faulty 48v Battery Charger

[memy]

Hi All,
Have lurked in this forum for a number of years now, lots of good info to absorb.

I have a 48v charger that worked for about 2 charges before it gave up.
I don't hold much hope for it but figured I might as well put it up here to see if any of the clever people have any bright ideas.

It always ran hot (has a fan).
Failed 5 minuets into charging a battery that was deeply discharged.
After it failed opened her up to find the soldered in fuse blown, been a lazy fellow I soldered a long piece of (4 inches) of 26 AWG wire across the fuse. Funnily enough this did not fix it, it just blew the thermistor NTC 5D-11.
At this point put in to hard basket.

It looks like 240V goes straight into a bridge rectifier with a 400V electrolytic capacitor across the output.
I could be wrong as to my limited knowledge this is not what I’d expect to see. (I expected thermistor to transformer with bridge on the low side).
Putting a multimeter across the capacitor (still in circuit) shows a dead short though I expect this could be other components.?

As far as I can tell, a Tl3842 (PWM CONTROLLER) drives the 10n60c (600v mosfet), so it looks like a switchmode power supply
This would make sense (to me) except the “transformer” looks very much like a transformer; rather I would have expected an inductor.??

In case the pictures are not clear the other IC is a LM324.

I guess that the high heat may have affected the capacitor, do electrolytic’s go open or do they short out if the electrolyte dries out?
Why does this circuit still employ a transformer?

Any ideas?

Thanks

 

[dnmun]

why are you running it on 240V?

you should really use a fuse, other form of current limiting is useful if you have it too, but you can swap out with another 5A-7A fuse, and solder it on, even 10-15A if that's all there is.

i have some thermistors i can mail you one, or two since you may blow the first one.

the transformer is still a transformer in a switcher, but the output voltage is higher and the current is regulated by the pwm IC current controller that manages the mosfet. the current goes out of the inductor through those 2 diodes on the right side and the 14 pin package controls the optotransistor in the 4 pin package between the input and output sections, and it also controls the leds, and keeps the power level down to levels the package can manage.

you can reduce the power in the input oscillator section by going to the source leg of the mosfet and changing the big power resistor to a larger value so not so much current will flow in the primaries, but that may not be needed.

they all get hot, that one doesn't look like it got too hot since nothing burned up. sometimes the output shunt resistor burns up. that is the big green wirewound precision resistor over on the right.

i have some cheap metal film power resistors, .2ohm that you can use to replace it if that one is burned up. check with your ohmeter, with the voltage disconnected. but it is so tiny a value you cannot measure it. but if it burned open then you will see high resistance.

 

[memy]

Hi, I’m running at 240V 50 hertz because that is what comes out of the wall in Australia.
Thanks for the offer of parts, it really is a pain sometimes to locate something suitable for a reasonable price much appreciated.

The fuse is not the issue.
As the fuse is on the high side I’d expect a 1 amp slow blow fuse should be more than adequate.
The charger would be 120 to 130 watts, a 5amp fuse would allow the high side to sink 1200 watts before it would blow. Only reason I soldered in the wire was to see if it was just a fuse issue.

As the thermistor blew after the wire mod, there must be a short elsewhere in the circuit.
I desoldered the 400V cap last night, and it’s not faulty.
As no components look damaged/burnt in any way I suspect the transformer itself may be the issue.
A fair bit of work to remove that and test and if it is the fault then source a replacement.
Or maybe I should start with the mosfet. ?
If the fault is on the low side then surly there would be burnt components.

I’m currently using this power supply to charge at work.
http://cgi.ebay.com.au/350W-48V-7-2A-Switching-Power-Supply-CCTV-Radio_W0QQitemZ360194075571QQcmdZViewItemQQptZAU_Security_Equipment?hash=item53dd3d6bb3

Problem is it only goes to 55V and its pretty dicky having this sitting on my desk. (It needs to be boxed). (I’m using a isolated DC to DC converter to achieve 60V)

I can buy another 2.5 amp charger on ebay for 40 odd AUD dollars, hence my feeling that trying to fix this one is not worth the effort.

 

[dnmun]

i sent leslie a thermistor so you can see if he will let you have it for a buck or two.

not sure why you decided to unsolder the cap or now have decided the inductor is bad if there is no sign of failure. usually it is obvious when a cap fails and inductors too.

all the cheap $40 ebay chargers seem to blow the thermistor which is why i have been providing a larger one for people to replace it with. but if it is not worth your time to learn what's wrong, nothing can be done about that.

it really is easier to just buy a new one. that's what everybody does. maybe not cheaper, but easier that's for sure.

 

[memy]

dnmun,
I checked the cap as I expected there is a short down circuit from thermistor.
Events:
Charger blows fuse, when connected to battery half hour into chargeing.
Charger blows thermistor with a load pop when fuse is bypassed and no load on output.
No burnt smell, or damage, (apart from thermistor).

Why would thermistor blow the fuse half hour into charge, it would not matter if the thermistor shorted at this point, it would blow on next start.
If thermistor went open then no blown fuse to begin with, just dead charger.
Its not that I can't be bothered its just that i'm not convinced the thermistor is the root cause.

 

[memy]

Although thinking about it some more. If the thermistor is the fault, it could have seen resistance rise as it went faulty blowing the fuse before the thermistor had a chance to let go. Restarting without the fuse would kick off its final dramatic demise.

 

[dnmun]

yes, the thermistor starts out with a relatively high resistance of about 5 ohms. when it heats up the resistance drops to about .15 ohms, so it stay shot, but if there is not too much current it should not blow up.

the problem is that the cheap chargers have these small thermistors which can't handle that much current. the 11mm size that is on this 48V charger will only handle abotu 30 joules of heating, but the 18mm 6.4A inrush current limiting thermistor i got from mouser will handle almost 70-100 joules.

when i took apart li ping's 5A charger that was what i looked at first. it was huge. 22mm in size so it can handle almost 120 joules, and so it will never blow, never.

so that is the difference between the ping charger and the $40 charger. he learned to overbuild them so that people don't send them back. also he can adjust the voltage on his charger so he can push it up to 60V to force all the cells to balance.

anyway, if you want leslie to mail you that thermistor, send him a pm.

he is building a single chip dc converter, using the 2576 converter, i just opened a new charger for me, the octopus type, with 16separate switch mode power supplies with 3.65V output on them stacked on two pc cards.

will do a thread on it yet. this is the charger for brandon's battery i fixed for him. before today i don't think i would have understood how the switch mode regulator was working, but i can see all the parts now on this charger, 8 separate transformers, 16 caps, 16 diodes with heatsinks, 8 transformers with 2 pair of windings each, all so simple, 8 controllers for 16 channels so the controller module is handling 2 channels each.

anyway, maybe with a bigger fuse, like a spade fuse for a car, solder the legs onto the spots on the pc board underneath the blown out fuse, replace the thermistor with the one i sent leslie, and check the resistance of that big power resistor on the charger negative return leg, with the silver band on it. that is the shunt. it should read .1 ohm, if it is open then you need to replace it too.

i have some of them too now.

 

[dnmun]

thinking of another tack to take to reduce the power consumption and the heat is to change the resistor in the source leg of the power mosfet in the switchmode front end. the power mosfet that is controlled by the pwm IC current controller.

you can reduce its overall power consumption by doing that, so it doesn't get as hot too..they are aboout .15 ohm normally and if you swapped out for .2 ohm, that's 33% cooler.

 

[memy]

Hi dnmun,
When I have time I'll go dig up a thermistor from one of the many computer PSW's I somehow seem to collect.
Reducing the PWM, I could also probably play with the feed back that the opto coupler is supplying, I’ll see how thermistor goes first.
Only reason I has thinking of reduced power was because I thought heat had killed it. If it is only the thermistor, think I’ll leave it alone.
thanks for the help, I’ll let you know how it goes.

 

[memy]

Hi dnmun,
Installed a CL-40, thermistor got hot and triped the house CB.

 

[317537]

I sent those parts onto alswiseowl on tuesday.

 

[317537]

First test would be to see if all of those rectifier diodes are fully working, blocking the current and passing it as they should.

 

[317537]

I can buy another 2.5 amp charger on ebay for 40 odd AUD dollars, hence my feeling that trying to fix this one is not worth the effort.

http://stores.shop.ebay.com.sg/HuaQiang-North-Road__W0QQ_armrsZ1