BMS Battery Alloy Shell Charger Problems

[GCinDC]

dr messed me up yesterday! followup appt, and he yanks it all over. hurt worse than day after surgery.

he's got me on steroids + perc now tho, so seems like a good time to get back into high voltage dc stuff...

and the fact that BMSBatt wants me to fix it, then turn it on with the top off - reminds me of that post, ypedal, where someone recommended charging lipos w/ a facemask? :lol:

anyway, tiny question: i probed that yellow insulation w/ the DMM checking for continuity w/ GND and even 1/3" in, i couldn't touch metal... and if you notice at bottom of pic, it and another wire are heatshrinked where joined to some component. should i just strip as much of yellow insulation as possible on that wire and try to twist splice a new wire in? doesn't seem like i have much of a choice...

 

[dnmun]

no, the case is shorted to high voltage. don't plug it in again.

 

[Samson]

You may have to replace the yellow ground wire as it may have melted inside. Also be aware that the case is anodized and therefore a good insulator. The connection to ground really should have the anodization removed around the ground screw to make a proper connection. Also the top of the case should have a separate ground connection as it as it to is anodized.


Do not plug it in until you fix the short circuit, and do check that the outlet has the neutral and ground connected correctly so the charger fuse can protect the charger.

 

[GCinDC]

Damn, I'm glad I asked!

and just to double check - I'm pretty sure I mentioned it before - your diagnosis is consistent with the fact that I was able to plug it in to the wall and the output to a fully charged battery twice before without explosion and have the green light come on each time?

The only time it popped was after i plugged the ac then output and it started trying to charge.

 

[dnmun]

you have a short from the high voltage that is on the output of the diode bridge. the current flows from that high voltage circuit into the case somewhere, and then it is returned to the ground terminal of your service panel through the ground wire. it stopped burning up the ground wire from the current surge when the few strands of wire in that lug burned up and left the ground open.

i am not sure where it is shorted, i figured it shorted through the insulator on the npn transistor. if you have a switch to turn it on and off, that switch is also on the high voltage so it could be shorted by the leads to that switch, but the wire woulda melted too like the ground wire.

 

[GCinDC]

you have a short from the high voltage that is on the output of the diode bridge. the current flows from that high voltage circuit into the case somewhere, and then it is returned to the ground terminal of your service panel through the ground wire.

ok, this makes sense. and the short is an error, missed in QC, that i have to detect before testing again.... seems like a replacement is in order. i don't have time for this... maybe if i could slide the board out, but the fact that it's glued in turns this into a pia job... despite being completely uncalled for...

fwiw, here's a side by side, not the best res:

the numbers on the big white thing in the middle are different... one side is EE55...

 

[dnmun]

yes, it is serious. if you can measure the resistance between the center leg of the npn transistor on the heat sink, between that collector leg and the case, i think you will find it is zero. if you take picture of that with the probes on the npn tansistor and case, i think they will realize what the problem is.

the rubber sealant is just on the end, you can cut it off along each side and tear it out with needle nosed pliers. but the other stuff has to be unscrewed from the case too. to get it all out to work on.

but before that you can test it by unscrewing the clamp that holds those transistors to the case, pull them away from the heatsink so they are not touching it, and then see if will run without shorting out this time. you can test it by using your voltmeter when it is plugged in then, measure the voltage on the case to the ground on the wall outlet. if it is still 170V then the transistors are not the cause. put the charger on a non conductive surface to do this.

 

[GCinDC]

BMSbattery has agreed to build a replacement and i agreed to pay shipping ($60) and hopefully it won't take the full 3 weeks...

i'm not supposed to lift anything heavier than a coffee cup w/ my left hand now, but when i get use of my hand again, i'll see if i can try you suggestions (and blow the whole garage.... lol)

seriously tho, now i gotta worry about getting a working garage door again! the opener/system i have is no longer for sale, so i might have to replace the whole thing!

and i discovered this weekend that my wife's bionx charger is dead, it had been plugged in at the time. that really sucks.

maybe time to check the circuit too...

ah, ebikes!

 

[dnmun]

you can wait for your arm to get better. not much to do in any case. it is a surprise you have lost other devices attached to that circuit. i am assuming something has happened to the electrical connections of the wires in that circuit. i don't see how a short on the ground could make them fail so it has to be in the wire connections imo.

$60 is too much for shipping. they can buy another charger and send it to you for that much. i thought several others got their charger replaced for free. but i am fairly sure your charger can be repaired.

 

[GCinDC]

According to tracking number, the 900W BMSbattery replacement is now in Chicago and should arrive Thursday.

Should I set up a live streaming broadcast of the first charge, in case we have the same fireworks?

 

[dnmun]

you actually were lucky before. the case reached 200V after the wire burned open on that lug. if you had contacted a water pipe or something else grounded while touching the case, it woulda been a severe shock. 200V DC is really a lethal risk.

i think it is clear that these metal alloy shell chargers should never be used without a ground fault interrupted circuit or outlet.

you can get the GFCI outlets at home depot for $12-15 i think. i still have a few i found in a dumpster a long time ago.

your short to ground has changed my mind about using GFCI on these chargers.

when you wire them into a circuit, as i understand it, if you use the gfci outlet in the first of the series of duplex outlets in that circuit, then all the duplex outlets beyond the gfci outlet in that circuit are protected too. as long as the ground returns back past the ground on the gfci outlet when returning to the service ground.

 

[GCinDC]

garage circuit panel was in a remote part of the jungle but i managed to bushwhack in, evade restless natives, remove panel and photograph it:

the main comes underground from the house and is on the top two 30A circuit, so 60A i guess:

so yeah, i need to get some gfi's going!

ps. btw, worst design ever, but i managed to replace the garage opener fuse and it's now working again.

 

[dnmun]

you can use a 20A GFCI breaker in that service panel in your garage. they never installed a breaker in that box. just don't touch it if it smokes like before. at least not while grounded.

 

[GCinDC]

Fyi

hello sir,

The charger has protection function , it must be connected to battery to activate it . So you can connect to the battery , then disconnect , to get the real charge voltage.

 

[GCinDC]

replacement charger charges like a charm.

spent a minute with the dud finally. replaced the ground wire with the big black wire here:

plugged it in against advice and no lights.

then i noticed the hidden fuse holder!

so i replaced that with a good fuse and plugged it in and no lights.

so i sliced thru the caulking that secured the board and took a peek. no giant burn marks:

long wire ends, and patched that were sanded. and the center slot unsoldered, but since there's no metal around the hole, should that be left unsoldered?

 

[dnmun]

yes, it should be left unsoldered. it is a locating pin to make sure it is installed correctly. that is the bottom of the relay on the output. there is diode next to it that allows current to go into the relay coil if the polarity of the battery connected is correct.

 

[dnmun]

look at the side where the transistors were seated on the rubber insulator and see if there is a burn mark. look where the rectifier diode bolted up too. duane actually shorted out his through the legs of the diode bridge to a loose ground wire.

the last charger duane brought me was so bad the fuse holder had melted into the socket, had to dig it out with my screwdriver.

in this case he had put the charger back together after repairing it and the choke on the AC input shorted to the metal stub on the back of the AC input plug. no clearance and the ceramic cap shorted to the AC, maybe totally outa phase too. bad.

 

[Alan B]

garage circuit panel was in a remote part of the jungle but i managed to bushwhack in, evade restless natives, remove panel and photograph it:

...

Greg, consider getting this panel reviewed by a proper expert. It appears to be wrong to me (though I am not an electrician). It appears to have only two connections, and grounds and neutrals are on the same bus bar. I don't believe this is correct, and it probably should have a separate breaker on each outgoing circuit as well.

Ground and Neutral are interconnected only in very specific places in an electrical system, and I don't think this is correct here.

Note about GFCI's - they operate by subtracting hot side current from opposite side/neutral current and looking for zero missing current, so they actually don't care about ground, just that the current isn't "leaking" out via some other path. They can protect downstream outlets if they are wired "through" the outlet in both hot and neutral. The grounding doesn't actually matter.

 

[dnmun]

it's ok, he said there is a breaker inside the house that supplies this box. but he thinks it is a pair of breaker when in fact it is 120 in the service box, not 240.

the box is wired to code. if he wanted to he can go back to the service box inside the house and rewire that circuit to 240 and bring 240 out to the box in the garage. he has space for 4 fuses, one that goes to the charger should be a GFCI.

so you are cool, but if you wanna rewire that, i will have to explain about the white wire coming in too, but as it is in the picture, there is nothing illegal, just no breakers.

 

[GCinDC]

plugged it in against advice and no lights.

never assume that your AC cord works... WTF??

the original was cooked as you may recall, so this was a replacement from my A30 PSU that mysteriously stopped working a while ago...

so with the ground wire fixed in the dud, i put in the fuse from the working charger and plugged it in for 3 seconds...

it should be noted that plugging it in as is was against advice. :| but at least i used a GFI kitchen outlet.

 

[dnmun]

i have a box of 100 fuses if you need them. did it blow the GFCI?

if you still have a fuse, try again but first loosen the clamps that hold the transistors to the side of the case, and pull them out away from the case so they don't make contact, then plug it in to see if the fuse blows then, and it it blows with the transistors not touching the case, they are not shorted.

to try testing the other high voltage part connected to the case, loosen the rectifier diode bridge from the case and pull it free. i figured one or the other is shorted to the case.

 

[GCinDC]

try again but first loosen the clamps that hold the transistors to the side of the case, and pull them out away from the case so they don't make contact, then plug it in to see if the fuse blows then, and it it blows with the transistors not touching the case, they are not shorted.

the whole enclosure was off last time i plugged it in and blew that fuse. maybe i should check continuity on the GND line attached to the board... not the one attached to the case.

so how does the fuse work anyway? it's connected to one red wire/one AC lead and the other side isn't connected to anything??

so the fuse is right below the AC leads, shown here from outside:

the bars can be seen here from inside but one AC lead isn't connected to anything???

the device uses only the ground and one AC lead? maybe i'm an idiot but i'm not sure i understand how this circuit works! :|

try testing the other high voltage part connected to the case, loosen the rectifier diode bridge from the case and pull it free. i figured one or the other is shorted to the case... <and from another post:> look at the side where the transistors were seated on the rubber insulator and see if there is a burn mark. look where the rectifier diode bolted up too. duane actually shorted out his through the legs of the diode bridge to a loose ground wire.

i have no bleeping idea what the RDB is. is it shown on any of my pics from before?

 

[dnmun]

ok, if the entire case was off, it is not shorting to the case. not sure how the current got onto the case and your ground wire then. the liitle metal bar on the side connects the fuse to one of the legs of the plug.

maybe it is shorted on the AC input, not the high voltage like i thought.

 

[miro13car]

regarding trimpots assignment:
I report that
blue trimpot marked as #1 on the picture is really for voltage adjustment, turing it counterclockwise dicrease output voltage
my EMC-900 has identical logic board and identical Volts/AMPS display boards.

 

[999zip999]

So how much is shipping a 900w from bmsbattery to USA