Preamp distortion or power amp distortion. How to tell?

[diydidi]

Hi Guys.
Lately been thinking about the topic of where a given amp creates most of its drive/ distortion.
If one looks at the older Marshall amps. They all develop most of their distortion in the power amp section. A Jcm800 on the other hand has its distortion coming from the preamp.
I think master volume amps are mostly preamp driven too.
Is there a way by just looking at the schematic to determine where the distortion and drive of an amp is generated?
Maybe I can measure it somehow??

[m1989jmp]

Hi Guys.
Lately been thinking about the topic of where a given amp creates most of its drive/ distortion.
If one looks at the older Marshall amps. They all develop most of their distortion in the power amp section. A Jcm800 on the other hand has its distortion coming from the preamp.
I think master volume amps are mostly preamp driven too.
Is there a way by just looking at the schematic to determine where the distortion and drive of an amp is generated?
Maybe I can measure it somehow??

You can measure it precisely with a sine wave generator and oscilloscope.

Check out this SLO100 preamp analysis too, it helped me understand some fundamentals:

https://www.ampbooks.com/mobile/clas...-slo-preamp-1/

[Chuck H]

There are a lot of different combinations of preamp/power tube distortion. I can tell by looking at a schematic, but I can't tell someone else how in less than a few courses. That is, it's not as simple as "master volume = preamp distortion and no master = power tube distortion" Most vintage and vintage style amps start clipping the power tubes first. Many, at some point as you advance the volume control, also clip in the preamp somewhat later. Some of these amps have master volumes and this can indeed reduce or eliminate the power tube clipping and leave only the preamp clipped aspect. That's one reason many players don't like master volume amps. There is too much alteration of the basic tone of the amp at different MV settings. Nearly all modern cascade preamps tube and master volume amps are processing the guitar signal for distortion and THEN amplifying it with the power tubes. In these amps it's pretty much all preamp clipping. There are some amps and mods for amps that are sort of between these two. Channel stacked type mods or amps like the JCM800 you mention are intended to allow for preamp distortion when the master is down, but most players prefer some combination of preamp and power tube clipping in amps like this and use the master only judiciously if at all.

Being able to tell just looking at a schematic can be tricky unless you're aware of circuit function for things like preamp tube bias, voltage dividers, tone control types, feedback circuits and also able to see them in the circuits even when it isn't obvious. But in general a vintage or vintage style amp will have one or two gain stages driving a phase inverter and a modern master volume type will have three or four. A problem with just counting stages is that you need to be able to tell the difference between a function stage and a gain stage. For example, playing the reverb channel on a classic Fender BF design has the signal going through three gain stages before the phase inverter, but the signal is heavily padded before the third stage to match it with the reverb signal. So this third gain stage is a mixing stage. Then, and also not obvious, the signal is heavily loaded by the tremolo circuit and THEN it goes to the phase inverter. In the classic Marshall non master circuit the signal also goes through three triodes. But one is a "cathode follower" which offers no gain at all. So there are really only two gain stages. This is just a quick example of how complicated it can be to explain to someone how to tell just by looking at a schematic.

And, as noted, a signal generator and a scope for signal tracing/watching are nice to have. But if you don't have access to a lot of amps, only schematics, I don't suppose it would help much.

[J M Fahey]

In general, a power amp starts clipping when power tube grids, normally biased negative, reach 0V , which is the point at which they can put out most current.
Thatīs what Class AB1 means (or ClassA1)
Grids can go positive and tubes will put out even more current and power, but then they start drawing lots of current which the PI simply canīt provide, so itīs not normally used.

So a classic Fender/6L6 amp will reach peak power and start clipping with +52V peaks from the PI and a Marshall/EL34 one with +38V peaks.
Or respectively some 35/28 V RMS

Is that taxing the PI?

Not really, a preamp tube fed from a 250V rail or a PIn fed from 300/350 (remember they lose some 50V because their cathodes are that high) can put out some 80V RMS ... what means that power tubes will always clip before the PI.

And earlier tuibes in the chain, even less, since they handle , say, 10 to 10V RMS, piece of cake.

So in a classic Tube amp, power tubes will be the first to clip.

Last preamp tube will not, because it will need to supply 1 to at most 2 V RMS to power amp, so "just by looking at the schematic" , if it follows this classic/standard architecture, power amp will clip first.
About the same with normal SS amps: power amps will clip with about 1V RMS at the input, while most preamps will clip with some 8 or 9 V RMS, again ample headroom ... which is fine.

If you want to clip a preamp tube you must drive it to beyond 80/90V RMS out, which is **a lot** so it usually requires at least an extra gain stage in the preamp, and then a Master volume at the end of the preamp so you tame that HUGE squarewave down to 1/2 V RMS or less, in a controlled way, so you make certain that you are using preamp clipping and not just slamming the power amp (with bleeding ears results)

Thatīs why first preamp distortion amps were Marshall JCM800: "a Plexi with an extra gain stage" and Mesa Boogie Mark1: "a Fender amp with an extra gain stage" plus added Master Volume.

Thatīs why just adding a Master volume to a classic Fender ... or even a Plexi, is a failure, those preamps do not have enough gain to (significantly) distort on their own.

[glebert]

Maybe I can measure it somehow??

Don't know if this is what you mean, but this is something I have done. This only works with amps with an effect loop, but you can feed a clean signal into the "return" (I use a stand alone preamp with lots of gain range for this) and increase the level until you hear (or see with an oscilloscope) the power amp distortion. Similarly to hear the preamp distortion go out of the "send" jack into a clean amp. I did this recently on my Hot Rod Deluxe.

[pdf64]

...Lately been thinking about the topic of where a given amp creates most of its drive/ distortion....Is there a way by just looking at the schematic to determine where the distortion and drive of an amp is generated?...

To try and summarise things, I think it should be fairly easy; as input signal amplitude increase increased from 0 -

For a non master volume amp, with any reasonable control settings, the power tube grids will be the first point in the signal path that clips, hence that must also be the point where the most distortion gets generated.

For an amp with a master volume, it will be completely dependant on the control settings; with the master up full it will be the same as above.
If the master volume is turned down below that sufficient for the power tube grids to clip, then if there's a single gain control, with most reasonable control settings, it's most likely that the last preamp gain stage will be the first point in the signal path that clips etc.
If there's most than one cascading gain control then it's going to depend on their relative settings, but the same principle can probably be applied.

[diydidi]

The main reason im asking this, is that I have been getting many requests to install fx loops in customers amps. My only concern is that if most of the amp’s distortion is generated in the power amp, I see no use for the loop as the customer might as well add his cho, mod pedals before his amp anyway.
Is my reasoning correct?

[Pedro Vecino]

According to that idea (power tubes will be the first to clip), how can we explain then the big difference in the distortion level that exists between a master volume installed in a Plexi previous to the phase splitter or after it with maximum preamp gain?
In a Plexi, using a post phase splitter master volume or an attenuator in the speaker output the distortion level is very similar with the gain set to the maximum.

[frus]

The main reason im asking this, is that I have been getting many requests to install fx loops in customers amps. My only concern is that if most of the amp’s distortion is generated in the power amp, I see no use for the loop as the customer might as well add his cho, mod pedals before his amp anyway.
Is my reasoning correct?

Of course. If the distortion is generated in power amp, there's no way to put some effects "after" the distortion (unless it's miked etc). If you use pedals before the (pre)amp, then putting some effects in fx loop is the same as putting them after the dist pedals. So no need for fx loop anyway.

[diydidi]

So this means that the PI is responsible for the clipping not the power tubes?
Thats is way a ppimv is better than a pre PI MV?

[Chuck H]

If you're amp uses more bias volts and has low-ish PI volts (reduces swing volts) you'll get a lot more distortion from the PI at relative VOLUME KNOB settings due to drive from the preamp. But the power tubes will still clip first if NO master is used. The reason a PPIMV sounds similar (in some amps to some people) is that in many respects a square wave is a square wave. So what if breakup starts one digit higher on the knob.?. You hardly notice. In the case of the classic Marshall circuits, the preamp does begin to distort soon after the power tubes and the PI is typically of low-ish voltage (relatively) so it kind of works. The Bassman/Marshall design actually has LESS gain than the BF Fender design due to topography and losses in the tone stack. There's more distortion in the Marshall design because the power tubes ARE driven to clipping and then the preamp, being of lower headroom, starts clipping soon after creating more of the scenario where you're clipping a clipped wave form. The VOX AC50 has a similar topography but higher losses in the tone stack. As a result most AC50 models used a paraphase inverter because using a long tailed pair would barely push the power tubes to full power before the preamp started clipping. It's possible the LTP AC50 was a short lived incarnation.

re: effects loops... Since NEARLY all vintage style amps clip the power tubes first there are considerations. Instead of effecting a distorted tone you may be distorting an effected tone. But this still works most of the time for most effects. Consider the Fender BF design. You can crank the thing into clipping and, if you're conservative, you can dial in some reverb and it's not too bad. That's a bold example, but effects that work just fine in loops feeding clipped power tubes includes delay, chorus, phase shifters, octave pedals, tremolo, etc. If you're using one of those Digiyuk module type things then your distortion is generated by the the effects unit and your amp is just a slave. Since you don't have to drive the amp to clipping for full power it still works just fine.

[Pedro Vecino]

I had to read it several times, but really a great explanation. Thanks!

[Justin Thomas]

If you're using one of those Digiyuk module type things...

Sigged. The next devolution after moving to sillicooties...

Justin

[Chuck H]

I hate to think it, but it's probably true that if there is some kind of guitar amp museum twenty years from now there will be a Rockman, a Digitech processor and a Line 6 included in the displays

[Pedro Vecino]

I used in the late 80īs and first 90īs a small white Steinberger (which I still have) and... ("Fanfare for the common man" intro): a Rockman Sustainor!