Driver Transistors Overheating

this amp works but the driver transistors marked in red are getting pretty warm,

we tried changing 100 ohm emitter resistors in preceding stage to 47 but no joy, still 0.5 V on top of 1 ohm emitter resistor which = 500 ma idle current,

wondering what controls that idle current so i can mod it,

this is the Webb steel guitar amp.

thanks for any help!

Transistors Q8 and Q9 are output transistors (parts of Darlington and Sziklai pairs) for small output power. Q10-13 kick in when the power is higher and are off at idle current. You should adjust idle current of Q8 and Q9 with diode string + trimmer in collector of Q3.

+1 to what moamps said...  The drivers are also contributing to output power... They are pushing 1/2A + before the power transistors ever cut on.

Their class A bias voltage comes from a conventional diode string, 3 diodes. The trimpot shunt across one diode would reduce class A bias, but that may not be the reason they are getting hot... they are driving the output too. 

Check the crossover distortion (clean small signal HF sine wave) and if not crazy good (too good for a guitar amp) let it get warm.

JR

Thanks very much! 

a part of the diode string is mounted to the heat sink, is that counter productive-where would be a better place?

in the Peavey version of this there is a 22 ohm resistor in the emitter of Q3, maybe i will try sticking one on there,

ok we replaced the 27 ohm resistor wit 22 ohms, the old one was carbon comp, had drifted up a bit, no we are stabalized at 466 ma which is better, transistors not quite as hot, Peavey runs 250 ma so maybe we will try 10 ohms across the diode,

10 ohms = 355 ma, i can keep my finger on the transistor, about a 3 Mississippi, thanks you guys! you bailed me out again! 

offset went down from 144 mv to 83,  ;D

CJ said:

Thanks very much! 

a part of the diode string is mounted to the heat sink, is that counter productive-where would be a better place?

Click to expand...

In fact in an ideal world all three diodes would be thermally coupled to heatsink, so their temperature is similar to transistors and will drop as they get hot, reducing the drive voltage to the transistors.

As the transistors get hotter their Vbe gets smaller, increasing the class A current, and getting hotter... this is how thermal runaway happens.  (a bad thing)

in the Peavey version of this there is a 22 ohm resistor in the emitter of Q3, maybe i will try sticking one on there,

Click to expand...

any reason in particular?

CJ said:

10 ohms = 355 ma, i can keep my finger on the transistor, about a 3 Mississippi, thanks you guys! you bailed me out again! 

offset went down from 144 mv to 83,  ;D

Click to expand...

spit on your finger and touch the transistor.. if it boils away the case is 100C or more.

JR

from Self's power amplifier book, 4th edition, p age 343:
"In the CFP the drivers have the most effect and the output devices, although still hot, have only one-twentieth the influence.  Driver dissipation is also much more variable, so now the correct place to put the thermal sensor is as near to the driver junction as you can get it."

for those wondering about CFP, aka Sziklai. essentially the lower section of the quasi-comp amplifier  in question.

thanks gridcurrent!  we stuck the dual diode on the driver and took off another 20 ma after settling time! 

here is the Peavey 300 which is dang close to the Webb, there are some differences in the area we just got done tweaking,