Lithium gem battery display

Hey everyone we recently did a lithium upgrade on a 2002 e825 gem using 24s prismatic lifepo4 cells. Everything works fine but the battery display is acting up. When we take it for a drive it drops rapidly even though the batteries are perfect and are still full. It is around 55 degrees Fahrenheit (I’m not sure if cold has anything to do with it.) Any help would be much appreciated.

What kind of meters are you using?

What kind of meters are you using?

Good point. The original post is vague.
Also- What do you mean by acting up?

Assuming you are looking at the original BDI installed on your car from the factory? Keep in mind that is 20 year old tech and I think there is a reason it was abandoned in 05 when the car was redesigned.

From what I understand that round BDI used on the early Gen1 cars is a Voltage based device that Guesstimates remaining battery charge via random peeks at battery V as you are driving around.
This is not a very accurate way of calculating battery capacity and there are other methods to do this now and might be adapted for use.

The BDI is calibrated for a set of batteries that is no longer installed in this car. You might be able to get into the controller and play with the calibration but you need the programmer, software, cable and trial/error time to get this tweaked to read correctly.

Not knowing much about your new battery setup I hesitate to point you in an alternate direction at this time. It sounds like you put together a custom pack of your own design. Are you running some form of BMS/monitoring system that would be a better access point to get battery status?

The simplest method is to just add a pack V meter and monitor it yourself. Know your bottom limits and make sure you get home before that.

The car has the original speedometer in the dash and also no bms connected to the batteries. The issue is that the indicator goes down so fast that even though the batteries are fully charged I’m concerned that the car will shut off while driving because it thinks it ran out of battery.

Also, when I shut the car off and turn it back on it resets itself, but still goes down rapidly.

The BDI does not actually control the motor controller. A zero on the display won’t shut down the car.

Since you are not running a BMS I’d be more concerned about the status of your pack. This means YOU are the management system. Hook your handheld meter to the top of the controller POS & NEG and monitor what it is seeing when driving around. Take note of the Resting V, then hammer the pedal and see how far it drops under load. If you have a hill it is easier to see load drop when it is sustained. If you have a connection that did not get tight enough it may be driving that BDI nuts!

Your statement:

even though the batteries are perfect and are still full.

Has me wondering how you are judging this. “seat of the pants” is not always very accurate.

How long ago did you complete this install? “Recently” doesn’t reveal much. Depending on your answer, it might be a good idea to do a check of your cell balances. How close to each other are they?
Also- What is your charger running them up to after a charge? Don’t cheat and do only one or 12. You need to do them all.

Are you way too busy to do all this? I suppose you can hire a kid to do this it or put on a BMS.

Lfp discharge curve is very Flat so 80% or 20% will be almost the same voltage. You need another way to monitor that pack.

Also, LIfepo4 discharge rate is not the best so I wouldn’t be surprised if voltage Ives getting a huge drop when accelerating.

Do what @AssyRequired recommended hook a multimeter and hammer the pedal to see voltage drop,

I agree. The only good thing is that LFP are so low cost that larger cells, and more than 24s. will cover up any problems.

Thanks everyone for the input, I’ll do the tests @assyrequired suggested and check all the connections. I’m hoping it’s just a connection issue.

How new is this install?
Can you guess how many cycles you have done on your new pack so far?
Did you have your controller programmed recently?
(just trying to track down anything that might have been changed)

I remember reading some chatter(elsewhere) about how this may be a “smart/learning” monitor that sometimes takes several cycles to adjust to the new set of batteries but (given the source) I’m not giving it that much credit. The PID (display) communicates directly with the controller so getting an external PackV monitor on what the controller is actually seeing at runtime may give us a clue.

Perhaps the controller is doing some calculating and will readjust after several cycles?
SWAG - If smart adjusting itself, You are running off a previously calculated parameter that was based on some tired old LA batteries. In my fantasy world this will re-adjust after 4~7 full cycles or that many trips to the ice cream shop(depending on how far it is away).

Do you have access to a Motor Controller Programmer?

The ah and “C” rating may be a clue.

" C-rate (C) = charge or discharge current in amperes (A) / rated capacity of the battery (Ah) Therefore, calculating the C rating is important for any battery user and can be used to derive output current, power and energy by: **1C means 100Ah*1C=100A discharge current available"

As long as we are on battery specs. You are in NY? Do you use car in cold weather? LFP doesn’t work below freezing.