PDA

View Full Version : GPR-S Battery pack question



Richard230
01 October 2010, 0844
Last night I decided to take all of the covers off of my 2009 "24-cell 72 V" sepex D&D motor GPR-S and check the battery voltages, as I have had continuing problems with voltage sag under power and the performance of the motorcycle is not quite what I expected with the battery/motor combination. Plus, after a run the charging cycle seems to go on and on for hours. After bulk charging at 7.5 amps, the charger goes on and off every couple of seconds at a charging rate that varies from 1.5 amps to .4 amps. Lately, this cycle of on and off charging seems last for many more hours and I was worried about wearing out the charger, or at least the charger's cooling fan. So I figured that if I took the outside off, I might find something interesting on the inside - which I did.

The first thing that I did was to grab my digital voltage meter and start taking voltage measurements of every cell that I could reach after the pack was about as charged up as it was going to get. Most of the cells were reading around 3.65 volts and the entire system showed 83.5 volts. When I checked them this morning they were reading around 3.4 volts plus or minus .5 volts and the pack was showing a total of 78.5 volts.

The odd thing though, is that I got no reading from one of the battery cells. This is the second cell on the left of the top set of 4 cells as shown on the attached photo. My volt meter reads 0.00, even when I connect it to the battery terminals located under the metal plate. The other odd thing (which may or may not mean anything) is that there is some solder located next to the positive terminal on the plate. There are no other solder smears on any other of the cell assemblies.

Can anyone offer an explanation why this one battery would show no voltage at all? It is like it is a dummy and stuck there just to take up space in the box and to make up a total of 24 cells. It sure doesn't seem to be adding anything to my "72 volt" power system. The odd thing is that it looks like it is hooked up to the system, but since the cells are connected in series, I assume that if it was really a dud and connected to the system, the bike would not operate. Can anyone explain what is going on here? :confused:

Thanks.

cycleguy
01 October 2010, 0938
Richard, 83.5Volts/3.65 = only 23 cells. It certainly looks like you have a dead cell, which would definitely effect the performance of the bike.
I'm confused with your reference to a 72 Volt system. 3.2V x 24 cells = 76.8 volts. I would definitely verify that the cell is dead and replace it.

billmi
01 October 2010, 1007
0 Volts from the cell, and it looks like the cell's terminal got hot enough to melt the solder on the BMS circuit board's contact pad. Sounds pretty dead to me.

Richard230
01 October 2010, 1458
The 72 V designation relates to nominal voltage. My original GPR-S had an advertised 60V system that actually showed 68 volts when fully charged.

What I don't understand is how a cell can be completely dead and still allow the system to function. Wouldn't batteries connected in series not function if one was completely dead? The other thing that comes to mind is the experience of a GPR-S owner in the LA area (whose handle escapes me at the moment) with the same system that I have. He said that EMS did not hook up one of his cells when the bike was assembled. I was wondering if that was what happened with my bike, as the fully charged voltage has remained the same from the day that I picked up my GPR-S from EMS.

magicsmoke
01 October 2010, 1653
What I don't understand is how a cell can be completely dead and still allow the system to function. Wouldn't batteries connected in series not function if one was completely dead?

Ahh, but there's dead, dead and dead.
The cell could fail 'open circuit' which as you say, in a series circuit would cause a complete failure.
The cell could fail 'resistive' which if the resistance was even as high as a couple of ohms would cause total voltage sag even at a few tens of amps and so would effectively cause a complete failure.
Then there's the 'short circuit' which is what I suspect you have here. This effectively acts as a wire across the cell, and acts as though you just had one less cell in the series chain.

As billmi pointed out, the terminal looks to have got hot enough to melt the solder on the BMS pcb, and if you look, you'll notice that the path of the heat is toward where the 'heatsink' of the pcb is diminished i.e. the green area.

If I was to put a forensics hat on, I'd say you've got either a slightly resistive terminal connection to this cell or a defective BMS. Either way, something that was progressively undercharging this cell and hence progressively over discharging it. When the 'short circuit' happened, it was likely in a discharged state otherwise I think the effect would have been a lot more dramatic.


... the fully charged voltage has remained the same from the day that I picked up my GPR-S from EMS.

Maybe they killed it on a test ride, or maybe you did on the way home?

If it's gonna be a while before you can get it sorted, I'd definitely consider whipping the dud out and replacing with a heavy gauge wire strap as the cell could very possibly 'un short' and become the dreaded Lithium ticking time bomb!

Definitely get the BMS checked as well!

Rob

Richard230
01 October 2010, 1755
Thanks Rob. What about replacing the cell with a new one?

magicsmoke
01 October 2010, 1813
Hi Richard, to be honest, the extra cell is gonna have very little effect anyway. You'll still get 23/24 of top speed and range won't be affected at all without it, but I would definitely remove it and bridge the 'hole' with a link.
I wouldn't normally suggest this without some sort of comment about what the BMS might make of what it thinks was a shorted cell, but it seems to be ok with the shorted one that's in there already.
There's a lot of anecdotal stuff on the net about mixing new with 'old' cells in a pack, but unfortunately it mostly is just that, anecdotal, and I certainly don't have any real world data.

Either way, get the BMS checked just in case it's responsible and does it to another cell.

Rob

Richard230
02 October 2010, 0751
Thanks Rob. Unfortunately I have the double-secret Modolis Engineering Revision 2 four-cell BMS. This is the BMS with no instructions or supplier contact , which has been discussed several times here. I don't think anyone really understands how it works or where you can buy a new one. Since I have no idea how to check the BMS, I think I will just buy and install a new battery and hope for the best. I don't see how it can be any worse than having a shorted battery.

Thinking back though, I am pretty sure that this battery was already dead when I first bought the bike, as the fully charged voltage was 80 the first time I turned the throttle and it had just been removed from an all-night charge by the retailer. If all 24 batteries were working , the voltage should have been closer to to 84 volts, which is the cut-off voltage for the BMS. I think.

guity
02 October 2010, 1453
Hey Richard, I think I am the L.A. guy you referred to. I didn't have a BMS, which made it easier to determine that the 24th battery had definitely been by-passed. In fact, when I changed the way the cables were hooked up so as to include that 24th battery, I had to go buy another heavy duty cable to run between battery boxes because the previous cable had been cut to run exactly to the battery just before the 24th battery and was not long enough to make it to the 24th. And my experience agrees with magicsmoke's statement that there was not a huge effect created by adding the 24th battery. But since it appears that your 24th battery was always included in the hook-up, then replacing it might help with the sag. Are you going to do this yourself, or take it to the dealer?

Richard230
02 October 2010, 1831
Hi Guilty. Nice to hear from you. My plan is to order a new battery cell and have a friend who maintains radio station transmitters install the new battery for me. Frankly, I am not all that confident in my dealer's ability to perform this work. Then there is the problem of getting the bike there and back. It can only go one way, without an all-night charge. But if my friend doesn't feel comfortable performing the battery replacement, I will have to head for my dealer. In the meantime, I attempted to order a new battery cell from an outfit in Salt Lake City, but their web site was not willing to take my charge card and requested that I email the company, which I did. More to follow, no doubt.

Today I put everything back together and plan to keep riding the bike until I can acquire a new battery. I am pretty sure that a new battery will make an improvement in maximum voltage and perhaps a slight increase in speed, if not range, as the stock 2009 GPR-S with the Etek-R motor has 24 40 Ah TS cells and I have seen 84 Volts on their CA when fully charged. My understanding is that they use the same BMS and charger as my bike does. If I screw things up, it will just be another one of life's little lessons for me. Live and learn, I say.

guity
02 October 2010, 1937
It might also save some wear and tear on your pack as a whole if replacing the battery helps keep the pack from sagging down below recommended voltage levels as often...

ChuckJ
22 October 2010, 0929
Richard:
I get 80 volts on my 20 pack in my GPR-S and it's 2 years old. Sounds like something is wrong to me.

ChuckJ

Richard230
22 October 2010, 1434
I just received a "new" used battery from Harlan and plan to replace the one that is shorted in my pack tomorrow. I can get 84 volts as soon as I pull the charger plug. But if I wait a day, the pack drops to 76.8 volts. By replacing the dead cell, I would imagine that I should see about 80 volts, once things balance out. I am keeping my fingers crossed and will report on the results as soon as I can install the new battery and have a chance to ride around. (I hope I don't break anything.) Now all I need is some dry weather. It looks like rain all weekend.

guity
16 January 2011, 1451
Hey Richard, how did all this work out?

Richard230
16 January 2011, 1632
Keep an eye out for "Sparky's Saga" that I am currently writing.