Overdrive/clipping in a valve amp

[MrBump]

Possibly a dumb-ass question, but in an amp with a MV, and a gain/volume setup, what's happening to produce clipping/distortion?

I vaguely get the idea of diode clipping in stompboxes, symetrical/asymetrical etc.  Is that different in a valve amp?  Would it use the valves to produce some sort of clipping?

I get this for a mild sort of breakup, preamp valves running hard into power valves - but what about the higher gain stuff?  Is that all soft valve clipping, or do manufacturers stick some more aggressive asymetrical clipping circuits into amps?

Mark.

[BigB]

Basically, it's preamp valves clipping - that is, you have a second gain stage - controlled by the 'gain' pot - that you can drive into clipping. High-gain amps usually cascade even more gain stages to produce that insane amount of clipping.

Some amps also include a diode clipping circuit - at least the latest channel-switcher version of the JCM800, the Silver Jubilee and the JCM900 Dual Reverb (the JCM900 SLX has an additional valve gain stage instead).

Now I guess some of our in-house experts will pop in and correct me :mrgreen:

[dave_mc]

i suspect the answer is, "it depends on the amp". I've heard of some valve amps which also have diode clipping in there.

obviously wait for the amp gurus before taking my answer as gospel. :lol:

[jpfamps]

All amplifier stages have a maximum voltage they can swing before they start to clip the signal.

This is simply because you run out of voltage.

e.g. in a pedal run of a 9V battery, theoretically you can produce an unclipped signal of 4.5 VAC peak-to-peak (3.2 Vrms), although in practice you can probably swing less than this.

If you try to amplify the signal voltage beyond the voltage rails the peaks of the signal will be clipped off, hence the term clipping.

In a valve amp there is also another way the signal can be clipped and this is by a valve going into grid conduction. This mechanism doesn't really exist in solid state amps.

One way of reproducing clipping is to use diodes, a la various distortion pedals.

Modern amps with master volume controls are designed so that the pre-amp will clip the signal with the drive level up, so that you can generate distorted tones in the pre-amp and then reduce the level going to the power amp, so that you can have a distorted tone at lower volumes.

As stated above there are valve amps with clipping diodes in them (JCM900 etc).

[MrBump]

Thanks, that's kinda what I thought.

Presumably the monster-distortion amps such as dual rectifiers do a little more than that?  Does the fact of having "dual rectifiers" somehow produce more distortion?

[Frank]

rectifiers aren't part of the signal path, they convert alternating current from the power transformer into direct current (valves only operate on DC)

[BigB]

Thanks, that's kinda what I thought.

Presumably the monster-distortion amps such as dual rectifiers do a little more than that?

Just more cascading preamp tubes, really - remember that a 12AX7 (the standard preamp tube) is a double triode - that is, there are really 2 distinct tubes in one bottle. This, and I guess selecting hi-gain preamp tubes  and some tweaking in the preamp circuit to push them as much as possible.

  Does the fact of having "dual rectifiers" somehow produce more distortion?

As Franck said, the rectifier is not in the signal path, so it doesn't add any gain. The point with rectifiers is that they impact the way tubes (power tubes mostly) work. Overly simplified (please in-house experts correct me if I say anything wrong): a tube rectifier will cause "sag" - that is, voltage drop when there's a quick peak in input level, voltage drop that kinda starves the tube, causing a soft, spongy compression on attacks, specially on low ends (which requires more power). Diode rectifier don't have this "problem", so they'll give you faster attacks and tigther lows. Usually, tube rectifiers are more "vintage" sounding and diode rectifiers more "modern" sounding.

[Dmoney]

Thanks, that's kinda what I thought.

Presumably the monster-distortion amps such as dual rectifiers do a little more than that?  Does the fact of having "dual rectifiers" somehow produce more distortion?

Cascading gain stage is one way of doing it. Dual Rec's have an extra trick though.

5150's & Dual Rec's and others, have a cold biased valve clipping stage. Its basically a triode stage (half a 12AX7) with a big resistor on the cathode. Usually 39K. I think its 39K in the 5150 and Dual Rec at least. I think both these amps got that idea from the SLO-100. If you can read amp schematics you should see it. I think its usually done at the 3rd gain stage. It works the same way all clipping works I guess, but as with other devices used for clipping, the tone varies between them. LED clipping won't sound like MOSFET clipping and neither will sound like that valve stage.

'sag' is indeed the thing with valve rectifiers. there IS an debate about changing 1N4007 diodes in SS rectifiers for UF5408's or other hyper fast switching diodes. people claim it cuts some very high end fizz out of an amp. not sure I believe that. Neither would add gain, like BigB says.
 

EDIT: I'd also like to add, that I think although the rectifiers and filtering after them isn't always considered 'in the signal path' (I think it is considered in the signal path by some...), if you have a bad rectifier or bad filtering it can and probably will impact on your tone a great deal by adding harmonics of your mains frequency into your amp.

[MrBump]

Awesome replies, thanks guys.

Mark.

[jpfamps]

Rectifiers ARE part of the signal path, albiet for only a small portion of the signal cycle when they are supplying the pwoer to the amp and are pulse charging the main filter caps.

The high gain sound of the Dual Rectifier is not due to the rectifier/s but due to the cascading of triode gain  stages in the pre-amp.

[BigB]

5150's & Dual Rec's and others, have a cold biased valve clipping stage.

What does the cold bias change here (if I may ask) ?

[BigB]

Rectifiers ARE part of the signal path, albiet for only a small portion of the signal cycle when they are supplying the pwoer to the amp and are pulse charging the main filter caps.

Do you mean the signal REALLY goes thru the rectifier ??? Or that the rectifier's behaviour impacts other parts the signals goes thru ?

Or is that a totally clueless question ? (trying to make sense of audio electronic and well, computer programming was way easier :lol:)

[Dmoney]

I don't mean bias in relation to power tubes.
Each triode stage has a bias. I'm not the best guy with the maths for this but I think its done between the relationship of plate and cathode resistor at each stage. not all 'stages' add gain. A Cathode Follower (I think im right in saying this) for example, usually used to drive a tone stack, won't be adding gain to the signal.

The 3rd stage cold clipping in the SLO & 5150 and more (again if im right) brings down that stages cut-off point. Or the point where the signal going through it starts to get clipped. This might be why the clean channel on an SLO is a lower level than the Overdrive. It chucks out gain in the earlier stages so that stage isn't driven into clipping as much (or at all).

If I'm wrong, or someone can explain it better, chime in.

[jpfamps]

Rectifiers ARE part of the signal path, albiet for only a small portion of the signal cycle when they are supplying the pwoer to the amp and are pulse charging the main filter caps.

Do you mean the signal REALLY goes thru the rectifier ??? Or that the rectifier's behaviour impacts other parts the signals goes thru ?

Or is that a totally clueless question ? (trying to make sense of audio electronic and well, computer programming was way easier :lol:)

It is not a clueless question.

Yes, the signal does go through the rectifier.

For current to flow through the load (ie the speaker) then you have to have an electrical circuit. For most of the signal cycle the AC current to the load completes the circuit flowing through the filter caps. The filter caps are pulse charged through the rectifier and whilst the mains filters as being chraged the signal current flows through the rectifier.

If you think of an amp as a device that modulates the mains to produce audio then this is more obvious.

[BigB]

If you think of an amp as a device that modulates the mains to produce audio then this is more obvious.

Thanks jpf - FWIW, I was just about to ask on some electronic forum if describing an amp as "a device that modulates the mains with the input signal to produce an analog 'biggest' output signal" was correct or if I was totally hopeless :lol: