Randall RH 100 G2 / UPDATE! Question:

[plexi50]

The transistors marked X in the pic smoked up after turning this amp on. I have ordered all new power amps and the Transistors that smoked up. Is there anything other than the 63V / 33,000 uf caps that could have caused this? Any tips would be welcome

[PRR]

> Is there anything other...

Q1-Q11 and some other stuff.

A direct-coupled amp is a house of cards. The original fault may be nowhere near where the smoke came out. The failure may have blown other parts.

If you have a good understanding of transistor-action, and can break the power supply to insert 100 ohm 10 Watt resistors between the main power and the actual amplifier, you can usually find faults by deduction.

If Randall has a rebuild service, that would be a wise path.

[plexi50]

Thanks PRR. I want to learn more about SS amps. I have not had time to check all the parts. Checking every part will take some time but im hopeful i can narrow the problem down. The amp was free so i have no real $ invested. I have $9.00 in parts on the way. I hear it sounds like a distortion factory. Not for me. I have a set of transformers and i may wind up using the chassis and cab to build a tube amp

[plexi50]

> Is there anything other...

Q1-Q11 and some other stuff.

A direct-coupled amp is a house of cards. The original fault may be nowhere near where the smoke came out. The failure may have blown other parts.

If you have a good  understanding of transistor-action, and can break the power supply to insert 100 ohm 10 Watt resistors between the main power and the actual amplifier, you can usually find faults by deduction.

If Randall has a rebuild service, that would be a wise path.

I finaly got around to working on the Randall RH100 G2 today. I found 2 4148 diodes open and both of the 22 ohm ceramic 5 watt  resistors were open as well

I removed them from the circuit to test them as i was getting odd readings with them in the circuit. I tested all the power transistors and they are shot dead as well


I cant see anything else bad on the board. The new transistors will be here this afternoon. Hopefully it will crankup tonight

[plexi50]

I have it working but the volume is on the low side of being loud. There is a breakup and fadeout for a second after making a bar chord

I used what resistors i had. Instead of using the 22 Ohm 5 watt resistors it had originally i used 100 ohm 10 watt

I am assuming these are the bias resistors?

That would explain every thing i am hearing. Sound like it is too cold on the bias. Then again SS is not my specialty

[PRR]

Those are/were 0.22 ohm.

They must be "large" compared to transistor emitter impedance (it is part of bias) but "small" compared to load. One ohm is high. Part-ohm is usual.

100 ohms is hundreds of times too high.

If you want to be cautious, go 10 ohms and then 1 ohm. It will still choke-up. However if it is still shorting-out, the 10-ohms will burn cheaply.

The value is somewhat critical. It is also part of the current-limit. Don't go 0.15 ohms, it will not shut-down as designed. 0.22 should be easily found. My bench had 0.27 ohms, and I would try that until I could get the right value.

[plexi50]

Thanks much. I dont have any thing expect for tube amp part values in stock. They were 22 ohm but they were open and not reading at all. I thought i had some 33 ohms but they turned out to be no good either. Thats why i tried the 100 ohms. None of the components get hot while its on and playing. Dam i suppose now i have to start stocking SS values as well. It's a good idea anyway. I have more of them comming my way these days. Thanks much PRR

[sluckey]

Those are/were 0.22 ohm.

They were 22 ohm but they were open and not reading at all.

Listen closely to PRR. Those resistors are NOT twenty-two ohms. They are 0.22Ω (thats zero point two two ohms). That's slightly more than two tenths but slightly less than one quarter.   

Look here for a similar power amp for comparison...  http://www.schematicheaven.com/newamps/randall_rg100es-rg80es.pdf

[plexi50]

I see my error in thinking. Thanks for pointing this out. SS is a whole new world

[PRR]

> Listen closely

Thanks.


Zero POINT Two Two ohms!
 

-Or- '... "small" compared to load. One ohm is high. Part-ohm is usual.'

[PRR]

> Look here for a similar power amp

Shudder. (not just the "SCHEMITIC")

_NO_ protection!

The fact that NFB is current-sensing is interesting. When Q2 clips you can "only" get 15 Amps output. The 2N6254 is a robust old-skool 2N3055, and effectively doubled here, but it still skirts 95% of every limit in the spec.

So let's not even consider the bootstrap, the poorly-scaled longtail, or the class-B final. I used to salvage Fisher-factory rejects of the same ilk, and they sounded "fine".

This thread's amp is a more modern beast, nearly double the transistors. That surely includes protection, and probably SOA-kinked. Such simple tricks improved hard-abuse tranny amp reliability a LOT. Still we blew them. Crown developed a micro analog computer to predict device stresses in real-time. The more common plan now is to use MANY more output devices.... no more 75 Watts per pair.

A power-transistor shop will have an ample supply of 0.1, 0.15, 0.22, 0.27, 0.33, and 0.47 5W-7W parts. NOTE: all less than one ohm! 10, 22, 47, and 100 are also common around the drivers, but you get these with a 500-part 50-values 1/2W assortment off eBay.

[plexi50]

I found some 0.33 ohms and put them on the board. I turned it on and it had sound for a 30 seconds and blew the mains fuse. So the short is still there some where. I am removing the Q's one at a time and checking them on a transistor checker