your mind, EV conversions are not for you. Small or large.

    - Doc

    ReplyDelete
  4. I would be dissapointed if you don't get to show "Revenge of the Electric Car' as I had really been looking forward to it. The guys at EVBuilders.com wanted to screen "Who killed the electric car" at a fundraiser but the licence/fee was just ridiculous. Ends up that the total amount raised was less than the screening fee would have been. No doubt you are up against similar issues?

    I recently received the "Racing Green" DVD and watched it all in one go, just re-viewed "who Killed the Electric Car" and I have already seen "Charge" - twice. Makes for a few late nights!. The EVTV friday show generally comes out Monday morning here (NZ)and I have usually managed to watch it before I leave work at the end of the day.

    So, as you can tell, I am thirsty for new and unseen video to watch! Here's hoping you can swing it, but failing that I will happily watch "Charge" one more time.

    ReplyDelete
  5. Well, we ARE a TV show. We have to get your attention somehow. If I do a Ranger over and over, I don't think that is going to work. We need to come up with new and different things. And of course talk about new components.

    It IS fun to kind of advanced projects. But we try to illustrate the principles and actually available components you can use for YOUR build. Like all good games - bridge, chess, and hand grenades, it can be played at a lot of different levels.

    We;re basically looking for projects that will be fun for us, entertaining for you, and perhaps illustrate something new.

    The Elescalade and larger vehicles. I'm afraid I have some bad news for you. It is not a LINEAR progression up in cost, more like an exponential one.

    The Escalade will wind up being 7200 lbs. It starts out at 5800. The bottom line is it takes a certain amount of energy to move a certain amount of mass. And a similar design concept as described above. It does a 7.5 second 0 to 60 now. Why would I want less. And so we wind up with TWO Netgain Warp 11's on one shaft and TWO Soliton's at 1000 amps. So now I need 2000 amps at 192v.

    So we go to 400 amp batteries and 4/0 cable. Note that just the cable goes from $8.57 to over $12 per foot. The cells weigh 1700 lbs and need an enormous box and we're adding heating.

    ALL the components then get to be more expensive. Larger DC_DC and probably an aux battery of 90Ah Thundersky's. Have to look at Ultracaps. Heating, cooling, auto transmission.

    The car cost $59K. The batteries alone are $34,000. The motors and controllers were $20K. I think I'll be lucky to get it all done for $160K all in, counting the cost of the car.

    It's kind of a demonstration that it COULD be done. A no compromises electric vehicle. That is not precisely to say it SHOULD be done. The costs grow exponentially with the growth in power and battery pack. It's just a function of mass.

    This is why we've been saying "weight and aerodynamics." It's because they become important. A Hummer or an Escalade CAN be done. But it kind of demonstrates the problem.

    So what is the ideal donor car you would LIKE to see done as a project?


    Jack Rickard

    ReplyDelete
  6. Just curious if you considered the TransWarp 11HV, Jack. Its supposed to be a special order motor from NetGain, but I don't really know much about it.

    ReplyDelete
  7. No we did not consider it very long. It is the same motor with a yoke for direct drive. We need and want the transmission and clutch.

    Jack

    ReplyDelete
  8. Jack said "weight and aerodynamics."... Allow me a moment to grovel.... Abserluckinglutley!

    On the evsource webpage there are a couple of calculators.

    On my project If I triple the AH then add in the extra mass, I get double the distance gain for what is a minimalistic vehicle. Obviously there are real world variables that will make these figures vary but its very telling and honest from the people who sell the parts.

    Therefore I will be including "hot pack" changing, (all in parallel) for when the large distances need to be covered.

    http://www.evsource.com/tls_battery_main.php

    The gearing calculator proved useful for me too as it includes tyre sizing.

    ReplyDelete
  9. I think jasg2's post--a little heavy on the cryptic rhetoric-- may be clarified by answering the following question.

    "So what is the ideal donor car you would LIKE to see done as a project?" -Jack

    That flat bed work truck we saw in the initial walk through tour of the new shop, optimized using factors 1)Range and 3)Cost. (but not golf cart performance, say 13s, 15s 0-60?)

    ReplyDelete
  10. What would I like to see done? A GT40 replica with a carbon fibre monocoque (Tornado claim 1850 lbs rather than 2180 lbs for their example). Take out 500lbs of Small Block Ford and put in 1500 lbs of batteries plus 150 lbs of motor and controller, and it would still weigh less than Redux.

    It is a big footprint and a small cockpit so there is room for a zillion batteries. Frontal area is about 3/4 that of Spyder and Speedster. Forget acceleration (you have a Redux for that): use an AC motor chosen for efficiency, and batteries chosen for energy density. Drive it to Nashville and back on a single charge.

    On second thoughts it might be less than ideal. Getting in and out is easy only if you are 5 ft 8 inches tall and weigh 140 lbs

    ReplyDelete
  11. Jack,

    I would like to get back to the range issue with the Spyder... Somehow I still can't believe that a 3.88:1 ring and pinion instead of a 3.44:1 should decrease the efficiency by 20%. Wouldn't this mean that the drivetrain losses in the Spyder should be around twice as high as in the Speedster?

    You didn't clearly say (or I missed it) whether the input shaft of the transmission in the Spyder turns in the same direction as in the Speedster (as seen from the motor) or not? Or ask differently, are you sure that the transmission Duane installed has the ring gear mounted on the opposite side?

    The gears inside the transmission are all helical gears and therefore they produce thrust along the shaft depending the direction of rotation. AFAIK some Porsche transmissions don't have thrust bearings on both ends, so running the transmission in the 'wrong' direction could lead to higher losses... I don't know if this is true for VW transmissions as well?

    One relatively easy way to check this would be to put Spyder on jacks, remove the wheels and measure the current required to turn the axle in the typical drive gear at constant rpm (perhaps using a 10 turn pot instead of the throttle pot to be able to adjust the rpm precisely). Then swap two motor leads to have the motor run backwards and measure the current again. If the current required for running the wheels backwards is significantly lower than running forward you know what's going on. The fix would be to open the transmission and to turn the ring gear to the opposite side...

    Regarding what car project I would to see: I would love to see a small modern sports car like a Porsche Boxster or Audi TT etc. trimmed for
    efficiency. Also it would be great to show what effect it would have to go direct drive (I know you like to keep the transmission and the clutch, but I think the gain in efficiency could very well be worth the compromise). To make it even more interesting, use two smaller AC motors and controllers and drive each rear wheel separately (which of course means you would have to get a custom reduction gear box, but one the other hand would eliminate the losses in the differential).

    Sorry for the long post,

    Markus

    ReplyDelete
  12. Markus, what you suggest is a soap box derby going backwards down the road.

    For instance to see if unloading the thrust bearing will free it up?

    ReplyDelete
  13. Andy, yes - going backwards but without the influence of weight, tire and alignment differences.

    ReplyDelete
  14. Correct me if I'm wrong but, I believe this controller has the ability to run the motor in reverse without reversing the leads!

    So, you could electrically power the car in forward an then use electric reverse (on a flat part of road) to test for the difference in amps or watts between the two?

    ReplyDelete
  15. Mark,

    The AC motor can in fact run both directions and can be done so quite easily but why? There is no need. The issue with the transmission is that the transmission is flipped and needs to have the pinion ring flopped. That actually keeps every working properly. The ring gear continues to work properly. Why is everyone stuck on the fact that the trans is turned around? It has not bearing on the issue. Tolerances do have bearing. The issue is pretty much done anyway.

    ReplyDelete
  16. Yeah, The soap box derby was pretty hairy as it was. Doing it backwards might just be that bit too much. What was telling was the way Deux initially seemed to race off and Spyder slowed yet it kept rolling as if it had treacle in the gearbox.

    12% velocity on the shaft should not equate to 18% more drag at slow speeds. The battery vehicle society published data on drag. Gearboxes are pretty flat but only rose slightly at high rpm. That would be the oil drag.

    If a gearbox/differential runs backwards under load, often they're quite noisy!

    Conclusion:
    My belief is The Spyder simply has a very tight differential and/or maybe extra drag between the gear synchro's. Nothing that a good 40k mile run won't kill or cure.

    ReplyDelete
  17. Greene,
    call me stuck, but my point was that it's hard to swallow that a tight transmission and/or a little bit different gearing should almost double the mechanical losses. Jack looked at every other aspect but wasn't 100% clear on whether the ring was mounted on the other side or if Duane skipped this step and just reversed the rotation direction of the motor. Do you know for sure that the ring gear was moved to the other side? And even if Jack thinks that the ring was swapped, the quick experiment I suggested should tell us if it was done or not.

    Just my 2 cents...
    Markus

    ReplyDelete
  18. that escalating cost is one of the reasons you should try a so called direct drive approach.
    70k$ conversion might be tolerable to you but it isn't to 99.99% of people interested.

    take your own advice and don't type yourself smart on direct drive. do it, if only as a test. think of the allure of the chance of proving me wrong for the very first time : )

    ReplyDelete
  19. Markus,

    Well if the ring was not flipped and it was being run in reverse then I might just have to agree with you on that issue. However if the ring was flipped as it should have been then all bets are off. One easy way to know that is ask Duane. Duh!

    Dan,

    .....for the first time....... Boy are you going to get spanked. I know Jack does not need to type himself smart. He already is. By the way. Its easy to advise when its others money. I could care less if his conversions are $70k the information will be good for those who do $5k builds too. Direct drive has issues that will only cost more. Use what is available and get over it. Your will be a happier camper if you do. Can't imagine having to re-engineer a vehicle that works perfectly fine with a transmission. Can't imagine spending the money for the re-engineering project. Go spend your money and prove Jack wrong. :)

    ReplyDelete
  20. GreenEV,

    I would just like to see the Spyder do the best it can.

    -If you used an inclined short piece of road and went up backwards (recording power used)
    - then repeat going up forward
    -and the difference turns out to be significant...We will have felt one more piece of the elephant!

    ReplyDelete
  21. My vote for the "would like to see" is a 1st generation Honda Insight. It'd be really neat to see what one of the most efficient hybrids ever made could do with electric drive.

    ReplyDelete
  22. Dr. Righteous,

    That would be easy to answer. You'd then have one of the most efficient electric cars. :)

    I think that would be a good project as well. I vote for AC drive hands down.

    ReplyDelete
  23. Mark,

    For how the Spyder is built right now who's to say its not running its best. Just because the Speedster coasts further does not mean that the Spyder is not at it's best. I still think the internals are tight and that they'd be tight enough to cause the short free rolling distances. Might get better under power now but they are still doing coasting tests. If the tolerances can be adjusted later and improve then it should be done so we all know about how critical tolerances can affect the range.

    Pete :)

    ReplyDelete
  24. To prove its not the brakes on the Spyder giving you a poor rolling resistance you could pull the calipers and tie them back. But make sure and put a temp plate of steel in the caliper in case someone touches the brakes.

    I have found disk brakes tend to drag a little more than a drum to keep them warm and working when going through a water puddle. The residue pressure valve also may be inside the caliper or the master cylinder.

    Thanks, Jim

    ReplyDelete
  25. It's not the brakes. It will be seen, heard and felt on such a quiet open top car. Brian will of spun and checked the wheels after bleeding the brakes as a matter of course. Everyone does this. Anyone here ever heard of brand new hydraulic disc brakes binding. After replacing all the brakes why should this problem persist?

    Dan and Pete you are both correct about gearboxes.
    It depends on power/weight ratio and what you are willing to lose in the process. Risks of high continuous motor loading at low rpm, lack of top speed and a band of efficiency within a certain speed range. However, at the cost of doubling up on motor/controllers on my project this is what I am aiming to do.

    ReplyDelete
  26. We are a group of volunteers and starting a new initiative in a community. Your blog provided us valuable information to work on. You have done a marvelous job!

    Gem Car Hub Caps

    ReplyDelete
  27. When Jack upgraded his controller on the Spyder he found he had to swap a couple of controller/motor wires because it was running backwards.

    An easy "fix" if the bevel ring was not turned round.

    Apologies if I'm playing the devils disciple.
    Andrew (m1aws)

    ReplyDelete
  28. OOPS!
    Forgot to say, like Dan, I can't, won't and not going to fit a gearbox so aim to venture into this eyes wide open..
    You could say I'll be stuck in third gear ;)

    ReplyDelete
  29. This is a joke suggestion for a conversion project (sarcasm is hard to convey through text)

    check this video for a concept vehicle geared to help cash strapped Americans
    (there is a commercial at the beginning)

    http://www.theonion.com/video/ford-unveils-new-car-for-cashstrapped-buyers-the-1,14381/

    ReplyDelete
  30. speaking of good donor cars, I've recently become aware of the Audi A2 as an EV donor car.
    it's all aluminium and has one of the best CdA's of all cars. and it's quite affordable.
    it is only available in europe afaik though.

    funny rumor is that the A2 was developed to be an EV for the california carb mandate but never made it because carb caved. not sure if that rumor is true though.

    it weighs 900kg and one converter pulled 250kg of ICE junk out of it so an EV version could be as light or lighter.

    ReplyDelete
  31. I love the Cobra. Always have... since long before I knew anything about electric cars. I can see this beast running a lipoly pack over 300 volts with a bad ass Zilla Z2k. But hey, it isn't my money so, it's easy to dream.

    As much as I like the LiFePo prismatics for conversions, this high performance car is begging for lithium polymer. It would go a long way in contributing to 2 out of the three challenges you're faced with. The higher energy density will take up a lot less space and should be easier to fit in to available spaces. The output power will be more than you need also. Your weight (I mean the car's weight) :) will drop significantly also. Depending on packaging, these could be a little more challenging to connect/box.. especially pouch cells. That said, it would be a very interesting process for many viewers who know a lot about integrating the prismatics, but haven't seen much with pouch cells. Certainly cost is another issue.... although, it hasn't seemed to slow u down any yet.. :)

    The other thing I have stuck in my little brain is the controller choice. For me, going with a proven high power controller for a high power conversion... this thing is begging for a Zilla Z2k hv. You know what you are getting, it has a proven track record and meeting your performance goals is only a software setting away. The Netgain control may be fine, but at 1400 amps and really an untested beast... I dunno.

    Either way, I'm a big fan and look forward to each week of EV activity. I'll see you in a couple weeks. :)

    Best Regards.
    Gary

    ReplyDelete
  32. Can I swap 12lb of rocking horse manure for a Zilla 2k? Oops! no Zilla's. Jack and lipo's? hmmmmm. .. Nice thought though.

    A2 Audi.
    Interesting and maybe a great choice Dan. Prospective EV'ers ought to read up the wiki on them though. I'd recommend the diesel for lower rpm and high torque resistant transmission.

    If you obtain such a car, write to Arvind for details about the large Diameter Agni motor which ought to be a very good match for it. They are making the first batch of 50 in a week or so. With such low inertia motor using brushes, would you need a clutch? This might almost fit inside the clutch space.

    Still wildly guessing but I'd consider 22 GBS 200AH (5.6"x2.55" each, 45cm^2 for all 22, 120Kg, ~$6k) cells stacked vertically to use the available height under the "bonnet". This car has a front service hatch - ideal for quick service items and plugging in.

    ReplyDelete
  33. "So what is the ideal donor car you would LIKE to see done as a project?"
    Jack,
    I would love to see a different kind of car. A Toyota Landcruiser FJ40. Perhaps first with clutch and Gearbox, later perhaps direct drive with a Epicyclic gearing, one or two motors, AC or DC....
    I think it would be nice to stay on one car, and see the difference by changing parts.
    Thanks
    Mathieu

    ReplyDelete
  34. Andyj said...
    "Can I swap 12lb of rocking horse manure for a Zilla 2k? Oops! no Zilla's."

    You apparently haven't looked at http://www.manzanitamicro.com/ lately.

    David D. Nelson

    ReplyDelete
  35. Hi David,
    Yes, it's been announced on here. Can't say if any have been made though.

    ReplyDelete
  36. David

    Ah. You found a WEB SITE with a description of a great new product. Oh well, a great old product. GREAT.

    I spend about three hours a morning looking at all the sexy hardware I CAN'T HAVE for my electric car. Because it only exists in web form.

    That brings up an idea. Let's give up on making these real cars, and just design REALLY MUCH BETTER electric cars, but only on the web. Without having to actually build any, we can do some awesome paint and interior, 600 mile range, 0-60 in 1.1 seconds, and 25 wH per mile. Trust me they would be MUCH cooler than the clumsy awkward beasts we make in the garage......

    Jack Rickard

    ReplyDelete
  37. Ahh, your avatar needs a car, does it?

    ReplyDelete
  38. I want mine with a gearless magnetic transmission!

    http://www.magnomatics.com

    ReplyDelete
  39. Here in the UK I have had a similar issue with unobtainium on a different level. Set myself out to spend over £2.5k ($3k+) on motor/controllers only to be told by the UK based importer 'oh, we only import their electric bikes'.

    This man is not hungry enough.

    ReplyDelete
  40. The Zilla Z2k HV can STILL be had. Even before the Manzanita Micro venture. I spoke to Otmar a few months ago, and he still had a few of them and was all ready to ship me one for somewhere around $5k. One step beyond (behind?) the old web search is the telephone.... :)

    ReplyDelete
  41. For all of you that jumped on me for questioning an unsubstantiated statement, here is the response when I asked if the Zilla controllers were in production:

    =================================
    from Manzanita Micro sales@manzanitamicro.com
    reply-to sales@manzanitamicro.com
    to "David D. Nelson"
    date Mon, Sep 12, 2011 at 9:25 AM
    subject Re: [Contact Us: Sales] Zilla in production?

    David,

    Yes they are in production and shipping. Currently the lead time is 1-2 weeks, maybe less depending on which options are chosen.

    Thank you,
    Clarice Matz
    Manzanita Micro LLC
    Phone: 360-297-1660
    Fax: 360-297-1905
    www.manzanitamicro.com
    =========================

    David D. Nelson

    ReplyDelete
  42. It seems the Spyder gearing must offer some advantage over DUH at some speed or instance, or it would be a buggy whip and not still be being manufactured. The same motor under different loads will display maximal efficiency at different vehicle speeds. Several of the constants in the equations used in recent EVTV episodes to describe motion and resistance are not really constant, e.g., total aerodynamic drag and rolling resistance are non-lineal subject to speed and other forces such as lift. The derby test needs to be performed using different hills with the cars evenly weighted [total and possibly front and rear distribution]: a heavier [within reason] car has more potential energy to convert to kinetic/forward motion and eventual range. Coast downs [from motor powered runs of 30mph, 20mph, etc.] need to be conducted in addition to the free fall derby runs. Energy use tests need to be conducted at 20 mph and 80 mph to rule out peculiar aerodynamic interactions such as negative lift given that the Spyder was engineered to be a race car while the Speedster was developed as a roadster. I was surprised to learn the Spyder has a higher cD than the Speedster. I was also surprised to learn the less aerodynamic looking 356A has a lower cD than the Speedster. The EVTV Spyder doesn't seem to have a racing "tarp" which envelops the driver's seat which I saw in a racing photo. Are EVTV Porsches exact reproductions of the originals aerodynamically speaking, or could the reproductions have different pans, whatever, which unexpectedly detract from the Spyder or augment Speedster? As to suggested conversions, I'd like to see Jack noblesse oblige tackle a project optimized for range and cost. Make it a more common car like David Cloud's highly aerodynamically modified Geo Metro which is reported to be able to travel at highway speeds for 200 miles using lead batteries. The Audi A2 would be very interesting. Don't know if European homologation = U.S. street legal. How about a Porsche 356 reproduction conversion given that the is supposed to have had a cD of 0.28. I think Jack mentioned his wife's Lexus. If that is a Lexus SC 430 convertible, that would make a beautiful conversion. I think SC 430s have a cD<.3. I have been researching like crazy and will not make it a habit of posting so many excerpts/links. Thanks for all your work and the great shows!


    http://www.thetruthaboutcars.com/2010/02/an-illustrated-history-of-au
    tomotive-aerodynamics-in-three-parts/

    An Illustrated History Of Automotive Aerodynamics – In Three Parts

    ...Either of these values put the “pillbug” at or near the top of the list of the most aerodynamic concept cars ever built, like the Ford Probe V of 1985, with a Cd of .137. Built on the chassis of the rear-engine Mercedes 170H, it was substantially faster as well as 20% to 40% more fuel efficient than its donor car.

    ReplyDelete
  43. I'd like to spent all day answering for you Perin, But you do have a LOT of errors.
    Brian weighs 200lb, "Hop in then". = No change.
    Air resistance does not matter at slow speeds.
    Porsche = Volkswagon drives.
    Not a Lexus, an Escalade.
    The best cd was the GM EV1.
    cd has to be multiplied by the frontal area for the cda = Total drag.
    If the lead battery powered car does 200 miles then why is it not in the Guiness book of records?
    Spyder and Speedster final tyre speeds not so far apart from each other.

    The Spyder with less weight, better weight distribution, a better ride, better cda, similar everything else except shows more drag in the drive.
    Very simple and nothing confusing.

    ReplyDelete
  44. I may be mistaken, but the act of beginning the free fall derby--releasing the brake or clutch, or any other of a number of variable start up factors could come into play. The cars may display different free fall derby outcomes using different hills. The weight balance of the cars could be important to the tire contact patch. Coast downs from slower speeds could also yield important information. Energy uses tests need to be conducted at 20 mph [lower aerodynamic drag] and at even higher speeds [80 and even 90 mph if possible] in order to determine if negative lift or variable cD might be peculiarly affecting the systems. There performance of the transmissions and the gearing are surely subject to variations per speed/load, as well. The Spyder was designed as a racer, the Speedster, a lower speed roadster. I don't think we know if the reproduction cars have the same cD as the originals. The EVTV Spyder may be especially weak in the cD department because there is no tarp which surrounds the driver. I've read the Mercedes project modeled from a boxfish had a super low cD. Reports of Mr. Cloud's vehicle are but a Google search away for you. I'm almost certain Jack mentioned his wife didn't like filling her Lexus with gas recently. If it is a convertible Lexus, I think that would make a great project. But as I said, I'd like to see a range and cost optimized project. Blogspot doesn't seem to like it when I share more than one link, so if you want a url, refer by number in a pm or in a post.

    ReplyDelete
  45. 1.) www.newton.dep.anl.gov/askasci/eng99/eng99640.htm

    ...Basically, there are far too many variables and unknowns to be able to calculate exactly what speed will give maximum mileage. Among the parameters that must be known are the following: Engine, transmission, and differential efficiencies at all speeds, detailed knowledge of aerodynamic drag at all speeds, rolling resistance of the tires and wheel bearings at all speeds, and so on.
    As you can imagine, in order to attempt to accurately calculate the optimal speed for minimum fuel mileage, this quickly turns into an impossibly complex problem. As a result, engineers do not attempt to calculate this. Typically with most cars the speed that results in minimum fuel consumption is between 50 to 60 MPH (80 to 100 km/h). This is generally the speed where the engine runs most efficiently, and the speed just before air resistance (or "drag") starts to exponentially increase. To determine the answer to a greater accuracy, actual tests are made. I won't be sharing a million links in future posts. Add "w" or "h" as needed.
    2.)Ferdinand Porsche once said it makes no difference where a car’s engine is located, so long as it’s light. The 1500 unit weighed a mere 160 pounds, while early 356s rarely exceeded 1,750 pounds at the curb. Fore/aft weight balance was around 780/970 pounds, but this wasn’t the drawback it might seem. In unofficial tests conducted by an unnamed Southern California aircraft company, the 356 body generated 175 pounds of front aerodynamic downforce that effectively equalized weight distribution.
    3.) 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."
    4.)...In doing this they were disobeying important principles in automotive aerodynamics, and were all the more surprised at the results: the Cd value for the car was 0.095. In aerodynamic terms it was just as good as the shape – as measured on the ground - considered ideal by aerodynamics specialists (Cd 0.09).
    4.)Dramatic streamlining demonstration video
    5.)...Aerodynamicists often say that a 10% improvement in aerodynamic performance is good for a 2-3% improvement in gas mileage, based on an EPA road speed of 48 mph. Therefore, an improvement in Cd of 5.73% could account for a 1.8% improvement in gas mileage.
    6.)Conclusions:Already at 50 to 80 km/h (depending on the car), aerodynamic drag becomes the greatest loss factor It's much easier to reduce drag than weight-related losses: reducing the Cd from 0.29 to 0.28 is like eliminating 100 kg vehicle weight

    ReplyDelete
  46. Perin:

    You've got a lot of swing thoughts in your golf game there. But you have picqued my interest there with the reference to the 1954 Spyder cd of 0.45 and frontal area of 1.03 square meters. Where can I find that reference?

    Assumng 3 being true then 4 doesn't follow.

    And 6 is just nonsense lacking context. Weight matters tremendously when accelerating and to large degree when decelerating, but hardly at all at steady speeds.

    As is OBVIOUS from the data shown, rolling resistance is clearly the prevailing factor at 50 to 80 kmh while aerodynamics plays an increasing role above that. The Speedster was 18% better at 40 mph but 10% better at 70 mph. But of course all values were worse at 70 mph. This showed the increasing role of aerodynamic drag at the higher speed.

    So at 40 mph, the majority component has to be rolling resistance. And indeed in the Soapbox Derby, we found a huge difference in rolling resistance.

    After eliminating everything we could between the two cars, including tires, wheels, bearings, calipers, rotors, etc. we are left with the transaxle itself.

    That's what's left. As to what IN the transaxle is the problem, I kind of have an interest diminishing with my ability to test. We'll use the transaxle we used in Speedster Duh in future builds.

    ReplyDelete
  47. Jack,

    Are you using good test procedures and using the same Amp-Hour meter on both cars.

    ReplyDelete
  48. Perin,
    For electric power, consumption is directly measurable as you see on Jacks graphs. Fuel engines are another matter entirely.

    The differences between Jacks two very similar cars free wheeling down the same stretch of road at slow speeds does not involve engines or air resistance.
    It's not complicated.


    Finding the CdA only needs the will to do it.
    CdA= F/(.5*Da*V^2)

    Cd= drag coefficient
    A= area in m^2
    F= force in Newtons
    Da= air density in kg/m^3
    V= air velocity in m/sec

    Da --> 1.204 Standard dry air pressure@20C

    Here we can measure F at any given velocity through power consumptions between the same wheel speeds unloaded on a rolling road and rolling. We can directly measure CdA by subtracting the rolling resistance and fitting the final elements in the equation.

    If anyone is bothered. The Spyder wins easily.

    ReplyDelete
  49. dochiller:
    The soap box Derby was decisive. No motor involved. Spyder lost.

    At speed, the Spyder was catching up on consumption because it is far more slippery.

    how complicated can it be?

    ReplyDelete
  50. Andyj:
    I am not worried about the cars.
    But more that the meters are reading Amp-hours the same. Use the same meter for both cars.

    ReplyDelete
  51. Jack, I posted the link you requested but the blasted blogspot deleted my post.Google "364 Aerodynamics of Road Vehicles 7.6.3 Feedback to the Design" and see Table 8.1 on page 6 of 25.

    Andyj, as I mentioned, we really do not know if the EVTV productions are as slippery as the factory originals. Plug in a cD of 0.28 for the Speedster and a 0.60 for the Spyder and see what you net. Carefully study and consider #1 in my earlier post.

    ReplyDelete
  52. Jack,

    I can't wait until it is winter time there so you can replace the wheels with skis and THEN do the downhill derby with Duh and the Spyder!

    Then we can move on to more interesting topics such as how far a layden swallow can fly while carrying a coconut.

    It is and was the gearbox as was stated before. There is no science required to understand this. There are no formulas that need to be calculated at all.

    At 210 miles per hour it may matter but at 2 tenths of a mile per hour it certainly doesn't. Stop beating a dead horse here guys.


    -Chris

    ReplyDelete
  53. dochiller,
    On the soapbox derby. Both of Jacks meters read the very same super accurate reading of zero.

    How accurate do you want that to be?

    ReplyDelete
  54. Perin,
    When a downhill race is less than 10mph. What role does CDA play when the vehicles weigh nearly a tonne, have a CD of a half and a cross sectional area of 1.76m^2 each? Apart from nigh on none.

    I read your post and was astounded to see you are missing the point of something so simple! Why are you attempting to overcomplicate the facts with obfuscations? Nobody is interested in straw man details like the variability of fuel engines.

    The experiments were not vague. They gave constant answers.

    Chris,
    I like to tenderise dead meat >:-))

    ReplyDelete
  55. Andyj, I have not indicated cDA has much of a role in the downhill tests. Why you hold the Speedster and Spyder have the same areas when authoritative documentation clearly indicates they do not is a mystery. A bit more than 2D chess is needed given the Porsches are fighting in a 4D space war. My position is that a sufficient amount of data has not been collected to know what is causing the strange results. Could it be the transmissions. Sure, it could. It could be something else or a combination of a number of things. I'm not convinced the gravity derby runs are eliminating enough variables. Jack weighs 78 or so pounds more than Brian. The Speedster is heavier than the Spyder. The Spyder's gearing may dissipate more energy from stop than the Spyder. Might the Spyder outperform the Speedster on a different hill? Yada yada. The gravity derby needs to be conducted removing as many variables as possible [tire pressures, weight,...]. More importantly, I think, roll down tests from powered runs need to be conducted. On the CdA front, tests at 25-30 mph and 80-85 mph might be very revealing. Why is the Spyder transmission not a buggy whip? It is also only assumed that the EVTV reproductions have exactly corresponding cDs as the originals. On all this I'm an honorary Missourian: SHOW ME!
    7.)STUDY OF AERODYNAMIC INTERACTION BETWEEN CARS AND ROAD Recent studies show that for a modern car the basic shape stands for approximately 45%, wheels and wheel-housing for 30%, and the floor and detailing for 25% of the total drag. As there can be observed a significant potential for further improvements are to be found in the underbody flow. Today, this represent the main research issues of the designers , the team members of project having also contributions. 8.)
    The Influence of Aerodynamics on the Design of
    High-Performance Road Vehicles
    Guido Buresti
    Department of Aerospace Engineering
    University of Pisa (Italy) pg 50 of 55 "The increase of vertical download generally causes an increase in drag"
    9.)The Aerodynamic Drag of Road Vehicles
    Past, Present, and Future by William H. Bettes If we compare the tradeoffs between weight
    reduction and aerodynamic drag, a ·10 percent reduction in the drag coefficient (with no change in frontal area) would yield approximately a 5 percent fuel economy improvement in a subcompact on the highway. This same improvement would require a 16 percent weight reduction, which is substantial. The same 10 percent drag reduction in an intermediate car would require a reduction of 22 percent in weight for equal fuel economy improvement. While further weight reduction on autos as we now know them will be difficult, there are a number of ways to achieve easily that 10 percent drag reduction and much more by aerodynamic design...It is interesting to note that the theoretical lower limit for forebody drag in potential flow is zero for a semi-infinite body, and for a body of finite length it is actually negative; that is, it can provide some thrust. Some examples of this are aircraft wings and yacht hulls. Even if we can't actually approach this theoretical limit with trucks, there is at least much potential for improvement in the front end of these vehicles.

    ReplyDelete
  56. This dead horse is beginning to bloat. Maybe I'm not the one who wants to puncture its hide.

    A long post for something so simple!
    They both have an area of 1.76m^2 and differ in teeny tiny details.

    Everything else, you are splitting the same variables that were used in comparison. WHY?

    Brian did hop in the lighter Spyder with Jack to make it the same weight as the Speedster. Waste of time. The Spyder exhibits a greater rolling drag.

    Once again you talk about aerodynamics at <10mph.
    FFS! Give reality a break!

    We all know the Spyder has a more slippery shape. So what? It's a shame it does not have a good free wheeling 3.44 ring & pinion.

    P.S.
    I'm fully cognisant in engineering & aerodynamics being a retired aerospace engineer.

    ReplyDelete
  57. Perin:

    That may be so. What you are apparently NOT cognizant at all is of the tests we've already performed and shown. It is clear from you comments you just haven't been following this series AT ALL and have jumped in with an incredible amount of rather blathering text that is mostly out of context.

    We did one of the rolls with both Brain and I in the car. Brain is 190 lbs and we have about a 200lb difference between the two cars. Nothing changed as far as the derby went. So subsequent tests with or without the weight don't either.

    Go back to the beginning in the videos and go through this again. It is ABUNDANTLY clear aerodynamics is not the issue. Being an "expert" and typing a lot doesn't get you very far in this venue. "Show me." Which generally means suggest a PERFORMABLE test we can do to eliminate something. ANd if it's something we've already eliminated NO INTEREST.

    In further developments, we AUDIBLY have a dragging brake in the rear - a recent development. We'll probably repeat the test without the rear calipers at all.

    Jack Rickard

    ReplyDelete
  58. Andyj, "The 1954 Porsche 550 Spyder has a cD of 0.45 and a frontal area of 1.03 m2." Just take a look at a side by side photo: http://www.constructorscarclub.org.nz/Resources/spyder2a.jpeg Jack's Spyder does not roll as far from the hill in front of EVTV-HQ as his Speedster. Whether the forces [weight distribution, vehicle lengths and gravity centers, tire contact patch area, transmission/gearing, etc.] which cause the Spyder FROM STANDSTILL to roll less distance than the Speedster would effect a similar outcome in roll down tests from motor powered speeds of 20-25 mph is an important, truth seeking question, in my opinion. Jack, Andyj wrote, "At speed, the Spyder was catching up on consumption because it is far more slippery." This conflicts with Jack's, "It is ABUNDANTLY clear aerodynamics is not the issue." [To be fair, Jack may be referring solely to the derby tests.] Andyj's, "Nobody is interested in straw man details like the variability of fuel engines." seems to allude to non-variable performance electric motors. I can see the powered roll down and 25 mph and 85 mph speed tests are not going to be conducted, and that's certainly the host's prerogative. I'm not afraid to confess ignorance. I would like to know if electric motors somehow are non-variable as to efficiency? And what of the Spyder's transmission/gearing: why is it not a buggy whip?

    ReplyDelete
  59. Perin, May I call you "cloth ears"?

    "Abundantly clear aerodynamics is not the issue".. *AT VERY LOW SPEEDS ON THE SOAP BOX DERBY!*

    I will repeat AGAIN! There are extremely few mechanical differences between the Spyder and the Speedster. Anything you come up with is counter to the Spyder losing out, so forget the similar parts.

    How simple can I tell you the Spyder DOES have a lower CDA than the Speeedster but the same Spyder has a higher rolling resistance.

    Everyone is now singing from the same song sheet here but not you. Ask yourself why.

    Did your teacher have a nervous breakdown or simply ignore you? I suggest you watch the video's and read Jacks blogs. Absorb the details within the graphs.

    You remind me with the issues with Redux. Battery's that can easily give 1000Amps; a motor that can not help but accept 1000Amps but only receives 700Amps at max throttle. Almost nobody would blame the man in the middle...
    The controller.
    It was set wrongly.
    We were hearing theories that were one step beyond recharging UFO's on here! And you seem to share this mindset.

    ReplyDelete