Tuesday, February 7, 2012

This week the Brain is off to see his parents in Southern California. They are not as young and pretty as they once were and indeed Lou is facing open heart surgery. If you are accustomed to prayer you might put in a word.

Leaving me loose in the shop by myself. Actually, I enjoy solitude and particularly in the shop. If I set something down, I can kind of count on it being there when I come back. With a single other soul in the building it is always remarkable to me. I could build a trick double throw me down quick disconnect air powered overspeed protected refrigistastitator flavis waven that no one on the planet but myself even knew what it was, much less have a use for it, and lay it down on a bench immediately on assembly.

If I turn and refill a tea glass, when I come back 12 seconds later, the damn thing will have disappeared completely. A search ensues. EVERYONE in the shop SWEARS they haven't touched a thing. 45 minutes later I find it on the SINK in the bathroom.

OH YEAH. Is THAT what that was? I was wondering? So I took it into the bathroom. NOW I know what you were talking about.

I actually had an incident this week, again with an object so nondescript NO ONE could have a use for it EVEN ME, in theory, but I did. It had been moved and when I complained I was told in no uncertain terms by the only other person in the shop that they hadn't done it. When I noted that there were only two of us, he became THOROUGHLY incensed that I would "call him a liar" and threw a total fit. Without missing a beat, he then noted that he had only moved it for "safety reasons" and because it was in his way. ???? So he was INCENSED that I would call him a liar, which I never did, and then confessed to lying about it, which he did.

I apologized profusely of course and noted that I wouldn't' for anything in the world have him offended in any way. And indeed to prevent any POSSIBILITY of a future occurrence, I invited him to leave the shop and our employ as quickly as he could assemble his train.

Alzheimer's is not precisely a disease in my family. Picture it more as a tradition. So I kind of feel like Helen Keller after her parents have rearranged the furniture for the twelfth time.

So I kind of had a good time this week, once Rod had been promoted to whatever he's doing now, and I'm rattling around the shop by myself. It has been wonderful. It's true I am not as good as those guys at fabrication and just being able physically to reach things and lift things and so forth. But I actually got quite a bit done, albeit in piddling ways.

Not really shown in this weeks' episode. I got some 4 AWG cables run from the Manzanita to some terminals under the truck tied in with the J1772 plug wires. This gives us much stronger cables to carry the current for our PFC-75. I'm so accustomed to chargers that put out 20 amps, that having a 75 amp monster is a constant reminder. The little 10 gage wires I had on it, normally overkill for charger duties, were getting very warm.

As many know, the Manzanita is NOT my favorite charger. But this particular unit we spent about $4500 on several years ago when it first came out and it is capable of 75 amps at up to 400 volts. That's pretty stout.
Despite Manzanita's assurances that they are all fully capable of 75 amps, we get 68 amps or so into this one from the wall. That's still pretty stout. And with a 400Ah pack, we need all we can get. At that high current level, it will still take six hours to charge this 76 kWh pack.

I've done something kind of goofy here and may pay the price. We've mounted the charger, and a DC-DC converter, on TOP of the polycarbonate lid of the pack. So now to get to my pack, I have to remove a lid with a n ever increasing array of wires and stuff on top of it. I added the little voltmeter we talked about recently from LightObject to it for example, so I can see the pack voltage at a glance. Turns out this little 5740TV voltmeter is really pretty accurate. I just love these things. I found an older JLD404 AH counter from these guys and got it set up in the lab and working and it is pretty nice. Voltage, current, hours, and AH all in one little device. Problem is, they don't sell it anymore. They sell a JLD404, but it does something else and no longer does AH. But we're talking to them about getting them again.

I also tied the charger to the pack INSIDE the box. This poses a little problem I hadn't thought through. I need a hall effect device and a shunt device for current measurement and in order to work, it has to be inside the loop of both the controller AND the charger. In this case, that now means inside the battery box. So my electric car is moving one piece at a time into the battery box in back. And I'm not sure I can stem the flow of parts into the area.

We did bleed the pack down using our Aurora Inverter - basically running the shop off of the 76 kWh pack for two days until the pack was drained. Then I used an old Thundersky 30 A charger I have laying around with big jumper cable clips on it, to individually charge each cell up to about 2.80v - then letting them fall back to about 2.75. We did this to all 57 cells until they all read 2.75 +- 0.05v. At that point, we hooked our Manzanita back up and charged it to 205 volts. This will be about 3.6v per cell and is high enough for my purposes. Again, I like to undercharge a bit and we have a huge pack here.

As noted in the video, we did build a little heater for the Vantage GreenVan. Brain had reported in from California that he had visited HPEVS and they were working with Vantage GreenVan on a LiFePo4 version of the van. We've been enjoying one for some time. But it has a little diesel heater that I'm scared to death to even turn on. So I wired my daughter up a little electric heater using two of the by now familiar PTC heater elements and a Kilovac relay.

And as promised, I show the damage done to our A123 module attempts and even pry open one of the cells to look at the very different looking cathode on this cell and talk a little about the patent disputes over this cathode.

I'm still mystified by our losses there. This week I've made TWO little modules with 6 cells in parallel more like our prismatics. They are champions. About 117 Ah - full spec 19.6 Ah per cell.

I DID notice something that is a little problematical in an electric car and kind of hard to test. If I fully discharge a set, and then immediately full charge it, it looks like ti has lost capacity and reaches a high voltage quite prematurely. If I let it then set overnight, I can add another 10 or 15 Ah to the cell the next day with no harm at all. This is NOT like our existing cells. It is quite strange behavior. And it might explain some of our damages in the modules. I was running it HARD to discharge it and pretty hard to charge it and doing it quite back to back based pretty much on what it OUGHT to take.

Of course, I don't want to have to leave my car overnight before fully charging it? That makes no sense.

IN any event, this week I'm working on a rubber mold to make cells that look like CALB 180Ah cells but of smaller size and HIGHER power of course. A 120Ah cell would have a current capability from our tests of about
2750 amperes. Imagine driving the Elescalade on 57 of those. 57 x 6 x $26.60 = $9063. That might seem steep but it is some less than the $25,000 we have in the back now. Of course, again that would be a scant 23kWh pack and we would probably be limited to 25-30 miles on such a pack. But it drives home the point of those high power cells, we could still drive the two Soliton1's to their limit easily with such a pack or an even smaller one of 90Ah for example. And so we can use less expensive battery packs for shorter ranges. Not my style, but an option.

The problem of course is that it would take quite a bit of "sweat equity" to convert boxes of individual A123 cells to our prismatics - including hardware, resin, and so forth. So long run, I'm not sure what would be saved. But many of our viewers aren't concerned about the long run. They're concerned with limiting expense in the right now. If they'll settle for less range, these cells appear to be an option.

The cells from VictPower seem to be testing much better than the cells received from OSN power. We're clearly up over 19 Ah with these. We are charging to 3.65 volts and discharging to 2.50. There is indeed some power between 2.50 and 2.00. We're content to let it remain there.

Jack Rickard


  1. Thanks for the show again Jack. Noted what you said about Lou: I have three prayers to a full stop in the last 90 days (FAR 121.339), so that's a runner.

    1. Kewel. Then you can take passengers too. Put in a word for me as well.



  2. Jack,
    I would also be interested in getting a JLD404 with AH capability. Let me know if and when you work things out.
    Thanks You for all your efforts!

    1. I've been playing with this one I have from a previous order I never really looked at. In spite of having no manual, it does use the same codes for the menus, it just has slightly different menus. I've fooled around and gotten it working.

      Voltage only has 100v and 500v resolutions so one digit (i.e. 3.4v) is as good as it gets. But it displays AH, hours, amps, and volts. If you hold down SET for three seconds, it goes into a cycle mode and displays each in turn.

      The amp meter and AH meter are surprisingly accurate for the price. The digits are large and bright. This could be on a dashboard and easily look better than the cheesy little Xantrex. The other JLD404's are about $35 while a Xantrex is more like $275. It has two alarm relays I haven't mentioned,, just like the AC/DC Voltmeter. ANd yes, you can alarm on AH.

      But they do not exist. I'm pestering Lightobject to bring them back.

      Jack Rickard

  3. Hi Jack

    Is It possible to buy one of the heating element from you ?

    Best regards Allan

    1. It would be time consuming to test them before shipping. You can get these from Randy HOlmquist at Canadian EV. Or the least expensive way is to actually buy the $18 heater at Walmart and take it apart yourself.


  4. Jack:
    Thanks for sharing your practical knowledge of Li batteries. I have learned what not to do as well as what to do from your experiments.

    What do you know about NiZn batteries? Do you think they could be used for low-range EVs?

    One company, that makes these batteries, claim they are a better product than Pb cells for microhybrids(start/stop technology), i.e., lighter, higher-power, longer cycle time, etc.


    1. I really have zero experience with NiZn batteries so what I think is largely immaterial. I guess I think a lemon a nail, and a penny are better than Pb batteries, if that gives you any indication of my sincere affection for lead acid cells. So a claim that NiZn cells offer advantages over Pb would neither alarm nor surprise.


  5. Thanks for the show... I am learning a lot each week. I am currently shopping for a car to convert. My choices seem to keep coming back to:

    Classic VW Beetle
    VW Karmann Ghia
    Ford Model A (There is just something about an electric Model A I Like )

    If any of you guys have any thoughts on these I would love to hear it...

    I think I will sell my 2008 Corvette to finance the operation and keep the wife happy. She won't let me touch her 2003 Beetle. I got caught with a tape measure under the hood and have been banned from the upper garage (Hers). I am getting bored with the 436hp the Corvette has that usually ends badly with a turbo or supercharger and broken drive line components. I think going electric makes more sense this time...

    Anyway, thanks for the show...!

    1. Fee, Fi, fo, fum, I smell the blood of a converted horsepower worshiper.

      Excellent news Jeff. Store bought Corvette's are for kids.

      I promise you will find both the process and the result of converting a car to Lithium electric drive infinitely more rewarding in all respects and a pleasure to drive for many years to boot.

      Beetles and Karman Ghias are done to death. THere's a reason for that. They are both just really THAT good converted to electric. And if you don't get sidetracked into a bunch of rust removal, very easy to do really.

      A 2003 VW beetle on the other hand, has environmental issues and opportunies for air conditioning, heat, etc that lead to a true all weather car. Kind of depends on just how bit a boy are you.

      In my world, the wife has already laid the ground rules. I would sell the vette, let her KEEP her old oil burner, pick up a little better 2004 model of exactly the same car, only slightly better, and convert THAT to a totally silent, gasoline free version with all the amenities.

      With a rolloff at the end and certain side wagers could have me feeling lucky before it was all over.

      But I'm hard on wives that way....

      Jack RIckard

    2. I hear you....

      My only worry with a newer car is the DMV here in GA. I live in one of the Clean Air Force counties. The ultimate irony is that they make it nearly impossible to get a tag on a car that is converted to run on electricity. The problem is that all cars that are newer than 1987 require an emission sticker each year. You have to get an exemption for an alternative fuel car to get you tag. That process is almost impossible to get through....

      If the car is older than 1987 there are no such issues.

  6. This comment has been removed by the author.

  7. Nice show. Very interesting seeing inside the A123 cells. And best wishes to Brian's dad for a speedy recovery.

    Well, my Winston 400Ah cells from the Czech Republic turned up on Monday and in the same sturdy wooden crates that you were doing your show on in the lab :D

    Just like in your previous shows, these came out of the box to within just 3mV of each other at 3.307V to 3.310V. They were all quite new with October 2011 date codes.

    I'm using the comedy copper laminated straps as the pack will just sit in my living room behind the sofa for the next 10 years (I hope) and so vibration and flexing stress shouldn't pose as much as issue as it does in the Escalade. My biggest load is about 120A and so thermal cycling isn't an issue either!

    I've seen you testing batts for a few times and you mention charging up to 3.65V at 0.05CA as the end point. But did this apply to the first charge too? The supplier (and Winston Battery, through a note on formation charging) say that these prismatics need to be charged to between 4.00V and 4.20V individually for the first charge ONLY before normal use.

    What did you do with the 57 400Ah cells on the Escalade? I saw/read you charge the pack as a whole set and then discharge it on the inverter and then bottom balance them individually. Did you do the first charge with all of them in series and if so, did you charge to 228V or the usual 208V?

    My problem is I've got a solar charger that can do 24-32V at 75A but only a programmable bench supply (to do CC-CV charging) that can only do 5A. That will take a couple of days per cell to do the initial charge if I have to do each cell individually. It can automatically stop and disconnect when it gets to a set point (4.05V) but my max charge current is already only 0.0125CA, much lower than the 0.05CA (20A) end point current you're using.

    Apart from the very long time it takes to get to the end point when charging at 5A, is a very slow charge to 4.05V actually a problem for the formation charge? Can it cause damage? From the charge curve for these things, I'm assuming it will take like 98% of the time to get to 3.65V and then suddenly (in a few minutes?) shoot up to 4.05V.

    My first cell has been on charge for over 12 hours and is only at 3.40V... This could be a long week!

    1. Sounds like you're a wee bit hampered by available equipment.

      First, the formation charge. Yes, one should be performed .... at the factory. You shouldn't have to perform one at all. If you do, it is news to me. We've never done them.

      Elescalade. Pack of 57 w as charged to 208 volts with constant voltage taper to about 10 amps. The Manzanita charger is really not cool. It tapers to about 1.5 amps and it NEVER quits. The documentation is absurdly bad and also technically incorrect. This charger never does properly terminate and represents a fire hazard as it sits frankly. The good news it puts out 75 amperes.

      We used the Aurora inverters to bleed the pack down at about 40 amps until we were down to 2.5 to 2.9 volts. We then used an old Thundersky 30-160v 32 amp charger with cable clamps to individually hand charge each cell UP to 2.80vdc. When we removed, they would typically fall to 2.75v. We bottom balanced the entire pack there.

      We then subsequently charged them to 205.5v using the Manzanita. Again, it gets down to about 1.5 amps, but never does actually terminate properly.

      You have a similar problem. With such a low charge level, I would charge to 3.600 and as soon as it hits that, terminate completely.

      I would never let those very expensive cells see 3.8v, much less 4.00v.

      Jack Rickard

      You are quite correct. IT takes like 99.5% of the time to get to 3.65 v and then suddenly, and literally in a minute or so, shoots up to over 4.00v.

      With such a wealth of 76kWh, we're charging to 3.6v x 57 cells or 205v volts. It will fall back to a fully charged 189.8v or pretty close thereby.

  8. Jack,

    Another good show. Best wishes to Brian's Father.

    On your new terminal block for the A123 pouch cells I noticed that you only put a Nordlock washer under the nut side of the clamping bolt. Thinking about this and then looking at Nordlock's recommendations you need to have one under the bolt head as well as the nut because there is nothing to keep the bolt from turning inside the nut without it.


    1. Doug:

      That's not entirely correct. The sides of this washer do grip the opposing faces, but that's not really what "locks" the device. It puts pressure on the threads of the bolt. Any attempt to back the nut off increases that pressure.

      But sure, you could have one under the bolt head as well.


    2. Doug:

      I reexamined our design and the NordLock instructions that actually come with the washers. I'm going to reverse myself and say that you caught us in a good one. On thru bolt applications such as this, we SHOULD have a nordlock on both sides. Some days I wish I were selling washers. But indeed, we do probably need them on both sides, just as you said.

      Good catch. We'll put them on there.

      Once they are poured in resin, I doubt it will matter much. But if that's the case, why buy the washers in the first place. We'll put one pair under the bolt head, and one pair under the nut.

      Jack Rickard

    3. Jack,
      As much as I would like to keep Nord-Loc in business, two sets of washers seem to be a bit much. How about using Loctite 263 without the washers? Looked up the specs on various Loctite products and 263 seemed to be best for this task. However I also noticed that SST bolts do not make as strong a bond as more reactive metals like steel or zinc compound plated steel. Locktite reacts with the surface chemistry of the hardware to produce the bond. The nickel/chrome in the SST inhibits this reaction. Using plain steel hardware would reduce the cost per cell a little bit.


  9. @Outtasight - you can't change the charge rate of the formula as the apparent voltage is a function of the charge rate.

    Charging a 400ah @ 5ah is essentially trickle charging, and LifeP04 cells don't need to be trickle charged.

    3.4V static is fully charged when rested (12 hours or more). I would stop there if you are only charging at 5ah and see what the resting voltage is. A resting voltage of 3.33 to 3.35 is a nice safe target.

    Do not charge to 4 volts at 5ah unless you want to see what happens as an experiment. Please capture on video :)

    Anyone chime in if I've made any mistakes in this info, as this info is from being a weekly viewer. No EV yet for me.

  10. This comment has been removed by the author.

  11. @Outtasight

    I'll be eyeing your blog with interest. I've been pondering 400Ah cells for a solar array on a Tiny House (or box truck conversion).

    - Doc

    1. I actually cannot imagine going to the expense of Lithium cells for a solar installation. I probably will, but only because I'll have a lot of remaindered or used lithium cells laying around.

      Lithium ion offers some serious advantages in size, weight and life cycle. But size and weight don' t matter very much for solar installations. The difference in cost is pretty substantial for cycle life alone.

      Just my two cents...

      Jack Rickard

    2. Lead work fine for home installations normally. The typical discharge rate is C/20 or lower so you get full capacity out of them and normally you never take them down more than 20-30%. They tend to last up to 10 years. However all this changes if you decide to use them to charge your electric car. The currents will be quite a lot higher and the Peukert effect kicks in. You can overcome this by vastly increasing the size of the battery. It gets lots worse if you want to do 3C Level 3 charging. I don't have enough space for the lead acid battery bank to support running my house for three days and doing a L3 charge on my car three days in a row. Lithium gets rid of all the maintainance issues with a lead acid battery as well.

    3. Obviously I'm a fan of LiFePo4 cells. The advantages aren't lost on me. But the difference in price is pretty huge. For batteries sitting on the ground, where I don't care what they weigh and where you typically get an 8 to 10 year life anyway, the question becomes why.

      In fact, if I were doing a mother bank to recharge cars at Level III, I would probably still use lead AGMs.


  12. Jack, much appreciated for your detailed response and advice.

    I've had 3 sets of lead acid and they just don't perform well in heavy daily cycling. Weight IS an issue, as my old bank weighed 350kg and so could only be on the concrete ground floor in the living room. The roof space wooden floor (ideally close to the PV on the roof) can't take the weight - it's just ceiling rafters and chipboard floor. A 111kg lithium pack is do-able without resorting to strengthening the floor and it's easy to lift the cells up the loft ladder into the roof space.

    The GWL guys posted a photocopy of a Thundersky note (in Chinese and English) on their blog making it clear that initial formation charging to 4.2V BEFORE first discharge is required. In another post the supplier said 4.0-4.2 was ok but it must be at least 4.0V.


    See the initial charging FAQ link on the page.

    On the other hand, you've had the exact same cells but never charged to more than 3.65V at 0.05CA cut-off and never had a problem... That counts for a lot as it is "real world" experience. Maybe moreso than whatever the supplier or even Winston say!

    So I've stopped charging for now (it was still only at 3.41V after 22 hours!) until I get a better handle on this and the slow charging connundrum. Better safe than blown up :D

    Incidentally, they've just posted a report from the Faculty of Transportation Sciences at Prague University that shows no degradation in charge / discharge performance of a 90Ah Winston cell after 13,000 shallow cycles (10% DOD). Test was charging to 4.00V at 1.5C and 0.03CA cut-off and then discharging at 1.5C for 250 seconds on a 16 minute charge/rest/discharge/rest automated loop 24 hours a day for 144 days.


    1. We have ordered many cells from both Thundersky and China Aviation Lithium Battery Company and never received this notice. I've asked everyone involved and no one seems to know the source of this information.

      There IS a formative charge. It is of course performed at the factory. I can find no mention of any necessity to EVER fully charge a LiFePo4 cell subsequent to this in any of the literature. No mention of "repeating" the formative charge.

      I think we've had some pretty excellent results over a number of years while basically abusing our cells by "not" having a BMS, not top balancing, etc. and a key component of that good result is we undercharge the cells.

      Thundersky, meanwhile recommended 4.2v charge level. Later 4.0 charge level. Most recently 3.8v charge level. We've been using 3.65 v for several years. I did have a small THundersky branded charger purporting to be for 24 cells which was hard wired to 87v. So in practice they were doing the same thing.

      Your mileage may vary. If you want to charge each of them right up to the brim at 4.2v, they are indeed your cells.

      Jack Rickard

  13. Thanks again Jack.

    I agree with you fully and plan to only charge to 3.55V in normal service. Just flummoxed by the insistance of the supplier that some kind of special initial charge is needed.

    So, following your tried and tested procedure, I can forget about initial charging each cell individually and just wire up the bank into the solar charger (minus the cell I charged a bit on the bench yesterday and so put out of balance with the rest of the cells). When that lot gets to 3.60Vpc (25.2V) I'll add in the no.8 cell (will have to finish up that one on the bench). Then use the 3kW inverter to drain the whole 8S pack down to ~2.60Vpc (20.8V) and do your trick of manually charging each cell UP to 2.80V (with the bench supply) to bottom balance the pack at 2.75Vpc.

    Then for added safety, I'll reprogramme the solar charger to quit at maybe 3.55Vpc (28.4V) on its "absorption" preset after maybe 5 minutes (it has no current trigger to end charging, only a timer). Then it can enter "float" mode at some low value like 3.32Vpc (26.56V) so that the charger is essentially "off" until the inverter drains the pack down below the pack's "rested full" voltage level and then the solar charger will only try to hold that level. Of course, it has to be sunny enough for it to even try to "float" the pack AND drive the inverter loads at the same time.

    At least lithium cells actually enjoy being chronically undercharged for years... UNLIKE my nearly dead 2 year old AGM and gel packs. Like in EVs, you don't get 10 years of daily cyclic life from lead solar batteries unless you only plan to do 20% DOD cycles. In that case an equivalent lead pack would have to be 1600Ah to match the Winston 400Ah pack that can do 80% DOD cycles every day for 10 years.

  14. @Outtasight:

    Have you considered a strategy for keeping this big cells alive long-term that maintains them around the 50% SOC level?

    It seems the cells are shipped at about 50% SOC, and the information Jack has shared indicates that's where the cells live the longest. If your solar bank is usually topped off, you may well be significantly shortening their life compared to, say a bigger bank that is nominally kept at 50% SOC, and only topped off in advance of an anticipated discharge into a vehicle pack.

    I predict that the first wave of "smart" EV chargers will adopt a similar strategy for large format prismatic cells. Having a programmable charger doesn't really make it smart. A smart charger would track Ah trending data, combine that with a robust user notification and control scheme (integrated to desktop and phone) and thereby do several things automatically that might significantly impact cell life by maximizing the time the pack is at 50% SOC:

    1. Initiate charge shortly before anticipated use, so that on a day selected for full capacity, the charger starts at 3 or 4 am and not at 6pm the night before. That reduces the time the pack spends at a high SOC.

    2. Learn the weekly usage pattern, and default to setting the car up every morning with the SOC necessary to complete the anticipated run and bring it home with the appropriate reserve, all focused around that 1/2 charge level. Then, upon plugging it in at dinner time, the charger would run the pack up to 50% SOC and leave it there, either waiting for user command for the next charge, or holding until the default charge the next morning.

    That would be a SMART charger, and over a long period of time, like 10 years, it could really extend the life and capacity maintenance of an AV battery pack.

    I wonder how significant keeping the pack at the 1/2 charge level is to cell life. If it is really important, that suggests a strategy of installing largest practical pack, and a corresponding charge/discharge cycle protocol of shooting for the shallowest cycles, centered around optimizing a nominal 50% SOC level, for longest battery pack life.

    I'm thinking it would be even more useful to do this with a stationary mother bank, which if suitably over-dimensioned and maintained mostly half-charged, might become a lifetime proposition.

    Just typing myself smart...


    1. Maybe the battery installation should be considered as a twin pack. The first one would dance up and down within the desired voltage levels and capacity, leaning on the back-up pack, where eventual overpowering can be drained to.

  15. Video Archive Trouble

    I was going through reconciling my stored EVTV videos making sure I had them all.
    As I was going through the 2009 archives I ran across several anomalies.

    First, there are videos titled "waterheater - computer" on September 14 and
    October 16 2009. They appear to be the same video. What is the correct date?

    Second, I have videos titled "gastankbox-1280", "firechief1080p -
    Computer", "thunderskybattery-1280" that don't appear on the archive list.
    What are the dates for these videos?

    Third, I'm unable to download "exhaustandtank - Computer.m4v" on Jul 26 2009
    and "driveshaftremoval - Computer.m4v" on Jul 28 2009. I get 405 access denied
    errors which I suspect is caused by the fact that the video files are not there.

    Can you please check into this and let me know the correct information.

    Thank you

  16. @TomA

    My off grid solar batteries are lucky if they spend much of their lives at more than 80% charged (averaged over 24 hrs). At best, it's 95% for a couple of hours in the afternoon on REALLY sunny days and then they're discharging again. Huge amounts of solar power are wasted because the lead acids can't accept the charge and sit for 30% of the charge cycle in the CV current taper mode. Lithiums should spend much longer in the CC mode as they can suck up incredible rates of charge. Charging in the CC mode means the pack is absorbing everything the PV array can throw at it. In the CV mode, the charger is throttling the power and so wasting it... "Use it or lose it".

    I have more trouble with the battery bank bumping along the bottom in the Winter. The almost permanent state of partial charge is a total killer for lead acids, but the Winstons should be fine with it.

  17. @TomA

    I think the 50% DOD cells living longest is a much bigger deal with the Tesla-style lithium cobalt 18650's. Tesla was so concerned about the issue, they studied ways to encourage owners to store their cars with partially full batteries.

    I don't think we have any information regarding LiFePO4 prismatics having a significantly shorter life at 80% or even 95% full compared to 50%. Given the information (or lack thereof), it seems like way too much expense and effort to shoot for a 50% full LiFePO4 prismatic solar battery pack.

  18. Not a lot of sun at 3 or 4 am. At least where I live.

  19. Well, yes and no.

    If it were my solar battery pack, I probably wouldn't let it go way down to 20% or even 35% SOC in the winter. There's no reason to do that, since I probably couldn't generate enough daytime energy from there to go into CV charging anyway. Once I figured out the most likely strongest energy day, I'd subtract that from full the full capacity of the bank, and then try to establish a winter, or maybe even a monthly, SOC (Ah discharge) floor at that point.

    Of course, hitting that point reliably and terminating the discharge on cue may be an elusively tricky procedure in a chronic and variable partial-cycle environment, but it costs nothing to play with it, and its likely of some benefit, particularly if it keeps the batteries off the SOC floor all winter long...

  20. Weight, form-factor and cycle-life are the primary concerns I have in a PV off-grid setup for a Tiny House.

    Weight concerns should be obvious when you build a small house on a trailer (or in my case, box truck)...

    Form-factor, I'd kinda like to hide the cells in a wall to take up minimum space, since space in a Tiny House is at a premium...

    Cycle-life and discharge rate are of concern since I'm planning to run a small air conditioner off the system (in addition to lighting and laptop power). Not a big deal in the winter since heat will be provided by propane, but in the summer the pack could see some serious DoD and current draw from the A/C. While the Tiny House will be very well insulated, this is still Texas.

    Not to mention this is an enclosed space. Venting hydrogen would be a pretty serious safety concern using lead-acid. Better to avoid it all together.

    Of course the unknown factor of all of this is how to use PV panels safely with a lithium pack. I don't know of a charge controller designed for use with lithium batteries. This might be a job for that programmable volt meter and a couple of contactors (one to disconnect the panels when full, one to cut off the load when empty).

    - Doc

    1. @TomA
      "Once I figured out the most likely strongest energy day..."

      Dude, the Met Office has been trying to do that for years and hasn't got it right yet :D

      I do look at the 12 and 24 hour weather forcasts in the evening to plan / moderate my battery consumption, but sometimes they say "blue skies tomorrow" and I go for it with the 28" TV and then it snows for a week. The UK weather is like that.

      At least with lithium cells, I have the luxury of getting the "power forecast" severely wrong and not trashing the bank from being stuck at 20% state of charge for days. In practice, real off gridders would fire up their diesel generator to rescue the bank. $20 in diesel to save a $5000 battery bank from rotting.

      Yep, charging lithiums with a PbA solar charger is a bit of a challenge but I've got these programmable Morningstar MPPT chargers that (I hope) will be flexible enough to do a reasonable job - with some external bodging (hence my blog's name!).

      A/C is a BIG ASK for off grid PV systems. That's huge energy demand over 24 hours and you'll be mostly running it on battery as the peak PV output is only for about 4 hours of a day. I can run my fridge freezer on PV from March to August but that's only a 1.4kWh 24hr load. Again, I'd say get a diesel generator to run the A/C in the night and cloudy weather. Or be prepared to shell out for an 800Ah 24V lithium battery and a MASSIVE PV array. It ain't gonna be a "Tiny House" if it's got a 6kW array on the roof!

      I'll be posting on my own blog over the next few days (rather than continue to hijack this one!). So we can take all this non-EV lithium chatter over there...


      Thanks again Jack for your patience and very trusted advice. :D

  21. I know Jack hates them, but sometimes I wish EVTV had forums. Just for eventualities like this.

    One last reply before I shut up... The A/C units I've been looking at are only 500W and I only plan on running it during sleep hours. I figure 15 minutes of runtime an hour for 8 hours would be about 1KWh. Well within the capacity of a 1KW (5x200W) PV array in the Texas summer sun. Just need a good pack to store power in.

    - Doc

  22. Jack,

    In setting a car up for J1772 charging how many signal and associated wires need to be run to the charger or internal circuitry inside the car and what guage should they be over a 6 foot length? I'm not talking about the wires that carry the actual charging current. I noticed in a previous show you had changed things a little from your previous setup.

    1. If your charger is actually capable of receiving and decoding the J1772 pilot signal you only need 1 wire about 22AWG. If you're using the Active Vehicle Control module by Modular EV Power no other wires between the charger and the J1772 connector are necessary. The relay on the AVC1 is intended to disable the drive but is not required. The proximity and the pilot wires don't have to go any farther than the AVC1 module 22AWG is plenty.

      In a previous video I noted that Jack and Lee were a little confused about the EVSE proximity signal. The proximity signal can be generated entirely in the handle so that only 4 wires have to go between the plug and the EVSE. That open-EVSE module Lee was trying to use doesn't even have a connection for the proximity.

    2. Don:

      None. But we use the Modular EV Power AVC1 module and so we typically run three light gage wires from the receptacle to the board, 1. pilot signal, 2. proximity signal, 3. neutral.

      22 gage is fine. None go to the charger. We have yet to have a charger that can actually make use of any of these signals.

      Jack Rickard

  23. Here is an article discussing solar storage comparing LiFePO4 to Lead acid.



    "Lead acid batteries are traditionally the energy storage device used for most photovoltaic systems. However, as an energy storage device, lithium batteries, especially the LiFePO4 batteries we used, have more favourable characteristics."

    Yue adds, "the research showed that the lithium battery has an energy efficiency of 95 per cent whereas the lead-acid batteries commonly used today only have around 80 per cent. The weight of the lithium batteries is lower and they have a longer life span than the lead-acid batteries reaching up to 1,600 charge/discharge cycles, meaning they would need to be replaced less frequently."

    So one must consider charge efficiency when storing solar energy, or one would need more solar to have the same useful energy. Kind of like it is smarter to change to compact fluorescents to reduce your load than adding solar cells and storage.

    Yes, cost is still the issue, but maybe not as big as what we thought.

    I guess that charge efficiency is something to consider is electric cars too.


  24. Charge efficiency is certainly something to consider in electric cars, and it is something to consider in solar installations too.

    The advantages of lithium are well documented. But I've had solar systems and actually stored the batteries under the redwood deck behind the house. Space and weight just weren't a problem. Actually charge efficiency wasn't either. They were all lead. They all died within 10 years which is a bummer as I had bought 50 kWh of lead.

    But 50 kWh of lithium is$20,000 or so. I just don't see it.

    I guess I think the way I'll get lithium cells in solar is just retired cells that didn't make capacity or were weeded out of good packs.


  25. Jack, I just watched your video on making A123 batteries from A123 cells. I see that you had a time compressing the pouches before pouring the resin. What do you think about inserting the assembled cells in to a polyolefin heat shrink pouch and then shrinking the polyolefin pouch? They're available on ebay with an 8” x 12” bag probably about right for your 6 cell pack.(http://www.ebay.com/itm/100-8-X12-Polyolefin-POF-Shrink-Wrap-Bags-For-CDs-DVDs-etc-Food-Safe-SALE-/110824242712?pt=LH_DefaultDomain_0&hash=item19cda3be18).

    Fold the open end down to adjust for the length you need and then heat shrink tight.


    1. Err, the issue isn't precisely compression and I don't need shrink wrap, although I have bought some and have it and assumed I would use it.

      With six cells on these 3/4 terminals they really don't spread out much. If I align with the tops of the tabs they really hang without strain.

      No that isn't the problem. The problem is if I lay them down flat in the mold, I do not get the needed space for resin beneath them. Ergo some little plastic feet to stand it off. Same on the plexiglass side. I just need to make sure I have a gap between the cells and the module wall for resin to fill in.


  26. 2010 Ranger,

    I was considering just that but did not find a suitable source for shrink wrap. You may not need much but just enough to make like a narrow band around the cells to hold them while molding them. Excellent.

    Thanks for the link.

    Pete :)

  27. 2010 Ranger,

    I was also thinking of using some scotch tape to hold as well. A few wraps near the base to hold them would be plenty as well. Easy to do. No heating required. Even clear package wrapping tape would work too. Just enough to hold the ends.

    Pete :)

  28. I'm still wondering why this* Cylon slick battery module isn't being embraced in Cape? Obvious flaws? Plexiglass allergy? And maybe there is some good news on the Alzheimer's front: http://medicalxpress.com/news/2012-02-fda-approved-drug-rapidly-amyloid-brain.html

    *[apologies for the longest url I've encountered] data:image/jpeg;base64,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

  29. Perin du Bulu,

    The longest URL and tiniest picture to go along with it.

    Got a magnifying glass?

    How about a tiny URL for the Tiny Photo:


  30. GreenEV, I didn't think of the tinyurl. I don't need any copyright headaches, or I'd capture a frame from this video http://www.youtube.com/watch?v=_wZ34syIEMU

    1. FWIW:


    2. Martin, Your link makes no sense to me as to why EVTV is not using a future perfect looking module like the one in the video.

  31. Perin,
    The Cyclon "J" cell for instance (12AH), weighs in at 0.84Kg. They have long tables of efficiencies at various "C" rates. They are better than lead but inferior to all Li cells.
    A big thanks to Team7 for showing us how cheap A123's can be. and I thought I was getting a good deal. now its back to a bit of haggling.

    Does anyone know if prices for shipping & handling vary across different sellers?


  32. AndyJ, Cylons and Cylon design: