Elevated heaters/power supply technical info

[trobbins]

If the heater DC supply is floating then parasitics control what is happening. Good valves with high heater-cathode resistance may not 'anchor' the heater, and it is plausible for some AC voltage to exist between heater and ground - which could allow hum from not only the heater-cathode resistance and capacitance, but also from capacitance leakage to the input grid. Poor valves with low heater-cathode resistance may better anchor the heater and suppress any AC heater voltage with respect to ground. This is obviously a murky topic, and may be of no issue if your input stage has a capacitor bypassed cathode bias.

As an aside, the Dynaco ST70 and Mk VI don't use elevated or DC heater circuitry, or do they DC ground the heater, they include a capacitor to ground and rely on the heater-cathode resistances of all the tubes, and the various cathode DC voltages, to charge that capacitor to effectively an elevated DC level with negligible AC voltage on the capacitor. They do use an AC heater so their is still the chance of hum to the grid of a valve, but this method also effectively increases the heater-cathode resistance as it likely elevates the heater away from the various cathode voltages.

Aggressive rectification/filtering may show up as transient glitches occurring at twice mains frequency - similar to B+ glitches from ss rectifiers - if that is not well managed then it can appear in the audio signal.

[Helmholtz]

As an aside, the Dynaco ST70 and Mk VI don't use elevated or DC heater circuitry, or do they DC ground the heater, they include a capacitor to ground and rely on the heater-cathode resistances of all the tubes, and the various cathode DC voltages, to charge that capacitor to effectively an elevated DC level with negligible AC voltage on the capacitor. They do use an AC heater so their is still the chance of hum to the grid of a valve, but this method also effectively increases the heater-cathode resistance as it likely elevates the heater away from the various cathode voltages.

Interesting concept, thanks for the explanation. It also explains why they didn't simply ground the CTs and used relatively low value caps (probably charging time concerns). It seems essential that the caps have high insulation resistance/very low leakage.
I wonder if some cap charge could be measured with a high resistance DCV meter (short time only).

I still think that 20nF is large enough compared to the parasitic inter-winding capacitance, so this will effectively provide AC grounding as well.

[trobbins]

Just chatting off-line with LT who pointed me to the PAM1 preamp that was used across the range of Dynaco amps, and was powered from the main amp. The PAM1 rectifies that AC heater to provide DC, and uses a humdinger pot and capacitor coupling to ground to achieve minimum hum and with no direct heater coupling to ground. The Dynaco amps were also described as being suitable for use with other commercial preamps, so just the amp itself had to be low hum. So it looks like Dynaco needed a way to manage a variety of heater situations, including their own PAM1, and settled on capacitor bypassing the heater CT as a universal catch-all.

[Helmholtz]

May I ask who is LT?

[Enzo]

To me it means Loud Thud, a regular member here. Often provides good experimental data.

[Helmholtz]

Thanks Enzo.

[g1]

I still think that 20nF is large enough compared to the parasitic inter-winding capacitance, so this will effectively provide AC grounding as well.

Just to throw in a similar example, Stromberg-Carlson AU-42 shows a 100nF for such a cap.

[trobbins]

A quick look at the Stromberg-Carlson schematics showed they used a typical humdinger pot for amps with intergrated preamp stages (certainly later 1950's models), but also did a capacitor to ground for one side of the heater for the AU34 that didn't have a heater winding CT.

But of greater noteworthiness is that they liked the idea of perching a turntable on top of their amps - very retro and a standout!
http://www.preservationsound.com/wp-...rlson_1950.pdf

[Hermanni]

Valves have a limit on the max voltage between heater and cathode.

There is also maximum permissible resistance between heater and cathode.

[trobbins]

That 1962 article doesn't elaborate on the reason for imposing any external heater-cathode resistance upper limit. Such an upper limit is not a typical datasheet entry. The Philips Jan 1970 ECC83 datasheet for example identifies the limit as related to the cathode to heater circuit resistance in phase splitting circuits, which indicates the reason relates to operational performance and hum levels due to hum current through the splitter circuit resistor to 0V. This isn't a concern for a circuit where any such resistor between cathode and ground is bypassed, and the assumption that the heater is grounded.

PS. The article appears to be effectively a cut/paste compilation from various valve databook descriptions and text sections.

[frusciante89]

Thanks everyone ! Now it's definitely more clear