Power control regulating only the screen voltage?

Mostro · Post by **Mostro** » Sun May 18, 2014 9:14 pm

Hi guys.

I´ve been experimenting with the MOSFET B+ regulator, currently regulating the power section only. Installed it in a AC30-ish amp, EL84´s in cath bias. Had some trouble with crossover distortion on low voltages, lowered the PI coupling caps but it didn´t dissappear completely, in the lowest voltage range it´s still there.

Then I thought: Why not regulate only the screens? lower screen voltage equals less power, right? And you can eventually make the amp silent if the voltage is low enough, right?

Just wondering if anyone has tried regulating only the screens...

kdmay · Post by **kdmay** » Sun May 18, 2014 9:21 pm

Check out the YJM power control. I am pretty sure it regulates just the screens and apparently works quite well. Never tried it myself though.

katopan · Post by **katopan** » Sun May 18, 2014 10:10 pm

Screens only changes where the load line sits relative to the knee of the characteristic curves. So as you dial down you'll get less screen compression, different distortion and a different tone. Both plate and screens together more or less scales down to keep around the same load line area.

Mostro · Post by **Mostro** » Sun May 18, 2014 10:53 pm

Well, I looked up the YJM Marshall. Wow!

It has a microcontroller controlled (ahem...) power section. Seems to be it controls the screen V, Bias V and something else at the same time. Not DIY stuff, I think. But very clever, it seems to be. Of course judging by a Youtube video is difficult, but seems to work well tone wise.

Katopan, so you say it will vary the out power BUT also the compression and tone?

There must be a way to do what that Marshall does but without the microcontroller...

ampdoc1 · Post by **ampdoc1** » Sun May 18, 2014 11:40 pm

Nobody here has tried this, but it's simple and effective.

Put a 25 W variable resistor in line with your output tube heaters (I'd suggest using DC heaters). Lowering the voltage there lets the entire amp, except the output tubes, to run exactly as stock. The output tubes don't attract as many electrons, this lowering the volume, while all the gain and PI stages are running normally.

I used a variable resistor when I made my amp 2004, but would use a VVR to control the heaters if I was building it today.

Mostro · Post by **Mostro** » Mon May 19, 2014 12:41 am

Wow, Ampdoc1, that´s a novel idea!

Unorthodox for sure. How about the tube life and/or bad consequences? I´ve read numerous times that the heaters V has to stay +-10% of 6V3 or bad things will happen...
And to get DC for power tubes, you´ll need some big ass regulator that can handle a few amps of current...

jazbo8 · Post by **jazbo8** » Mon May 19, 2014 9:32 am

Mostro wrote: Unorthodox for sure. How about the tube life and/or bad consequences?

+1, I don't know how low you need to dial down the filament voltage in order to get a decent reduction in output power/SPL, but I suspect it is not a linear relationship... while running the filament slightly below rated voltage can improve the tube life, running it too low will greatly reduce its life. So this seems to be a risky proposition.

Leo_Gnardo · Post by **Leo_Gnardo** » Mon May 19, 2014 12:52 pm

I have some customers who run their whole amp on a Variac, always simple amps that is without channel-switching low voltage supplies. Old Marshalls and HiWatts, for instance. One dials his 100W HiWatt all the way to 60% of line voltage. Haven't seen any damage nor heard any complaints, except "I have to let it warm up at full voltage otherwise it takes all day to warm up."

In these cases the hi voltage and bias and filament all track down together.

I measured the HiWatt turning out 25W at clip, AC dialed down to 60%. Still plenty loud, but smooth not spiky tone like you'd expect.

teemuk · Post by **teemuk** » Mon May 19, 2014 4:53 pm

https://www.google.com/patents/US4286492

Then there's Blackstar's patent (GB2446614) but I don't know how to link to it. It can be found with searh at http://worldwide.espacenet.com, though.

andrew · Post by **andrew** » Sat May 31, 2014 6:53 pm

I have a vvr I would like to use on EL34 power tube filaments. I tried it with a rheostat once with some Kt66s but I couldn't get much range from it. The preamp tubes are on DC. Do I need to heat sink the mosfet? Thanks

teemuk · Post by **teemuk** » Sat May 31, 2014 9:58 pm

andrew wrote:Do I need to heat sink the mosfet?

How much will it dissipate power? Can the device handle that power without heatsinking? If not, what is the minimum heatsink required?

These can be all calculated. You probably should know the details for the calculations as you are the designer.

Most often power transistors will need a heatsink.

Tube Amp Archive

Power control regulating only the screen voltage?

Power control regulating only the screen voltage?

Re: Power control regulating only the screen voltage?

Re: Power control regulating only the screen voltage?

Re: Power control regulating only the screen voltage?

Another way to Control Output

Re: Power control regulating only the screen voltage?

Re: Power control regulating only the screen voltage?

Re: Power control regulating only the screen voltage?

Re: Power control regulating only the screen voltage?

Re: Power control regulating only the screen voltage?

Re: Power control regulating only the screen voltage?