Diode clipping mystery

[tepsamps]

Hi,

i built a new amp, based on a little prototype i've made with diode clipping. The proto worked fine, so i decided to build a more versatile amp.
So here is the schematic. Just look at the second channel with diode clipping. I put four diodes on 2 switches to be able to switch between four possibilities.
Basically it's a fender twin channel with a 2 gain stages in front, and diode clipping between these 2 triodes, to simulate a distorsion pedal before a fender twin.
I used a 1 knob tone control and a volume also. So except the switching between diodes, the proto is the same and works fine. In the proto i can switch between 2 diodes OR no diode at all.

In the new amp, when i move the 2 switches, nothing changes. Same tone and same volume. Strange! After some time of debug i found out that the diodes don't clip nothing actually, if i disconnect them with a switch, the tone is the same.

Only difference, the 2 switches are relays for layout concerns...

So i really don't anderstand.

[Helmholtz]

At least the 2 center contacts of the switches need to be connected to provide anti-series wiring of 2 zeners and to allow for a current (to ground).

[tepsamps]

At least the 2 center contacts of the switches need to be connected to provide anti-series wiring of 2 zeners and to allow for a current (to ground).

Yes they are connected, just my schematics, sorry

[tepsamps]

Actually, it works, but i had to take diodes with lower values, like 6V or 3V and a little red led. At the second stage, i don't have 20v, even 13v...
So now i can play with my toy...

[Helmholtz]

Actually, it works, but i had to take diodes with lower values, like 6V or 3V and a little red led. At the second stage, i don't have 20v, even 13v...
So now i can play with my toy...

Makes sense.
I was actually going to ask how you determined the (high) zener voltages.

[Enzo]

And I was going to ask what the signal level was at that point.

[Gtr0]

Safe to assume that the signal level wasn't strong enough to clip at 18v?

[tepsamps]

About zener voltages, if you look at this schematic, first top channel (ch3 actually), between first and second gain stage, there are 2 16V Zener diodes to ground, intended (i think) to lower the signal before the second gain stage and add distorsion. Have we got 16V at this point ?

[Chuck H]

About zener voltages, if you look at this schematic, first top channel (ch3 actually), between first and second gain stage, there are 2 16V Zener diodes to ground, intended (i think) to lower the signal before the second gain stage and add distorsion. Have we got 16V at this point ?

As far as I can tell the Zeners will only conduct on hard picked peaks with humbuckers. Slamming chords that include the E and A strings. Not at all with single coils and only maybe on single string attack with a hot humbucker on the E and A strings.

[Helmholtz]

About zener voltages, if you look at this schematic, first top channel (ch3 actually), between first and second gain stage, there are 2 16V Zener diodes to ground, intended (i think) to lower the signal before the second gain stage and add distorsion. Have we got 16V at this point ?

From the test signal voltages in the schematic the gain of input stage (V1-A) calculates as G=57. So with gain control fully open it needs a peak input voltage of more than 280mV to produce a signal peak of 16V at the zeners.
Such high guitar signals are untypical and might only occur during the first milliseconds of the attack. So I exclude that the purpose of the zeners is to produce distortion. Maybe they are required to protect the Mute FET.
High input (hum interference) signals of several volts can be produced when the instrument cable is plugged into the amp first and then connected to the instrument or if the player touches the tip of the plug.

[Chuck H]

From the test signal voltages in the schematic the gain of input stage (V1-A) calculates as G=57. So with gain control fully open it needs a peak input voltage of more than 280mV to produce a signal peak of 16V at the zeners.
Such high guitar signals are untypical and might only occur during the first milliseconds of the attack. So I exclude that the purpose of the zeners is to produce distortion. Maybe they are required to protect the Mute FET.
High input (hum interference) signals of several volts can be produced when the instrument cable is plugged into the amp first and then connected to the instrument or if the player touches the tip of the plug.

Here's something I found regarding guitar output voltage. It's consistent with tests I've done (but never made records for).

http://tomsguitarprojects.blogspot.c...ge-levels.html

I remember DiMarzio put out a silly looking humbucker in the late 80's that had huge black hex head slugs for pole pieces. It was reported to put out a full volt on peaks.

So while those Zeners may be there to protect the mute circuit they will clip the signal under some circumstances. But it's probable in the Fender amp that this happens well after the signal is clipping in later preamp stages.

[tepsamps]

Dimarzio claims some humbuckers have an output of 500mV, the Tone Zone i think...
A few years back, i tried to clone this channel (6 gain stages, wow) without the zeners, and i ended with a jam full of larsens, untamable, and now i thought it was because of the no zeners in my layout. But i must be wrong...

[Chuck H]

Dimarzio claims some humbuckers have an output of 500mV, the Tone Zone i think...
A few years back, i tried to clone this channel (6 gain stages, wow) without the zeners, and i ended with a jam full of larsens, untamable, and now i thought it was because of the no zeners in my layout. But i must be wrong...

That high output claimed for those pickups would be SLAMMING AN OPEN E CHORD. And then only for a few milliseconds, immediately decaying to a much lower level. Look at the link I posted. It's pretty accurate in my observation. The sort of clipping those diodes in the Fender EVH amp would create isn't very useful musically. So they must be there for another reason. Certainly the designers didn't believe that only the hottest humbuckers would ever be plugged into their amp and that payers would only want hard chords to distort.

[Helmholtz]

It was reported to put out a full volt on peaks.

Zollner has published histograms (statistical distribution) of PU peak voltages repeatedly slamming chords on a R9 LP. They show that very few peaks of even 1.5V were produced. (BTW, input amplitudes above around 1V produce grid conduction clipping in the first tube.)

Problem with PU output measurements is that there is no standardized method and the results depend on a number of influences (PU height, PU position, string gauge, pick, player etc.). So values are not comparable.

High peak signals only occur for a few periods after the attack and I doubt that clipping them will be perceived as distortion, as the following hundred(s) of periods will pass unclipped. Especially as clipping by the following stages is very likely to mask such short term effects.

Also if the zeners were intended to produce some distortion it would not make sense to choose that high zener voltages which will leave lower output PUs mostly unaffected even at full gain.

[tepsamps]

When i measure the output of the most wound PU i 've made, i get no more than 300mV that i can read on my multimeter. And it's pretty hot yet!

About the zeners, this first gain stage is pretty hot also, except the 470pF coupling cap, there is no signal attenuation before the second gain stage, anyway it sounds great...
On youtube...

Diodes clip the signal, but don't clip the noise, so when i switch between different clipping, the noise level is the same, but the output level vary.