Monday, June 28, 2010

Speedster Part Duh Range Testing

Went on quite a drive today. Everyone is concerned about the range of electric cars and fear it will be insufficient. At least when test driving, it is MORE than sufficient. It was again unseasonably warm, and two hours in the seat was a bit more than I had sought to enjoy.

I drove the beast out state highway 177 to country road V, thence to 177 to join US Highway 61 North. Highway 61 is one of those old scenic highways that was once a national artery, bypassed by the Interstate largely. It's a lot like Route 66 but about five less.

This is hilly, winding blacktop road through scenic country. Speed limit ranges from 40 to 65. I drove it between 50 and 60/65 almost entirely. I did finish at 70 mph for a 10 mile stretch on I 55 coming home. The route took me to Perryville Missouri, which has a reasonable facsimile of a Kentucky Fried Chicken. Cape Girardeau's KFC is notorious mostly for food poisoning.

The route convered 171.9 kilometers or 106.83 miles. We consumed 173.5 of our 180 theoretical ampere-hours. The Xantrex LinkPro made this quite easy to monitor. Of course, therein lies a tale. The result is 1.009 AH per KM or a little over 1.6 amp hours per mile. This works out to right at 190kWh per mile for the car. We also weighed it last week at 1986 lbs. This is nearly 200 lbs lighter than Speedster Senior at 2190. We did NOT have the hood an bonnet on part Duh when we weighed it though.

There are a couple of things notable about this. This drive is within 2/10ths of the longest drive in Speedster Senior. We get virtually identical AH per mile. The AC induction motor, 3 phase controller, and pressure tranducer driven regenerative braking, all put together, amount to just about NOTHING.

We have heard this chant about regenerative braking for years now. I did plenty of braking. We went down plenty of hills. We slowed into plenty of curves. The supposedly more efficient AC motor and the regenerative braking CANNOT BE DETECTED WITH TEST INSTRUMENTS. It's all a fraud. A lovely theory. A hope for efficiency. But a fraud.

You can niggle this to death however you want. More in the city. Stronger regen settings. Weaker regen settings. Regen on neutral and regen on braking only. We had regen. We started. We stopped. We did hills. No impact. Nada.

Yes, we'll do further experiments to see ensure apples to apples and oranges to oranges. The Speedster Senior is nearly 200 lbs heavier for example. But comparing these two vehicles, I find one thing VERY interesting. A lot of effort to go to a 3 phase controller, ac induction motor, etc. It would appear that if you are bemoaning the limited capabilities of the Netgain Warp 9 motor, you've been reading too much of too many people typing themselves smart. The old DC series motor and controller work fine. And the improvements you'll get with AC induction are mostly in your mind. Ok, you don't have to change brushes. Beyond that, methinks much over little.

The reports of 20% gains from regen are just total bullshit - off the table forever. Doesn't happen. How does this get started? Well some of these people are arithmetically crippled for one thing. The only thing I can imagine is they are recording 120 AH out, 20 AH regen, net 100 AH and concluding a 20% efficiency. First, the math is wrong. But second, the thinking is flawed.

We count amp hours to determine how much energy is used to go forward. The point is to go forward. Not to manipulate amp hours. It might be that coasting is a better use of the energy than all that regening.

Truth to tell, I had turned off the neutral braking regen for this trip and DID coast. I assumed I would get SOMETHING from braking. Apparently not much. I can't find it in the rounding errors.

Here are the finished cell voltages:

Cell 1 2.973
Cell 2 3.007
Cell 3 2.979
Cell 4 2.991
Cell 5 2.993
Cell 6 2.944
Cell 7 2.956
Cell 8 2.978
Cell 9 2.956
Cell 10 2.992

Cell 11 3.142
Cell 12 3.165
Cell 13 3.151
Cell 14 3.166
Cell 15 3.135
Cell 16 3.166
Cell 17 3.168
Cell 18 3.178
Cell 19 3.170
Cell 20 3.181
Cell 21 3.163
Cell 22 3.140
Cell 23 3.156
Cell 24 3.170
Cell 25 3.150
Cell 26 3.168
Cell 27 3.151
Cell 28 3.161
Cell 29 3.163
Cell 30 3.165
Cell 31 3.168
Cell 32 3.161
Cell 33 3.155
Cell 34 3.137
Cell 35 3.153
Cell 36 3.169


Anyone notice anything beyond the skipped line I inserted to call your attention to it? The cells are remarkably well "balanced" I thought for this end of the discharge curve. Clearly the cells have more than 180AH capacity. At 173 we really aren't down to the magic 3.00 volt level. There was probably another 10 miles in the car. Very pleased I didn't drive them.

This largely because the FIRST 10 cells are significantly lower than the rest. Why? Well, we have the Xantrex LinkPro connected to them. And while it features a very modest drain normally, I really like the backlight and have it on all the time. Unfortunately, the first 10 cells are what we use to power the Xantrex. And it is on 24x7 of course.

We've only been driving 3 weeks. Over the course of a year, this could lead to a serious difference in these cells, and ultimately the death of some. We can't turn the car loose on nonEV enthusiasts without fixing it. And I've no idea how to fix it. In fact, I'm having more instrumentation ideas all along, but they all come from the concept of powering from the traction pack to avoid isolation issues with the 12vdc system.

We HAVE to have an AH counter. And we CAN'T have an AH counter. I don't know.

The daughter box, which has been reduced to a daughter potentiometer and a couple of wires, worked marvelously. It came and went for about two miles. It would run normally, and then trail off as the throttle divider cut in. Then it would regain somehow and go some more, and then cut in again. After five or six iterations, it just stayed cut in and we had to limp home. It would limp ok. But it was clearly a sign the car was hurt and we had to creep home. I had it set a little generously because I wanted to see how far the car would go. WE'll simply set it for a battery size of 144 AH and use the 0% level to trigger. This will limit the cars range to 144 km of course, about 90 miles.

As noted in the video, we had a problem with the tachometer cutting out at 3500 rpm. It was a simple fix. On the theory that the thing was turning into a generator and overdriving our tachometer, we installed a 7.5kohm resistor between the signal lead and ground lead. Problem gone. The tach is rock steady and confirmed in accuracy by the weeny little numbers on the Curtis 840 display.

The heat exchanger wasn't getting rid of enough heat. We installed the old oil cooler that came with Speedster Senior in its place. Works great now. A little noisy.

If we are in third gear, and we press on the brakes deploying regen, it pops the transmission out of third gear. We've tested the shift mechanism extensively. The guys at Long Street agree that it has to be the transmission. So we have to pull the transmission, seal it in a manila envelope, and mail it to them - using an extra postage stamp I suppose. Easier said than done. Does ANYONE test ANYTHING before they ship it these days? Do you have any idea how many components we receive from American companies that are DOA out of the box? In fact, it is about a 1 out of 6 chance it won't even be the item we ordered at all.

Do you know how many we've received from China? None. Never. Nothing. Well.... we DID receive a charger from Thundersky several years ago with the voltage adjust knob broken off and missing. I couldn't figure out how to adjust it. Then I looked at the product photos online and saw what it was SUPPOSED to look like. We had a replacement in 3 days.

In any event, I'm very pleased with the car. It's a 100 mile car. We'll set daughter mode at 90 miles just for grins and to protect the cells. But it will do 100 miles driven for pleasure. We made no attempt at "hypermiling" or anything like that. It wll undoubtedly do more in town. And it will undoubtedly do less all freeway. But the winding country blacktop highway we took it on is just perfect for a Porsche. And yes, I'm a little tired from the two hour drive in the sun, but it was really quite enjoyable all the way around.

As to range, I guess it's a matter of seeing a glass half full. I'm amazed it will go this far this fast on batteries. I may never get over it. It's all the driving I want to do in a day. And it does it well.


Jack Rickard

20 comments:

  1. Very interesting.

    I think the weight difference is probably less than it appears. At least 40lbs of it is in the battery configuration and packaging.

    The efficiency comparisons bring me back to gearing. You already have (arguably) much more ideal gearing options with the AC setup, provided you are using it ideally, could really test that and were interested in doing so.

    With the DC there seems to be room for improvement. You could use taller gears. Not available in this chassis economically, but the 9" series motor has so much more torque than the AC that if you had just two or three widely-spaced gears and a final drive about 35%-40% taller, (and a big ducted blower to protect the motor at lower RPM,) the DC efficiency could compare even more favorably with the AC/regen.

    More than ever I'm thinking that ideal gearing and DC power is really the way to go. Its also currently the only way to go with a gross weight over 2500lbs or so, yes? I think its all too easy for everybody to get swept away in the OEM PR smokescreen about AC and unobtainable technology and lose sight of what is good and practical right now. If we just had a gearing and voltage map of the 9" and 11" motors plotted by efficiency across a range of vehicle weights that are realistic for the cars most people want to drive every day, (and some gearbox strategies to match them,) these series motors would run cooler and farther enough that the AC option would go off the table even where it isn't already.

    It would be tedious, but not terribly difficult to get that information. You could do it in a Vega (for the lighter curb weights,) or Malibu (for the heavier weights) wagon with a Lenco 6 speed planetary transmission (for the wide ranges between, and interselectability of the ratios,) a quick change rear axle with a few cog sets, a Soliton1 programmable for the various voltages, good instrumentation and a fair amount of time. One thing is for sure- the OEMs have never bothered to gather this data. I'm not aware of any university work on it, either. It is undone basic EV research. Where are these EE doctoral fellows looking for grant money for their labs? Just a ranting thought...

    Jack, you men are slaying shadow dragons and debunking shibboleths week in and week out. Thanks so much.

    Tom

    ReplyDelete
  2. Now you've got me second-guessing my AC50 choice and wondering if I wouldn't be better served by a Warp 9. You're happy with the car so far, and are getting the range with relatively fewer batteries, so I think I'll stay with the AC50. I'm probably a week away from ordering.

    As to the Xantrex, have you looked at the 1:5 Prescaler Kit on the Evolve Electrics web site? (http://www.evolveelectrics.com/E-Xpert%20Pro%20Accessories.html). It looks like it would be wired to the entire pack, eliminating your "first ten batteries" effect.

    ReplyDelete
  3. Hi Jack and Brian.

    I would put a small DC/DC converter to power the xantrex - the standard 1:5 prescale would most likely do the trick and is only $75.
    Tried to buy a LinkPro and wrote to both the US and the European e-mail address's at their website, but didn't get a reply. So I sent an e-mail to TBS and had an answer the next day.
    I'll wote for the original hub caps.
    Have completed charge/discharge/charge for 30 of my 90Ah cells so it is going (slowly) forward.

    ReplyDelete
  4. And here you had me considering bidding on a Curtis 1238-6501 on eBay. In any case, it sure didn't take you long to put this iteration through the wringer!! Amazing work once again. Have you heard that Dave Cloud's Metro has gone 201 miles on a single charge-mostly freeway- on lead/acid? Imagine what that baby could do on LiFePO4? I look forward to your next project now that this one is practically done....

    ReplyDelete
  5. Yes, I originally had the Xantrex on a "prescaler" of two resistors. The voltage accuracy was not good. So we just went to the first 10 cells. Hadn't thought it would load it enough to matter. What I fail to recognize is that the car sits there even when I'm not in the garage. 24/7. So 35 ma adds up eventually.

    We'll go back to a resistor divider to scale the voltage down. Maybe i'll throw in a potentiometer to try to adjust it correctly.

    No, I think the AC50 is a good package. The Netgain is definitely stronger, depending on your controller choice. But if you are doing it all for regen, I'm not seeing the gain. There may be one in the noise somewhere, but at that point, whats the point.

    Jack Rickard


    Jack Rickard

    ReplyDelete
  6. I wonder what would happen if we get rid of the 12v all together and feed the car with some higher voltage. Lamps, accessories and all can be found on 110v.

    ReplyDelete
  7. Hi Jack, after this first comparison, what is your personal DC/AC conclusion?
    "Take the cheaper motor/controller combination, advantages or disadvantages are insignificant smal" ?

    ReplyDelete
  8. I think I probably like the AC a little better. The Netgain feels more powerful, but it is larger in the compartment and poses us some serious battery box issues. The brushes are a bit of an issue. The AC motor just shouldn't ever need any maintenance.

    On the other hand, the Netgain feels more powerful - understand that we have very different gearing in the two cars so it's hard to tell how much of this is additional leverage and how much is the motor. And the Netgain motor runs cooler. The AC 50 gets quite warm - even with the additional fan we have on the end of the motor.

    There IS one element of regen that applies to both the Mini and the Speedster. The regenerative braking gives the feeling of power brakes. Since we have no vacuum pump on the MIni Cooper, this is a bit of a gig. The regen provides the "power" in our power brakes.

    Similarly, the Speedster, which has no power brakes, and which I can't imagine ever needing any, DOES feel pretty good with the regen on them.

    There has to be SOME gain from capturing energy in an actual braking situation. But it doesn't seem to show up in range or overall AH per mile usage.

    The AC 50 package is $4500. The Netgain is only $1700 but you do need a controller. The Kelly we use is about $2400. So you're at $4100 anyway. Not much to pick between them pricewise. A new Soliton1 for example is $2895 - putting you at the same price point.

    Jack Rickard

    ReplyDelete
  9. Jumping out of third happened to me on a Ghia once. You and Brain can fix it easily, at least compared to some of the other tasks you have done. You need to preload the gears before adjusting the linkage. Use a piece of plywood with a hole for the pinion bearing and 4 pieces of running thread to compress the gears to their running positions. Measure this. I was dismayed when you began to bad mouth regen but I see you have had a slight reversal of that opinion. Regen should give the car a more "normal" feel and allow the brakes to last longer. Also, why not use current limiting for valet mode? I'd like to build a car myself, but I'm 80 so I have to live vicariously through you. I hope you don't tire of this soon.

    Bob

    ReplyDelete
  10. Another big advantage of AC also isn't available in this chassis- simplified drivetrain requirements.

    The OEMS are using AC because a reversible, well-sized motor would be great with a two speed gearbox that's 1/3 the weight and size of the transmission you still need for a DC motor. I know you're trying to get one of the Getrag 1eDT330 units for its monster torque, Jack, but if Vera could spare you one of the smaller 2eDT180 two speed units, it would be ideal for the speedster.

    You'd save 45+lbs over the AC50/adapter/coupler/clutch/VW gearbox, with better weight distribution and more continuous power, liquid cooling, and a higher top speed. It makes me think that the available AC systems are still a little too bleeding edge to be demonstrating their advantages and real potential- There still aren't enough controller or motor choices; The things that are really good, like the Getrag units and that Sevcon "self-charactorizing GP240-Series controllers, are just not here yet. Close, but nothing yet...

    At the same time, we are seeing real maturation with DC. The Soliton1 is a really, really nicely finished piece, and the Warp11HV is an order of magnitude better highway EV DC motor than everything else to date. In Sepex trim it could do regen with a Soliton1, if George and Qer were stimulated to do it by, say, a customer like yourself who could see it through to make one work and publicize it. I'm suggesting that should be your next power system and project- you could easily pave that road, and its a great one for guys who want to back it up with a Powerglide in a RWD chassis. Full disclosure- that's the drivetrain I would use to do a true daily driver EV for my family.

    Consider yourself nudged, and keep up the good work.

    Tom

    ReplyDelete
  11. Jack,
    Had the same problem with running the battery monitor from the lower cells (6 in my case). Upgraded to a victron BMV600HS. Its basically the same as the xantrex and tbs units (probably a rebadge) in its capabilities but works from 72v to 350v.

    Very nice work on the speedster!

    Damien

    ReplyDelete
  12. I can't speak to the real world range impact of regen as I have nothing to compare it to. I'd imagine unless you had long, steep hills where you can use regen instead of brakes to hold your speed regen probably won't give you much. I do know when I first hooked up my Curtis and AC31 the factory settings for acceleration and regen were very week, both a bit disappointing. I borrowed a 1311 programmer and tweaked everything I could to make both more aggressive. The car became much more fun to drive, and since my regen is only hooked to the throttle it's mostly a single pedal driving experience, and I can control the amount of regen fairly easily. Generally I leave the car in second from 0-65mph, the higher motor speed giving me greater fan speed since I'm only using the stock fan. My motor temps have been better than my controller temps.
    Forgetting the range issue, I see AC having 3 other potential benefits:
    1. No brushes to wear out as has been mentioned.
    2. A more interesting driving experience, similar to driving a high compression sports car, with strong regen on the accelerator pedal.
    3. Greater safety as there is a much lower possibility of a controller failure causing a runaway motor.
    I'm still hoping for a higher power AC controller in the future as I think this motor has more potential. Even just 144 Volts and 700 amps would really wake things up.
    JRP3

    ReplyDelete
  13. Interesting JRP.

    I had regen on the accelerator and it seemed to actually hamper range. I removed it. We use a pressure transducer for the brakes and they feel very good with regen.

    I may have been hasty in my comparison. I drove Speedster Senior yesterday and did NOT get 1 AH per kilometer - more like 1.13. Today, weather permitting, I"m going to drive both cars on the same route for a direct count on both.

    I much prefer the free wheeling feel to the one pedal control you get with aggressive regen on the neutral braking band of the accelerator.

    But your message puts me in mind to test it all three ways on the new Speedster - no regen, regen braking only, full regen to see what the actual numbers are.

    Jack Rickard

    ReplyDelete
  14. Damien:

    I find the Victron BMV600 on eBay but I'm having a little difficulty locating the BMV600HS. Can you provide a source or link for this device?

    Jack RIckard

    ReplyDelete
  15. It takes some practice and some getting used to with regen on the accelerator pedal, some may simply not like the way it feels. I have mine set to 40% regen, (neutral braking), which gives me up to 200amps of regen in second gear, normally I see between 50-120 amps. If I want to coast I can hold it at or pretty near 0 amps, though that's harder now with my failing pot.
    JRP3

    ReplyDelete
  16. Jack,
    Have a look on this page and download the datasheet for "precision battery monitoring".
    http://www.victronenergy.com/battery-monitors/bmv-600%20and%20bmv-602/
    It lists all of the BMV series specs including the 600HS. Company I work for are a victron dealer in Ireland but i'm sure you'll find something closer. If not drop me an email we always keep one in stock.

    Damiem

    ReplyDelete
  17. Damiem:

    I can find the datasheet alright. I can even find a 600 on eBay. I can't find ANYBODY who carries the 600HS. I don't have your e-mail address. But I'd love to find out what one of these would cost and where I could order one - mjrickard@gmail.com

    ReplyDelete
  18. I could find the price in the pricelist... +€500 for and isolated DC/DC converter - ouch... or am I reading it wrong?

    ReplyDelete
  19. Its powered from the traction pack. No dc dc is required.
    Damien

    ReplyDelete
  20. The Victron BMV 602 is about US$ 240 (€ 190 Euro). You can get it at marine/boat suppliers.
    Datasheet: http://www.victronenergy.com/upload/documents/Datasheet%20-%20Battery%20monitoring%20-%20rev%2001%20-%20EN.pdf

    ReplyDelete