New power control (YJM AFD)

[Guitargearfreak]

It's hard for me to understand how it really works... As I understand it's not just like an onboard powerbrake, it reduce the actual voltage running through the amp... but what does that mean?

Does it mean that the power tubes are really cooking as if you were really playing at the crazy level you can set the master... or you get the sound of a cooking power amp, without really blowing your tubes that much...

In other words, I'm wondering if the power tubes will bie as quick as if you were really playing at this volume on the master... if yes, I don't see the difference detween that and a standard attenuator...?

 

[indeedido]

I don't understand it either, but it seems to sound far better than any outboard attenuater. Those suck tone so badly, the demos and descriptions I've seen don't appear to have that same loss of tone.

 

[big dooley]

the power control actually decreases the headroom of the poweramp... and by doing so, it decreases power as well...

yes you can run the tubes at full clipping, by turning the master all the way up, but the tubes aren't passing on large amounts of current anymore, so you get the full tilt sound from them, without cooking them

 

[MKB]

Please keep in mind the following is speculation and educated guesses on my part, some of it may be way off base as I haven't seen a schematic yet.

If I understand it correctly, the YJM/AFD power control actually lowers the voltage to the screen grids of the power tubes. To simplify it a bit, the screen grid is a part of a power tube that regulates the current going through it. By lowering the voltage the maximum output power of a tube will be lower as well. This is basically lowering the headroom.

I believe the bias voltage has to be lowered as well, this also requires that the signal from the preamp be lowered also. The power control has to manage these three things (screen voltage, bias, signal to output tubes) in order for it to work correctly and sound good. In the past this was unheard of as it is a hard job to do with regular parts; but I'm certain Marshall is using a microcontroller in the circuit which makes it much simpler to do. Marshall has been spraying microcontrollers all over the place lately.

The difference between this and a power attenuator is the power control lowers the amount of power the amp produces; the attenuator dissipates excess power as heat instead of sending it to the speakers. The power control is a more efficient way of doing the job; the question is does it sound good?

 

[big dooley]

Please keep in mind the following is speculation and educated guesses on my part, some of it may be way off base as I haven't seen a schematic yet.

If I understand it correctly, the YJM/AFD power control actually lowers the voltage to the screen grids of the power tubes. To simplify it a bit, the screen grid is a part of a power tube that regulates the current going through it. By lowering the voltage the maximum output power of a tube will be lower as well. This is basically lowering the headroom.

I believe the bias voltage has to be lowered as well, this also requires that the signal from the preamp be lowered also. The power control has to manage these three things (screen voltage, bias, signal to output tubes) in order for it to work correctly and sound good. In the past this was unheard of as it is a hard job to do with regular parts; but I'm certain Marshall is using a microcontroller in the circuit which makes it much simpler to do. Marshall has been spraying microcontrollers all over the place lately.

The difference between this and a power attenuator is the power control lowers the amount of power the amp produces; the attenuator dissipates excess power as heat instead of sending it to the speakers. The power control is a more efficient way of doing the job; the question is does it sound good?

the assumptions are for a great deal correct...
the power control indeed affects the screen voltage and the bias.. this is fully automated, hence the automatic biascontrol on these amps, with a very wide range, as both the AFD and YJM can accept almost every known octal based pentode / beam pentode

furthermore the negative feedbackloop to the PI is affected as well, so i guess this is the part that affects the amount of signal

indeed marshall uses firmware to control all these things... a lot of people were afraid it would made the amp a digital nightmare, but santiago (the designer) revealed, that the circuit is in fact more simplistic then the idle speed control on electronic fuel injected cars, which has been done since the early 70's

as for the sound... that's in the ear of the beholder

 

[MKB]

the assumptions are for a great deal correct...
the power control indeed affects the screen voltage and the bias.. this is fully automated, hence the automatic biascontrol on these amps, with a very wide range, as both the AFD and YJM can accept almost every known octal based pentode / beam pentode

furthermore the negative feedbackloop to the PI is affected as well, so i guess this is the part that affects the amount of signal

indeed marshall uses firmware to control all these things... a lot of people were afraid it would made the amp a digital nightmare, but santiago (the designer) revealed, that the circuit is in fact more simplistic then the idle speed control on electronic fuel injected cars, which has been done since the early 70's

as for the sound... that's in the ear of the beholder

Thanks for the confirmation on the guesses. It makes perfect sense to put microcontrollers in tube amps, in the past they would only be there to manage channel switching or manage signal levels (in more complicated preamps with multiple channels that share controls). Marshall has them in the Haze series, which is why it can do all the cool functions such as storing effect levels between different channels, and use of the 4 button footswitch.

My hats off to Marshall for pioneering the use of microcontrollers to manage power amp functions like the power control and automated bias. I'm waiting for footswitchable power control settings; since they have a microcontroller in there they only need to add code and a line or two for external control.

 

[big dooley]

Thanks for the confirmation on the guesses. It makes perfect sense to put microcontrollers in tube amps, in the past they would only be there to manage channel switching or manage signal levels (in more complicated preamps with multiple channels that share controls). Marshall has them in the Haze series, which is why it can do all the cool functions such as storing effect levels between different channels, and use of the 4 button footswitch.

My hats off to Marshall for pioneering the use of microcontrollers to manage power amp functions like the power control and automated bias. I'm waiting for footswitchable power control settings; since they have a microcontroller in there they only need to add code and a line or two for external control.

yeah the digital age isn't killing tube amps... combine them and you get killer features from our beloved old technology... just take a look at the stock JVM footswitch... just a guitarlead to connect it to the amp and yet you have all different kinds of way's to use the unit... it's just awesome

 

[Riffraff]

Thanks for the confirmation on the guesses. It makes perfect sense to put microcontrollers in tube amps, in the past they would only be there to manage channel switching or manage signal levels (in more complicated preamps with multiple channels that share controls). Marshall has them in the Haze series, which is why it can do all the cool functions such as storing effect levels between different channels, and use of the 4 button footswitch.

My hats off to Marshall for pioneering the use of microcontrollers to manage power amp functions like the power control and automated bias. I'm waiting for footswitchable power control settings; since they have a microcontroller in there they only need to add code and a line or two for external control.

It sounds like a game changer to me. There was an earlier post asking if the days of tube amps were numbered, I'd say no. Maybe the days of modelers as we know them are numbered because it sounds to me that it's feasible to design tube amps that can actually become the circuits that modelers have been trying to imitate. Not in a one amp does everything from a Fender Champ to a Mesa Dual Rectifier but at least within variations of a circuit like 1959, JCM800, JMP100, 1987, JTM45, Fender Bassman & Traynor YBA-1.

 

[Guitargearfreak]

wow very interesting guys, thanks for all your posts!