For a pair of power valves working in a push-pull situation, often there is a bias pot arranged so that the bias of one valve is adjusted in relation to the other so that the two are in perfect balance. Of course this will set the dc operating conditions in perfect balance but would it not also affect the output stage linearity because each valve would then be operating on a different part of it's curve than the other? The basic cause of the problem is unequal cathode emission at a given bias voltage, so instead of adjusting the effect by changing the bias, why not adjust the cause by changing the cathode emission by raising or lowering the heater voltage slightly? If the heaters were run from a regulated dc supply this would be quite easy. In fact, as the cathodes start to age you could have a dc servo loop that gradually winds up the heater voltage to keep the emission constant.
HEATERS.
Hi,
Varying the heater voltage to control cathode emission can be done but you risk cathode stripping at too low a voltage.
The workable range is only +/- 10 % of the rated heater voltage.
The best workaround is to have independent bias supplies IMO...and, this may surprise you...often a slight imbalance of the output tubes gives better sound.
Especially in class AB1 designs.
Cheers,😉
Hi,
Varying the heater voltage to control cathode emission can be done but you risk cathode stripping at too low a voltage.
The workable range is only +/- 10 % of the rated heater voltage.
The best workaround is to have independent bias supplies IMO...and, this may surprise you...often a slight imbalance of the output tubes gives better sound.
Especially in class AB1 designs.
Cheers,😉
The tubes aren't working on different parts of their curves, their curves are different altogether. They may or may not be in their best spot.
The obvious answer is to get well-matched tubes, reducing distortion in the first place. 😉
Tim
The obvious answer is to get well-matched tubes, reducing distortion in the first place. 😉
Tim
Imbalance in output tubes comes from a large number of sources, not just differences in cathode emission; one obvious set is the various physical tolerances inherent in mechanical assembly. What's worse is that cathode emission does NOT vary proportionately with heater voltage or current. The curve of emission versus temperature is strongly nonlinear, and trying to keep a tube stable on the high slope part of that characterisitc is a major challenge.
There's an easy way to balance tubes.
There's an easy way to balance tubes.
It works!
I have to agree with Circlotron. Tuning the heater voltage will work. It shifts the entire slope of the curves. Basically to match gm. Then you would adjust bias to match dc. Obviously, matching in both dc and ac is good.
The effect of heater voltage on the plate curves is easy to see if you have a VacuTrace or other curve tracer.
Sure, the adjustment range may not be too great, but you can pull together a moderate match into a perfect match. Besides, any pair of tubes that is perfectly matched probably won't stay that way forever. A clever adjustment scheme like this will keep an amp tuned to max performance.
jh
I have to agree with Circlotron. Tuning the heater voltage will work. It shifts the entire slope of the curves. Basically to match gm. Then you would adjust bias to match dc. Obviously, matching in both dc and ac is good.
The effect of heater voltage on the plate curves is easy to see if you have a VacuTrace or other curve tracer.
Sure, the adjustment range may not be too great, but you can pull together a moderate match into a perfect match. Besides, any pair of tubes that is perfectly matched probably won't stay that way forever. A clever adjustment scheme like this will keep an amp tuned to max performance.
jh
It's called "separate bias controls." Adjust for equal idle current or, if you want to be fancy, lowest 2nd HD at a desired output level. That chosen level is a matter of religion.
Heater Voltages
Allowing heater voltages to vary more than 5% is generally considered to be bad for the valve.
This is covered in "Getting the Most out of Vaccuum Tubes" (Robert Tomer 1960)
It is not the way to adjust the electronic characteristics of an individual device. This must be done (in a common cathode stage) by adujusting the cathode characteristics and in a differential stage by adjusting the cathode balance.
7N7
Allowing heater voltages to vary more than 5% is generally considered to be bad for the valve.
This is covered in "Getting the Most out of Vaccuum Tubes" (Robert Tomer 1960)
It is not the way to adjust the electronic characteristics of an individual device. This must be done (in a common cathode stage) by adujusting the cathode characteristics and in a differential stage by adjusting the cathode balance.
7N7
Perhaps *where you are on the curve* is simply a function of the current [density] which is generally referred to in terms of grid bias. So maybe the absolute bias doesn't matter, only the current does and how far the bias swings from the set point whatever that may be. Restated - the recommended bias is just a starting point; the important part is the current.
- Status
- Not open for further replies.