Marshall MG250DFX - Smoke Generator!

[Andy1960]

Hi 52Bill. The problem im having is identical to the one posted by JoKers about 15 posts back. Yes new chips blow straight away, both outputs the same, checked everything JoKers did. No load apart from the first time but Im told these will run with no speakers attached as long as you dont attach a load while powered up. The chips blow instantly its not liek they overheat. this problem reoccurs all over the forum and in fact the internet regarding these TDA7293's in various amp models and while some have managed to repair them there doesnt seem to be any rhyme nor reason to the repairs, its almost sheer luck.

[52 Bill]

So it will power up fine without the power amp chips connected? All of the supply voltages are correct, including the standby voltage? And you have tested all of the parts on the power amp chip boards and the input circuitry that they connect to? How about the speaker jacks and wiring?

Are you using some sort of limiter when powering up, like a variac or a light bulb fuse saver? If both power amp chips blow, then it should eliminate some things and lead you to some others. Where did you source the new chips?

[The Dude]

Also, make sure the heat sink is isolated from the chassis.

[Andy1960]

52Bill, Yes without the boards connected it all powers up fine and the preamp works fine. Supply voltages are all correct. Standby voltage was previously at 12 due to the zener being open but is now 9 but either is within safe limits. Speaker jacks are ok with no shorts on the output. All caps on the amp boards check out fine and Dude, yes the heatsink is isolated by the fan with no shorts to chassis.
the chips in both channels blew but ive only been testing it in one since it is just two identical outputs, not bridged etc.
You are about the fourth person to ask where I got the chips from and I have been suspecting it could be the chips but they are ST branded and I didnt get them from China which is what everyone seems to ask.

[Jazz P Bass]

I have to ask.
Where did you purchase the IC's?

(They may be bogus.)

[Andy1960]

Got them from ebay but a UK seller. They are ST branded. I have some more from a different seller but again a UK one. Is there any way to check without fitting them?

[DrGonz78]

Yeah it seems there is UK, Chinese, AU etc... source for that part on FleaBay. It appears, by the wording, that each of the sellers are about the same thing. Perhaps it is time to order from a good source. I am pretty sure this is the right part.
TDA7293V STMicroelectronics | 497-2170-5-ND | DigiKey

Edit: oh by the way here is a UK version of that same page...
http://www.digikey.co.uk/product-det...70-5-ND/599502

Also to anyone that knows is there any difference with this model of TDA7293 with the "V" at the end of the part name?

[Andy1960]

yeah the V and the H are just vertical or horizontal pins. Just spoke to Marshall again and he also said that as far as he can tell everythign I have done is correct and he also says suspect the chip

[Andy1960]

IT LIVES!!! Well, finally! I had just about convinced myself it must be dodgy chips causing it and after speaking to the Marshall engineer (I emailed them and the guy actually rang me which was good) I was about to order some direct from them (£4.80 + VAT + postage) but I thought I'll just check yet again as I had a couple of chips left so thought I might as well blow them up in the name of experimenting. I checked all my voltages yet again and got the same as Ive always got which all concurs with the schematics. Just as an afterthought I recall reading a thread on here where a guy doing an AVT150 had changed the electrolytics after noticing one was a 105 degree rated one and he had success. Well I had measured the three 47 mF caps on the little boards several times with my Fluke meter and all were within about 4 mF of what they should be. Just for the hell of it I changed them and Bingo! all working, no explosions, no magic smoke and its been sat here for about three hours playing to itself. I can only think that it may be time I invested in an esr meter as these caps still check out ok on the Fluke.
Only other thing I found was that reg3 had broken solder pads which I noticed when wondering why the channel wouldnt switch but I dont think this would have had anything to do with the outputs.
So, if your having the same problem give it a go. for the sake of 6 caps its worth a try.

[g1]

That would be C2, C3, and C5 on the power chip boards.
C2 & C3 are supply filters for the plus/minus power rails. Although they are small value, they could hold a charge so it is necessary that they be discharged when the chip is replaced.
Any chance you were discharging the main supply but not these caps?

[Andy1960]

no, not a chance I made sure there was no voltage anywhere! I even shorted all the pins of the ic with a clip before soldering it in and before connecting the cable each time. It can only have been the caps but as I say they all measured ok. Ive kept them just out of curiosity in case I get chance to measure the esr sometime.

[g1]

Then I don't see any reason C2 or C3 would do it, maybe someone else has an idea.
That leaves C5, which is between the output chip terminals "bootstrap" and "bootstrap loader". Symptoms of low capacitance there are listed as "signal degradation at low frequency".
Maybe someone could speculate how/if the bootstrap cap could be responsible?

[Andy1960]

I dont know either but its the one and only thing I changed. The original post where i got the inspiration from is here:- http://music-electronics-forum.com/t31030/
Post #7