I was hoping that I could tap into your knowledge to help troubleshoot a rebuild I am have been doing on a 2017 GEM. I received the GEM as stock and took out the 48V PbA pack and replaced it with a LFP pack. I also replaced the existing Delta Quiq charger with a Delta IC1200 charger. Everything was working fine and I ran the pack for 4 years.
Something happened and I had to build a new pack to replace the old one. I took out the old LFP battery and put in the new one. The problem now is that the pack will not charge anymore. I tried looking for a wiring diagram for the system but it seems that everyone has the older generation and not the ones that cover this vehicle. Looking through the forum, I tried compile the information that most people as for. I have listed it below. If anyone has any thoughts they would be great appreciated.
Any help or thoughts you all would have would very appreciated as I fight these ghosts.
Thank you
-Problem: New LFP battery pack will not charge with Delta IC1200 charger that was working prior with other LFP pack.
-What does work: I can unplug the charging cable and operate the car in drive mode fine. It goes forward, backward and the speedometer indicates speed.
-Does the speedometer indicator have a wrench symbol?:
Yes, when I turn on the car the wrench will show up.
When I toggle through the display with the car unplugged I get one screen that says “Veh Fault”.
The screen after that only shows something “AC E 00”.
All other screens show standard Range, ODO, Trip, Time.
When I plug in the car I the display screen changes from “N” to “C” to indicate charging.
I now get the one screen that changes to “AC E 0.1”. This one seems to reset periodically back.
I also get a new screen that is after the “Veh Fault” that says “ETR28:47”
Symbols on during this entire time are the brake symbol, the orange question mark in a triangle and the passenger seat belt light.
I’m pretty sure people here are asking why was this stated this way instead of explaining it a bit more. After all, it’s conceivable that it does relate to this:
Seems pretty straightforward. If charger is connected to battery, it should charge.
Check for 48v at charger.
If yes, look for charger fault code.
I’m more familiar with the original quiq charger that the ic series.
I assume the quiq went bad, or you could try connecting it again.
Fair enough. The GEM gets used very infrequently and so we have it constantly connected to a wall outlet. The pack itself has a BMS that constantly monitors the pack voltage but has no ability to disconnect the pack from the GEM. During the year the GFI on the charging cord tripped but no one noticed it. The parasitic load of the system was enough that the main battery self-discharged down to around 8V (completely over-discharged).
So had to build a new pack. This time I wont be using the cord with a bad GFI. Also plan on making it easier for me to disconnect the main battery from the GEM so it cant overdischarge due to parasitic load.
When we check the voltage on cables coming up to the charger we see no/low voltage. So for some reason it is open in the circuit but we cant track down where. The pack has the voltage back at the main terminals and the motor is seeing the voltage it needs when operating. The charging isnt seeing the voltage for some reason.
Lithium batteries have a very low self-discharge rate so wonder if it wouldn’t be better to charge it and then disconnect via the breaker. The BMS shouldn’t load it down much but it can all be calculated so you’re not running into the GEM systems being a major drain and killing the battery if power to the charger is lost.
@dougl
Good suggestion on how to arrange it going forward. With the BMS I am seeing a total system drain of 0.3 to 0.4A when sitting idle. Not alot but enough that when it was left along for long enough it caused an issue. Using a breaker would definitely help solve that problem going forward.
The original charger was swapped because it was still using the PbA program. Back when I was trying to get it working the first time no one wanted to try and reprogram the original charger. I got a new one to get past that.
When we put the new charger in I tried to match the charger pins with the pinouts of the original charger. Should I tried to do this another way?
Yep, we are working on fixing the issue with that now. Once I can get all the parts back up and working where it was before I can start adding in new system elements. Wanted to simplify troubleshooting for now until we are back up and running. You mention jumpering the charger rely. Are there instructions on the best way to achieve this? It definitely sounds like this could be the source of my problem.
I don’t have a 2017 but my 2008 has a battery master disconnect which is an 80A breaker in the fuse panel area(also called PSDM ) so I would expect maybe the 2017 models have a similar way to disconnect the battery from the GEM electrical system. Once you throw the master disconnect, look at your BMS to see what kind of drain you see. it should be only due to the BMS and hopefully is in the low milliAmps.
No master disconnect… you have to disconnect the battery. I put in a Chevy volt master fuse in mine to do just that. Or you can turn off discharge on the bms. Assuming you have a bms and that’s an option.
I’m surprised the bms let the batteries die. The ones I’ve played with with power off if they get to a dangerous level
Your old charger would have worked fine as it was without reprogramming or rewiring. 90%
I assume that the ic series charger changes algorythms by bump method, just as the quiq.
If not, it should not have worked
The Gem will bump the charger to the preprogrammed profile before allowing car to charge. Check profipe in current charger. You maybe surprised
The programming/bms relay is zip tied to vertical member near charger. It holds B+ open coing to charger.
If current charger does not agree with bms/vcm, it (relay) will click for a while after applying power.
@Erniea15 - Yep, thats the problem that I am planning on working towards. The idea of using the Chevy Volt master fuse. I have one of the disconnects from their packs. Once I get everything up and working that sounds like a great idea.
@Inwo - It is a full BMS (OrionBMS Jr2), but you are right in that I have only given it the ability to monitor and balance based on how I have it hooked in. I can try going back to the old charger and see if I have any luck. I am a bit skeptical. Since the LFP pack will sit at a higher voltage than the PbA, will the BMS close the rely even if the GEM BMS shows at 100%? This is why I have originally only been using my BMS as a monitor/balance mode because I didnt want the two BMS’s fighting with each other on which profile take control. Again, thank you/
You have some mis-conceptions.
Do you need to charge over the flooded voltage about 58v? I can program up to 66v.
The relay is Used mainly to bump profiles. Did you locate it?
That is an expensive complex bms for what you are doing. The switching is built into the more common bms and can be controlled in app,.
For storage turn the outputs off… Or push a switch to put bms to sleep. A fuse, master switch, and relay, not needed. Short circuit protection is built in.
@ Inwo Nope, dont need to go over the 58V in this case. Looking in the front I do see a white 12VDC box. Would that be the relay that you are speaking of?
An additional piece info that puzzles me. On the car side going into the charger there are the two power wires. In this case they are orange and brown. With everything hooked up I am reading 0V across the wires on the car side. I would have expected to see the voltage of the pack on these wires. Is this not the case? This is with the key in the on or off position. Or is there a switch/rely that doesnt engage?
Related to the BMS, yes agree that it is more high powered than I need. In this case it is what I have available. As you mentioned it is only monitoring so I dont think it is what is causing my issue with getting it to charge. Once I can get it back to charging I can work on making things simpler. Main focus for me now is getting a working system as a baseline that I can move forward with.
Let’s try this again, Your GEM has a charger which is NOT connected directly to the battery pack. Between the charger and the battery pack is a relay box which has 2 functions. Function 1 is to allow the GEMs onboard control system to observe which battery charge profile the charger is in because these chargers can contain about a dozen different charge profiles for different types of batteries. Your GEM was shipped with a certain type of battery so your GEM controller was set to observer your charger is set to this specific battery type charge profile and “relay box” is how theGEM control system does this. It will even try to find the factory set charge profile and if it does not find it then the charger is not allowed to get connected to the battery pack.
Function 2 of the relay is to connect the battery pack to the battery charger. When the battery charger senses the battery voltage it will start the multi-stage battery charge profile which was factory set by GEM/Polaris.
Hopefully you now understand why your “new” charger didn’t work. And yes, what people like Dave and Ernie have been saying about the BMS, etc has been to help you solve the problem which got you here in the first place, ie killing good lithium batteries by incorrect BMS usage/setup. 2 problems needs 2 solutions and because these guys are also helping a bunch of other people, they are addressing both problems since we all would like to not see you back here with dead batteries again.
Hello All. I wanted to thank you all for the help in troubleshooting the issues on this. I finally did get some resolution on the problem. It turned out to be two issues. The first issue was one of the 30A fuses that sits between the charger and the pack was blown. Took a little while to find a fuse that was able to accept the 48V/30A but finally found one. The second issue was the charger. It looks like the charger continue to look for a signal from the GEM indicating that it could charge. After discussions with Delta I was able to upload a file to the charger to wipe the charger back to initial state and then reinstall a new profile. After that it works nicely.
It is back to a working state and so now I get clean up some of the loose ends that were mentioned as suggestions. I found a Volt disconnect that I will install inline to help isolate the pack during down times. I will also need to put a disconnect on the 12V but that shouldnt be an issue. I am also going to start working on identifying a relay that my BMS can use to open the charging circuit.
@dougl - You mention function 1 of the GEM onboard control system. Do you know if there is a way to access this control system so that I can reprogram it to match the new battery type? I have an auxillary battery SOC gauge but it would be great if I could read it straight off the GEM display.
