Well I don’t have great news. I was able to balance voltage to .003 in all cells. I charged to 95.9v and went out for a ride. I watched all the cells on my tablet and they were all dropping to 2.5 and slightly below. After about 1 mile to 2.25 under load.
Start voltage was 95.9
End voltage 92.7
I don’t know what to think. Talked to mike today running 25 of the same cells. He said that they do not drop that much under load.
Either the batteries are not full, the gem is drawing too many amps, or something is wrong with the batteries.
If you can run them empty, about 3 volts and track the charge, should take >60ah to fill them up.
I’m thinking they may have never been charged fully.
Should be safe to charge if it goes through bms. The chargery will not fully charge 29 cells though.
You may have to go back to 28 for testing.
Ok, Im going to drop it back to 28 cells. Ill set the charger to life, what should the end voltage be set at? Sorry this is becoming a nightmare! I just want to get this finished at this point.
Charge to 3.6v and 10% and see how many ah it takes.
Set bms up to 3.75v so it doesn’t kick out early. Or keep an eye on it so none go over 3.75.
My hope is that they are just dead. Other than that, You can send me a sample to test.
It’s not freezing where you are charging is it? I forgot where you are.
I can see one or two failing, but I’ve sold >1k of these cells and only 2 failures other than yours.
I wish Michael would check in with a fresh look.
So I just switched up to 28 cells. The BMS was reading 1AH? Its RAPIDLY climbing though now that its charging. Looking back to our conversations about the 11%, Im guessing that was capacity based off of 60AH. Keeping my fingers crossed.
Me too, my only hope is that the batteries were never charged. I ship them completely empty of course.
That would explain the jump in performance with 29 cells. Michael runs 40mph for 20 miles using only 25s.
I emailed him for help. 
Im very confused too. Everything went down hill when I went to 29 cells. Im keeping my fingers crossed here. I changed Battery Physical Capacity to 60AH so hopefully it is reading correctly. I honestly think they werent charging correctly for whatever reason. Im up to about 6ah total now according to the app on my tablet.
29 cells wont work . your charge voltage is not high enough to charge that many . Target between 2.45v & 2.55v per cell . watch at the end of your charge cycle that they are all very close in voltage , they must be close
(top balanced )
Then take your cart for a long drive and deplete . end voltages should be close . if one is out from the others it is bad .
Tnx @LithiumGods, need a fresh look. Uncharged batteries are the only thing that would explain all cells dropping so low.
I hadn’t thought of that previous, as I assumed any testing would start with charged battery. It’s second nature for me to not run lithium batteries low.
I was using a Chargery charger set to Power Supply, 95.9v, 15a. Was charging 29 cells to exactly 95.9v, but the bms was showing 11% in storage. It really didnt make sense. I still dont understand it. As of this morning, It looks like I’m fully charged with 28 cells and at 59.9ah according to the BMS.
Thats why Im so confused. When I was running the LIFE setting on the charger at 28 cells, it was great. Switched to 29 cells and ran power supply mode, the cells werent getting the AH they needed. Power supply mode should act just like CC on a power supply.
Im going to go out for a quick ride here shortly and let you know.
when you say fully charged , what is the finished voltage showing per cell ?
Yes sir, they were hovering around 3.30v when fully charged with power supply mode at 15amps (29 cells)
We had a little issue balancing a known dead cell, but the avg was 3.295v, the 11% is where my eyebrows started to go up.
After 3 miles barely a drop in capacity! Ok solved that problem. All batteries while accelerating were about 2.75v, #12 was about 2.6v. Does that @LithiumGods is that what you are seeing as well?
Maybe. After you run batteries low, and they take close to 60ah, I’ll call them good.
LiFePo are strange with constant voltage throughout.
The only way to fill them is to sit on 3.5v-3.65v until amps drop.
I test at 3.5v/10a and end test at 1 amp.
So at 15a end at 1.5v or 10% setting. Not critical. 20% may be fine.
See what Michael does.
I think 29s will work. Needs a 106 volt charger.
Some may have been hurt by running low. I can replace a few more as you find them. I want this to work for you! 
This is a big weight lifted off my chest at this point. Something strange though, the buzzer on the converter was going off, I dont think that it is related to what we have been doing. Always something! Its the old style and I was able to salvage a new style converter but havent tested it yet. Im having a hard time locating a GEM dealer that will give me a new converter from the recall without having them put it in.
Oh, I thought this was the 2014 car.
I have chargers with built in dc-dc converters.
Going forward, keep it simple?
These are inexpensive shunt balance boards. They do nothing until a cell gets up to 3.65v.
It then puts a resistor across that cell to bring it down to 3.6v.
Simple insurance, if batteries are balanced it will never come on line. 1 wire to each cell.
This is my 2002! My friends is the 2014. Its most likely a faulty parking brake switch. Ill work on that tonight, should be an easy fix.
Thats nice and easy, but the BMS is nice, once you understand the parameters it gives you lots of info, maybe too much…