BMS Battery Alloy Shell Charger Problems

[GCinDC]

reposting from other bms thread...

brand new 900W charger. plugged it and POOF:

then a minute later, still some poofing:

and the AC cord so hot it was melting:

ps. i have an identical 900W charger which is fine. this was for the guy who's buying my GT. he came by to check it out moments after the magic smoke came out...

 

[GCinDC]

pron for dnum

melted/broken ground wire?

board is caulked in place so will wait till i hear back from bms before attempting to slide out:

i've seen worse caps:

mini board attached perpendicularly:

lots of black caulk everywhere, presumably to prevent movement:

fets look ok:

nothing obvious to me but ground wire, and that might have been side effect of whatever blew, considering how hot the AC line got. no shorts in the AC wire tho.

 

[heathyoung]

UM.

Wow. Excellent QC there. Check line to ground, I'm probably betting that its got a short from a device to the heatsink.

Be glad you were not holding it when you turned it on.

 

[cwah]

My bmc 400W is down after connection. Fuse was melted. I changed it but it's still not working

 

[dnmun]

greg, since it shorted to ground and burned the ground clamp off, but did not blow the fuse it must have shorted in front of the fuse. so the leg in that fuse holder that goes down to where the fuse is must have shorted to the case. it should have blown the circuit breaker in your garage first though before it got the wire so hot it melted. it might have shorted to the wires that go to the switch too. that switch is actually on the high voltage on the other side of the rectifying diode bridge, and at 200V.

i am curious as to why the fuse did not blow. if it shorted in front of the fuse, the rest of the charger may be ok, but it might have shorted at the npn switching transistors on the side too because that would return current through the ground. see if the wire that melted the AC cord was the ground wire or if it was one of the two line wires, black or white. if it is green then it may be the switching transistor that shorted but not enuff current to blow the fuse. in that case you should be able to measure the continuity between the tab or collector of the transistor and the case.

 

[GCinDC]

it should have blown the circuit breaker in your garage first though before it got the wire so hot it melted.

it's weird. since the poof, my garage door opener doesn't work anymore.

but it's got current. i had a radiant heater on at the same time i plugged in the charger, but that continued to work after the Fry Zap, as does the power strip the charger was connected to and all the other electric.

would a ground short account for all that smoke tho? the inside cover doesn't show the typical cap spew, and beside the GND wire, i see nothing else burnt...

BMS is asking me to take it apart and send pics, so that won't be a problem. will report back..

 

[dnmun]

i am thinking now it has to be shorted from the switching transistors to the case. the current not high enuff on the main line breakers to blow the fuse but high enuff on the ground wire that it was overheating. there is no circuit breaker on your ground wire so it never blew the circuit breaker.

 

[999zip999]

GCinDC I had a fire in my garage and my bike was fine but a little soot. It worked the first day but the next day it seems the halls are out also the hyperion cellogs are wierd now ?. So soot is a electric mayham. So yes ?

 

[heathyoung]

I'm trying to remember what the fault with these was, the fact the ground burnt up has reminded me of something. It has something to do with the way they sense the 120/240V input voltages, measuring phase to ground for some reason.

I'm amazed your RCD didn't trip (you do have them don't you? Or is it not law in the US?)

I'll have to see if I can dig it up. Its a known issue IIRC.

 

[dnmun]

no, that is a separate problem that the kingpan chargers have now that they started using a thyristor daughterboard so they can use just one of the phases when it is 240V input so the charger doesn't have to use a switch for the rectifier diode inputs between 120V and 240V AC.

my theory is that the thyristor can not get a ground to switch against when it is a 2 wire 240V circuit like cargo tom had happen to him. i did the same thing by using a 2 wire extension cord on 120V, and the ground was not present on the right pin, so it could not get a ground reference to switch against. that's how i interpreted it. ping said he got a lot of recalls on these chargers with the daughterboard when kingpan did this to him.

he still uses a 120/240 switch on the higher powered kingpans. i told him he might wanna consider going back to using the switch on the low powered chargers again. but they want it to be 'universal voltage', it must be a strong selling point to some people.

 

[heathyoung]

Jaysus. What a shit design.

Why not go a proper PFC design? It isn't that much more expensive, and you can get rid of those god-awful IRC's on the input with a proper soft-start.

I suppose you get what you pay for. The TC / elecon chargers are much better, but big and heavy.

 

[Lenk42602]

Funny you ask, i talked to cell_man about those chargers the other day and he said that he orders his from a different company, and hasn't had a lot of problems with the ones he is currently selling.

Apparently these 'alloy shell' cases can contain boards from different manufacturers. Sorta like meanwells VS knockoff meanwells.

the units he sold late 2011/early 2012 had these same problems with reliability. he got them from Kingpan. these are both from him: http://endless-sphere.com/forums/viewtopic.php?f=14&t=45770

Dnum, i just found this thread, I will read through it all and get back to trouble shooting my units...

Len

 

[Stevil_Knevil]

Bummer! Vast quantities of magic smoke - and especially the frowny face above the bike says it all

 

[dnmun]

i think these things need a ground fault circuit breaker. this coulda happened anytime. it just changed my attitude about using GFCI breakers.

 

[amberwolf]

it's weird. since the poof, my garage door opener doesn't work anymore.

Is there a breaker in the GDO itself? On the outside of the cieling portion?

 

[dnmun]

is it possible your garage door opener shared the ground wire?

maybe it used the same wire for neutral that was the ground wire that melted the cable. it might have opened that circuit somewhere further along, like a loose wire nut or screw connection. is the circuit still functional except for the garage door opener? other outlets?

 

[Samson]

It suggests to me that the charger fuse is in L2 not L1 of the AC line. This can happen if the house outlet is improperly wired (reversed hot and neutral) or the extension cord hot and neutral is reversed or the chargers AC line cord is reversed. If a short occurs between case Ground and L1 then there is no fuse in the circuit and high current flows through the ground lead and the power cord if there is a short. Also likely that the short to ground was after the diode bridge. If so the components may have limited the current enough that the house breaker did not trip (it has a fairly long thermal time constant).

You need to check to see that the L1 L2 and ground wiring are correct. It is not uncommon to find Hot and neutral reversed somewhere along the line.

I have the same charger and the wiring is correct on mine, that is the fuse is in L1 the AC hot side and the house outlet and power bar are correctly connected. The AC line fuse in my unit is a 15 amp fuse BTW.

I have added a 5 ohm inrush limiter to my unit to reduce the start up spike to a reasonable 20 amps or so. This reduces the wear and tear on the plug connections and will increase the capacitor life. Another way is to use a two position switch to precharge the capacitors for a few seconds through a 10 ohm 10 Watt resistor and then switching to the ON\RUN position to bypass the resistor.

 

[GCinDC]

woah, i missed all these responses!

i hired a dude off CL to wire the garage after i built it. after this happened and someone first asked, i noticed that the circuit panel didn't have a door, and then i started thinking back - the clown may have hard wired the garage w/o a breaker, in which case it's on a 50A circuit off from the house. and i'm not sure what that's on, but it has to be on a circuit there... doesn' t it??

the garage door opener AC plug is in a recepticle on the ceiling, so yeah, i assume same ground.

i'll try to pop the panel to see what's in there, but it's hard w/ shoulder now and wife just went out of town for 3 days. :x

also, i'll pop open my 'identical' working 900W charger and see if anything's different!

fun, fun, fun

 

[heathyoung]

Ah - that makes sense, US doesn't run a MEN system (multiple earth neutral) - neutral is bonded to earth, so active/earth short blows fuse, as does active/neutral.

Messy.

 

[dnmun]

nope, not really what i was thinking. many older homes have just two wires that carried 120V AC. it could be a bad ground somewhere too. but i am now thinking that the reason the ground burned but did not blow the 20A fuse in the charger is because it was shorting from the high voltage on the front end. from the rectified 170V DC that goes to the collector of that first big npn transistor in the oscillator. the one attached to the heat sink under the clamp. so it coulda been 35A+ without blowing the fuse in the charger. i think the 35A woulda been hard on that 18G ground wire. so i think that is how it smoked up the garage.

so if it is just a short between the switching transistor and the heat sink, then it could easily be fixed by replacing the two transistors and the rubber insulator where it shorted. maybe only one, but they are kinda matched, i have some of those too. i bot some on ebay.

fairchild J13009 in the TO-3P package.

thinking some more about this, if it is only the rubber insulator being bad and the DC was shorted directly to the case, the npn switching transistor would be ok. it may only need the rubber.

 

[GCinDC]

reply from bms

Hi Sir,
Firstly,please check if the fuse is OK . If it's OK ,please change a AC wires(thick one) ,and connect that yellow wire (that one your finger showing) and then try again.(you can leave the cover open)
Best wishes!
--
Best regards

Sincerely yours,
ShunWu Yu 余顺武

Company Name: Ecity Power Co., Ltd 广州亿城
...

 

[dnmun]

were you giving them the finger?

 

[GCinDC]

haha

 

[migueralliart]

Any options to different suppliers that might give us a better luc?

 

[dnmun]

if you can, get the voltmeter probes onto the legs of the switching transistors behind that capacitor next to the ground lug where your finger is. the left leg facing it is the base, the middle leg is the collector. measure the resistance from the collector to the heat sink. for each one.

that will help them understand.

http://www.fairchildsemi.com/ds/FJ/FJP13009.pdf