Friday, June 17, 2011

Feast and Famine in the land of the Electric Car.






It IS Friday, and I'm just now blogging LAST Friday's show, which I have to do in a hurry so I can film THIS Friday's show, so it isn't so late that it becomes NEXT Friday. Brain started this Friday thing.

We have gone from standing around summarizing things with no parts to work with to having four projects in the air at the same time.

This week, we received our motor for the Cobra (no wait, that's THIS Friday's show which I haven't filmed yet).

We also received our 400Ah Winston batteries from China after a loooonnnnggggg wait. We actually have a good bit of the Cadillac done , it's just strewn about the shop and none of it is in the car.

Arnulf Larrsgard deposited two vehicles with us in shutting down his efforts to convert cars for Norway. It would appear that since the fire on the Pearl of Scandinavia ferry, Norway has BANNED all conversions. The only electric cars allowed to come IN to the country are OEM electrics. No idea how this effects personal conversions IN Norway yet. But Arnulf's plans to make them in Davenport Iowa and import them to Norway have been legislated out of existence.

We consequently have inhereited a Ford Edge and a Chrysler Town and Country. The Chrysler doesn't even have a motor and controller since apparently Arnulf was scammed out of a small fortune by Richard Hartfield of Altantic EV or some such. It does have some Thundersky batteries that have never been hooked up to a motor.

The Ford Edge runs. But it is atrocious. Originally converted by Paul LIddle in West Palm Beach, it didn't get any better in Iowa either. The batteries are not in boxes, but rather lead acid style racks that lip over the battery edge. Oh, they're secure. But the rack is about 1/8 inch from the terminals. The DC-DC converter was actually resting touching one of the terminal straps. A box underneath was faced FORWARD with the terminals open to the elements - beneath the car. The TBS Expert Pro was disconnected and they had lost 10 cells already trying to power that.

The vehicle is massively heavy at 4600 lbs. It uses a Soliton1 and Netgain Warp 11 to drive a Ford automatic. Apparently no effort to change the shift points on the transmission so it hunts and slams at random. The Soliton was one of the early beta models - red annodize in fact and the software was last updated in 2009.

The cutoff switch is a loop of 1/0 in an Anderson plug under the hood. And the thing uses about 575 wH to move a mile. A 300 v battery pack the width of a single Thundersky 90Ah. It doesn't quite fit our design scheme if you know what I mean.

We're adding a second string of 90's, lowering the voltage to 250, losing one of the undercarriage racks, repositioning some things. I don't know what we'll do about the transmission.

But it seats four. Has plenty of cargo. The Soliton idles the motor well and it has great air conditioning and heat. And the radio works. But the cell voltages were everywhere and balancing the now 144 cells has been a little bit distracting.

We are actively working on the Cobra battery boxes at the same time. We had a local fabricator make us up three for hte Cobra and one for the Ford Edge and he missed by exactly 1/4 inch all around - very consistently in EVERY dimension. So those had to be redone.








Most frustrating of all is the Mini Cooper Clubman Electric. We are using the Rinehart Motion Systems controller. We are trying to get a hydraulic pressure transducer to work with this controller to control regenerative braking. Actually it works pretty well. But anytime we have it connected, it causes the controller to fault out. Doesn't matter if we are in the proper brake mode or not, just CONNECTING the transducer causes a random fault of various types at 3500-5000 rpm. Below 3500 rpm it works and indeed the brakes modulate regen very well and it is most pleasant to stop the car now.

But the controller shuts down randomly at anything above 3500 rpm. I THOUGHT I had it fixed with some aluminum foil on a noise issue. Apparently not.

We had used a wire braid shield on this, but ot put it on, we had to disconnect the phase cables from the controller. In doing so, we shorted one of the outputs so that when we fired it up, we fried the controller with a loud pop.

But unlike the TIMS600, we do have a bit of support. I can e-mail Rinehart at any time and usually get a response pretty quickly - and they are in Oregon. So we sent the controller back for repair and they had it back to us in LESS than a week. Having a support function is a big deal for us after our ordeal with European component suppliers who are simply useless.

But we got the controller back and wired it up with the shielded cable. We STILL have the IDENTICAL problem we started with.

But I did put the controller in and out three times - which has gotten to be a pretty arduous task. I read their HV connection document thoroughly and it appeared to imply that the steel gland nuts should have a proper ground and that this somehow grounded the shield on the cables. I took the thing apart and examined it carefully and came to the conclusion that there was NO WAY that shield would ever contact the case or make ground. There was a plastic capsule from the gland nut shrouding the copper foil taped ends of the braid, and they jammed into a phenolic stop that cannot conduct anything.

When I mentioned it to them, they responded that my controller has the EARLY gland nut design, and that they have NEW gland nuts that ground the shield. Clearly this is an ongoing and developing problem. IT has been about six months since we've had the controller, and already we are an EARLY model.

Why they did not replace the gland nuts when they rebuilt the whole power section AND replaced the controller board rather escapes me. But they did overnight me the new gland nuts, and if I get a day sometime in the next six months, I guess we'll go through this whole thing again.

Meanwhile, we drive the car using the brake light switch to control the regen.

Worse, I DID ground the OTHER end of the braid and it didn't appear to have any effect on the problem. I now again do not believe noise is the cause of the problem. But I don't know what is. And they cannot reproduce it at their facility. Impasse.

We did add a season switch to our cooling/heating system so that the controller and motor provide heat in winter, and in summer use the radiator and fan for cooling. It's manual. And it's not daughter friendlly. But it works.

Anne Kloppenborg and company visited this week from Amsterdam and brought some Dutch Gin with them. We had a famous time talking EV's and drinking gin.

We also did an interesting experiment where we compared the voltage of the newly received Winston 400AH cells to some cells I had still in the original unopened box that were manufactured October 11, 2008. The voltages were within three THOUSANDTHS of a volt between all of them. I have said it many times, there IS no "self discharge" shuttle mechanism in this cell chemistry. IT's not just that it doesn't perform this lead acid feat of legerdemain, it CANNOT because it has no such internal shuttle mechanism to support it. Without self discharge, you cannot have cell drift. And that is the final peg that the BMS guys all scurried off to to hang the BMS hat on. I think we've finally found a way to debunk that. But I have no doubt we'll continue to hear it chanted and quoted on into the night forever.

In any event, you have seen it with your own eyes. There is no "proper care" for these batteries in storage. These have seen temperatures well over 100F and down to 0F through three seasons with no humidity control. The boxes were covered in dust and looked as if insects had been at them. They have been in no particular orientation. All I can say is they were never rained on or flooded. And they come out of the box within a couple THOUSANDS of a volt of each other and a brand new battery.

ANd so all of the THEORIES about all that are no longer interesting to me. We've done this with CALB cells and now THundersky's - both aged in the barrel room of your local illicit distiller for years of careful aging. They should taste great. But in fact, they are just batteries. They do not apparently self discharge at all.

Regards;

Jack Rickard

23 comments:

  1. Great show, so Anne Kloppenborg and company came all the way from Amsterdam. Impressive.

    JMS

    ReplyDelete
  2. Lots of interesting stuff going on. Should be a good show this week. Kind of sounds like you need to hire a few extra people temporarily. I say "hire", but no doubt some enthusiastic viewer will want to do it for free.

    Didn't your Cadillac Escalade battery box come back being slightly the wrong size? I see a pattern here. ;-)

    I liked Mr Ball's video last week. Better camera work. Oh! talking about camera work i was wondering if you guys ever considered making a DIY steadycam? I have been thinking for a bit it might be something you can use and with your fabrication skills it really shouldn't be very difficult at all for you guys.

    Just as an example take a look at this youtube video:

    http://www.youtube.com/watch?v=-MQAPmsABnE

    it shows some of the guys footage using his DIY steadycam and a few shots of the device. If you just do a video search on youtube for "diy steadycam" you'll find a load of stuff on how to construct one.

    Here are a load of tutorials and various ways to do it:

    http://www.instructables.com/pages/search/search.jsp?cx=partner-pub-1783560022203827%3Anpr2q7v5m6t&cof=FORID%3A11&ie=ISO-8859-1&q=steadycam

    Hope you find this useful. :-)

    ReplyDelete
  3. A boat is an interesting proposition. I have a nice old Glass Correct Craft. I could mount a motor and have it mid mounted as this is a center mount setup with the prop out the bottom of the boat. It is light and quick to get to plane. It may actually be an excellent platform for an electric boat. Small but large enough for loads of fun. They are 16' boats. Not sure if I will do a boat but it is a consideration. I have an extra motor. .

    ReplyDelete
  4. Hi Jack,

    Great show with the guys from Amsterdam. One can think of a lot of ways a silent -FAST- boat could be used. Hope they are able to finish the project.

    On the Mini grounding and shielding issue, noise can be one of the most frustrating problems in the electronics business. After several decades chasing such problems I have come up with a few "rules" I use to help eliminate them. (Notice the weasel word "help".)

    1. Shielded cables should be only connected to a local ground at one end. I prefer to connect them on the end with the most sensitive equipment -- i.e controller boards as opposed to battery packs. In chasing the noise problem you may have to experiment with grounding the shield at the opposite end. Connecting at both ends sets up the potential for ground loops inducing even more noise.

    2. NEVER leave any extra wires in a multi-conductor cable open and un-terminated. They make excellent antennas for noise. Just connect the unused wires to ground at one end. This is one rule that I do not violate.

    3. Always have a pocket full of 0.1 uF caps of the appropriate voltage rating to jumper across suspect terminals or to ground to try to localize the noise source. In this case, with the high voltages AND high currents, it might not hurt to try some noise suppression components like Transorbs or MOVs. This is where engineering design ends and the black art begins.

    As for the Ford Edge, I liked that vehicle for the roller, but could not find one cheap enough. I ended up with an older 1997 Ranger.

    From the sounds of the one you have, I would think the "bulldozer rule" should be under consideration. Removing all previous attempts and starting from a clean slate may be better than trying to fix faulty design and poor execution.

    Looking forward to today's video.

    Roger...

    ReplyDelete
  5. Boats.
    In the UK, people who have canal boats are speaking up on full vehicle tax to their diesel fuel. These people who live on them really do live up to their name "Water Gypsy's" and will go to any lengths to avoid paying any taxes.

    EV power boats will give impressive acceleration with a big prop and so easy to keep and own if engineered properly. Something to watch!

    Sproggies,
    Much Radio Amateur equipment tends to use a soft ferrite block (toroid) for their DC input terminals. Wrapped together to kill off lopping great amounts of loose RF.
    We also see this method on computer peripheral cables. As a block attached a couple of inches from the plug. This is only a fix for the signal and supply cables.

    0.1uF caps are great. I'd fit them on their own. Might saturate too easily in an EV power environment though.

    The whole idea killing stray noise is if both cables pick up exactly the same interference between both polarities and in the same phase then the difference in voltage is zero. Cables are best tied together in parallel to their earth. Maybe with ferrite rings.

    3,500~3,800 rpm on a 350V motor. Is this where the back emf is starting to bite?

    Andrew

    ReplyDelete
  6. Hi Jack,

    Too many projects! (although I am equally guilty)

    I really like the thought of a Ford Edge as an EV.
    That automatic is electrically controlled,
    so it CAN be made to obey...Easy?, not so likely.

    The Mini's symptoms are consistent with noise/EMI issues, but they probably are better addressed at the inputs to the controller.
    Do you currently use twisted pair shielded cable for the inputs such as the throttle, speed sensors, temp sensors and the like?
    If not, it will be worth looking at too.
    And the advice already given regarding grounding the shield at one end is spot on.
    Usually most effective when grounded at the controller end only.
    I'd also experiment with filtering the power inputs with chokes and caps on the 12vdc input (if it has a 12vdc input...) and ferrite cores and film caps on the traction voltage into the controller.

    Good luck in your pursuits!

    Mark Weisheimer

    ReplyDelete
  7. I was going to use your TS/Winston cell comparison to prove to some people that these cells don't self discharge, but then I realized they'd probably just say we didn't know the initial shipped voltage of the 2008 cells so it doesn't prove anything. Did you happen to measure and record any of the 2008 cell voltages when you first received them?

    ReplyDelete
  8. Hey JP, I think the reasoning is that these type of cells are always shipped at the same Storage Voltage.
    The 90Ah cell pack was sealed and had been stored since 08.
    That is one hell of an experiment. Perfect timing having an outsider to verify. Not that we need one as from what I have seen on the shows, Jack would have read out the results no matter what.
    Drawing a comparison to Lead Acid,NiCad or NiMh, one could argue that cell drift is more pronounced when the cells are connected in a pack. For some reason that I do not understand, there seems to be some truth in this for cell chemistries that suffer from self discharge. In 2001, I laid to rest all my Sanyo NiCad packs and switched over to LiPo's. I had split one of the NiCad packs into single cells with the explicit intention of checking whether there was some truth in this. What I can say is that the voltage of the packs dropped to Zero within a year while most of the single cells still had 0.2V after four years. This experiment did not have the proper controls but does give some indication.

    ReplyDelete
  9. I really like the idea of electric boats and though my budget would dictate something along the lines of a kayak with a Torqeedo drive system (http://www.torqeedo.com/?id=258&L=2), my favorite so far is the Czeers MK1. Although sadly, their website seems to have disappeared, there are still pics and video of it in action. (http://alturl.com/kkcdk)

    They used the same drivetrain as Electric Cars Europe (http://www.ececars.nl) so if Anne wanted to contact Czeers people, perhaps these guys can put him in touch.

    Another company with electric motorboat plans is Epic Boats. Apparently still just a computer rendering but maybe they'll actually produce something soon. (http://epicboats.com/232se.html)

    Thanks for a great episode.

    Also, I am glad I was not drinking milk as the end credits came up.

    Domenick

    ReplyDelete
  10. The Epic is real, but a plug in hybrid using Flux Power batteries, not a pure electric.
    http://www.youtube.com/watch?v=KxygByCdI0I
    http://www.fluxpwr.com/products/batteries/

    ReplyDelete
  11. Thanks for the answer to my last question, you've been helpful. My next question has to do with "miles per cell" as opposed to "miles per gallon". How far will the average car go on a given cell? Will a car go farther on a 200 Ah cell than a 100 Ah cell? If so what is the difference in distance? I think these are questions you have to ask yourself before beginning a project.I have a 1981 Corvette that I want to convert, but I'd like it to go 100 miles on a charge. Needless to say, Corvettes don't have much room. Do you have a formula for figuring out how many cells I need to reach the goal of 100 miles per charge

    ReplyDelete
  12. Glenn,

    It really isn't miles/cell you want. It is Watt-hours/mile you want. If you are talking LiFePO4 cells I have found that 3.2V is a good working voltage for estimating energy capacity. A 100Ah LiFePO4 cell will have 100Ah * 3.2V = 3200Wh of energy storage. Naturally a 200Ah cell of the same type will have twice the energy. If you are talking about a different type of battery then a 100Ah battery of that type will have a different amount of energy.

    I did a quick search on evalbum.com and found one Corvette that had a believable 425Wh/mi figure. If you want a 100 mile range you will need 42,500Wh or 42.5kWh of useable energy storage. If you only use the top 80% of your cells you will need a 53.125kWh pack. This translates to about 16,600Ah worth of cells. If you go with 200Ah cells you would need 83 cells to reach this energy storage and the nominal voltage would be 265V. If you went with 180Ah cells you would need 92 cells for a nominal voltage of ~294V. My 12.8kWh pack weighs about 320lbs so that 53kWh pack would weigh about 1325lbs.

    HTH,

    David D. Nelson

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

    ReplyDelete
  14. I posted without checking my own facts. The average car will go 1.2 miles on a 200Ah cell, and 1.09 miles on a 180Ah cell. I don't see my project being finished soon, so I'll get the newest batteries when I get to that point. What are battery companies going to be offering when my project is ready?

    ReplyDelete
  15. Glenn:

    The average car doesn't go anywhere on a 200 Ah cell. In fact, the average car uses gasoline.

    And electric cars won't go anywhere on 200Ah cell either.


    It' s not terribly complicated, but you will find it a little easier to deal with range issues if you think in watt-hours per mile.

    A watt is of course the basic unit of power and 1 watt equals 1 volt at 1 amp of current. One of these cells is nominally 3.2v and a 180 Ah cell would then contain about 576 wH of energy.

    We then come to how much power is required to move the car. As a rough rule of thumb, you can take the weight of a car and divide that value by 10 to get wH per mile. So a 2200 lb car will take about 220 watt hours per mile while a 3500 lb car will take 350 wH per mile.

    So if your car weighs 2800 lbs and you have a pack voltage of 192v and you are using 180Ah cells, you are looking at 192 x 180 = 34,560 wH and you will most likely use about 280 wH per mile.

    34,560 / 280 = 123.42 miles.

    I do hope this helps. A "miles per cell" rating really doesn't make much sense..

    Jack Rickard

    ReplyDelete
  16. I'm using the "miles per cell" as a way of judging how many cells you can fit in a car. If you really like the car, but it won't hold enough cells to do what you want, it's a waste of time. Storage of electricity is a problem, but storage of all the cells is most important in the building of the car. I really like my 81 Corvette, but if I can only put enough cells in it to go around the block, that kinda defeats the purpose. I'm in the very beginning of my project, finding places to put cells is sometimes a little frustrating. Jack, there hasn't been a thing you've told me that hasn't helped in some way. Here's an idea of my project. I bought an 81 Corvette off ebay, chasis, and body, that's it! It's almost the same as one of your rollers. My plan is to finish it so that no one can tell it's electric like you've done with the 356's. There just isn't much room for cells. I'd like to have a 100 mile battery pack. Now I have to find the space to put enough cells to go 100 miles on a charge. This is where the "miles per cell" comes in. Do I have enough room in the car to put enough cells to go 100 miles on a charge? I'd rather not have cells in the seat next to me. Space is finite!

    ReplyDelete
  17. Glenn,

    All the information you need is in Jack's and my posts. You just have to decide what cell size you are going to use and then use 3.2V as the nominal voltage for a cell. Calculate the Wh in one cell and the Wh/mi you expect to get with the car you want to convert and then divide the Wh/cell by Wh/mi and you get miles/cell as your answer.

    We are just suggesting you work it the "normal" way where the total pack energy is considered. Determine what cell you are going to use, determine how many cells you can fit of that size, calculate the total pack energy and divide by the Wh/mi figure for your car and you have your range. If it is big enough great, if not, try a smaller cell size but more of them and see if you can get it to work. If not, compromise on something.

    For my Gizmo I wanted to keep my max current below 3C so I chose a 200Ah or 180Ah cell. I couldn't fit the proper number in my battery box so I "compromised" and went with 100Ah cells but buddy-paired them for my 200Ah pack. Pick your method and use it but when comparing cars it is useless to say something like 1.09 miles/cell. Use the kWh of the pack and the Wh/mi of the vehicle.

    ReplyDelete
  18. You're not getting what I'm saying. If I wanted to do an MG Midget, and I wanted it to go 100 miles on a charge, I'd have to put X amount of cells in it. For that car to be able to carry 2 people, and go 100 miles let's say 60 cells. There isn't much room in an MG. The cells have to be placed in the car so they don't over stress wherever it is that you are putting them. It would really suck if as you were going down the road one of your battery boxes fell out because of how or where you put the box. Over loading the car is hard on the suspension, tires, frame, and the drive train. It will also effect handling. It is anything but useless to determine miles per cell. If you want 100 miles on a charge, but your car won't hold enough cells to go that far, you're wasting your time. I'm sorry to say this, but there are way too many cells in the Cobra for it to handle even decent. The forces working on a car as it goes down the road are sometimes extreme. If cells are not secured properly, you might see your battery pack laying in the street in your rear view mirror. Jack, I think your Cobra will do 120 miles on a charge, but there better not be any cornering involved. I want my Vette to go 100 miles on a charge, but I still want it to handle. Will it handle as good as it did when it was new? I'm starting with the suspension first. Then I'll replace things like the gas tank with an equivalent weight in batteries. Big block springs in the front will carry more weight, so I can replace the engine with even more batteries. By my calculations I need 60 cells to go 100 miles. Placement is very important. If I can get 60 cells in my car without having to make some major changes to anything, and without loosing too much handling, great! In reality, I can make the car go 1000 miles on a charge, but that would be the ugliest car to ever roll down the road. I really do think all the cells you put in the Cobra are going to make that car "unfun" to drive, or at least not as fun as it could be. I'm sure you can get it to go, and go well, but I think it will handle like a mini van. I guess I'm talking about the physics of placement. Where you put a cell, how it's placement effects handling, and how much stress is that cell putting on where it's attached. I've been working on cars all my life. Whatever you do will effect everything else. I want my Vette to perform, not just go in a straight line. I'm not trying to figure Wh/m, I'm trying to figure cell placement.

    ReplyDelete
  19. You may be right Glenn. Obviously I think I can do it and have a very nice handling car. But I could be wrong.

    Tell me, how many of these electric cars have you done?

    What we were pointing out is that it is unusual to think of miles per cell. You normally convert it to watt hours and estimate watt-hours per mile. That will give you your total pack size.

    Then you have to break that down by what voltage you want, what current you want, and as Gizmo points out, what can fit where. 200Ah cells are nice, but they don't do curved bodies very well. 100 Ah cells are a little more "granular" with regards to how you fit them in, but there is a lot more strapping to deal with.

    And so it goes. If you have your kWh pack size, probably next is your voltage, which is determined largely by your motor/controller choice. At that point, you simply divide the kWh by voltage to get your current size. And then pick a cell close to that.

    Battery sizing, configuration, and placement are easily 50-60% of the total effort on converting a car. The rest is relatively easy.

    Jack RIckard

    ReplyDelete
  20. To answer your question, this is my first try at electric conversion. I like a good puzzle, and converting is a great one! Gizmo seems to be doing a lot of electrical math, and I'm trying to do placement math. No offence Gizmo, you've actually helped.

    When I look at a car I try to guess how far I can make that car go on electricity. After a certain amount of cells you start adversely affecting performance, handling, and ride. I hope I'm wrong about the Cobra. It would be a shame to see a Cobra handle like a mini van.(I consider mini vans an abomination in the car culture. They are usually driven by an obnoxious woman, with 2 even more obnoxious kids. No, not all women are obnoxious, just that the ones who are obnoxious seem to want to drive mini vans.) I wonder how many people I pissed off with that statement?

    50-60% huh? I think this is the most important part of the build.

    Have you beefed up the suspension on the Cobra? How much do the cells alone weigh? How much more will the car weigh than stock? I don't remember if you mentioned if the brakes have been upgraded. All that weight has to be set in motion, controlled, and stopped. A friend of mine once said, "Going is good, but stopping is better."

    ReplyDelete
  21. Glenn:

    We've covered all that in earlier videos. Now let me get this straight, you've never done an electric car, you don't know how much the cells weigh, you don't know how many cells we're using, you don't know how much we will weigh over stock (it's a kit car, there is no stock) you don't know anything about the brakes, and with all this information, experience and expertise, you're offering us a little condescending advice that you THINK we've got it all hosed up. And after Gizmo has gently pointed you in the direction of conventional engineering terms such as watt-hours and amp-hours, you are going to bless us with a new and better metric, miles per cell, never specifyng which of 200 different available cells.

    Every day is an education for me here at EVTV. I just never get enough of it.

    I DO hope we are able to construct a vehicle to your satisfaction anyway...

    Jack Rickard

    ReplyDelete
  22. I'm sorry, I didn't mean it to come across that way.

    ReplyDelete
  23. "Gizmo seems to be doing a lot of electrical math, and I'm trying to do placement math. No offence Gizmo, you've actually helped."

    No offense taken. I've been a HS math & science teacher for 23 years and now I work in a Juvenile Detention Center. Periodically my captive audience doesn't say nice things to me.

    The reality is that you can do placement math and then you have to do electrical math to figure out what you got. I went back and forth between the two just to figure out what to put in my relatively easy to do conversion from lead to LiFePO4. I did, however, start with my max battery current and divide by 3 since I wanted to stay under 3CA. This meant I had to go with a minimum pack of 167Ah. That gave me the 160, 180, or 200Ah sizes to use or a buddy pair of some smaller ones. I started with those sizes and found I couldn't get my minimum of 16 to fit so I went with pairs and found that pairs of 100Ah sizes would just fit without going over my $5000 budget. I then had to figure out how many kWh the pack was to get an idea of how much energy I was using so I didn't kill off my cells. With a 12.8kWh pack I knew that when a drive took me 9kWh from the wall I was still ok. I later got a CycleAnalyst from ebikes.ca and wished I had gotten one a long time ago!

    Anyway, have fun with your build.

    David D. Nelson

    ReplyDelete