How to Calculate BIAS on Transistor amp ?

[MoonsTaBass]

Hi folks !

Simple question, maybe already answered but can't find it...
How To calculate the Bias of a Transistor amplifier ?

I have a Schneider ST1444 hifi amplifier (pretty rare) with dead short power transistors because of a short on speaker cables....
But I can't find any schematic or manual so...
How can I set the bias properly after replacing the transistors ?

I believe (if I remember) there is some kind of formula for calculate.
If I know the voltage (40V), the power (35w) and the output resistor (0.33ohm)... Can I calculate the voltage I need to read directly across output (emitter) resistor ?
Is that OK to mesure directly across resistor OR there is some other spot to mesure ??

transistors are BDT 96 and 95

Thanks !
Antoine

[g1]

Measure DC millivolts across the emitter resistor and solve for current. No signal applied.
This will tell you the idle current through the output transistor.

[MoonsTaBass]

Hey !

Thank you, so you answered my second question
But now, how can I know this current I mesure is correct for this amp ?
My first question was, How to calculate the bias ? not "how to mesure it"

[Helmholtz]

There's no general formula for every SS amp.
Using sig gen and scope adjust for lowest current without significant crossover distortion.
Often something like 10mA is sufficient.

[MoonsTaBass]

Ok thanks !

I thought about that... So just start with the lowest bias as possible and then scope and adjust bias to minimise crossover distortion and it's good ?
100HZ or 1kHZ ?

Thanks !

[Helmholtz]

100Hz or 1kHz shouldn't make a difference.
I like to use a 400Hz triangular signal, because that shows nonlinearities best.
But a sine signal works fine.

1) Make sure there's no DCV at the output.
2) Connect suitable dummy load and scope.
3) Feed test signal, avoid clipping. Crossover distortion shows best at low output.
4) Increase bias current for best linearity. Avoid overbiasing.
As this is a hifi amp it might use a higher bias current of maybe 20mA.

[MoonsTaBass]

Ok many thanks !

I'll just do that so

So there is not some kind of formula to calculate bias in amplifier ?
I'm pretty sure I've read this somewhere in a book years ago...

How manufacturer calculate this in some servicing manual ?
I fixed a Gallien Krueger years ago and the servicing manual said that bias was 8ma...
Where's this value came from ?

[g1]

How manufacturer calculate this in some servicing manual ?
I fixed a Gallien Krueger years ago and the servicing manual said that bias was 8ma...
Where's this value came from ?

Probably what they found to be the lowest level (coolest operation) that will still always get rid of crossover distortion. Also depends on the particular output transistors in use.

[Jazz P Bass]

Enzo had posted a wonderful way of biasing SS amplifiers.
Adjust the bias control for lowest bias.
Monitor the overall amplifier current draw. (ie: Kill-A-Watt meter)
Slowly turn the bias control until the amp starts drawing current.
Bump it up a tad.
Done.

[Mick Bailey]

I've never needed to calculate bias. Even if you did, with component tolerance you'd end up using this as a starting point and adjusting to suit for that particular amp. Where a manufacturer provides a figure I usually set the amp to that, though. Remember to re-check once the amp is thoroughly warmed up to operating temperature as there can be quite a bit of drift from the cold setting. You need to get it where it doesn't distort when cold but doesn't pull too much current when hot.

[TomCarlos]

Enzo had posted a wonderful way of biasing SS amplifiers.
Adjust the bias control for lowest bias.
Monitor the overall amplifier current draw. (ie: Kill-A-Watt meter)
Slowly turn the bias control until the amp starts drawing current.
Bump it up a tad.
Done.

JPB: Perhaps you are referring to this thread - How to adjust (bias) solid state amplifiers ?

See post #16. And there are other replies in that thread with good info.

[J M Fahey]

In general transistors are way more linear than tubes, also *always* there is heavy negatve feedback involved, way more than in any Tube amp, so in most Bipolar transistor amps, usually "above 10-15mA per transistor" is fine.
Or Helmholz´s 20 ma.

Of course checking waveform on a scope is even better.

[R.G.]

For more than you ever wanted to know about biasing solid state push-pull output stages, read "Audio Power Amplifier Design" by Douglas Self. He describes in depth not only why it's hard to do, but why it can't be done perfectly. Ever. There is also a lot in there about thermal compensation.

[x-pro]

I'll add an important one.
Perhaps you are interested not only in bias current, but also in the voltage across the thermostatic transistor, the voltage between its emitter and collector (aka between the bases of the upper and lower arms of the first output stage), then the calculation is simple. How many transistors are in the output node? How many emitter repeaters are in it?
Suppose one of them is at the top and the other is at the bottom. The voltage at their b-e junction is 0.55-0.65 V, so they are slightly on. Two transistors are 1.1-1.3 V. If there are three transistors each, on the top and bottom sides, that's 3.3-3.9V. Don't forget about 10-15% variation and deviations.

[Mick Bailey]

......but why it can't be done perfectly. Ever. There is also a lot in there about thermal compensation.

I haven't read the book, but with cheap and powerful processors such as the $5 postage-stamp sized 133Mhz RP2040 Seeed Studio it may be possible to attain near perfection (or at least as good as it gets) by closely monitoring and constantly adjusting an amp's parameters.