My EM3EV battery will not turn off? Help! What should I do?

[inkeeper77]

Question: I just got a new battery from EM3EV, a samsung frame mount battery with 29E cells, 48V, and 11 AH. This is the battery: http://em3ev.com/store/index.php?route=product/product&path=35&product_id=161

After charging it today, the charger automatically cut off, and the battery is at 54.2V according to my cycle analyst. The battery will not turn off though- the LED stays on in the on and off positions, and the battery provides power no matter what the switch is at. Should I be concerned about a fried BMS, or is the BMS burning off extra charge?

 

[mlt34]

It's not going to ruin your BMS, don't worry about that.

It will very slowly bleed off energy though. If you left your bike for a month or so, you could come back to it and find the battery dead.

It's probably as simple as easy as replacing the switch (I got a battery with a bad switch once or twice) but you should contact paul at em3ev first.

 

[inkeeper77]

That's good to hear- I was afraid that this could be a symptom of a malfunctioning BMS rather than a malfunctioning switch....

I've emailed paul, but I was a bit worried about charging/discharging the battery in this state. Thank you!

 

[mlt34]

For now it's just fine, but yea you'll want to fix it so that it's fine in the future if you ever have to let it sit for a few weeks on vacation...

 

[inkeeper77]

Now the second (identical) battery has the same behavior.. all I did was charge it up, put it on the bike, and ride a bit.. perhaps it's a bad batch. I'm anxiously/eagerly awaiting Paul's email batch for today....

 

[dnmun]

not sure what you mean about 'burning off the extra charge' with respect to the led. the balancing of the cells with excess charge is totally separate from the discharge circuit of the battery and would have nothing to do with this observation you have described.

from reading his description in that sales brochure it appears he has adopted the practice of turning off the circuit current of the BMS in order to turn off the battery. when the mosfets have no voltage at the gate from that circuit current then the mosfet is turned off so the negative lead of the battery is disconnected and the battery is 'off'.

that was what he meant about not needing a precharge circuit and that the discharge current does not go through the on/off swith since the discharge current goes through the mosfets and not the BMS's control circuitry.

the led would be on the post BMS section of the negative wire to the battery. the positive to the led would be through a resistor from the top of the battery.

if your charger is still attached to the battery when you are testing the on/off then the led may remain lit even if the BMS mosfet is turned off.

 

[inkeeper77]

not sure what you mean about 'burning off the extra charge' with respect to the led. the balancing of the cells with excess charge is totally separate from the discharge circuit of the battery and would have nothing to do with this observation you have described.

if your charger is still attached to the battery when you are testing the on/off then the led may remain lit even if the BMS mosfet is turned off.

I didn't know if it was a possibility for the BMS to keep itself on even when sold to switch off, perhaps for purposes of balancing unbalanced cells.

The charger is not attached to the battery- both have been used and tested, and simply do not turn off (The switch truly does nothing. The LED stays lit at all times, and the battery will power the bike in either switch position.) - This means I cannot turn the battery off before connecting it to the bike, allowing arcing to potentially occur (I am careful to turn off the controller before removing or inserting the battery, but it still isn't ideal) For a little while the first battery behaved as expected, turning off with the button and not allowing any current from the main connector.

 

[dnmun]

in that case you may have shorted the mosfets or the switch itself no longer functions and is shorted. did not know the switch had worked originally. the spark to the controller will only occur if the battery is on. the circuit current switch on the controller does not stop the spark from connecting the live battery to the controller.

 

[mlt34]

Can you access the backside of the switch? I'd do a quick continuity test (the beep mode on your digital multimeter) to see if there is continuity in both switch positions. If there is, that means the switched failed 'on', which basically means it's like there's no switch there at all. It's better than failing 'off' which would be like a broken wire, but either way you should eventually have the switch replaced. If you can determine that is the problem then I'm sure paul would just send you another switch.

[edit: it's important to remember to remove the switch from the circuit before testing it with your digital multimeter]

 

[dnmun]

the switch would be in an active electric circuit so do not use the ohmeter of your DVM. it will burn up.

 

[inkeeper77]

EDIT: Ignore this post, I somehow in a morning not quite awake haze forgot I could get to the connector to test continuity.. nothing to see here

 

[dnmun]

if the battery does not turn off when you disconnect the sense wire cable then you can worry about it. do not put the ohmeter on the circuit if the sense wire cable is plugged in.

 

[mlt34]

I would also wait for Paul's response. Especially if warranties are involved...

 

[dnmun]

there is no reason not to test it to see if the mosfets are shorted but it is unfair to tell someone who is totally inexperienced to put an ohmeter in an active electric circuit. i tried to help him but he wants to do whatever you tell him to do. why do i even try? people don't even care to pay attention and then there the ones who insult me about how i don't know what i am talking about. or my stalker who demands that 3.65V has to be the LVC for lipo. as though he knows anything.

 

[inkeeper77]

if the battery does not turn off when you disconnect the sense wire cable then you can worry about it. do not put the ohmeter on the circuit if the sense wire cable is plugged in.

That is the case.. and I checked continuity after disconnecting the power switch- the switch is fine. (I was being not quite awake earlier and somehow forgot I could just check it at the connector itself.. I blame a lack of caffeine at that point. Not my proudest public moment )

there is no reason not to test it to see if the mosfets are shorted but it is unfair to tell someone who is totally inexperienced to put an ohmeter in an active electric circuit. i tried to help him but he wants to do whatever you tell him to do. why do i even try? people don't even care to pay attention

Don't worry! I didn't use the ohmmeter on an active circuit, I disconnected the switch before checking. Your warning was a good point, especially for anyone who finds this thread in the future, though I wouldn't have put an ohmeter across active voltage. (I did that once a few years ago while quite inexperienced- replacing fuses in my DMM made sure I would remember to be careful in the future! I didn't note this in my response because I thought you were just making the prior post more clear and telling me that I of course should disconnect it before testing continuity. (which I read, thought "yup that sounds right" and then did the test) (See above the explain my confusing post about insulation. I'm not always a morning person.. )

Results: the switches are functioning fine, the BMS units simply do not respond to them on either battery. Apparently this is a bad batch of BMS's or a bad charger or something.

Paul responded- apparently he was on a business trip explaining the later than normal response, which is of course totally fine (He still responded in about two days- quite good compared to many businesses! I have been quite happy so far, even though I am new to anything with a BMS thus this thread) Based off this information he thinks the BMS units themselves are bad, and we will get them replaced. I need to determine now if they are still working to protect the cells, just not switching off, or if they are somehow completely fried.

 

[mlt34]

there is no reason not to test it to see if the mosfets are shorted but it is unfair to tell someone who is totally inexperienced to put an ohmeter in an active electric circuit. i tried to help him but he wants to do whatever you tell him to do. why do i even try? people don't even care to pay attention and then there the ones who insult me about how i don't know what i am talking about. or my stalker who demands that 3.65V has to be the LVC for lipo. as though he knows anything.

you're absolutely right. I was assuming, but I guess I should have been more clear about disconnecting first. It's good you put the warning there though, and I'll edit mine to make sure anyone in the future sees the warning right away.

Why try? because you do a great service to the ebike community with your guidance and assistance!

 

[mlt34]

very interesting, I have to say I'm surprised it was the BMS and not the switch. I'm sure Paul will take care of you though. He's one of the good ones!

 

[dbIsMe]

I ran into the same problem as innkeeper described above. My frame mount battery from em3ev was manufactured around the same time (ordered mid Aug, arrived early Sep).

Paul from em3ev will be shipping out a replacement BMS for me also.

In case anyone else runs into this, I'll just add on that Paul said "In the meantime you can use it, but do not discharge the pack deeply, as it could damage the cells. Power on the battery when using it (switch in)".

Thanks for sharing the info to start with. It was nice to know it wasn't just me, especially being a noob.

 

[dnmun]

but you can test the output mosfets on the BMS and then we would know if the BMS was not functional. when paul observed that you should not over discharge the pack, (it was because the symptoms were that of shorted mosfets so just simple test would show that.) because then it was possible that the BMS would not turn off at the LVC.

 

[3Dkiller]

I have the same issue i contact Paul and he did send me a new BMS