There is only one gear ratio. It’s a gear reduction/diff not a transmission.
Yes it only identify a pair that are sagging but then easy to diagnosis from there. Here is a possibility, on a classic 72v Gem the Deltaq charger goes into balance mode and slowly raises the pack to 96v, 16v per battery. Doing this helps balance the 6 batteries in series. I dont know if the Deltaq on the new Gem does the same thing. If it does it could be triggering one of the bms’s to disconnect, therefore stopping charging. If the batteries get out of balance One pair wont fully charge and could be sagging under load. Another thing to do would be to charge the Gem and then individually charge each pair with external charger.
Quick update here. The same thing happened this morning after charging overnight, so the batteries had a full charge. I called the Gem dealership, they to thought it would be the batteries. I decided to go ahead and unplug the main disconnect from the Gem and reconnect, sure enough after I did that the acceleration is perfect now! Does anyone know what could have triggered this? I mean it is a workaround, but if I can avoid doing that anytime this happens again, it would be great!
I did a real quick scan of your thread and I couldn’t seem to find what I needed so I will just go from my rapidly fading memory. Correct me if I’m wrong. I realize these are not your actual numbers but I simplified for explanation. I do not pretend to be an expert and am learning too.
You have your 8) 12v 100Ah packs wired up as 2p x 4s.
This means you have 12v(100Ah+100Ah) + 12v(100Ah+100Ah) + 12v(100Ah+100Ah) + 12v(100Ah+100Ah) = 48v(200Ah).
Theory:
If you have one cell off balance the BMS should shut it down to protect the battery and take it out of the loop.
This now gives you 12v(100ah) + 12v(50Ah+50Ah) + 12v(50Ah+50Ah) + 12v(50Ah+50Ah)= 48v(100Ah). I’m guessing the remaining pairs will just step down and share the current duties.
So this essentially creates a choke point in the string where ALL the current is trying to pass through the active BMS in the broken pair. It sounds like the BMS is current limiting instead of triggering an overload condition and breaking the connection.
You cant tell which battery it is with a voltmeter because you are measuring them still paired together and getting a reading off the battery still active.
I think you might need to pull off one of the parallel wires in each of the pairs and a volt check might show you the missing battery. It depends on what is triggering and resetting the BMS. You might have to wait until the issue happens again, or do individual load tests on each battery to see each drop. A more detailed capacity test may also be in order but that is very time consuming.
I am confused at your description that solved your problem today.
" unplug the main disconnect from the Gem and reconnect, "
Did you unplug the 120v charger cord and reconnect? This essentially put it through another charge cycle and may have reset the BMS on the problem battery and brought it back into the loop.
You definitely caught on fast 
You jotted my setup correctly. I better understand now why it may be an issue. I would think that the BMS would simply cut off, I initially tried running the cart with only 4 x 12v and when I floored it the BMS simply cut off and the cart stopped. That is why I had to go with 8) 12v 100Ah packs wired up as 2p x 4s.
I unplugged the main disconnect to the GEM, that is located by the front right wheel.
But you can tell with an ammeter.
I asked previous if all the series parallel connections were made, with equal conductors. In such a sketchy battery system it may make a difference.
In plumbing it’s called reverse parallel, I think. A way to get equal flow through parallel loops.
sal101011- I would think that the BMS would simply cut off, I initially tried running the cart with only 4 x 12v and when I floored it the BMS simply cut off and the cart stopped.
This is where my theory somewhat falls apart. I don’t understand why the second cell in the pair will still allow a high current draw and not shut down and kill the cart completely? (or maybe it will, but you haven’t pushed it hard enough in this limp mode?
This was a snip from the battery description from the link you provided.
“Includes built-in PCM (protection circuit module) which provides internal cell balancing and management, protection from overcurrent, undervoltage (overdischarge), overvoltage and over temperature.”
Inwo- But you can tell with an ammeter.
Ah, Exactly what I was thinking! I was going to make this point but without knowing how the 8 cells are cabled I couldn’t offer up a logical place to put a DC clamp meter that would reveal anything. Then you would need to run the car in the “problem” state with the meter in place. and not everyone has a DC clamp meter in their toolbag.
Interesting reverse flow diagram. I’m going to have to study up on that theory a bit more. I wouldn’t have thought the extra bars bridging the packs would make a difference. I probably would have run either the red or the black jumpers. I wouldn’t have even thought equal length either! That might be tough in a cart with limited space for 8 packs.
Yes, equal conductors were used. I used 4/0 wire across all parallel connections as well as the series connection. I will try to use an ammeter, it will be hard when I am driving.
But what still behooves me is that when I unplugged the main disconnect of the gem, and reconnected the disconnect, I got the acceleration back.
That will be tough with out some special meters or battery monitors.
I agree, reseting the Gem should not reset the batteries. A battery issue may put the Gem in limp mode though. Does the bms reset after tripping?
When I trip bms on a Sevcon system it triggers a fault that is reset by a power cycle. But the key will power cycle controller. Did you notice a Gem fault or a controller fault ever?
I have a battery monitor, but it is at the entire bank level. On Load, voltage stays at around 48v.
What makes things tough is that my speedometer is dead. Does the Sevcon controller have any lights to it? I havent looked to be honest. The only way I was able to reset the BMS at the individual battery level is to disconnect the leads to that battery and reconnect them. The main disconnect wont reset the BMS on the battery.
Yes, sevcon blinks error codes. I may be able to repair display.
The pain in removing the display…I tried once and it was a pain. any special tools that I need?
It just unplug. I forget what holds it in. I think it pushes in through a big rubber grommet.
Just wanted to let everyone know, once I reset the Gem the acceleration has been fine. No drop in acceleration yet…will keep everyone posted.