The peavey transtube thing...

[brett]

I wonder if this is one reason why "super-starved" systems can sound so good.  Joe Davissons's bias network on the Blackfire is a classic, with 4.7M resistor from V+ to base and a 1M from base to ground.

[Steben]

I wonder if this is one reason why "super-starved" systems can sound so good.  Joe Davissons's bias network on the Blackfire is a classic, with 4.7M resistor from V+ to base and a 1M from base to ground.

Well I'm not sure.
Maybe with BJT's, if something can pull the power down to 4,5V, 1/6 of the power could reach a bias below 0.8V, which brings you in knee voltage area... in fact it is useful with FET's too...
What about a compressor idea, but instead of pulling the AC input down, doing it to DC power line?
That could bring the biases of the trannies into soft clipping, combined with compressing amplitude loss.
Nice...

[PaulC]

What about a compressor idea, but instead of pulling the AC input down, doing it to DC power line?
That could bring the biases of the trannies into soft clipping, combined with compressing amplitude loss.
Nice...

That's what I was talking about with the maven peel circuit.  an audio detector circuit is used to control a variable regulator in the pwr supply.  The circuit has two section - one to set a pre fixed B+ voltage drop, and the other to set a sag amount to this drop.  There's a patent on this, but like talked about here there's several ways to do this. 

There's a bunch of ways you can get sag type of effects.  A simple way that will get some "punch" is to have a comp built after an OD circuit, but to have it keyed in at the input of the OD so it's controlled by the clean/uncomp'd signal.  That way you can have the compression of the overdrive circuit going on, but add a little bit of attack back into it at the output.  This part of what goes on with the pwr amp stages.  The signal is clipped, but the pwr supply sag sort of adds an attack back to the signal.

As pwr supply sag happens the headroom decreases increasing clipping at the same time as there's an output level loss.  You could add a another type of signal variable circuit with the first one to add a little more gain to get more clipping as the output comp adds punch.

You could add a voltage ref compliance circuit to a diode pair that is signal variable to get the diode threshold to change.  You could sort of think about it as having a pair of led's at low signal levels, and as signal increases they turn into a pair of Silicons ets...  If it's timed just right they could add a little bit of level drop after the attack acting a little like power amp clipping.

You could add an LDR, and have the resistance as a shunt in the pwr supply.  diode lights up from a detector, and the B+ drops.

Lots of ways to do this.

Later, paulC
Heritage amps/Tim & timmy pedals

[Ben N]

There's a bunch of ways you can get sag type of effects.  A simple way that will get some "punch" is to have a comp built after an OD circuit, but to have it keyed in at the input of the OD so it's controlled by the clean/uncomp'd signal.  That way you can have the compression of the overdrive circuit going on, but add a little bit of attack back into it at the output.  This part of what goes on with the pwr amp stages.  The signal is clipped, but the pwr supply sag sort of adds an attack back to the signal.

I know this pedal part is only a side note to an amp discussion, but:
I like using a compressor after OD, to get touch sensitivity with even signal levels. This idea of a post-OD compressor with clean sidechain would also have the benefit of reducing the old problem of OD-comp chains: noise. 
(It is also somewhat like what I believe was the EA Fuzz (anyone recall? link?)--a one transistor fuzz wrapped in a 571 compander. That, IIRC, preceded the clipping stage with heavy compression and followed it with a variable dynamics stage that could be set for expansion, controlled by the uncompressed signal in a sidechain.)
Ben

[Steben]

Noise: that's why I think shifting the bias towards cutoff at the base (or gate) of transistors is better than the post-compressor. The only set back is there is only compression hand in hand with distortion, but hey: isn't that "sag" after all!?
I'm thinking of a circuit I will draw out a little.
There are several options:
- a single stage followed by another one, so you have both sides compressed.
- A mu-amp style stage with the power supply variable.

I guess the mu-amp has a very nice potential, since:

- it's clipping is very tubey
- It's self-biasing, creating symmetrical response no matter what power (push-pull emulation)
- you have the possibility to create compression without the distortion too. You only need to amplify your dry signal first into the audio detector...

FUN! A possible tube-killer! (read: push-pull power tube killer).
If you really want to emulate a push-pull amp, I guess you only need one mu-amp stage, since there is only one power amp.
Besides, High Gain sounds are not that bluesy tubey, so no need to bother.

Patents: pff, i even haven't seen it. You can not forbid electronic theories.

[PaulC]

Actually with the post od comp you wouldn't need to have any noise issues.  A lot of these circuits are already pretty loud with the signal being knocked down at the output to keep from killing an amp.  It's a simple thing to have an ldr/jfet or whatever as a small shunt on the output.  All you're looking for is a small dip in output - not any type of heavy comp.  Just something to add a little bit of attack emp back into the compressed od.  The thing that's different about this way is it's keyed in before od compression so the output comp is controlled by your picking dynamics.   Since a lot of pedals are to loud before output padding to begin with you don't need any type of gain recovery circuit to make up for the loss in the comp circuit.  But unlike real "sag" there's no distortion increase to go along with the level decrease.

i think I saw a squezzer type comp on here before using a MU amp, but I can't remember for sure.  You could do the pwr suppy sag/comp to the mu amp, and you can also sag/shift the bias voltage - then you get asym clipping out of it.  If you set up a bias supply like R.G. does you can have the detector circuit hooked up to the bias supply making it shift.  you can do this with mosfets and other things that use a bias voltage.  You could tap the detector pre or post the mu amp for different results.  Lots of fun geeky ways to spend a weekend with this stuff!

(It is also somewhat like what I believe was the EA Fuzz (anyone recall? link?)--a one transistor fuzz wrapped in a 571 compander. That, IIRC, preceded the clipping stage with heavy compression and followed it with a variable dynamics stage that could be set for expansion, controlled by the uncompressed signal in a sidechain.)
Ben

Damn - didn't know it was done before.   Maybe someday I'll have an original thought!

Later, PaulC
Heritage amps/tim&timmy pedals

[Jay Doyle]

From what you guys are talking about, and I haven't read a whole lot of this thread, just the last couple of posts, looking into servo controled biasing of transistor stages maybe something you guys might want to do.

That way you can have control circuitry to tell the servo amp what to bias the transistor to...

Kind of thinking out loud here...

[puretube]

supply-voltage-
reduction in SS-circuits
might not sound as advantageous
as in Tube-circuits...

[PaulC]

supply-voltage-
reduction in SS-circuits
might not sound as advantageous
as in Tube-circuits...

True - but given the right circuit it's just one more way to get ss stuff to behave a little more like tubes.  That's the whole idea for the transtube circuit - trying to get triode like bias shifts with transistors.  B+sag/bias shifts/headroom changes - just things to play with to get some dynamic changes in ss circuits like what you'd find in tube circuits.

Fun stuff

Paulc
Heritage amps/Tim & timmy pedals

[puretube]

I meant: tube-amp SAG is not just: less loud...

[PaulC]

I meant: tube-amp SAG is not just: less loud...

You're right - and we talked about that in this thread.  With the drop in B+ you get a reduction in output level, and you get a decrease in headroom - more distortion.  The operating points change which will change the harmonic content of the stages.  Many things happen with sag.

The idea of the back end comp with the front end detector is just one way to get some of the sag element.  It's a  way to bring some "punch" back into a compressed distorted output.  The things about dynamically control clamping thresholds and bias shifts are other ways that can all work together to get things closer.

Just theory and ideas.

Later, PaulC
Heritage amps/Tim & timmy pedals

[Jay Doyle]

I think for just B+ sag that a buffered and doctored level detector would do the job while keeping the stages common source/emitter.

Let's say we have two JFET stages in series and we want the V+ on those stages to sag with the strength of the note but keep all else the same, no extra circutry in the signal path.

We buffer the input of our circuit and run our buffered output to our JFET stages and, for the sake of argument and the fact that I know it well, the Ross Comp rectifier circuit (everything but the last transistor with the Sustain pot on it's emitter). Sitting on the integration cap is a reference voltage that goes from 9V to ground with the strength of the input, can be controlled with attack and decay circuitry, the extent of the drop of which can be controlled with diodes and/or the setup of the rectifier (putting resistors on the emitters) and the size of which can be controlled with a simple pot. We can then use the voltage on that cap, buffer it with a good stiff opamp, like a 5532, and then run the output as the V+ voltage for our two JFET stages. So instead of controlling the gain of something (compressor), or the frequency cutoff (auto-wah), we have our level detector in this circuit controlling the V+.

No need to compress the signal or use a regulator, we just use a level detector to control the V+ on the JFET stages.

I would suggest using a level detector a tad less rudimentary than the Ross though...

This same idea can be used with servos to control sag of the BIAS POINT of each of the JFET stages...

[slacker]

The Tone God did this with his punisher. It's an external "sag" simulator that you can add to any pedal.

[puretube]

Jay: do we need a CV-derivate, to adjust the jfet bias, too?

[PaulC]

The Tone God did this with his punisher. It's an external "sag" simulator that you can add to any pedal.

Now that's a cool idea!  A box that you could use with other pedals to get them to sag.  I always had to build that type of thing into whatever project i was working on.  Never thought about building it into a box that could sit on the shelf like my pwr supplies.

The only problem with this way as I can see it is if there's a lot of filtering in the pedal being used then the rise and fall time of the sag can be really slow.  If this was going to be used with stand alone pedals I'd maybe put the pedals main filtering on a switch so I could remove it if I used this box.  That's not a mod to the punisher - it's to the otherpedals.

I did build a keyed in comp along these lines that I used for that output thing.  It had parallel key in jacks so you could plug your guitar into it to get the key in signal, and then bring it right back out to go to an OD.  Then you'd plug the OD into the comp in and outs just like you had a comp following it.  Same thing as plugging a y cord into any comp with a key in circuit...

PaulC
Heritage amps/Tim&timmy pedals

[Jay Doyle]

Jay: do we need a CV-derivate, to adjust the jfet bias, too?

No, you don't NEED to, but you can. I haven't looked at servo stuff for a while, but you use an opamp's output to drive the gate/base of a transistor until the two inputs match (basically the transistor circuit is in the opamp's feedback loop) one of the inputs senses the bias voltage (connected straight to the drain/collector) and the other input is the voltage you want the servo to force the transistor to bias to. That voltage can be the CV or just a simple voltage divider.

I think the rule is that an inverting stage needs a NON-inverting servo (meaning the + input would sense the bias voltage and be connected to the drain) and a non-inverting stage needs an INVERTING servo.

[Jay Doyle]

The Tone God did this with his punisher. It's an external "sag" simulator that you can add to any pedal.

Actually his varies a resistance in series with the voltage to drop current and voltage. Mine just drops the voltage, with a stiff opamp the current supply will not change or at least be taxed.

Also, is the punisher confirmed? Because unless it is meant to run from a dual supply, I don't see it working without a bias on that first opamp.

[Steben]

The Tone God did this with his punisher. It's an external "sag" simulator that you can add to any pedal.

That's it! great. saves a lot of time...
Now plug this thing into a mini-booster... another step towards great stompbox "tonamics"...
The only suggestion I would make is using the FET in a voltage divider, not in series. Puts out variable voltage too, instead of current.

Also, is the punisher confirmed? Because unless it is meant to run from a dual supply, I don't see it working without a bias on that first opamp.

It goes without saying...