TC Electronic problems...

[gyraf]

I've sent an investigator team to TC Electronics, and I hope to get an answer sometime this week..

Jakob E.

 

[Svart]

more testing data:

The uc3842 (THE generic SMPS PWM IC!) is only getting around 14v at it's VCC pin. the UVLO voltage is 16v. The IC is not starting up and the Vref is not outputting, therefor not supplying the error amps with anything.

Pulling the IC allows the voltage to rise to 17v on the through hole for VCC but taking the IC and using an external powersupply to power it, I get around 20ma of current draw and the IC turns on.

So the IC is not bad( an educated guess..), it seems to be pulling down the VCC to a level below UVLO. I also found that one of the two caps that were not replaced does show a high ohm resistance between the leads, somewhere around 5M but it does pull down the voltage another volt.

So one cap is bad so far.

I should have more time later today to work on it.

:thumb:

 

[drpat]

Man, I'm glad you have the patience! Excellent work so far :thumb:

 

[Svart]

Ok, last week I wasn't able to work on this but today I got to work on it for a little while.

It's now WORKING.

The 100uf 25v cap that had ESR/short problems was the cause of it all. The cap did not allow the circuit to charge to the required 16v UVLO threshold. This circuit seems very sensitive to the amount of bulk capacitance on the VCC rail. When the 3842 starts up, the output stage pulls the VCC rail down considerably as it and if the cap value is insufficient then the IC fails to start the PWM cycle due to the Vref output being too low. If Vref is too low then the RT/CT (timing) circuit does not start and the PWM does not start as well. If the PWM does not start then the FET does not charge the HV transformer and the PWM IC does not get powered from the system and the IC stays in UVLO mode.

It's a vicious cycle!

I would suggest that this cap is changed to a low ESR type, as well as moved up in value from 100uf 25v to 220uf 35v. The 100uf cap seemed to allow too much ripple from switching the output drive this is why I would go to a higher value if possible.


here is the wiring pinout for those who want to know:

Yellow: -15v
Blue: return for +/- 15v rails
orange: +15v

Black: Return for +5v, +6v rails
Red: +5v
Purple: +6v (and pre pass transistor)
Brown: power good signal(?)

This is the troublemaker:

The system is pretty much a barebones SMPS without any truly fancy features. It has a overengineered feedback system but thankfully it's still very simple and I don't see why this system would cause problems.

That being said I don't see any type of protection for the VCC side of the PWM IC should there be a crazy line spike, just the high value/wattage resistor that bootstraps the startup voltage. I would have liked to seen some kind of protection for ESD on the VCC pin but most manufacturers don't see the sense in protecting a .25$ part with 1$ worth of parts. The FET seems worthy, a 2sk1118 Toshiba 600v 24AIdp part.

 

[drpat]

Did I miss a small electrolytic? Crap, I've got low ESR 220's all over the place. Even though I never use it, I'll yank the SMPS out of my Finalizer and try to revive it in the same way.

THANKS SVART!!!

 

[Svart]

do this:

with the SMPS out of the unit, flip the SMPS upside down and with the IEC power connector to your left.

the edge of the PCB closest to you should have a couple rows of pins, one with 4 pins and the other with 6 pins. both of these rows of pins go to the daughterboard containing the control circuitry.

The pin farthest to the left on the 4 pin connector is the VCC bootstrap pin. this supplies a voltage to the PWM IC so that it can start up. The next pin to the right is the return pin for that voltage rail. Measure between these and you should have around 16v. If it's low then there is a problem.

When recapping the SMPS don't forget any of the 'lytic caps no matter how small. If they are small values that you can get film caps to replace, do it.

 

[Michael Krusch]

 

[Svart]

EXCELLENT!!

I'm glad I could help the group.

The cap that came out of the unit was a Rubycon 105deg.

This spot on the switcher gets bashed by a lot of ripple due to the output drive of the PWM IC charging a FET gate as well as the output of the transformer that feeds the IC once it is started hitting it with rippling DC.

 

[Svart]

Think this should go into the troubleshooting meta?

 

[kvintus]

Great thread! My M-one gave up on me last weekend. I have to check if the PSU is the same.

/Anders

 

[drpat]

 

[amorris@home]

The 100uf 25v cap that had ESR/short problems was the cause of it all.

Thank You Svart! we have literally a pile of these units and half a dozen dead supplies. replacing this cap is bringing them to life. and being the impatient, lazy, instantly gratified american that I am, I just replaced this single cap (I believe it is C2) and voilá!

testing out of the box only fires up +/-15, but with the power good line wired into the unit, all came up!

 

[Svart]

Awesome!  I'm glad everyone is getting these working again!

The cap in question takes a hard beating from the powersupply.  It's sort of a cheap-out on the design side but better caps might also help them last longer.

 

[verbos]

I have an M2000 on the bench right now. I tried recapping the supply with no success. Is there a schematic for the supply? I'm at a loss.

Thanks


Mark

 

[alexc]

I just did the 'Svart fix' on my newly dead Finalizer and it returned to life!

Thanks man!

Alex

 

[zayance]

Dem A couple of years ago i had problem with Mone and M3000, and i was not in DIY stuff or whatever, just a cap ridiculous, the M3000 repair was 200 euros :-(, of course they kept the PSU that got replaced, if it really got replaced ;-), wow 200 euros for a Cap!
Thanks a lot Svart, who knows this will i'm sure, happen again, don't want to but seem to be always a failure on PSU.

 

[s2udio]

I have an M2000 on the bench right now. I tried recapping the supply with no success. Is there a schematic for the supply? I'm at a loss.

Thanks


Mark

Try replacing the PWM chip on the sub pcb........had a GForce which i recapped and still refused to play....
A replacement chip UCC28C42 cured it.........a UC2842B is Also an equivalent device

 

[deuce42]

And for that matter I have a crapping out G Major (a known problem child with almost every guitar forum on the web full of complaints from users about the unit freezing all the time).

Tried to check the power supply but couldn't locate anything obvious.

The unit works for a about thirty seconds then everything freezes. The information on the LCD screen stays the same but none of the buttons repond and the signal drops out.

Any of you guys that are much smarter than me have a possible answer or tip?

 

[s2udio]

Internal battery ok.......
Scope the PSU rails for ripple ?

 

[livingnote]

Wow. That is impressive.