Creative Gigaworks S750 PSU

[torress1990]

Hello, i want your help. I bought those speakers with damage. It won't start up, green power indicator was blinking/flashing. All outputs like +-24V, +12.9V, +9V; -8V; +4.4V is working, but +-70V is not working. So i replaced all capacitors, transistors all "IRF740" and "2SA92". Where i live it is impossible to get all part same as i found on board, so some capacitors like polyester are with bigger Voltage - is it OK? (that was the first question). I turned it ON, and i almost works: after transformer i see ~58V. But after bridge rectifier i don't get +-70V. So i ordered bridge rectifiers. I think i will get +- 70V. But there is another problem. IRF740 are heating like (the transistors base become black while i was measuring voltages, it took about 1 minute maybe 2 ) ... i don't know how to compare with... what should i do to eliminate that problem? could you help me ?

All the circuits i found:

http://www.tgahinfo.homecall.co.uk/S750 PSU/

 

[(*steve*)]

The transistor's base becomes black?

Can you show us a picture?

Is anything getting hot?

 

[torress1990]

The transistor's base becomes black?

Can you show us a picture?

Is anything getting hot?

So how i was saying "IRF740" transistors base for me is the case, where is the hole for the screw. Usual it is made form the copper and it is shiny (am i right?), but when i turned on, after 1 or max 2 minutes, those transistors became very very hot and how i said base is black now. Now it is impossible to take a picture, so i marked on other transistor where became black.

And i forget to say that, i also changed all diodes.

http://speedy.sh/CCy2j/tip31a-npn-transistor-to220-60v-3a-pack-of-4-327-p.jpg

 

[(*steve*)]

If it's just a bare patch of copper that goes black, I wouldn't worry about it.

I'd worry about it if the whole tab goes black.

 

[torress1990]

If it's just a bare patch of copper that goes black, I wouldn't worry about it.

I'd worry about it if the whole tab goes black.

But it heats up very fast without any load. I think it isn't normal or like used to be ?

 

[(*steve*)]

Heating up very fast with no load sounds like it has an output shorted or it's oscillating.

Disconnect the power supply from the rest of the circuit and see if everything stays cool.

 

[torress1990]

Heating up very fast with no load sounds like it has an output shorted or it's oscillating.

Disconnect the power supply from the rest of the circuit and see if everything stays cool.

You mean disconnect transformer ?

 

[(*steve*)]

No, disconnect the amplifier boards from the power supply.

You have identified the power supply as getting hot. We need to determine if this is due to excessive current draw from the amplifiers.

When we disconnect the amplifiers, if the power supply still runs hot, we know the fault is likely to be in the power supply, otherwise it's in the amplifier,

Next step (assuming the power supply cools down, is to connect the amplifiers one at a time (if this is possible)

 

[torress1990]

No, disconnect the amplifier boards from the power supply.

You have identified the power supply as getting hot. We need to determine if this is due to excessive current draw from the amplifiers.

When we disconnect the amplifiers, if the power supply still runs hot, we know the fault is likely to be in the power supply, otherwise it's in the amplifier,

Next step (assuming the power supply cools down, is to connect the amplifiers one at a time (if this is possible)

1) PSU is disconnected from anything, and after bridge rectifier there is no voltages.

2) When it is disconnected it heats up very very fast.

 

[(*steve*)]

At this point a photo would be really useful.

It sounds like the problem is in the power supply and that is often good news because linear power supplies are often not too hard to fix.

 

[torress1990]

At this point a photo would be really useful.

It sounds like the problem is in the power supply and that is often good news because linear power supplies are often not too hard to fix.

http://www8.zippyshare.com/v/20224329/file.html
http://www8.zippyshare.com/v/72033908/file.html
http://www8.zippyshare.com/v/9727926/file.html
http://www8.zippyshare.com/v/94299402/file.html
http://www8.zippyshare.com/v/24868601/file.html

 

[(*steve*)]

OK

There appears to be more board behind the rear heatsink that I can't see.

Also an image of the bottom of the board would be useful.

I note that you(?) have removed several transistors. why?

Also, I fear this is not a linear power supply

Have the symptoms changed at all since you started replacing components?

Can you tell us (by component reference and/or by marking up an image) which components you have changed out -- especially the diodes.

Please also point out the transistor(s) that are getting hot.

 

[quantumtangles]

Interested in seeing how you pinpoint this one Steve. What is your best guess based on available information so far. Nice thread.

 

[(*steve*)]

Interested in seeing how you pinpoint this one Steve. What is your best guess based on available information so far. Nice thread.

I note a touch of skepticism...

I can't help but feel that it is at least a little warranted.

Yet, toress1990 has persisted with this problem for a while and seems to be well enough informed to try a few seemingly sensible things that I think pursuing it further is worthwhile.

edit: what is my best guess? I've really not got one yet. Interested in what is getting hot (and yet not failing). Hot and not failing in an unloaded SMPS is odd.

 

[torress1990]

OK

There appears to be more board behind the rear heatsink that I can't see.

Also an image of the bottom of the board would be useful.

I note that you(?) have removed several transistors. why?

Also, I fear this is not a linear power supply

Have the symptoms changed at all since you started replacing components?

Can you tell us (by component reference and/or by marking up an image) which components you have changed out -- especially the diodes.

Please also point out the transistor(s) that are getting hot.

How i said, this is auto-generating power supply, not linear.
I tried to write on photos information, i think it is helpful.
Here are the pics:
http://www29.zippyshare.com/v/21529142/file.html
http://www29.zippyshare.com/v/605541/file.html
http://www29.zippyshare.com/v/10560347/file.html
http://www29.zippyshare.com/v/72198422/file.html
http://www29.zippyshare.com/v/15230643/file.html
http://www29.zippyshare.com/v/25774760/file.html
http://www29.zippyshare.com/v/78610088/file.html
http://www29.zippyshare.com/v/51402582/file.html
http://www29.zippyshare.com/v/7876588/file.html
http://www29.zippyshare.com/v/52219679/file.html
http://www29.zippyshare.com/v/34951563/file.html
http://www29.zippyshare.com/v/95922038/file.html
http://www29.zippyshare.com/v/90924367/file.html

 

[(*steve*)]

In the third image, you say that these 3 terminal devices should be bridge rectifiers.

That doesn't make sense -- bridge rectifiers have 4 leads. Maybe they are rectifiers.

However, if these are rectifying the output of the transformers that those "heating up fast" transistors are connected to, it may explain why they are heating up so fast.

If I am right, those rectifiers need to be there or you will be doing bad things to the transistors driving the transformers.

Do you have the original part number(s) for these?

 

[torress1990]

In the third image, you say that these 3 terminal devices should be bridge rectifiers.

That doesn't make sense -- bridge rectifiers have 4 leads. Maybe they are rectifiers.

However, if these are rectifying the output of the transformers that those "heating up fast" transistors are connected to, it may explain why they are heating up so fast.

If I am right, those rectifiers need to be there or you will be doing bad things to the transistors driving the transformers.

Do you have the original part number(s) for these?

Rectifiers have to "FEP16DT"

So you say that when i insert rectifiers, everything should become good? But if it will not ? (i am waiting for the new rectifiers). Because without rectifiers, transistors "IRF740", which drives transformers, becomes hot very fast.

P.S. sorry for my bad english

 

[(*steve*)]

I'm not sure everything will become good, but to operate without them can kill the IRF740's

They're not dead yet though (probably) because if they were, they would probably fail short circuit and then you'd be blowing fuses.

The transistors put energy into the transformer. This energy is then allowed to escape through the rectifier diodes. Without the rectifiers, it has to do somewhere, and that may be through the IRF740's while they're in their off state. This is an avalanche condition which heats the transistors up a lot.

In addition to this, it is typically the DC output voltage which is used to control the amount of power pumped into the transformer. Without the diodes, this voltage is zero and the regulator is trying to pump all the energy it can into the transformer. That's not going to be helping either...

 

[torress1990]

I'm not sure everything will become good, but to operate without them can kill the IRF740's

They're not dead yet though (probably) because if they were, they would probably fail short circuit and then you'd be blowing fuses.

The transistors put energy into the transformer. This energy is then allowed to escape through the rectifier diodes. Without the rectifiers, it has to do somewhere, and that may be through the IRF740's while they're in their off state. This is an avalanche condition which heats the transistors up a lot.

In addition to this, it is typically the DC output voltage which is used to control the amount of power pumped into the transformer. Without the diodes, this voltage is zero and the regulator is trying to pump all the energy it can into the transformer. That's not going to be helping either...

2013.04.22-23 i think i will get my new parts, and then i hope everything will be good. Thanks for explanation.

By the way, did you looked at the schematics that i attached in the first message ?

 

[(*steve*)]

I looked at them at the time you posted them, and I would typically only go back and look at them again for reference purposes.

If you have changed them:

1) I receive no notification
2) I can't compare them to the old ones
3) It makes my existing comments in the thread wrong, misleading, or questionable.
4) If told, I have to search back to find them.

In general, it's a bad idea unless you are updating some sort of FAQ thread where the comments are not especially important.