For those who can read Dutch (although the schematics are pretty international):
Bias automaat - forum.zelfbouwaudio.nl
Bias servo that works in class-A and class-(A)B1, designed by an elderly Belgian gentleman with nickname Ah!buis with some help from other zelfbouwaudio.nl forum members, and tried and tested by many zelfbouwaudio.nl forum members.
Bias automaat - forum.zelfbouwaudio.nl
Bias servo that works in class-A and class-(A)B1, designed by an elderly Belgian gentleman with nickname Ah!buis with some help from other zelfbouwaudio.nl forum members, and tried and tested by many zelfbouwaudio.nl forum members.
Thank you for suggesting 'inverted B'.
What would be the total grid resistance in 'inverted B' ?
The output impedance of thre emitter follower is, compaired to the 100k, practicaly zero.
So the answer is 100k.
Mona
alujoe2,
I did not answer your earlier question.
I have looked at a few versions of the 'Garter' bias circuit.
The disadvantage of the Garter bias, is that it is Positive DC Feedback.
If one tube in the pair goes gassy, or for any other reason that tube starts to conduct more current, it will cause the 2nd tube also to conduct more current, which causes the 1st tube to conduct even more current, and on, and on, and on.
That could cause the whole output stage to have current run-away.
Current run-away is something that I dislike.
In earlier times, there were no automatic tube matching testers and data base software for matching tubes.
Now we have excellent abilities to match tubes; particularly important in push pull output stages, as well as in parallel single ended outputs.
There is nothing simple that can not be made to be more complex.
Double your solder connections, double your parts, double your chance to make an engineering mistake.
Double your chances for failure (just a generalization).
"All Generalizations Have Exceptions" - Me
"You should make things as simple as possible, but no simpler" - Albert Einstein
I did not answer your earlier question.
I have looked at a few versions of the 'Garter' bias circuit.
The disadvantage of the Garter bias, is that it is Positive DC Feedback.
If one tube in the pair goes gassy, or for any other reason that tube starts to conduct more current, it will cause the 2nd tube also to conduct more current, which causes the 1st tube to conduct even more current, and on, and on, and on.
That could cause the whole output stage to have current run-away.
Current run-away is something that I dislike.
In earlier times, there were no automatic tube matching testers and data base software for matching tubes.
Now we have excellent abilities to match tubes; particularly important in push pull output stages, as well as in parallel single ended outputs.
There is nothing simple that can not be made to be more complex.
Double your solder connections, double your parts, double your chance to make an engineering mistake.
Double your chances for failure (just a generalization).
"All Generalizations Have Exceptions" - Me
"You should make things as simple as possible, but no simpler" - Albert Einstein
Last edited:
Thank you for the link. I am a little bit lost which version you suggest. Could you please be so kind and tell me more, especially if you think that automatic balancing of the currents in PP is worth the efforts. Thank you.
The final circuits are shown in the opening post.
The first schematic in the opening post (Auto Bias Anne & Co) is the most generic one. It's a stereo version, you need only half of it for a mono amplifier.
The voltage across D5 is the reference voltage (about 0.65 V). Each output valve gets a resistor between cathode and ground, and the feedback loops of the Auto Bias try to regulate the voltages across those resistors to the 0.65 V reference. When the amplifier is driven out of class A, the other diodes in the circuit clip the voltages to twice the reference voltage to keep the integrators from runaway.
If the high voltage supply is turned on suddenly after the valves have heated up, Re1 must be used to short the reference voltage until the high voltage is enabled. This is to ensure that at start-up, all valves will get maximum negative grid voltage rather than a grid voltage close to 0.
The second schematic in the opening post (CTA5.2 - CFB-Biasautomaat) is a variant designed specifically for an amplifier with local feedback at the cathode of the output pentode and with a pentode that needs less than 12 V of negative grid voltage. It can also be used in amplifiers without such local feedback, as long as a small negative grid voltage suffices. The following schematics in the opening post show how to connect the CTA5.2 - CFB-Biasautomaat.
The first schematic in the opening post (Auto Bias Anne & Co) is the most generic one. It's a stereo version, you need only half of it for a mono amplifier.
The voltage across D5 is the reference voltage (about 0.65 V). Each output valve gets a resistor between cathode and ground, and the feedback loops of the Auto Bias try to regulate the voltages across those resistors to the 0.65 V reference. When the amplifier is driven out of class A, the other diodes in the circuit clip the voltages to twice the reference voltage to keep the integrators from runaway.
If the high voltage supply is turned on suddenly after the valves have heated up, Re1 must be used to short the reference voltage until the high voltage is enabled. This is to ensure that at start-up, all valves will get maximum negative grid voltage rather than a grid voltage close to 0.
The second schematic in the opening post (CTA5.2 - CFB-Biasautomaat) is a variant designed specifically for an amplifier with local feedback at the cathode of the output pentode and with a pentode that needs less than 12 V of negative grid voltage. It can also be used in amplifiers without such local feedback, as long as a small negative grid voltage suffices. The following schematics in the opening post show how to connect the CTA5.2 - CFB-Biasautomaat.
Thank you !
Can I order the (Auto Bias Anne & Co) somewhere ? The Gerber files are all for the (CTA5.2 - CFB-Biasautomaat), seems I cannot find Gerber files for the generic version.
Ah!buis just answered the question, see this post at the end of the thread: Bias automaat - Pagina 39 - forum.zelfbouwaudio.nl The PCB of the second photograph has a mistake in the front legend, the transistors are drawn the wrong way around.