I think my charger is dying...anyone know a possible fix?

[joshseitz]

I have a 2.5 year old 48V Ping w/ the 60V 5A charger that shipped with it as an upgrade. Lately it has been behaving erratically. I'll plug it into the pack and it will start charging for 30 seconds or so, then stop for about 10. Today, it charged, then paused, then charged some more, then stopped for good. This happened once before a couple of months ago and I disconnected everything and let it sit for an hour. Upon re-connection all was well. Not so this time.

There are two LEDs on the charger: red indicates that it's powered on, and green indicating that it's charged. The green LED is also illuminated during these periods when charging pauses.

The charger is vertically mounted on my garage wall, with the exhaust fan facing up. It is exposed to some standard garage dust and it can get pretty hit in the garage.

In addition, I have hit LVC on the pack a couple of times in the past couple of months. The first time (and this happened about 2 years ago too), the fets on the BMS got super-hot upon recharge, melting the plastic shrink wrap and generating a ton of heat on every re-charge.

Is it possible that my BMS is causing the charger to think the pack is charged? Should I replace the BMS?

 

[dnmun]

nope, your pack is out of balance so much that the BMS is having a lot of trouble balancing it. nothing wrong with the BMS.

put the battery on the charger and measure the cell voltages for each cell, and list them here. measure while charging. do not disconnect the charger and wait until the cell voltages drop as some people think it has to because they think that is balancing. measure the voltage while charging. post up the list here so we can see how bad it is.

post up pictures of the BMS here too so we can see how burned up.

how long did the pack sit around unused? did you disconnect the sense wire cable when it was stored?

 

[dogman dan]

My bet is that your charger plug is wearing out, or rather, more likely just the wire where it solders onto the plugs. It could be on either the charger itself, or the battery side.

Have a look at that first, before you spend on a new bms. Unplugged gives the same green light as the battery is full. It could be just that simple.

If it's not that, then it's time to take a closer look at everything, When a battery is out of balance, it can go like that a long long time while it balances. On off on off, for hours or even weeks. If you have a really out of balance cell, it could be you have a problem that the bms is trying hard to deal with.

If that is the case, all this on off stuff would be happening at the end of the charge, not when you first plug it in. That's why I think you just have an intermittent connect at the plug, since it's happening early in the charge.

 

[joshseitz]

I'll try to post some pics of the BMS - it's pretty toasted, but when the charger worked, it always shows all cells peaked.

I've never left it unused for more than a couple of weeks at at a time. I always left it in a charged state, but left the sense wires connected.

I'll try replacing the charging connectors, as that is the path of least-resistance, but I have a feeling something else is going on...

thanks for your help.

 

[dnmun]

post up a picture of the signalab BMS. once you get it out in the open where you can see the charging mosfet, then it will be obvious if it has melted the solder off the pcb.

you can measure the voltage on the gate of the charging mosfet to see if it is turned on, and then when you disconnect it you can use a voltmeter to see if the mosfet has failed.

pm me and i will help by emails. i have parts and other BMSs too if you need.

 

[dogman dan]

Melty plastic on the bms doesn't mean it's fried, it could have just toasted the plastic. Melted solder would be bad though.

Usually when you open a bad plug, you see the problem right away. Looky there, that wire is broken off and makes contact from time to time.

 

[joshseitz]

I've inspected the XLRs on the battery and charger, and both are still solidly soldered.

Here is a shot of the toasted BMS. It has been in this condition for a while now with no issues. I suppose I should probably replace it regardless.
The first one I fried a couple of years ago melted the solder, but still worked.



Does anyone know why these get toasted when recharging following LVC? Is there a current spike from the charger when the cells fall below a certain threshold? Is 5A too much charging current generally?

Do any of you guys have an extra/used 48Vcharger to sell?

 

[joshseitz]

Any thoughts on something like this? http://www.ebay.com/itm/48V-3A-Charger-For-LiFePO4-Battery-ebike-DIY-220V-110V-By-air-mail-/140763911555?pt=Battery_Chargers&hash=item20c62ed583

Junk?

 

[Burtie]

The hot charging FET is an indication of a problem on the BMS.

Below High Voltage Cutoff, the BMS should turn the fet hard on (virtually zero resistance), so during charging at 5amps, it generates no heat.

If one or more cells reach HVC, the BMS should turn the fet hard off (infinate resistance), so it stops the charge current and generates no heat.


The only way the fet gets hot, is if it is partially switched on (cased by fault in BMS HVC circuitry), in this state it has significant resistance, so the charging current heats it up.


I had a Signlab BMS fail in a similar way to this. In my case, I found several of the opto-isolators in the BMS had failed, -become leaky.

You could, of-course bypass the BMS temporarily to prove if your charger is ok


A common charger problem is broken conductors in the cheap leads where they get bent a lot going into the charger case, easy to check for if you have a meter.

Burtie

 

[dnmun]

i have noticed that some of the opto transistors can eat up a lot of the voltage that starts at the begnning of the opto chain in a way similar to what burtie mentioned. on the v1 but the v2.5 has a different design for the charging opto transistor series.

measure the voltage on the gate leg of the output mosfet and the charging mosfet.

your picture does not show it has burned up. what are you referring to when you said the BMS burned up?

 

[joshseitz]

You could, of-course bypass the BMS temporarily to prove if your charger is ok

How exactlyu would you recommend that I bypass the BMS? Just charge throught the power output leads? Thanks.

 

[dnmun]

sure. why not? you already know its the charger that is dying.

just connect the charger up directly to the battery without the BMS.

currently you have some cells that are charging up to 3.9V and shutting off the charge for HVC.

you can disconnect the BMS and charge them up some more to prove if your charger works.

use your voltmeter and measure each of the cells while charging so you will know when they climb over 4.2V so you can turn the charger off. if you can't stop the charging above 4.2V then the cell will end up losing a lot of capacity very quick.

good luck.

 

[joshseitz]

If the charger was over-charging certain cells, wouldn't the LED on the BMS light up red, as it usually does when charged to indicate they are peaked?

Just to clarify, when charging without the BMS, I am still charging the pack as a whole (at 60V), not charging each cell independently (at 3.7V) right?

 

[joshseitz]

measure the voltage on the gate leg of the output mosfet and the charging mosfet.

So I believe you are asking me to measure across the terminals....if so, the charginge fet (the one that gets super-hot) measures virtually nothing when charging or sitting idle. For the sake of reference, the remaining four FETs measures about 13V.The overall pack voltage (according to the CA) was 54V, so that just about makes sense.

I was able to get the charger going for a bit tonight. I turned off the charger, connected the battery (tiny "pop" as connection was established), flipped the switch on the charger, and away we went...for about 2 minutes or so. I repeated this overe the course of the evening in the hopes that it would "catch on" and just keep charging. No dice.

I'll order a new charger from China and see if that does the trick. Valuable riding weather is fading away....

 

[dnmun]

if the voltage on the gate of the charging mosfet is zero then the mosfet is turned off. that is why the charger light is not on.

if you measure the voltages of the cells while on the charger you will see which cell is high and turning off the BMS.

 

[RICK]

I just replaced the charging fet on my signalabs bms (ping 48v 15 ah 2a charger). It was hot enough to desolder so i figured better to replace and then make sure it turned on properly. Used a 4110, now it runs cool (really cool) - also soldered on a little tab of sheet brass as a heat sink. If those fets are fully on their on-state resistance should only be 10? mohm, so it should only have to shed 50 milliwatts for the 5a charger. However maybe they are chinese knockoff fets, rejected parts, who knows? The on I took off had the numbers sanded off so that doesn't inspire confidence. One thing for sure, if it starts generating heat there is nowhere for it to go on that board.

Replacing the fet wasn't too hard; but i had access to an electronics repair bench and had done similar things before. You'll need a powerful soldering iron, some solder wick to remove old solder. A buddy with some experience with printed circuit boards would be a big help too. I'm sure there are plenty off youtube videos to get a general idea.

This was all coincident with putting on a new BMS, when I figured out I had a low cell by bouncing off the LVC well before I should have. Separate story here, I had damaged the old BMS and one cell ended up badly imbalanced. With the new BMS installed (where the balance LEDs were all operating properly) I ended up charging one cell separately to bring into balance (details below). Did the imbalanced cell fry the FET on the new BMS? I don't know. You'd think it should be on or off and either state shouldn't produce much heat. Maybe the new BMS would have been fine if one cell wasn't way imbalanced, but that doesn't strike me a good design.

Does your battery balance? Do all the LEDS light up and do they go off shortly after using it from a charged state? If one or more cells are badly imbalanced you can charge thru the balance leads (with a separate DC power supply of course). This is a corrective measure to bring the battery back into a balanced state, not something you should need to do typically. I had to bring up a unbalanced cell (resulting from a damaged BMS), pumped about 3 A into it for an hour and a half to bring it up to 3.9V. Now the battery balances fine. Be (really) careful not to exceed max voltage of 3.8-3.9V; or fire and destruction of battery may result (and maybe garage/home?). Like I said - really careful.