Transistor used as bias diode

[robjkmannering]

During my recent radiogram repair, I encountered a transistor in the dc coupled amp with the base emitter junction used as a diode.
This I assumed to be for biasing. As the transistor tested okay I did not replace it.
The transistor is a 2SC1129 wich is listed for vhf use.
As the amp now uses BC558s ect. I wondered if I need to change the bias 'diode' so that it will have the same temperature characteristics as the rest of the transistors used or will it have little difference?

Thanks,
Rob

[julie_m]

What you have here is one of those "it shouldn't work, but it probably will if you don't tell anybody" situations.

Try warming the amplifier board with a hairdrier and cooling it with freezer spray, and listening out for any obvious distortion.

[G8HQP Dave]

Why shouldn't it work? All BJTs follow the Ebers-Moll model fairly accurately.

[AC/HL]

Transistors were often used in place of diodes for stabilisation. I suspect that the 2SC device would not be the original one from a Thorn radiogram, but should be as good as any other silicon junction. It's only the thermal characteristics that are being harnessed.

[kalee20]

As long as you're replacing small-signal silicon with small-signal silicon, I'd expect it to work, as G8HQP Dave and AC/HL have observed.

If you were replacing with a 100A power transistor, then you may not get quite enough bias voltage generated. But that's not the case.

As it's just used as a two-terminal device, you could even swap between NPN and PNP, with polarity reversal.

May well be used because it's easier to 'clamp' a transsitor case ot the heatsink rater than a diode.

[G6Tanuki]

Originally Posted by merlinmaxwell

May well be used because it's easier to 'clamp' a transsitor case ot the heatsink rater than a diode.

Indeed: I'm prepared to bet that the quality of the thermal-bond between the 'bias sensing' transistor/diode is rather more important than the precise characteristics of the transistor/diode itself.

I have memories of a Redifon-designed HF linear amplifier which had a couple of dirty great stud-mount power rectifiers fitted on the heatsink right between the output-transistor quad. Along with two OC35 transistors passing a couple of amps, they worked quite well to stabilise the bias up to heatsink-temperatures around 80 Centigrade.

[robjkmannering]

In this case, the transistor is not mounted on a heatsink, the ouput pairs have individual aluminium fins. The transiator does not even sit close to another semi-conductor so heat transfer is not an issue.

Rob