Ping battery problem - need help

Hi, I've have a few problems with my ebike lately, and I'm wondering if it wouldn't be a battery problem.
About 2 weeks ago, ebikes.ca mailed me a 40A controller as they had originally sent me a 35A one, and I had ordered a 40A.
After the swapping the controllers, I started seeing some strange behavior with my bike, and although I don't think my problems are related to the swap I still wanted to mention it, just in case.
So, after a couple days, I started having problems after a full battery recharge, where the battery would shut down itself as soon as I would hit the throttle. At this point, I would unplug the battery from the controller and replug it (on / off on the controller wouldn't fix the problem), current would come back, then I'd hit the throttle, battery would die again, and so on. Eventually, after a few attempts like that things would start working again. Every day, things were getting worse (I would have to unplug the battery more often after a full charge), up to this morning where I had to wait up to 30s - 1 minute before the battery would come back, and I finally had to gave up on my bike to commute to work after like 40 minutes of attempting to get it to work. Also to note that as I was wondering if the problem could have been the controller, I had switched back to the 35A controller yesterday evening, that my bike hadn't been recharged at that point, and that I did a test run which went perfectly ok, so there's something about the full charged state ...

Went back home (I wasn't really far ...), and did some basic testing to try and see what was going on :
1. it is really the battery which shuts down, not the controller (I know that from the fact that my light also goes off, and it's powered from the battery, and the voltage on the charger port of the battery drops to only a couple volts and then slowing down towards zero when this happens)
2. The battery at no load shows 57.5V (so apparently it is fully charged -- I see 60V for about 30s when it's right off the charger --)
3. At some point, right after it shut down, I was actually measuring a voltage of something like 48V, which seemed really strange, so I went on and measured individual cell voltages. My pack is a 48V 20Ah made out of 2 blocks of 8s each, and interestingly, one pack had all its cells at something like 3.7V while the other pack had all its cells at 2.8V ... again, just moments before, the overall voltage of the full pack was 57.5V ... this state lasted for as long as I kept the 2 packs disconnected (by unplugging the big wire that usually goes between the 2 packs)
4. So my guess is that the BMS does kick in because it sees too low voltages on a bunch of cells, but that doesn't really explain how this happened almost everyday of last week, and eventually things would work again after a few attempts. If the cells really were discharged, then how would I have been able to drive 10 miles, using 6-10 AH every time? the last time I actually didn't charge from work because I had had problems in the morning, and didn't experience any problem when going back home, totaling more than 14 AH that day, while I had this problem really badly the same morning?

I've got a bit more than 800 miles on my battery.

Would anyone have a clue has to what could be happening? what kind of tests / measurements should I perform to better diagnose my issue ?

Gudy

Gudy, I hope I'm wrong but it sounds like the 40amp controller has totaly misbalaced the pack. Ping cells are 2c so 20amp pack being dicharged at 2c+ is tough on those cells as well as many others. Even the 35 amp controller is high. What motor are you using? I think Dogman will tell you about the same thing but in better terms.
Put the battery on the charger and leave it fo at least 24 hrs and see if it balances. Is the battery a 2.5v with the balance leds? If they are all lit there is other problems with the BMS I would guess.
Dan

yep, the batteries are the 2.5V, and the BMS has the leds ... all LEDs are on after the charge, and the CA did read 60V when I initially plugged the battery to the controller after a full charge. But it seems that pulling power from the battery brings the voltage quite low. I usually charge overnight, and keep the charger plugged until I leave in the morning (so the bike gets 11-12 hours of charging at 2A, while I usually use less than 8Ah per trip), and I also charge during the day at work (again, usually about 8 hours of charging, but this time at 5A, same 8Ah. This should give the cells plenty of time to balance right ?
Regarding discharge rate, 2C for 20Ah isn't that 40A ?
Also, even when at full throttle, the CA never shows more than 35A on my 40A controller, and 30A on my 35A controller, so unless ebikes.ca didn't print the right RShunt value for its controllers, it looks like I'm not pulling too many amps from the battery.

The other "weird" thing is that I did a test ride (granted it was only .5 miles) yesterday evening after changing the controller, and that was before my battery was fully charged, and it worked perfectly, but it seems that the problem mostly happens right after a full charge when I just disconnect the charger (although this morning it didn't even work 30 minutes after ...)

I'll go measure the voltages of the cells again tonight, I've left the pack unplugged for the day, and will report back.

So, I did a quick video of what I'm seeing tonight.
I topped my battery (only took a couple minutes) then I plugged the voltmeter into the charge port and plugged the controller.
A few seconds in the video I turned the throttle, BMS kicks in and shuts down. As can be seen voltage quickly drops towards 0. I then unplugged the controller, voltage comes back at 44V, most LEDs are still red showing that the BMS still believes that most cella are fully charged. Then I unplugged the voltmeter, and plugs it back, and now the voltage keeps changing between 44V and 57V ( and it's still doing that more than 10 minutes later). I've checked the packs independently, and they both seem perfectly fine. So I'm guessing that the BMS is completely borked.

Here's the link : http://www.youtube.com/watch?v=LpHYqIX0_TU

Do you have a inline key switch that might be having contact problems?

No, I simply use Anderson pole connectors.

Only some tedious testing is going to give you the real answer. I suspect a bms problem too.

But it could also be that the very high amp controller for a ping has resulted in some toasted cells. Some of the paralell groups have dead cells, so when you pull 30 amps or so from the 5 ah remaining cell, thats what, 6c? Ouch, voltage drops like a rock and bms trips. But you never got far enough to discharge that lonely remaining good pouch, so when you reset the bms the voltage looks normal. In short, your pack is behaving normally for a 5 ah pack of 2c cells on a 40 amp controller.

Some lower amp testing of capacity will reveal if you have dead pouches in there, or telltale swollen pouches. If it tests out for full capacity when discharged at 5 amps, then you do have a toasty bms. The blinky lights are great, and show you full voltage, but they don't tell you anything about capacity. If cells are not swollen, maybe some carefull balancing of individual cell groups with a slow trickle of 3.7v will bring em back. And then chuck the 40 amp controller for a 20.

Dogman,
Thanks for your answers. You're always helpful.
I'll try the slow discharge again, but if I remember correctly even my 10W lumenator triggers the Bms to shut down after a couple minutes. That's 0.2A at 50V... And why would the battery voltage constantly switch between 44V and 57V under no load other than the voltmeter which has a resistance of 1MOhm irc ?
I'm assuming that the cella are in series, so what could cause the voltage to switch between these 2 values ? Plus when measuring the voltage directly from the cells leads back to the BMS they show a perfectly constant voltage. It really looks like something in the BMS is kinda fried and causes an intermitent voltage drop of 13V. As I don't know anything about BMS I have no idea if this is even possible.

Make sure all of your BMS and other connections are tight. Give the battery a full charge then post the individual cell voltages.I had a similar problem with a split 48V20Ah pack and it was all the result of a bad BMS causing one cell to always be undercharged. It eventually caused the low voltage cutoff to constantly activate.

Did a full charge (didn't take long ) and then i measured all the cells except the first one because the only lead to measure it is from the ground, and it goes through the BMS which frocks up things... More on that later.
so here are the voltages :
3.59
3.60
3.59
3.59
3.59
3.60
3.60

3.60
3.60
3.59
3.59
3.59
3.61
3.59
3.60

Looks fairly good to me.
While trying to read the first cell, I realized that when the small cell voltage sensors wire are disconnected from the BMS the overall voltage pack as read from the BMS output is only 44V compared to 57.5V when it's connected. This prevents me from correctly reading the first cell voltage without opening the bms plastic protection.
So I went on and tried to see if using my light as the only charge would still trigger the BMS problem. waited 5-6 minutes voltage dropped to 55.9V, and the BMS was fine with this 10W charge.
At this point I thought : well if when one of the sensor wire is disconnected from the BMS I'm only reading 44V then a flakey connection would explain the BMS voltage going from 44V to 57.5V continuously. So I made sure everything was correctly connected and tried with the controller. Sure enough I was able to withstand full throttle with the brakes on resulting in 35A of using my motor as a heater
I them put everthing back into my triangle case and went on for a short ride. Worked great! So I'm thinking : great problem solved it was simply a bad connection somewhere.
Connected the charger back, went for diner, and then just to be sure everything would work out as expected, I tried the bike again... And sure enough, on the first start the BMS kicks in.
I'll try to see if I can reproduce the work around of discharging the cells with my light for 5 minutes, but now I feel like I'm back to where I started, which is that my BMS fails when my cells are fully charged.

Well using the light triggered the BMS again, and it was back in this 44V to 57.5V oscillation mode...
Disconnected one of the group of 8 wires coming from the cells, and plugged it back => same oscillation, tried the other one, same. Tried both BMS back to 57.5 V.
Light back on,BMS triggered again after a few
minutes, and now I can't get it out of this oscillating mode.
I give up for tonight, this is too frustrating, and it really looks like a BMS problem to me now : even if that first cell was bad ( which apparently it's not given the full pack voltage when the BMS doesn't screw up) that would not explain 44V as in my maths, 15*3.6 is 54V ( unless of course by some magic this cell would have a -10V on it... Grrr)
I'll see what ping can do for me.

I agree, I think you have narrowed it down to a bms issue. Just a new one for us. I had a hard time really taking in the whole story to diagnose it right, while you are there seeing it all. No way a 10w load should pop the bms on even a single ping cell charged to 3.6v.

Good news actually, replacing the bms is cheap and easy. But you are still cruising for a bruising with that 40 amp controller.

I've replaced the controller back to the 35A one, and will keep things that way for now.
Ping is goon to send me a replacement BMS, so hopefully I'll be back on my bike in no time. Till then, I have to figure out how I'm going to commute...

Gudy ping will help you though it. But keep us updated on the fix. You said you have a CA turn it down to 25amps. My neg. wire on the bms was only half solider and inside the 8 pin connecter the pins did not touch I pushed the wires in to for a good connection then hot glued each side of the plug so the pins stayed in place. After charging I let the bms bleed down the cells till the led blink so the battery will then be balance. Sounds right to me. What motor you running.

I believe it's the controller. I had exactly those symptoms before with the top of charge voltage being right at the limits of the controller, and over a certain point the controller would shut down after just a pulse of the throttle. Granted that was occurring at 83v with a 72v controller and two Ping 36's in series, but it was the exact symptoms you report. It's an obscure issue, but a frustrating one to track down if you don't know about it.

That was back in the days before my batt pack was permanently on my bike, so I just used a light bulb for a minute to remove that top of charge surface charge to reduce the voltage. If that's inconvenient, just turning your bike on for a few minutes without touching the throttle while you have breakfast or coffee may do the trick.

John

999zip999 I currently have a 9C 9x7 although I've made the delta/wye mod on a 9C 6x10 so I might switch when I get my battery problem solved, and fix some mechanical friction problems due to the added relays in the new hub
John, I really don't think it's a controller issue at this point, as the BMS shuts down the battery on 2 different controllers, and even on my light only.

The easy solution then is to bypass the BMS for discharge and use it for balance charging your pack. I did that very early on with my version 1 ping packs because the BMS tripped too easily with my 15ah packs in series.

Ping is sending me a replacement, but you're right that I could robable get away without the BMS for discharging... However I like the idea of having a short circuit protection, which I guess the BMS provides, right ?
On the charging side is it the BMS or the charger which has the HVC ? Are the ping BMS actually doing cell balancing?

the HVC is active for each individual cell in the signalab BMS you get from ping. if any cell goes above 3.9V during the charge, then the comparator built into each channel signals the optoisolator for that channel to shut off which cuts off the voltage on the charging FET gate. this turns off the charging FET and stops current from flowing into the battery until the high cell drops down below that level. this is why it is hard to balance a pack when it first arrives since there is a wide variation in the amount of charge the individual cells have and it takes a long time to charge up the low cells when the high cells keep shutting off the charge. in these cases, you can drain charge out of the high cells manually by shorting out the high cell with a small power resistor, i use a 5W 5 ohm ceramic encased power resistor for this.

This may be an unrelated question, but if I wanted to use a switch ( or a relay ) to disconnect my battery from the BMS ( instead of plugging and unplugging Anderson connectors, should I put it on the ground or positive wire ?
While I can't use the bike I might as well finish building it the way I want

Ping sold me a 60amp. bms he knows I running a 600w.h.s. BMC and a 40amp. crtstalyte with the v2.5 48v20ah split pack. I will buy a CA soon to turn it down to 25 amps., but he wants (s.f-ebikes) my controler too, so he can calibrate them. I though it was just plug and play I guess I'm wrong again. Hate to give up my bike for a week. And why do I need a 60amp. bms its twice as big in size. Yes I remenber now I keep tripping my 40amp. bms I though it was the controller till ping told me to put solider on the shunts to up the cut off, but I keep tring it on the hills good luck a mo. later I blow 4 pouchs and replaced them its still works find with my 48t-11t. = 34mph. ha ha

Gudy the ping does cell balancing when you uplug it the red lights will start blinking one at a time till they all bleed down to 3.31v. or something like that. After a few light charge discharge it may be closer to balance. good luck. On the 16pin connecter keep plug to salvage pins and wires.

Fyi,
My current workaround consists I'm not charging the battery completely ( I time my charges and then monitor till I get between 55 and 56 V)
I guess that's not ideal for cell balancing, but it does trick while waiting for the replacement BMS.

I use a 2c rated 48 volt 20 amp hour battery (v-power). I run a 40 amp crystalyte controller and have been very happy so far with the whole setup and there have been no issues to date. I really don't believe that you have too much controller for the battery. Like you said, you have informed Ping of all of the particulars of your setup and didn't get a no go. Sad that you have a problem and hope you figure it out soon. You haven't pulled all your hair out yet, I hope!

Replaced the BMS with the new one that Ping sent me ... and it didn't solve the problem
Turns out one of the wire from battery had a bad connection somewhere, which I believe I found, re-soldered it ... and that didn't fix the problem either
Anyway, I went to work with it today, started charging it using my 5A charger, until our assistant told me that my bike was smelling kinda burnt ...
Opened the battery case, and the heat shrink wrapping around the BMS had completely melted around one of the (what I believe is) MOSFET ... why it happened, I have no idea ...
I call it a bad day, someone stupid decided not to follow a stop sign and destroyed the front aisle of my wife's car... it's on its way to the repair shop ...
So, my bike has a serious problem, and our car needs a repair ...

... and I'm still not sure what was initially wrong with the bike, but now, in addition to that, I have a 2nd problem when charging ...

... and a brand new (gas) scooter would be about the same price as my electrical bike, would go faster, probably be safer (with the crazy car drivers not paying attention to bikes) and much less hassle I'm sure ...