Tubes in lieu of clipping diodes

[dido__15]

Hello folks,

I'm a hobbyist amp/effect builder, but I'm not the most technically minded person so was hoping to know if what I'm about to suggest is technically possible before I start blowing up tubes.

I'm sure most pedal builders/tinkerers are aquantied with the effects of using a transistor or FET in place of a clipping diode. After playing around with the circuits at runoffgroove, using FETs in place of triodes in a tube amp circuit, and was wondering if it would be possible to use a triode tube in the same way as an FET clipper?

If it is possible it could be fun to use three triodes (perhaps 12ax7 + el84) to create a discrete op-amp style circuit, with another 12ax7 used with each triode side as a clipper in the op-amp feedback loop.

I'm probably way on the wrong track here, so was hoping someone with a little more knowledge could shoot me down with a technical explanation of some sort.

Cheers

James

[Cliff Schecht]

You're not going to be able to clip a diode connected triode stage with your typical op amp and 9V supply. Unless you somehow bump the 9V up to 40-50V for the op amp stage and have an op amp that can drive outputs large enough to actually saturate the diode connected triode stage, you're not going to get too far with this. I can't give you much more detail without running some simulations and posting the results but in essence it's not going to sound much different than any other forms of clipping (some low-pass filtering with germanium diodes would get you close). Since you aren't operating the triode stage in the linear region I'm not even sure it follows the 3/2 law that gives triodes their unique distortion sound anyways..

[dido__15]

I think you may have misunderstood my meaning. I meant building a tube amp in this way, using the tubes to build the discrete opamp. A circuit sort of like in this link http://gpds.uv.es/elec_anal_2/archiv.../project07.htm , not tubes + opamp in a pedal. I figured the sound wouldn't be some amazing new discovery, just wasn't sure if it was possible.

[printer2]

Why go through all the trouble of building an opamp whos purpose is to cancel out distortion (by using a lot of gain and then using feedback to reduce it and linearize the frequency response). And on top of that put a tube in the feedback network to make it non-linear?

[celeste]

Could this be what you are looking for? Real tube diodes tasked as clipping diodes in a tubescreamer type circuit.
http://www.ax84.com/bbs/dm.php?thread=437984

[dido__15]

Originally Posted by printer2

Why go through all the trouble of building an op-amp who's purpose is to cancel out distortion (by using a lot of gain and then using feedback to reduce it and linearize the frequency response). And on top of that put a tube in the feedback network to make it non-linear?

For exactly that reason. The op-amp stage can be used to provide lots of perfectly clean gain to overdrive a power section, or a pair/triplet/quad of clipping triodes can be blended in to provide distortion. Series resistance could be added with the diodes (another pot perhaps) to tailor saturation. Opens all sorts of possibilities such as blending in a high pass/low pass filters on the clipping circuit so that all frequencies are amplified to the same level but only certain ones are clipped by the feedback triodes/diodes.

Originally Posted by celeste

Could this be what you are looking for? Real tube diodes tasked as clipping diodes in a tubescreamer type circuit.
http://www.ax84.com/bbs/dm.php?thread=437984

This looks very interesting, had considered tube diodes but figured their response wouldn't be too dissimiliar to solid-state diodes. Could be worth a try though.

Originally Posted by Cliff Schecht

Since you aren't operating the triode stage in the linear region I'm not even sure it follows the 3/2 law that gives triodes their unique distortion sound anyways..

Two possible ways of getting around this. Either 3 triodes/valve diodes on the positive side and two on the negative side (or vice-versa), or connecting plate/anode to grid on both triodes and using this connection to connect to the circuit with the two tubes joined at the cathode. Would probably need a blocking diode in parallel with the plate/grid connection (Due to there being no body diode as in a JFET/MOSFET).

[BiggerJohn]

Originally Posted by celeste

Could this be what you are looking for? Real tube diodes tasked as clipping diodes in a tubescreamer type circuit.
http://www.ax84.com/bbs/dm.php?thread=437984

Yep, 6AL5 is a tube double small signal diode. there is also a triple diode tube but i forgot the number.

[printer2]

Originally Posted by dido__15

For exactly that reason. The op-amp stage can be used to provide lots of perfectly clean gain to overdrive a power section, or a pair/triplet/quad of clipping triodes can be blended in to provide distortion. Series resistance could be added with the diodes (another pot perhaps) to tailor saturation. Opens all sorts of possibilities such as blending in a high pass/low pass filters on the clipping circuit so that all frequencies are amplified to the same level but only certain ones are clipped by the feedback triodes/diodes.

We can provide lots of perfectly clean gain as it is. Using a feedback loop results in a clean signal until all the gain is used up and then you hit a brick wall. Might as well use SS diodes. Blending in a non-linear element(s), you can do that in the standard signal chain as it is. You can do everything you suggest above in the signal path rather than in the feedback path. Other than having a different input you have your standard tube amp.

If you want to go that route do a search on the circuit.

http://www.glass-ware.com/tubecad/TC...criptions.html

Take a look at these compressors, they modify the feedback to provide compression, not a bad starting place.

http://www.soundonsound.com/sos/sep0...ompressors.htm

Some of these

http://www.odysseyprosound.com/thermionic_culture.php

A host of others once you start digging.