DIY Electric Car Forums banner

Experienced Very-High Headway Self-Discharge?

6K views 31 replies 8 participants last post by  frodus 
#1 ·
Hi all,

I have many Headway cells which arrived ~2 months ago and nearly all measured 3.3V (the bad ones were thrown away). Shortly thereafter, all the good ones were charged to 3.65V in parallel. Never exceeded 3.65V. Then they were stuck in a box with foam insulating each cell.

I just checked today and many range from .7V to 3.20V.

They are fairly evenly spread in three groups:
3.32-3.33V (expected)
3.26-3.30V
0.70-3.20V

My only hypothesis is high self-discharge. I am going to do some cycle tests on the ones above 1.1V to see if they meet their Ah rating.

Anybody else notice Headway cells with this problem?

Thanks,
Cory
 
#4 ·
Can't speak of Headway, but have experience with Thundersky and A123. 3.3x would be normal for a reasonably charged battery that has been sitting. I have seen good batteries from 2.28 to 3.32 with 3.30 and 3.31 being the average. When you charged to 3.65, did you allow adequate time for all to attain a full charge? A problem with parallel charging is some of the batteries may be pretty full while others are still low.

I have just under a hundred cells that have been in crates for months with the factory charge and I measure them periodically. The lowest is 3.29 and the highest 3.31 They have remained unchanged. I have yet to put a full charge on them. When I do charge them, it will be under BMS control and I will know which ones are doing what.

The ones with .7 volts sound pretty bad. Did you really chuck them, or send them back under warranty. When taking delivery of new batteries, you should immediately check the voltage and notify the vendor of any low cells at the time of delivery.
 
#6 ·
When you charged to 3.65, did you allow adequate time for all to attain a full charge?
Yes, they were left on a charge of 3.65V until the current was below 2.6mA per cell.

The ones with .7 volts sound pretty bad. Did you really chuck them, or send them back under warranty. When taking delivery of new batteries, you should immediately check the voltage and notify the vendor of any low cells at the time of delivery.
Hmm, this batch actually didn't have any bad ones fresh off the boat (previous orders had only a couple and it wasn't worth shipping). Sorry for the bad information.
 
#7 ·
Hi coryrc... we have seen this with Headway cells, too. It seems to be a factory defect type of problem - a contaminant in the electrolyte and/or solvent that causes the batteries to self-discharge in short order. A properly built LFP cell has essentially no self-discharge at all. I have an old school TS 200Ah cell that I have checked every week or so over the last 18 months and the voltage hasn't budged a bit.

Unfortunately, these cells need to be tested for capacity, internal resistance and, apparently, self-discharge rate at incoming inspection. Needless to say, once you've done all of that - as a business, anyway - you then need to charge A123 prices to recover your labor expenses.
 
#9 ·
Hi coryrc... we have seen this with Headway cells, too. It seems to be a factory defect type of problem - a contaminant in the electrolyte and/or solvent that causes the batteries to self-discharge in short order.
Bummer. In our latest batch, even the "good" cells averaged 10% loss per month (measured by charging until 3.65V -- took 2 hours at 0.1C).
 
#10 ·
I had some self discharge, but that was after a year of sitting without any charge. they came back, and apear to have acceptable discharge capacities.
We have some bought a year ago (from EV Components FWIW); of four that were never used, two are still >3.3 and the other two are above 3.0V.

Where did you get them? It's possible they sat around.
They came straight from Headway via boat. Hmm, in our original batch from EVC, about 5% were bad (<3.25V). But this time *none* were. If they sat around at EVC for even a short time, perhaps that was a decent-enough self-discharge test. This time, about 10% were terrible after a couple months -- not too far off from the original.
 
#14 ·
Hmmm, good thought. I assume the middle 4-character thing is the date code? Is the last number a serial number?

The "perfect" ones (3.32V two months after charge) were a mix of JK19 and JK18. No JK19-series were less than perfect.

All the other categories were a mix of JK17 and JK18.

The two worst cells (under 1V) were JK17.

Hmmm. I guess we'll demand replacements of all the JK17 at least. Hopefully the JK18 as well.
 
#18 · (Edited)
Almost true!
Lithium Polymer does tell its SOC by its voltage though.

Odd fact is that they refer to them as 3.7V cells, but in reallity they are 4.2V when full and empty at 3.6..3.7V. The voltage slope is linear in between between the two states.
Not really correct.... RC people use Voltage because they're pulling such high C at a constant rate that it apears to be less of an S curve, with the middle part being more linear. With EV people using them under varying discharge rates, with stops and goes, SOC is not indicated by voltage. All batteries have an S curve. If it was linear, they'd be more like capacitors. I've never seen a battery that had a linear discharge curve.

A test I did:
Kokam (lipo) 7.5Ah test at 5C (I've got higher discharge rates, I think up to ~15C which was the limit of my setup)

Ran at ~37.5A for ~12.5minutes (I ran some before and some after for reference).
Got to ~113degrees F
Started at 4.21V and ended at ~2.7V
 

Attachments

#19 ·
Well the big flat spot in the middle with LiFe is not present with LiPo's.Perhaps due to the large voltage difference between full and empty.

Back on topic,
I am interested in how this Headway battery self-discharge issue turns out. Just ordered 400+ 38140S cells and this post is making me a bit worried :D
 
#20 ·
Well the big flat spot in the middle with LiFe is not present with LiPo's.Perhaps due to the large voltage difference between full and empty.
Actually, that big flat spot is present in lipo, and other chemistries of batteries that I've seen, but it all depends on your rate of discharge on how flat it is. At 1C I could see a very flat part in the middle. True that it does act differently than Lifepo4, but it's still an S-curve, and it's not linear. The higher discharge rates, the closer to linear falling curve it gets, but for how we use these cells, voltage is not an indication of SOC, and your assumptions only work for cells under constant load. As soon as you let off the load, your voltage comes back up.

Back on topic,
I am interested in how this Headway battery self-discharge issue turns out. Just ordered 400+ 38140S cells and this post is making me a bit worried :D
I wouldn't worry too much, but I would get some sort of discharger or some sort of IR measuring device so you can match cells. If you match cells in parallel you should be just fine. Weed out the few cells that might not meet your spec. That's what I've done, albeit slowly.
 
#22 · (Edited)
Haven't heard that before, if you've got more info on SOC related to the resting voltage of a cell for lifepo4, I'd love to read up more on it.

Well, get with headway. Most of mine that have been sitting for a year, without any initial charging, were above 3.3V each. I only had maybe 4 out of 200 that weren't. Maybe they can give you replacement cells.

Good luck, let us know.
 
#26 ·
Huh? The factory acknowledged they sent us a bad batch of batteries and has already sent the replacements. As well, others have chimed in acknowledging this as a problem.

So, to answer your loaded question, an earlier model of this rack-mount, super-expensive 0-8V power supply set to 3.65V and not compensating for voltage drop (so they never exceeded 3.65V):
http://www.sorensen.com/products/xg1500/XG1500_Overview.htm
 
#28 ·
100% self discharge in two months is worthy of total prostration, a written apology and personally checked cells, hand delivered.

Will you let us know how the replacement cells perform?

Worrisome, I was considering buying a case of 10, 60V 50AH headway packs for my trike project. :mad:
 
#29 ·
Very odd. I've been running a 48s4p pack of 16ah cells in my car for the past 3 months. No bms. No balancing. If they drifted even a small percentage i'd have had my own personal fouth of July by now:eek:.
 
#30 ·
I've had only a couple cells that ended up self discharging that much out of 230 or so cells.

Realize also, they don't ship them fully charged.

I think it was a bad batch, I know many many customers that have had no problems like this high self discharge.
 
#32 ·
that was from a group over a year ago, and I left them sitting, in a garage, and hadn't touched them the entire time. Only a couple below 2V, and they came back just fine. Slow self-discharge didn't seem to hurt. I'm not putting it in my pack, but I'll be using it for my 12V aux battery.

Just make sure you buy from a company that has a warranty.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top