PSI -- LiFePo4 -- Testing.. 1 ... 2... Testing..

Actually, if you are talking about the LiFeBatt VMS boards, they do not use FETs for LVC. All they do is turn on a TTL-level signal (and an orange LED...) if any one cell goes too low. In a larger EV application, all of the VMS boards, in all the HPS packs, are daisy-chained together, and then connected to a vehicle computer system. I notice now that Dave, at Electricrider.com is now going to start offering LiFeBatt-provided 3620 and 4820 HPS packs, to replace the Eonyx packs they were trying to do using various Chinese offerings. PSI/Lifebatt is providing a small module that they will integrate in with their Crystalyte controllers, which inverts the TTL LVC signal, so it can be used to connect to the same ebrake signal line in the controller that we use with our LVC/BMS boards.

I have two 12-cell LiFeBatt/PSI-based packs that I use in a 72V/10Ah configuration on my wife's Townie, which routinely gets run down to LVC cutoff, and these cells seem to be as strong today as they ever were. They don't yet have more than a few hundred cycles on them, but I've seen no ill effects of using them this way. Basically, if we stop right when the first LVC "hit" happens, which is usually while under full throttle, going up a hill, the cells will still have about 10% left, and will take about 9.0-9.1A back in. If you back off the throttle, to about half, you can get another couple miles before the cutouts happen even with the slightest throttle. At that point, the cells will usually take 9.9-10.1Ah back in. I've only drained them down this far about 10-12 times. Most of the time we stop right after the first hit, or just before it hits. Anyway, like I said, they still appear to be going strong.

-- Gary
 
That's very cool. Lego blocks for batts. :mrgreen: Seriously, this will make assembling packs for these a "snap". It looks like they would go together like Anderson PowerPoles, so you can make pretty much any size configuration you need.

-- Gary
 
Gary,
I usual excellent post, thanks for sharing experience.
What I found amazing with these cells how they "recover" .
When you draw less than 10A they seems to last and last so I am not surprised that in your application you can still milk like 1Ah more after first LVC cutout.
So what circuit do you exactely usefor LVC?
MC
 
Ypedal said:
Something i received from Raymond a few weeks ago, should be available mid october, i am ordering a few hundred of them along with a few more goodies..

I like those clip-together jobbers for the Lifebatt/PSI batteries.

Are you going to give us a price breakdown when these become available?

Or is PSI going to sell them with thier batteries.
 
I like the concept, looks good on paper that's for sure.

IF they work as i hope they will, it would be wise for anyone ordering cells to get them unless you have a plan already in place..

Like all things, sometimes reality and concept colide.. i'll be sure to let you all know what i think when i get them !! :wink:
 
miro13car said:
Gary,
I usual excellent post, thanks for sharing experience.
What I found amazing with these cells how they "recover" .
When you draw less than 10A they seems to last and last so I am not surprised that in your application you can still milk like 1Ah more after first LVC cutout.
So what circuit do you exactely usefor LVC?
MC

It's the same one we've used in all the LVC and BMS circuits. Basically a 2.1V version of the TC54 voltage detector that drives an optocoupler via an 820 ohm resistor. When the voltage dips below 2.1V, the TC54 output goes low, which turns on the opto output.

-- Gary
 
Well, since I have both the same PSI/LiFeBatt-based packs, and a123s, I can tell you the PSIs are a close second to the a123s, in terms of performance on most ebike setups. For max currents up to 60-70A, the voltage drop, under load is pretty much the same as a typical "4p" 9.2Ah a123 pack, which is quite good. The differences don't really come out until you get up over 90A. For example, I was able to pull 104A on my last 4110-mod'd Clyte controller and a 5304 on my Townie, with a 24s4p a123 pack, but the LiFeBatt 24-cell setup topped out at about 93A. With my other controller, which topped out at about 70A, I didn't really notice a difference.

From what I understand, dealing with PSI directly is not a problem. The only real issue I see is that there just is not a very affordable way to get small orders shipped over here. Basically, you have to have them shipped Express, by air, which will cost more than $100. I think ypedal said with all costs figured, he's paying about $40/cell, or about $1.25/Wh.

You might also try what our good Doctorbass has been doing, which is to find a "local" DeWalt service center, and see if they have any returned 36V packs they will give you. I personally haven't had much trying this here at any of the services centers in southern California, but Doc seems to continue to score quite a few of these. I just got five more from him this week, as a matter of fact. :)

-- Gary
 
Wrobinson: This is a hard call, but to keep this as real as possible, it's like this. I've delt with all sorts of people over the last 4 years , trying and testing everything i can get my hands on that i can afford ( actually going a little too far at times lol ) and after a while you get to understand people's business practices..

Dealing with PSI has been a good experience for me so far, just like taking part on this forum exposes you to all walks of life, doing business with China directly has it's challenges. Keep communication short and to the point, avoid jokes and slang. Remember that time zones are 12 hrs apart so what you type one day gets read the next.

Prices change, and so do offerings, right now i'm waiting for a 48v pre-built pack from Headway with the yahoo group, and some single cells as well.. for those who run 20 amp controllers i feel that it may be a less expensive option, but I don't know that for sure.. yet. Stay tuned for that chapter as the boat should leave in a few days from China for those babies..

10ah is enough juice for a good ride, but to run an X5 or more you need to provide 40 amps and not many batteries can do real 5C.. let alone 10C !!

If you understand what you are doing, and take responsability for your mistakes, buying a box of 50 cells and building your own pack is a fun project, but not for everyone obviously.

I don't want to hype these things overly until i get a chance to beat on them for a while, i've had the experience in the past of randome people reading about my projects and get bit by bad vendors . that sucked.. so buyer beware...

These PSI's are the nicest things i've had my hands on so far tho !!!!
 
I saw some A123 cells in 4 cell packs on Ebay yeserday. Just do a search for lifepo4. As for the seller, welllll, it's not Ping. And they could be A123 cells inside. Anyway it could indicate more avaliability soon. Wish I had money for a stack of PSI's though.
 
joshua said the container will be sealed and shipped after october 5th. you can still order battery packs and individual headway cells with the screw cap directly from headway and they will put it in the container with the group buy, and it will be in toronto by the end of october. seems like the best deal around right now, and headway has earned a good reputation by responding to defective cells quickly. plus it appears this latest batch of cells has very low internal resistance numbers, along with the screw cap which makes them easily replaced, and a full 10.1Ah too. 48V20Ah pack is $630. shipping is the trick, they are consolidating everything in the group buy into one shipment to save freight and shipping it rather than air freight, deal!

http://tech.groups.yahoo.com/group/lifepo4-purchase/
 
Ok.. so i took a day off work.. called in sick.. lower left abdominal cramps from hell.. man.. this sucks... :x

Soooo.. since i'm home .. may as well work on the packs.. :wink:

Before i get into the G10 stuff.. i want to prove the concept with material that i can afford to screw up, found some transparent plastic sheet 1/8" thick at work sitting next to a barbage can... :D
 

Attachments

  • plate1.JPG
    69.6 KB · Views: 1,212
GGoodrum said:
Looks good, but I have a question. Why do you have leads going to both ends of cells that are connected with copper strap?

lol.. i was wondering how long it would take before someone asked that.

I have my collection of single cell VP chargers individually Powerpoled, i also use these chargers for my 12ah packs with the chopper, having one tap per junction would require an extra harness to hook it all up..

It's a temporary setup, once your BMS is complete and tested good, i'll rewire the bms to the packs permanently later on.

I have a dozen 24v 5 amp Soneil chargers waiting for me at the post office..

After sitting on the bench for a week, all cells are within 0.001v of each other.. very cool..
 
Them are some gorgeous lookin cells you got there bud. Possibly the most desireable cells for ebikes in the world.

The cylindrical li-mn cells in my Bionx kit had a "hot glue" bead between each cell and were staggered tight against each other. Not sure how much space saving that is, longer but narrower.
 
This is only the begining, next step is to support the cells against each other, pack1 gets hot glue.. pack2 gets silicone . Then the " Great Stuff " and then an aluminum box to cover all that.

This 48v stack is going to be a backpack battery, so the cube shape is ok.

For the 09 bike, I may call this one the " Grin Reaper " ..

The 72v, 10ah pack is gettin built again in 24v sections, but 2 x 2 x 2 shape... 2 wide, 2 tall, 2 long.. should make a 90mm x 90mm x 340mm, one on the downtube and 2 on the rear rack.

I wish these cells were any other shape than round, securing round cells is a pita.. But i have high hopes for these babies..
 
rolls on floor laughing at the "Grin Reaper"... great name!
 
Back
Top