TGM8 - my best amplifier, incredible bass, clear highs, no fatigue (inspired by Rod Elliot P3a)

[Bigun]

Good news ! - are you pleased with the results ?

 

[RCruz]

Finally had the opp to listen to it and i am really pleased with the sound.

 

[Bigun]

that's great!, any photos ?

 

[auriga2001in]

Thanks Bigun

Thanks for the wonderful amp. I have two amps (BJT, Latfet outputs) running wonderfully for a year now.

 

[Bigun]

Fantastic, it's good to know that it works well with the Latfets.

 

[pinnocchio]

I'll eventually rebuild a new channel. I must have a passive component busted since all active have been replaced on the defective channel. I will measure all passive in-circuit and compare values with working module. Too bad this happened to it.

Do

 

[Bigun]

I was even wondering if there is a defect in the bare pcb from the manufacturer. I imagine this is very rare on simple two-sided board though. But if you have a spare pcb and have the patience to start-over or transfer parts across then a clean-build might help ??

 

[pinnocchio]

I have a few spare PCBs so I might just do that

Do

 

[pinnocchio]

I disassembled the defective board part by part and seems like the culprit is R19, 10R 1W resistor that went kapout... It was reading 325K but seemed intact to the eye... Now the PCB is way past rebuild, I mean I could re-populate it since all SMD are still on it but many pads have gone so it is best to order new parts (I'm keeping what I salvaged since it was all tested ok).

What went so wrong that R19 got defective? It is a 1W resistor, at this power level I think I would have seen some smoke? It is not even a little dark/burned...

Do

 

[Bigun]

Well R19 (if I remember) is the Zobel resistor. For this to overheat there likely would have been some h.f. oscillations which passed a lot of h.f. current or else the capacitor in series with R19 is shorted out and passed a lot of dc current. It's hard to say why this happened but likely was a result of your 'torture testing' using your 'suicide speaker' when you had a metal particle short. I have not tested the amp to it's limits, i.e. I didn't try to break it by going into hard clipping into a load or driving square wave at 20kHz into load etc. so I am not sure how the amp will show a failure. Well, partly I'm lazy because I don't want to be rebuilding the amp! Anyhow, it's really nice that you found the failed part. Why it isn't showing burned signs, well that depends on the construction and materials, not all resistors are made the same way. What type / part number did you use for R19 ?

 

[RCruz]

that's great!, any photos ?

Soon

 

[pinnocchio]

I believe it was this part CPF110R000FKEE6

I would have to check the order but I'm pretty sure it is.

Do

 

[Bigun]

I believe it was this part CPF110R000FKEE6

I would have to check the order but I'm pretty sure it is.

Do

Well there's you answer, this resistor is specially made to be "Flameproof, high temperature silicone coating". No wonder it didn't show obvious signs of stress.

 

[pinnocchio]

All parts ordered!

Do

 

[Bigun]

 

[pinnocchio]

And one channel rebuild!

Only missing the output devices which I will add tomorrow, then test everything and see if it works.

Ciao!
Do

 

[RCruz]

 

[Bigun]

 

[pinnocchio]

module tested and singing beautifully!

Now I need to burn it in on a suicide speaker and then test in real life!

Do

 

[Bigun]

Don't break it again !