BMS Battery Alloy Shell Charger Problems

yes, the 4140 and 4139 are stable mates. different current ratings.

http://www.sanken-ele.co.jp/en/prod/semicon/pdf/2sc4140e.pdf

and i found the 4139 on page 92 here:

http://www.ibselectronics.com/pdf/ac/Allegro/Allegro_Power_Transistors.pdf

and the difference seems to be 8A for the 4139 and 10A for the 4140.

i use the big FJP13009 in the TO-3P package for the npn switching transistors because i got 10 for $25. on ebay, cheap cheap cheap. my first middle and last name.
 
now i'm talking about a 750w kingpan, as the 900w models are all working again, and have been modded with a amp/voltmeter.

this is the board:

CIMG6446.JPG


the charger is not outputting any voltage. the (normally) green led is RED, and the normally red led is dead. there is some buzzing noise - which all chargers emit to a certain degree depending on the output voltage.

what's weird it, that all red circled parts get HOT. even w/o outputting anything. it's not possible to measure any voltage as my voltmeter can't measure AC/DC. it flips between AC and DC showing different values. voltages at the 4-pin on in the upper left corner are ok - compared to the working 750w charger i have.

those are the parts that get hot (apart from the transformer)

CIMG6447.JPG


CIMG6448.JPG


any ideas?
 
dnmun said:
use the diode tester function on your meter and test the schottky diodes and post up the forward bias for each pair.
Click to expand...
the FMG36S schottkys seem to be ok. closing in one direction, and 370/390mV in the other direction.
next step i could test the two 2SC4140 transistors? what do you think?
 
were all four diodes the same? none of them was shorted? it only takes one. you only have two readings. and they show a dramatic dispersion at 1A. but they are in parallel so i guess that is all to expect.

the npn transistors are usually obvious when they blow up but you can use the diode tester on the C-E and look for the base to emitter to collector bias, and reversed to see if they are open, except the base emitter will have a base resistor that shows up.
 
by 4 diodes you mean 2 diodes per schottky. correct? as i said they are all the same. closed in one direction, 370 and 390mV in the other direction. i could measure the working one of the other charger and see i those work.
btw: the schottkys where desoldered while doing the testing.
 
you don't have to remove them to test them. but i realized after i asked for four that they are already in parallel so they are ok. just surprised the bias was not identical to .001V. but if the switching transistors are bad they will show it. they will be black and have smoke on the sides. you don't have to unsolder them either to test them.
 
i never know which ones to remove to test. sometimes there are cap or resistors connected to them that can give misreadings.
tested the transistors as well. both measure ok.
what's next? why are they getting hot w/o load? something "behind" them must be broken that produces load?!
 
i never remove parts to test them. sometimes the other elements in the circuit will affect the readings but you can just observe if there is a resistor or other passive element in the circuit you measure.

that is true here for the base/emitter diode test when you would expect the reverse of the p/n junction should be open circuit but you see about 2.2ohms resistance on the meter. so you can see the base resistor and it is 2.2ohms so you just put 2 and 2 together. a pun it seems.
 
i look for the 340V from top of the collector on the hi side transistor with the black probe on the emitter of the lo side transistor and then look for the base voltage on the lo side transistor, then the emitter voltage on the hi side and then the base voltage above that on the hi side transistor. but i don't know if that is a way to test.

i am just learning this stuff, electronics, and most of the people who really know never post up help for little stuff like this, not what they do.

if there is no voltage on the two transistors then go look at the rectifier diode bridge and see if there is 340V from top to bottom and if there is 170V in the middle too.

or you can test the rectifier diode bridge with the diode tester on the voltmeter. you should see the diode on some leg combinations, always seems to be two, and then open on the others. but if they are shorted then that is bad. it will show up as 0V on the diode tester.

i have found that before too. shorted rectifier diode bridge.

i measure resistors and diodes in the oscillator circuit in front of the big npn switchers and i check the two small signal transistor in the primary oscillator circuit next to the TL494.

there is a circuit diagram of the kingpan charger circuit here and the BMS battery circuit is identical:

http://www.avdweb.nl/solar-bike/electronics/portable-lightweight-lifepo4-ebike-battery-charger-800g.html#h0-1-2-5-the-battery-charger-circuit

thanks to albert vandalen for this. otherwise there is no circuit diagram i have found.
 
thanks for posting the schematics. tough i can not read it ;) it still makes sense by cleary identifying and naming parts on the board.
the rectifier B1 looks good. i measured the same voltages and diode readings as with the good one.
then i measure voltages of the TL494 and found that even pin13 and 14 showed 5v. pin12 was not showing 20v but 16v.
as well U3 LM7812 for fan power was weird.
input/ground pin1/2 was 12v, but output pin2/3 was 18v instead of 24v.
i don't know which two transistors you're referring two. maybe you could use albert's schematic's naming as well. you must know " 340V from top of the collector on the hi side transistor" means very little to me. i have no idea what the collector would look like, and what the "hi side transitor" looks like. i try to compare values of the good charger with the broken one, and try swapping parts. i can identify cap, resitors, transistors, diodes, you name it, but i following all the traces and knowing what does what is way beyond my knowledge. i know this is very little, but i try improving over time as well :)
thanks a lot.
 
rather than lecture you, you can read what wikipedia says about a bipolar transistor.

there is a pn junction between the base and emitter. that pn junction is a diode. when you put the probes on the base emitter the diode will conduct and you read the forward bias. the collector-base will also conduct, with a little lower forward voltage because of the way the epitaxial silicon layers are used to make the transitor.

read about the transistors so you have some basis to think about them, then find a transistor loose in your junk pile and start measuring it.

those big npn swittching transistors on the side next to the rectifier bridge are the ones i was talking about. the collector is the middle leg and tab is also collector, the base is the left leg and the emitter is the right leg.

for the little SOT-23 transistors, the single leg sticking out is the collector, and with the collector facing away, the leg on the left is the base and the one on the right is the emitter.

the TO-92 packages are not so uniform. it can be ECB or CBE, you have to check the data sheet.

that's the only way i can tell you to learn it, just like i did, looking and measuring and reading the data sheets. and analyzing the circuits too. by following the traces from component to component and figuring out what each does and why.

go read the wikipedia stuff on transistors, learn the color codes so you can read resistor values, and how to recognize what is a diode and what is resistor or cap and why a pnp transistor is used in one place and not another and vice versa for a npn transistor and why the mosfets are oriented with the drain of the n channel mosfets at the higher voltage than the source, always. for n channel mosfets, and the drain is lower voltage for p channel mosfets. so you can guess what kinda mosfet it is by its location in the circuit.
 
thanks dnmum. even though it was not helpful at all. i'm not going to learn how to design a charger, and how to build one by myself. i like to play around with things, and if i can repair them that's fine. but no way am i gonna invest weeks or month of reading to learn electronic engineering. i DO know what resistors, cap, diodes, transistors look like. i DO know how what the colored rings means and how caps are measured and identified. i also know how to test a transistor. and i also wrote that i did test them for good.
if repairing those chargers is not easy enough, then it's a waste of time. i'm fine with that. won't bother anyone anymore. :)
 
it is not a bother, it is just that the wikipedia article is better than i can do about explaining how transistors work. i do feel it is worth the effort to figure out how things work, since that is just what i like to do. it has taken me years too. makes me wish i had understood it earlier in life when it have helped more. not when i am almost 70.
 
i totally agree with you. i prefer to learn doing thing by myself, instead of staying a dummy. but some things are "worth" finding out, others are too far away, and the outcome not worth the effort. i've a master in technical engineering. i know quite a lot about mechanics and stuff. i'm working as sales for a storage vendor. i do electronics on an "executive level" ;) for fun. but reverse engineering a charger is way too much off my knowledge, and my spare time left over from my job and familiy with 3 kids it too little to do all the research necessary. so if i can get some help telling me: measure voltage between C1 and R2, and compare to 20V, then i can do it. remove fet Q3 and replace it. i can do it. and I'M STILL LEARNING by doing that. not all i need to know, but still enough for me :)
 
it may or may not matter, but i have a masters in solid state and nuclear physics, and i took the last two years of the EE curriculum when i was getting my masters then, 35 years ago. i am published, made major contributions to development of the current techniques used to manufacture VLSI parts from working with an IC manufacturing process development group for H-P, but i never had time to learn electronics so this is my way of teaching it to myself.

the Switch Mode Power Supplies are actually kinda simple to understand, but how to know what to look for at each individual part is over my head. i just keep analyzing until i hit something i can test.
 
dnmun said:
alan, why does your charger have voltage on the output when these don't?

they seem to give good warranty support. i think everyone has gotten a replacement who had problems.

i think greg's charger is the only one we could not get replaced, (italics mine - JKB) but even his charger might have been replaced if we could show a good reason for them to replace it. he had used it for awhile, but it would be worth a try, imo.

there was no reason for him to come here and hijack the thread to prove his oneupmanship. we didn't go puke on his threads.

i don't think the failure rate is as high as it seems either since these are just the people with problems but a lot of the chargers have been sold to sphere dwellers since the doc posted up about them 2 years ago now. these are big chargers. it still seems they are the best deal available. jmho
Click to expand...

You've GOT to be kidding. I got NO HELP from BMSbattery on either of two failed 900 watt chargers.

Mine were just under a year old, pampered and used lightly. I've got less than 600 miles showing on the CA display. The chargers failed, same way, within 2 weeks of each other. I contacted BMSbattery and the "customer support" (could have been your "Jack", I don't know "Jack" :shock: ) offered me a 10% discount on a new charger, that's all. I decided to pass that offer up, since I have no clue what "10% discount" means when you can make that 10% back up on shipping.

My ebike has been down and unusable for weeks now. I blame Kingpan/BMSbattery for that. I even considered fixing my failed chargers (and posted here for some help) but as the broken charger tales began to show up more and more often, I realized it might be a waste of time even if I did fix them, as there's no telling how long before the charger will blow up in some other section...

I'm sore about this...really sore. BMSbattery can kiss my <colorful metaphor>.
 
the charger that BMS battery resells is not the kingpan charger. the charger they sell is manufactured by EMC.

we can try to help you fix your charger if you want. i suspect they did not feel they were liable for the failure of your charger because you had used it for a long time. the replacements were all for chargers that had failed when they were first used.

we have parts and know some stuff so we can help you, but as i remember, your charger had blown up in the hi voltage oscillator and burned up a lot of components. i had never seen that type failure. i think this is where we heard about the flux walking but never learned what it was.

it terms of observed reliability, of the three big EMC-1000 chargers there was only one that actually failed because of a manufacturing defect. one had a human induced failure caused by leaving a loose wire inside that shorted, and the other initially did not work then started working for no reason after i fiddled with it by moving the trimpots back and forth once because i think that charger had been under water and got corrosion on the trimpot wipers. it started working after i did the trimpots.

you can buy the kingpan chargers from greetime too. you don't have to buy from BMS battery, and the small kingpan chargers now have the thyristor daughterboard for 'universal voltage' and that fails regularly too. so i would say that both sources, kingpan and EMC are equally reliable, but the kingpan can be fixed easily when the thyristor board fails.
 
I have the 600 watt 84 volt version. I was running 24s but recently added a cell. Was pleasantly surprised that with just a 1/2 turn on the Voltage pot I got 88 volts and a full charge with 25 cells. I'd add another cell except Im now at the top limit of the Kelly Controller. After a full charge I have to wait about 20 minutes before the voltage drops enough for the Kelly to engage.
 
dnmun said:
the charger that BMS battery resells is not the kingpan charger. the charger they sell is manufactured by EMC.

we can try to help you fix your charger if you want. i suspect they did not feel they were liable for the failure of your charger because you had used it for a long time. the replacements were all for chargers that had failed when they were first used.

we have parts and know some stuff so we can help you, but as i remember, your charger had blown up in the hi voltage oscillator and burned up a lot of components. i had never seen that type failure. i think this is where we heard about the flux walking but never learned what it was.

it terms of observed reliability, of the three big EMC-1000 chargers there was only one that actually failed because of a manufacturing defect. one had a human induced failure caused by leaving a loose wire inside that shorted, and the other initially did not work then started working for no reason after i fiddled with it by moving the trimpots back and forth once because i think that charger had been under water and got corrosion on the trimpot wipers. it started working after i did the trimpots.

you can buy the kingpan chargers from greetime too. you don't have to buy from BMS battery, and the small kingpan chargers now have the thyristor daughterboard for 'universal voltage' and that fails regularly too. so i would say that both sources, kingpan and EMC are equally reliable, but the kingpan can be fixed easily when the thyristor board fails.
Click to expand...

OK, EMC. Got me there, for sure.

So, wait: if I made a list of the parts from that section, could you sell me the set of parts so I wouldn't have to go all over creation and spend huge $$ to order parts from about a half-dozen places?

Thanks for your level-headedness. I still hate those chargers and BMSbattery, but I appreciate your kindness dmun....

JKB
 
i have the 2.2 ohm base resistors and the transistors but not the diodes. i have a buncha little daughterboards with small caps but i bet that is 60V at least. but i may have one in the junk piles. the diodes are on ebay and the caps too. except it is like buy one for $2 or buy 25 for $4 free shipping.

i just had one fail on me the other day. it was working for awhile after it arrived and while i was tweaking it it just decided not to work anymore. no smoke or any bad noises.

it is the 600W model with the switch on the base current so i hope i can figure out what happened.

except i made it worse since i was probing with my voltmeter, from the 340V at the top of the rectifier bridge because that is the pin on the end i can reach, using the red on the 340 and probing with the black, in the middle of those diodes and i heard some little crackling sounds. i think i toasted the diode with leakage through my meter.
 
i got this charger to examine and it worked initially and then just quit. no obvious problem. fuses ok, voltage in the front end, nothing going through the transformer. dead back end.......

so i was measuring the pn junctions in the big 4140 transistors. got the 570mV for the B-E junction and then 550mV for the B-C because the collector is so much more heavily doped. one transistor had .2V though so i checked the C-E bias which was open as it should be on the one transistor with normal B-C forward bias and it was .590 or something like that on the one with the .2V bias,so it was shorted i figured.

i had to remove the clamping bar to get it outa the way so i could measure the transistor's legs with my voltmeter. it was hard to reach the legs because it was so tight but removing the clamp allowed me to measure them in the case. so i was able to confirm that the high side npn switching transistor had failed. great i thought, fix coming.

soldered in a pair of the big fairchild transistors and put the entire thing back together with luck and when i plugged it in there was high voltage on the output, intially 160V then i tried to dial it down but went the wrong way on the trimpot and it skyrocketed over 200 outa meter range so i tweaked it all the way down to the end of the trimpot and it will not go under 116V on the output.

i really wish i could understand how these things work when it gets like this. i am assuming that the pwm controller the TL494 is running full speed and the normal feedback trimpot cannot match the current leaking onto the feedback loops that is driving it to max output. sometimes the mental questions, the why is that resistor there or where is the feedback coming from and why did it fail. this is the hard part about learning electronics from scratch. dmm
 
You must log in or register to reply here.

Similar threads

T
warning about ENERPROF.DE / ENERDAN battery&accesories
2
Replies
33
Views
2,903
tomtom81
T
Lucifer
Want a battery spot welder that works for £20?
2
Replies
26
Views
7,736
Daijoubu
D
E
BMS bypass problems
Replies
0
Views
562
EectraTechGlobal
E
Watchamacallit
Burnt BMS (Pics included) Please help?
2
Replies
31
Views
7,163
Watchamacallit
Watchamacallit
electroneg
New to LiFePo4 and New Member - Battery, charging and BMS Questions.
Replies
24
Views
5,110
john61ct
J
Back
Top