Powersuply sag emulation in a stompbox??

[Stellan]

Does anyone know of a reasonably simple way to emulate powersuply sag in a stompbox? I gues it is not that simple or it would be a popular thing to do?

[The Tone God]

Punisher

Andrew

[Andi]

5k pot in line with the power supply.

[Mark Hammer]

Look over the EM Fuzz article here: http://www.generalguitargadgets.com/index.php?option=com_content&task=view&id=36&Itemid=27

In particular, look at the manner in which the LM334Z current limiter is used.  This has always struck me as an underexplored project and topic.

[frequencycentral]

http://www.beavisaudio.com/products/devolt/index.html

[Stellan]

http://www.beavisaudio.com/products/devolt/index.html

I gues I should have been clearer. The effect im looking for is the sag of a rectifier tube beeing used in tube amps. This is just lovering the voltage going to an effect, which is for some reason also being called sag.
The use of an LM334Z chip might fit the bill, though not dead on, since it is not sensitive to attac. However, I have no clue how to implement that on a different circuit.
Edit: It seems to me that it is a circuit specific effect. Is that correct??

[Mark Hammer]

I gues I should have been clearer. The effect im looking for is the sag of a rectifier tube beeing used in tube amps. This is just lovering the voltage going to an effect, which is for some reason also being called sag.
The use of an LM334Z chip might fit the bill, though not dead on, since it is not sensitive to attac.

Well this is, to my mind, the "holy grail".  I'm not all that sure that the various DIY or commercial "dying battery" simulators capture the manner in which the power source attempts to provide the current requested by the initial attack transient, but ultimately fails.  Older batteries of the carbon-zinc variety are a bit like old men trying to get up off the couch in a hurry.  They lurch forward, but fall back down because they don't have enough strength to get them all the way.  Dying batteries have a momentary current-delivery capability that is very short lived.  This is what people like about dying batteries in many of the older 2 and 3-transistor variety fuzzes.  It is that dynamic aspect that I find missing in so many attempts.

If you find anything that nails it, let me know!

[Stellan]

The new byoc screamer 2 claims to nail it    This is from their site: "And we've come up with a new mod that increases the overall output and distortion of
the tube screamer circuit and adds some compression and sag like you'd get from a real tube rectifier making the pedal even more tube like" Anyone know what mod that might be??

[The Tone God]

I guess I'll reiterate:

Punisher, Dynamic Power Supply Sag Simulator

Andrew

[flo]

Link Punisher - Dynamic Power Supply Sag Simulator - Tonegod:
http://www.geocities.com/thetonegod/punisher/punisher.html

Btw:
What's wrong with "just" using an old battery, for instance a rechargable that you just don't fully charge?

[flo]

The new byoc screamer 2 claims to nail it    This is from their site: "And we've come up with a new mod that increases the overall output and distortion of
the tube screamer circuit and adds some compression and sag like you'd get from a real tube rectifier making the pedal even more tube like" Anyone know what mod that might be??

Schematics are in here but I don't see the "sag" circuit:
http://buildyourownclone.com/overdrive2instructions.pdf

[Mark Hammer]

Link Punisher - Dynamic Power Supply Sag Simulator - Tonegod:
http://www.geocities.com/thetonegod/punisher/punisher.html

Btw:
What's wrong with "just" using an old battery, for instance a rechargable that you just don't fully charge?

Nothing, really, except that the sag people want is really the kind provided by a carbon-zinc battery.  Other types do not "age" the same way.

[petemoore]

  Old battery is too much trouble compared to the short lived fun.
  Starting with something more reliable and adjustable allows time to find and tune the audio circuit and sag supply to cause the cool sag sounds.
  I built a Punisher for that purpose, I need to build a Punisher because that one quit Punishing for unknown reason.
  BTAIMay, find a circuit that responds to dropping voltage in a predictable or erratic way, and tune the bias or whatever else...
  I liked using Jfet and having the sound go soft when it should be hard, and was getting really nice ducking and surfacing attenuation type effects, some circuits will lose HF's first which adds an interesting flavor, others can have an attack-trem type effect [fairly fast attacks]...all in all a great place to start out or end up for sag.
  At the very least the env. filter is worth taking a look at for 9v sag circuit effects.
  There are many many ways in which the Punisher could be applied when used to emulate sag type effect.

[km-r]

I guess I'll reiterate:

Punisher, Dynamic Power Supply Sag Simulator

Andrew

Mr andrew, ive seen your design... but is this implementable to almost all distortion circuits?

from what i know the effects of sag is not only the "mushy""-fying the attack but there is also a sort of tonal shift from compressed to open sounding...

[petemoore]

from what i know the effects of sag is not only the "mushy""-fying the attack but there is also a sort of tonal shift from compressed to open sounding...
   We've been working on cramming that into a small signal analog circuit since...somebody wanted tube sag in a stompbox.
   There are many forces at work in a tube output / speaker that involve complex algorythms which alter the sound input and are 'very speedy' at shifting frequency gain and...
  They're interactive and kinda hard to explain and harder to emulate, RG didn't write an article...
  "How to get tube sound in a Stompbox"
  http://www.geofex.com/

[km-r]

sub-miniature tube rectifiers, anyone? if such thing existed! 

[R.G.]

I'm not all that sure that the various DIY or commercial "dying battery" simulators capture the manner in which the power source attempts to provide the current requested by the initial attack transient, but ultimately fails.  Older batteries of the carbon-zinc variety are a bit like old men trying to get up off the couch in a hurry.  They lurch forward, but fall back down because they don't have enough strength to get them all the way.  Dying batteries have a momentary current-delivery capability that is very short lived.  This is what people like about dying batteries in many of the older 2 and 3-transistor variety fuzzes.  It is that dynamic aspect that I find missing in so many attempts.

If you find anything that nails it, let me know!

Ask and you shall receive.

What you describe is very, very complicated to add to a dying battery simulator. It takes one more capacitor. 

First, for the purists and beginners, who will both have problems with this, albeit for different reasons: power supply sag is a complex issue. Like tube distortion, it will be amenable to emulating certain aspects, but probably not all at the same time.

To a first order, any power supply may be modeled as either a voltage source in series with an impedance (the Thevenin model) or a current source in parallel with the same impedance (the Norton model). This is how battery makes model their wares, as well. As a battery ages, two things happen. First, the open circuit voltage gets lower, and second the equivalent series resistance inside the battery rises. As you note, Mark, simply doing a variable regulated voltage with a variable resistance in series with it does not correctly emulate the time variance of the impedance of the battery. In a real battery, the electrolyte near the terminal conductors gets used up first, and at the end of life, the electrolyte furthest from the terminals is what is being used, so the expended electrolyte is in the way of the electricity getting out - a higher resistance, slogging through all that electrolyte mud to get out. But when no current is being drawn, the internals charge up the areas near the terminals with a standing charge. This is what is exhausted rapidly, and leads to the spring-forward/fall-back action, I think.

So, let's make this easier. Let's start with the now-industry-standard dying battery emulator which was taken from Geofex years ago: a regulated voltage with a resistance in series with it. Dialing down the voltage emulates the lower battery voltage (!?!) and a variable resistance in series lets you add a further voltage sag with higher currents, just like the innards of the battery. But it will not have an initial pulse of current that declines rapidly to whatever can be sucked from the now-in-line used electrolyte. How to emulate the short-term ability?

Easy - put a capacitor on the output. With the cap there, at times of no or low load, the cap charges to or nearly to the open circuit voltage. When a pulse of current is needed, the cap supplies it, but this rapidly falls to the regulated voltage minus the I*R loss in the series resistance. The cap discharges until the power supply takes over and stays there until the load drops and then begins charging back up.

I have several ideas for dinking with charge/discharge time constants and so on to emulate things more closely. But like with the simple dying battery emulator, each step of refinement gets you less total gain.

Give it a try. I think you'll like it. 

[petemoore]

  A load that varies with AC.
  The load varies with the AC signal is caused by many things including speaker cone weight and associated varying inertias, cone suspension, and air, among other things.
  That is the short story.
  When the load varies at signal AC, curved lines representing frequency gain cause the sag sound, those curves are moving up and down at the rate of the signal or some other offshoot phase deal...
  That is the long story, made very very short, the current/voltage going through the speaker rises and falls at the rate of the AC output of the amplifier.
  The speaker resistance rises and falls depending on the current/voltage at that time.
   Rectifiers made with tube and SS...rectify and supply raw DC. After filtering, if there's enough load on the supply line [preamp loads are much lower than speaker] tube will sag before SS. There are SS recto designs that emulate tube recto sag when under load. Rectifiers wont' sag without a load.
  Back to the Punisher...of all the Solid State 'tube sag' circuits I've heard, the only one that sounded as interesting that I've tried was in an Epiphone Triggerman 60, some other SS amps and digital emulators I think offer a 'sag' tone.
 

[R.G.]

There are a number of patents on variants of sensing the signal size and causing some distortion, compression, or both in response to the size of the signal.

[DougH]

If the OP is looking for "tube amp sag"- that implies a class a/b output circuit that dynamically increases current demand when there is an input signal. This draws from a "higher than it should be" impedance power supply which will dynamically drop the supply voltage on an attack transient, then restore it as the note decays. Zooming out a notch- the overall effect is compression. Simply using a compressor to emulate this sort of behavior seems to work pretty well.