That's an interesting question. It reminds me of the first amp I ever built which used two AD149's.
I'll see what I can come up with 😀
I'll see what I can come up with 😀
Essentially it seems it will work OK. The 1K pot (R169) which adjusts the mid point volts is a little near the end of its range (getting toward 900 ohm) with silicon but that's easily fixed with a 2K2 preset if needed.
The bias is the main worry. The 1K preset and 500 ohm thermistor will give zero bias current and hence crossover distortion. Resistor wise its nearer 700 ohms that's needed for the two combined but that's no good on its own due to thermal runaway of the output stage. It really needs a simple Vbe multiplier adding to replace those resistors. That's dead easy. One TO126 packaged transistor with a preset across C and E with the wiper to the base is all that's needed.
The original amp doesn't have a zobel network. With this design of amp you may not need one even with silicon devices and their higher speed, but, adding one is easy. Just a 0.1 uf cap and 10 ohm in series from the output of the amp to ground.
The bias is the main worry. The 1K preset and 500 ohm thermistor will give zero bias current and hence crossover distortion. Resistor wise its nearer 700 ohms that's needed for the two combined but that's no good on its own due to thermal runaway of the output stage. It really needs a simple Vbe multiplier adding to replace those resistors. That's dead easy. One TO126 packaged transistor with a preset across C and E with the wiper to the base is all that's needed.
The original amp doesn't have a zobel network. With this design of amp you may not need one even with silicon devices and their higher speed, but, adding one is easy. Just a 0.1 uf cap and 10 ohm in series from the output of the amp to ground.
You probably need to modify the bias circuit R175, R180.
Can you measure the voltage across R175?
If it exceeds 1.5-1.8V then it (the bias circuit) can be left as it is. If not, place two small signal silicon diodes (1N4148) in series with the lower end of R180 (cathode towards R180), short the wiper to the upper end and remove R180 or replace the entire circuit with a VBE-multiplier. The two diodes and the transistor of the VBE-multiplier should be in thermal (not electrical) contact with the heathsink.
Another posibility: Replace R175 with a 3-4 diode series string and increase R180 to 2k2.
Can you measure the voltage across R175?
If it exceeds 1.5-1.8V then it (the bias circuit) can be left as it is. If not, place two small signal silicon diodes (1N4148) in series with the lower end of R180 (cathode towards R180), short the wiper to the upper end and remove R180 or replace the entire circuit with a VBE-multiplier. The two diodes and the transistor of the VBE-multiplier should be in thermal (not electrical) contact with the heathsink.
Another posibility: Replace R175 with a 3-4 diode series string and increase R180 to 2k2.
Last edited:
Thanks for the ideas.Just what I was looking for.The voltage across R175 is only 0.6v so looks like I will have to modify the bias
A transistor(ge) in the power supply shorted and took out the four AD150's and a few other transistors.The AD150's are £36 for four but the 2955's are 50p each so I thought I'd try the silicon route first.It's from a Uher Royal deluxe tape recorder.
The 200pf cap could usefully be lowered as that component severely restricts bandwidth... but... it could also be argued that its "all in keeping" with the design goals of the original. As it stands the response is pretty much exactly 3db down (output voltage) at 20Khz. Values as adown to around 82pf should work well and extend the response.
Replacements for AD150
If you're not pleased with the "silicon sound" try AL100 - AL103 or ASZ15 (1015). The ALs are the best sounding Ge transistors I've ever tried (fT then times higher than ADs or ASZs). They can be found relatively cheap on ebay. (I paid 5 Euros/pair of AL103, factory matched.) In the power supply I would use silicon.
If you're not pleased with the "silicon sound" try AL100 - AL103 or ASZ15 (1015). The ALs are the best sounding Ge transistors I've ever tried (fT then times higher than ADs or ASZs). They can be found relatively cheap on ebay. (I paid 5 Euros/pair of AL103, factory matched.) In the power supply I would use silicon.
- Status
- Not open for further replies.