![[-]](https://theultimatetone.com/images/collapse.png "[-]")
12-11-2025, 02:34 PM
Hi Guys
The bias regulator used in our Power Scaling kits requires a higher raw bias supply voltage than many stock bias supplies provide AND a lower impedance. Bias supplies derived from the plate supply through high-value resistors DO allow an outrageous voltage to be attained, up to the absolute value of the B+. However this is at high-impedance. To make the bias regulator happy, the dropping resistor has to be made much lower in value, usually by paralleling many more resistors. This leads to an excess of heat in the chassis. The output of this supply needs a zener clamp as a minimum to protect the pass element of the bias regulator.
Other situations allow the stock bias winding to be separated from ground and a voltage doubler circuit to be implemented. This works reasonably well in Hiwatts, some Fenders and some Marshall amps.
Using an auxiliary transformer wired backwards and powered by the heater supply eliminates all of the concerns with the other methods, while providing a high-voltage at medium- to low-impedance. Our RBX Raw Bias Auxiliary Supply kit provides -84V with a typical bias-set network attached to the bias regulator output. With no load its output can pop up to -110V, but this is okay for the BJT in the regulator.
Readers of the TUTs (The Ultimate Tone series books), this forum and our FAQ will know that it is always a good idea to have excess sweep of the bias controls to allow complete turn-off of every tube sample. Usually, having 15% of the absolute screen voltage is sufficient. For example, if Vs=500V then -Vb should extend to -75V; for 400Vs, -60Vb, and so on. The target bias voltage for large-bottle tubes and 6V6 is around 10%, but we do not want to make the mistake of some amp companies of ONLY providing that amount of bias voltage as some tubes will red-plate and some will be stone cold.
From the numbers above we can see that there is a limit of compatible B+ for RBX. The bias regulator BJT can go to having zero volts across itself with -84V output. This would correspond to a screen voltage of 560Vs. For screen voltages higher than 560V RBX is inadequate for the task and the tubes will be overbiased and likely red-plate. The solution is to use a larger PT in the RBX format, i.e. go from 6VA to 12VA, then the output voltage will rise to almost double. An amp with 750Vs needs -113Vb with standard tubes to assure a proper control range of the tubes. Even at 600Vs, as in a Marshall Major, we need -90V for adequate controls, so RBX is not quite good enough there.
Note that the RBX PCB is sized only for 6VA transformers, which is suitable for the majority of amplifiers. For Vs>560V you would need a separate PT and RBX-LT with higher-voltage caps than usually provided.
Going to the larger auxiliary PT allows us to reconsider the decision of wiring the new PT forwards or backwards.The forward wiring has no loss per se and the output voltage will be as one would expect, allowing the use of the same 6VA PT provided this will support the bias-set network et al. We would need a 115Vac or so secondary. The primary can be a single or duals - with the 229-series from Hammond every PT has dual primaries and secondaries out of necessity of the design. The primaries are wired in series or parallel, as required for your mains, and the secondaries are wired in parallel. We now have enough bias voltage for Vs=800V. For higher Vs we wire the secondaries in series and regulate it down to a reasonable range less than 200V so as to protect the bias regulator BJT
Have fun
The bias regulator used in our Power Scaling kits requires a higher raw bias supply voltage than many stock bias supplies provide AND a lower impedance. Bias supplies derived from the plate supply through high-value resistors DO allow an outrageous voltage to be attained, up to the absolute value of the B+. However this is at high-impedance. To make the bias regulator happy, the dropping resistor has to be made much lower in value, usually by paralleling many more resistors. This leads to an excess of heat in the chassis. The output of this supply needs a zener clamp as a minimum to protect the pass element of the bias regulator.
Other situations allow the stock bias winding to be separated from ground and a voltage doubler circuit to be implemented. This works reasonably well in Hiwatts, some Fenders and some Marshall amps.
Using an auxiliary transformer wired backwards and powered by the heater supply eliminates all of the concerns with the other methods, while providing a high-voltage at medium- to low-impedance. Our RBX Raw Bias Auxiliary Supply kit provides -84V with a typical bias-set network attached to the bias regulator output. With no load its output can pop up to -110V, but this is okay for the BJT in the regulator.
Readers of the TUTs (The Ultimate Tone series books), this forum and our FAQ will know that it is always a good idea to have excess sweep of the bias controls to allow complete turn-off of every tube sample. Usually, having 15% of the absolute screen voltage is sufficient. For example, if Vs=500V then -Vb should extend to -75V; for 400Vs, -60Vb, and so on. The target bias voltage for large-bottle tubes and 6V6 is around 10%, but we do not want to make the mistake of some amp companies of ONLY providing that amount of bias voltage as some tubes will red-plate and some will be stone cold.
From the numbers above we can see that there is a limit of compatible B+ for RBX. The bias regulator BJT can go to having zero volts across itself with -84V output. This would correspond to a screen voltage of 560Vs. For screen voltages higher than 560V RBX is inadequate for the task and the tubes will be overbiased and likely red-plate. The solution is to use a larger PT in the RBX format, i.e. go from 6VA to 12VA, then the output voltage will rise to almost double. An amp with 750Vs needs -113Vb with standard tubes to assure a proper control range of the tubes. Even at 600Vs, as in a Marshall Major, we need -90V for adequate controls, so RBX is not quite good enough there.
Note that the RBX PCB is sized only for 6VA transformers, which is suitable for the majority of amplifiers. For Vs>560V you would need a separate PT and RBX-LT with higher-voltage caps than usually provided.
Going to the larger auxiliary PT allows us to reconsider the decision of wiring the new PT forwards or backwards.The forward wiring has no loss per se and the output voltage will be as one would expect, allowing the use of the same 6VA PT provided this will support the bias-set network et al. We would need a 115Vac or so secondary. The primary can be a single or duals - with the 229-series from Hammond every PT has dual primaries and secondaries out of necessity of the design. The primaries are wired in series or parallel, as required for your mains, and the secondaries are wired in parallel. We now have enough bias voltage for Vs=800V. For higher Vs we wire the secondaries in series and regulate it down to a reasonable range less than 200V so as to protect the bias regulator BJT
Have fun
