Nope - no schematic Fernando.
I've subscribed to the Marshall forum so there's now at least three of us there with an issue. Not good for Marshall is it?
Indeed not good.
From what gwEm says, sounds like it is a well-known problem with this sort technology, which would suggest that Marshall don't understand it fully.
I've logged into the "super secret" Marshall schematic service facility and there's no schematic there. I expect there won't be until it goes out of warranty.
Regarding the "Power Scaling", what I have gleaned is that the screen grid voltage and bias voltages are varied, rather than the plate, screen and bias voltages, and that this is controlled by some digital which has a number of stepped power levels, and this seems to be where the problem is.
As ever, using "cutting edge" approaches often reveals unexpected problems.
Hello.
The thing is the following:
1)First of all AFD and YJM use the same EPA (electronic power attenuation) circuit: I have made comparison by looking PCB and also by reading in the manual the explanations on autobias, on the use of the POWER pot in AFD (called LEVEL in YJM) I am quite sure is exactly the same circuit.
2)It is all based on a patent of the 80's that expired few years ago (therefpre Marshall use it without infringment). This patent was based on moving bias (control grid) and screen grid voltages in order to keep their ratio constant. If you do that the characteristic of the tube keeps the same shape (tone remains the same) but the clipping point moves to lower levels.
3)According to Marshall designer Santiago Alvarez they made some difference respect to the patent. By examining the AFD circuit (missing the digital part schematic) there are 4 regulators which provide a negative voltage for the bias and one High Voltage regulator, which provides screen voltage.The bias ones are following a voltage called VBC1 while the screen voltage regulator follow a voltage called VS. I don't have the schematic of the digital part of this amp (nobody has posted it yet) but I am quite sure that VBC1-4 and VS come from a digital control (like a microprocessor that digitally drives the voltages trough DACs) according to POWER potentiometer position. Moreover the LEVEL/POWER control acts at our ears like a "step" volume and not linear. This may confirm that there is a digital circuit controlling bias and screen voltage according to certain equations/laws or a simple LUT (look up table). Instead the original patent of the 80s was keeping screen voltage/bias ratio approximately constant with a continuous range (full analog!).
4)Finally remember that current in Pentodes (power tubes) do not depend on plate voltage but strongly on screen voltage (and of course also on control grid).
5)The POWER potentiometer is double (shared axes): one acts probably on the (missing schematic) digital circuit that, on his turn, regulates VS and VBC1-4; the other one acts on the power stage feedback in a way that it tries to compensate (as Alvarez stated) the power stage when changing POWER level. This should be done because when you change bias/screen voltages the stability characteristic of the circuit change, therefore you have to adjust the feedback to try to keep them constant.
4)Conclusions: this is the best post phase inverter volume reduction because it does not use any kind of potentiometer/circuits in the analog path; it reduces the power maintainig the characteristics of the power tube constant. It is the ultimate mod for any NON master volume amp. What is difficult to achieve is to implement the bias/screen regulating law using an analog circuit instead of microprocessor (if it uses micro according to my supposing...again, missing the digital part schematic that would reveal if my supposing is true).
The EPA circuit works also on EL34...the manual of YJM and AFD tells that trough a bias table!
Hope I gave an help with this.
Regards
