Need Help with Headway DIY Pack

atatistcheff

Solar Enthusiast
I need some help from someone with experience building LiFePO4 packs from used cells. Here is my situation:

I recently bought 8 Headway 8Ah LiFePO4 cells to build into a 4s2p battery for my UPS. I got these from batteryhookup.com. These are "engineering samples" and have a 2017 manufacture date. They advertise something like 90-102% capacity on these cells. When I first got them I measured the voltage on each cell. The high was 3.332 and the low was 3.240 so not crazy different - that's a 0.092V range across all cells. I hooked them up in parallel for a few hours and then built the pack and connected the 4s BMS and let them sit overnight. The next morning I measured the voltage across the four pairs of cells and they were as follows:

1 - 3.330
2 - 3.301
3 - 3.289
4 - 3.331

The range was then 0.042V which is well within the BMS spec of +-0.05V

I then did a capacity test at about 25A (3C) rate. The BMS cut off the power after 9.2Ah was used according to my hall coulomb meter attached to the battery positive wire. At 100% capacity this pack should have 16Ah @ 12V. That's only 57% of the rated capacity of the pack.

In testing voltages I found that pair 3 was at 2.5V which is why the BMS cut off the power. The other cell pairs were between 3.16 and 3.25 volts.

I tried bottom balancing the cells by bringing pair 3 up to 3.2V but again the BMS cut the power while the other cells were well above the 2.5V cutoff. So it seems pretty clear that I have one or two cells that are not cutting the mustard.

So my questions are:
Is my methodology solid?
If I do have a bad cell or two what is the best way to determine each individual cell capacity? I can power an inverter with 12V but trying to do a test with 3.2V is a bit problematic. I'd like to confirm each cell but I need a good way to do that.
Other than ask batteryhookup to replace the cell(s) which are under capacity is there anything else I can do to help this pack perform better?

Thanks in advance for reading this lengthy message!

20191127_083534.jpg
 

Supervstech

Administrator
Staff member
Moderator
I need some help from someone with experience building LiFePO4 packs from used cells. Here is my situation:

I recently bought 8 Headway 8Ah LiFePO4 cells to build into a 4s2p battery for my UPS. I got these from batteryhookup.com. These are "engineering samples" and have a 2017 manufacture date. They advertise something like 90-102% capacity on these cells. When I first got them I measured the voltage on each cell. The high was 3.332 and the low was 3.240 so not crazy different - that's a 0.092V range across all cells. I hooked them up in parallel for a few hours and then built the pack and connected the 4s BMS and let them sit overnight. The next morning I measured the voltage across the four pairs of cells and they were as follows:

1 - 3.330
2 - 3.301
3 - 3.289
4 - 3.331

The range was then 0.042V which is well within the BMS spec of +-0.05V

I then did a capacity test at about 25A (3C) rate. The BMS cut off the power after 9.2Ah was used according to my hall coulomb meter attached to the battery positive wire. At 100% capacity this pack should have 16Ah @ 12V. That's only 57% of the rated capacity of the pack.

In testing voltages I found that pair 3 was at 2.5V which is why the BMS cut off the power. The other cell pairs were between 3.16 and 3.25 volts.

I tried bottom balancing the cells by bringing pair 3 up to 3.2V but again the BMS cut the power while the other cells were well above the 2.5V cutoff. So it seems pretty clear that I have one or two cells that are not cutting the mustard.

So my questions are:
Is my methodology solid?
If I do have a bad cell or two what is the best way to determine each individual cell capacity? I can power an inverter with 12V but trying to do a test with 3.2V is a bit problematic. I'd like to confirm each cell but I need a good way to do that.
Other than ask batteryhookup to replace the cell(s) which are under capacity is there anything else I can do to help this pack perform better?

Thanks in advance for reading this lengthy message!

View attachment 2611
I’m trying to follow your question, but you said bottom balance, and brought up to 3.2v...

Bottom balancing would be to bring all the cells down to 2.5v...
 

atatistcheff

Solar Enthusiast
Yes, you're right. In trying to bottom balance using resistors it takes forever to get these cells down to 2.5V so I "middle balanced" by bringing the dead one up to 3.2 and testing again. Do you think that bottom balancing might actually be the missing link here? Because if so I can take the time to wait for the various pairs to drain using resistors. I just didn't want to spend the hours monitoring this if the problem is with the cells. If you think bottom balancing might be the key then I will try that.
 

Supervstech

Administrator
Staff member
Moderator
My favorite way to bottom balance is hook up a regulated load like a cheapo battery bank. They will pull a single cell down to 3.0v if you hook to a usb fan or similar load... it’ll take days to get low, but it won’t need to be monitored.
 

atatistcheff

Solar Enthusiast
Well, the good news is the second capacity test was much better. Got almost 13Ah out of the 16Ah pack. I wonder if these cells were just sitting for a while an needed to wake up. Bottom balancing now so we'll see how it goes after that.
 

atatistcheff

Solar Enthusiast
After bottom balancing to 2.5V per cell my latest capacity test was much better. Got 14.4Ah (90%) out of the pack before the lowest cell hit 2.5V. That works for me!
 

JAC

New Member
After bottom balancing to 2.5V per cell my latest capacity test was much better. Got 14.4Ah (90%) out of the pack before the lowest cell hit 2.5V. That works for me!
What method did you eventually use to bottom balance your cells? I picked up the same cells as you have.
 

atatistcheff

Solar Enthusiast
What method did you eventually use to bottom balance your cells? I picked up the same cells as you have.

First I discharged the entire pack with a 12V load until the BMS disconnected (lowest cell was @2.5V) then I used several 10W resistors to bring the others down to 2.5V. Just connected the resistors and set an alarm on my phone to check back every 30 minutes or so.
 

JAC

New Member
atatistcheff,
Thanks! I guess my 100W ceramic resistor is overkill for fine tune depletion. Time to pick up some 10W
 

krby

Solar Addict
I also just got these https://www.amazon.com/gp/product/B07RW2XKRT/ref=ppx_yo_dt_b_asin_title_o01_s01?ie=UTF8&psc=1 which make it easier by alerting when a set of parallel cells gets to 2.7V. It's adjustable to alarm between 2.7 and 3.8V.


Ok, but you really should each one of these against a multimeter or two and adjust the alarm based on that. There a bunch of that same form factor all over Amazon and other places. I used to keep one of these in my pocket while flying R/C aircraft, but at some point an "old guy" in the hobby (heh, I guess I'm one of those now!) asked if I knew how inaccurate they can be and how far one in a pack will be from another even from the same vendor bought at the same time. I've found some 0.2-0.3V high or low from "actual". This is big deal on a LiPO which we typically use from 4.2 full to 3.7 "empty". It's an even bigger problem with LiFePO4 where the voltage range is much flatter.

If you can manage it, measure each against two multimeters, hopefully those "agree" down to at least 0.01V, then measure with one of those guys and mark down the offset. When I had mine, I just put a label on the back and set the alarm to my alarm voltage + or - the offset.
 
Last edited:

Max Bender

Solar Enthusiast
I've had no luck with BatteryHookUp's 38120 Headway batteries - bought eight of them and all were below two volts except one that was below one volt. Brought each one up to 2.5 volts then connected in series and balance charged them to full capacity (except for three that absolutely would not reach a full charge). So... whatever - I'd rather risk buying from China than ever buy from Battery Hookup; and so far, after quite a few purchases, the Chinese have not disappointed me. Meanwhile, shoved BHU's crap batteries in an old jump-start thing, stuck a volt meter on it, slapped an inverter onto the side, and use the thing for... well... not much - those crap batteries needed a charge after only having charged a cell phone. LOL... I'm stll angry... I'll NEVER by ANYTHING from BatteryRipOff.
 

Attachments

  • 20191224_164523.jpg
    20191224_164523.jpg
    111.8 KB · Views: 3
  • 20191224_163830.jpg
    20191224_163830.jpg
    110.3 KB · Views: 3

JAC

New Member
I purchased 8 also, and have had no problems. I was reading that may still be able to resurrect batteries that are al least above .5V,but you won’t get the same life out of them. You should be able to return the batteries. What has been your experience with that process so far?
 

Max Bender

Solar Enthusiast
I purchased 8 also, and have had no problems. I was reading that may still be able to resurrect batteries that are al least above .5V,but you won’t get the same life out of them. You should be able to return the batteries. What has been your experience with that process so far?
My experience became irrelevant when it turned out that the battery that came in at .73v was the strongest among the eight when charged, and the most stable among the batteries as well. That's when I decided to stick with buying new batteries. There just seems to be a better chance of new batteries being fully usable. Another thing that I avoid, even new ones, are the aluminum clad LiFePo4 batteries. Plastic might not be eco-friendly, and plastic doesn't dissipate heat very well, but when it comes to something with the electrical capacity of a LiFePo4 battery, I don't want one that's susceptible to punctures, dents, and corrosion. Not only that but plastic is forgiving to the effect of expansion and contracting caused by thermal differences. Plus, though time will tell, does the potential exist for chemical reaction between aluminum and the lithium salt solution in a LiFePo4 battery?
 
Top