Power amp PSU sag emulation for rack preamps?

[roberto]

Hi,

I was thinking about how to emulate the sag of a full amp in a rack preamp.

- process a buffered outout (with a pot to set the emulated sag effect) through an AD737 ( https://www.analog.com/media/en/technic ... /ad737.pdf )
- scale the output to modulate the base (gate in case of mosfet) of a voltage regulator to lower the output voltage as the signal increases.

It's just an idea to be developed, but someone could be interested in developing it, or use the idea for other applications as well.



EDIT: added "PSU" on the title to better explain the point.

[maxkracht]

Interesting idea. My only experience modulating effects voltage is as a way to increase/alter distortion in fuzz pedals and similar. Starve for voltage, cause pedals to glitch or change the fuzz character. I guess a somewhat similar thing is happening in a tube amp, but no idea if the perception would be the same without a poweramp.

I’ve had an idea to put a “clap on” circuit in a pedal. Call it a very extreme version of the same idea, but it let’s the drummer have some control.

[Stevem]

I would think that to best emulate the dynamic sag of a complete tube amp that is driving a speaker that you would want your last gain stage to only have as much gain as the average output tube.
This is much less then a preamp tube .

[roberto]

I don’t want to emulate the sag of a power amp.
But… why you would set the gain to the same value of the final stage? And why you say a triode has less gain than a pentode?

[Stevem]

In your first paragraph you posted “ full amp” so I took that to mean output section also.

I recall the average pentode output tube having the gain of like a 12AU7 I believe.

[teemuk]

If I understood correctly, the chip is practically just a precision rectifier and it forms a part of an envelope follower that tracks the audio signal.

You can't drive the power supply voltage regulator directly with the envelope follower's output. ...or you can but then the circuit is nothing but a high power buffer and your power supply voltage will be identical to envelope follower output. Not nice.

You actually need to extract the envelope follower output signal from a suitable reference DC voltage, then use the result signal to drive the voltage regulator.

I have - like many others - simply used the result signal as a dynamic reference voltage for plain diode clippers, it effectively gets the same results than power supply regulator plus amplifier circuit(s) but with much less hassle.

And why don't you want to emulate the power amp? That's where most of effects from sag manifest and that's also what causes the sag to happen in the first place. Low current draw of preamps doesn't cause sag as is and there's a considerable delay before sag in HT supply actually "passes the barrier" of filtering of the lower voltage supply nodes. Those filters are designed to keep supply voltage steady in the first place.

[bepone]

Hi,

I was thinking about how to emulate the sag of a full amp in a rack preamp.

but sag is much less in preamp stages, if any, and is filtered a lot?

[LOUDthud]

but sag is much less in preamp stages, if any, and is filtered a lot?

Think so ? Try running your preamp from a regulated bench power supply. Your amp will FEEL different.

I assume your preamp can create significant distortion and it will feed a power amp through a Master Volume control such that the power amp operates relatively clean. The trick will be choosing where to pickoff the signal and at what level. You may end up with a switch and a gain control at least until things are finalized. Other parameters are attack and decay time constants although most of that will be in your power supply decoupling string. I suggest a starting point of 10% change in B+ for full rail to rail clipping. You may want more or less or maybe a pot.

My last amp build has two power supplies, one is soft, the other is a switcher and rock solid. The switcher doesn't sound bad at all, but you really don't hear it until the power amp is driving rail to rail. Anyone traveling to Dallas PM me for an audition.

[bepone]

but sag is much less in preamp stages, if any, and is filtered a lot?

Thing so ? Try running your preamp from a regulated bench power supply. Your amp will FEEL different.

easy to put 470uF/450V instead of 22uf everywhere in preamp and check

i meant sag from the power amp, is felt much less in preamp section because of multiple RC filters to the first tube.
so also when you play clean at room volume not loud, there is also good tone in good amp, but sag is not present.

there was some study about guitar amp tone, powered from the DC suplly, or from normal RC chain, 22uF and so on, standard caps, and on oscilloscope was seen that residual ripple (100Hz/120Hz) on filter caps were considered "musical" and DC supplied sterile. study of dynamic sag inserted in preamp supply vs sag in power supply+stiff DC supply for the preamp i didnt see, but i assumed also shuld be a good tone.

speaking of "sag" on higher volume, i was in the past monitoring DC voltage on first cap on my JTM45 amp.. normal quiet operation was at 420V or so..430V. after strong riffs during high volume playing, voltage dropped in those short transients to 385-390V , so we are speaking of sag of 30-40V in power amp. and this is max how the preamps voltage want to be shifted, but they are on another RC time constant, and this 40V drop is "filtered" out, suppressed.

[bepone]

to expand the theory more, another test in reality which can be done

2 same amps, type of JTM bassman or so, named A) and B) with fx loop and master volume..

A) loaded with power Resistor 250W instead of speaker, and turned close to max power, making the sag.
B) using for the test, always the same volume

1. test. amp B) using the "saggy" preamp from the A), but B) power amp kept in clean /* sag in preamp
2. test amp B) using own preamp on the same gain position like preamp in A) before, and B) power amp kept in clean / * no sag in preamp

switching A/B,
here should be clear, how is sag effecting the preamp tone..somebody can make and return the results to us

[LOUDthud]

Sag doesn't affect preamp tone in a big way. The tone will be slightly "browner" as B+ goes down, but you'll hardly notice that. The biggest affect is on sustain and compression. That's what you FEEL.

Know how to calculate gain from plate curves and a load line ? Know how to calculate change of gain in dB ? Draw a load line at say 300V and calculate the gain. Draw another load line at 270V and calculate the gain again, then the change in gain. Most stages will change gain by 0.5 to 1dB with 10% change in B+. That doesn't sound like much but almost every stage does it except cathode followers and cold clippers. It adds up. Each stage of filtering has it's own time constant. Usually at the first stage you see the B+ go down when you hit a note, then rise between notes. Sort of a bobbing up and down. The change in gain isn't big, but you'll notice it if it isn't there.

[bepone]

understand your point totally, but never tried by myself. this is the way of testing upper, maybe someone will do before me

[imjonwain]

Are we talking about a tube or SS preamp?

[roberto]

Are we talking about a tube or SS preamp?

Tube. SS has usually really tight PS, especially nowadays with SMPS.

[roberto]

The tone will be slightly "browner" as B+ goes down, but you'll hardly notice that.

I agree, and that's why I don't think the point is calculating the gain of the amp.

Know how to calculate gain from plate curves and a load line ? Know how to calculate change of gain in dB ?

This is a very simplicistic POV, not representing the complexity of a dynamically voltage supply modulating the signal that comes from the preamp, IMHO.