Are there any studies, if the impedance of power supply feeding heater has any impact on distortion?
I assume here DC heating power supply.
In other words, is heater radiation always constant, no matter how much power is going via cathode and is cathode load impacting on the heater temperature ?
I assume here DC heating power supply.
In other words, is heater radiation always constant, no matter how much power is going via cathode and is cathode load impacting on the heater temperature ?
I don't know if there have been any studies, but it seems to be generally accepted that the impedance of a DC heater PSU should have very high impedance to minimize interactions between the heater circuit and the power supply.
You have to design the heater PSU to provide a constant power to the heater. Yes, the cathode emission does have a slight effect on the cathode itself, however, it doesn't change the design goals of your power supply.
You have to design the heater PSU to provide a constant power to the heater. Yes, the cathode emission does have a slight effect on the cathode itself, however, it doesn't change the design goals of your power supply.
Tube manuals specify heater power way of specifying a design center heater voltage, such as 6.3Vac as an example. Supply that voltage, and you will get the heater power, knowing the heater current, which is also specified in the manuals.
Heater to ground impedance should be very low to avoid common mode heater noise and unwanted coupling between stages.
For better advice I need to see the circuit.
For better advice I need to see the circuit.
My udnerstanding is that cathode temperature should be kept as stable as possible to have constant electron radiotion. If the temperature changes due to cathode load the electron radiotion/flow is not expected and thus electon flow what anode sees finally is imapcted. In a way cathode temperature modulates amplification but because of the mass of the cathode and heater, there is delay and result is electron flow modulation which means distortion.
I tend to hear the effects of all parts in the heater supply and certainly the obvious: AC, LC filtered DC, regulated DC, current sourced DC. Briefly tried an ultrasonic AC and hated it.
In subjective conclusion: current source (high impedance) generally better than the rest, but upstream component quality not less audible. A friend prefers mains derived AC to any type of rectified or regulated supply.
In subjective conclusion: current source (high impedance) generally better than the rest, but upstream component quality not less audible. A friend prefers mains derived AC to any type of rectified or regulated supply.
Heater is heater, it's not modulated by signal. The load is stable so I can't really imagine how different supply impedance can change anything. My bet's on, contrary to other "smart" solutions, firm voltage source. The heater voltage is, after all, the main parameter describing the tube heater, and this way you are probably going to make the amp universal for different tube types or makes.
Last edited:
^^^^That!
The heater itself is low impedance, so nothing to be gained from a high impedance supply.
Moreover most tubes are designed for and perform best with constant voltage (not constant current) supply.
The heater itself is low impedance, so nothing to be gained from a high impedance supply.
Moreover most tubes are designed for and perform best with constant voltage (not constant current) supply.
Thanks Rikaro. I'd like to add that isolation and filtering/regulation can be always overdone and maybe should be sometimes? Also it depends on the condition of your mains. In my area I can say I can have reasonably quiet amp with heaters done old school style directly from PT with artificial center tap via a 100R trimmer. Somewhere else I can imagine you need to spend a lot on heater supply. Haven't tried smps for heaters yet, but want in the future. I like idea of saving pt iron that way.
Last edited:
An ac heater can modulate the electron flow within a receiving style valve, and that can be just noticeable as a form of hum ingress into an input stage high-gain pre-amp. The EF86 and 7025 ushered in helical wound heater filaments to lower that effect. That effect was due to magnetic coupling, not temperature induced fluctuations.
An indirectly heated cathode's temperature is significantly 'isolated' from heater temperature fluctuations by power flow's T^4 radiation transfer relationship.
There are many many ways for an ac heater to couple into an audio signal within an amplifier !
If you're keen on such physics and valve related technical issues then I'd recommend wading through the RCA 1962 Electron Tube Design book that is on-line.
Directly heated cathode valves are another story, and there has been testing of DHT output stage modulation due to heater AC.
An indirectly heated cathode's temperature is significantly 'isolated' from heater temperature fluctuations by power flow's T^4 radiation transfer relationship.
There are many many ways for an ac heater to couple into an audio signal within an amplifier !
If you're keen on such physics and valve related technical issues then I'd recommend wading through the RCA 1962 Electron Tube Design book that is on-line.
Directly heated cathode valves are another story, and there has been testing of DHT output stage modulation due to heater AC.
There are a few rod pentodes in which heater current is comparable to pate current. For the 1J24B pentode, the nominal heater current of 12 mA should be reduced to 10.4 mA at the maximum plate current of 1.6 mA.
A high-impedance supply will increase susceptibility to hum coupling into the audio circuit from the heater, but it seems unlikely that it could have any effect on ordinary harmonic distortion. If the cathode is that the right remperature, that's all that matters.
Electrons emitted from the cathode surface take some energy with them, and so there is a cooling effect; but in a DC-heated indirectly-heated cathode, I doubt that any notable instantaneous effect on cathode temperature or gm would be seen. And no external circuit could do anything about it, realistically.
Directly-heated Filamentary cathode often use current-drive (high-Z) heating supplies - but this is to reduce or eliminate the corruption of the cathode signal-current by the heating supply, and not for this effect.
Directly-heated Filamentary cathode often use current-drive (high-Z) heating supplies - but this is to reduce or eliminate the corruption of the cathode signal-current by the heating supply, and not for this effect.