extreme distortion with two triodes

caburron1ero · Post by **caburron1ero** » 13 Apr 2015, 20:03

hi,i have this thing in my mind ,why is not posible or maybe posible to obtain maximun distortion with only two triodes ,like a fuzz face or zonk machine,am a electronic tech but no experience in design .i ve seen a 100 schematics of valves preamps ,and all look the same ,only the ones with more valves sound good,or the ones with a solid state drive sound good(am a death metal lover,the scoop sound).i hope that someone has and idea of how to do the design or suggestion(i was thinking of a valve schmitt trigger).

okgb · Post by **okgb** » 13 Apr 2015, 20:21

For a 12AX7 tube one tube equals two triodes, don't recall if it is per half or whole tube
for a gain of 100 / 40 db , even the cascaded 80's Marshall 800's only had three preamp tubes
and one was for the phase inverter , good crunch but not extreme gain. some later Marshalls had
either diodes to ground or in the cathode circuit, that would give you more, probably not clean up well.

caburron1ero · Post by **caburron1ero** » 13 Apr 2015, 20:41

but like seymour duncan twin tube mayhem spokeperson,says that using clipping diodes is cheating,that they only use two pentodes with one as a triode,no solid state stage.fuzz face has only two transistors.

ppluis0 · Post by **ppluis0** » 14 Apr 2015, 01:43

Hi,

If you like severe distortion tones, try to build a clone of this: https://www.freestompboxes.org/viewtopic ... s&#p180066

The couple of silicon diodes are BAT41. I discover this after to publish the schematic.

Cheers,
Jose

DrNomis · Post by **DrNomis** » 14 Apr 2015, 02:42

You could try having a look at Darkrain's GTFO project thread, the GTFO is basically a clone of the GTO Supercharger High Gain Tube Overdrive Pedal which was designed and manufactured by Soldano, the GTO/GTFO is a design which uses two 12AX7 Twin Triode Tubes which are supplied with about 260V DC of High Tension, there are four Triode Gain-Stages which are cascaded so the amount of gain on tap is quite substantial, I'll post a link to the thread shortly.......

https://www.freestompboxes.org/viewtopic ... 28&t=15954

caburron1ero · Post by **caburron1ero** » 14 Apr 2015, 05:48

Genius is not all about 99% perspiration, and 1% inspiration - sometimes the solution is staring you right in the face.old solderhand says.i have a hotshot electronic engineer friend ,i said to him, what have you invent or engineered,and he said nothing,becouse all has been invented theres nothing to invent right now.he does not like to go to the trouble of inventing or thinking.

okgb · Post by **okgb** » 14 Apr 2015, 18:47

Right! there is only so much gain you can make a particular tube produce , and if it were easy likely someone would have done it, but
the pentodes can have a higher gain then the 12ax7 , you may have to investigate , the 6DJ8 I believe is a pentode & triode in a 9 pin package
pentode with gain of 500? but them you have problems with microphonics, so broaden your search and keep looking!

MoonWatcher · Post by **MoonWatcher** » 14 Apr 2015, 23:29

okgb wrote:For a 12AX7 tube one tube equals two triodes, don't recall if it is per half or whole tube
for a gain of 100 / 40 db , even the cascaded 80's Marshall 800's only had three preamp tubes
and one was for the phase inverter , good crunch but not extreme gain. some later Marshalls had
either diodes to ground or in the cathode circuit, that would give you more, probably not clean up well.

Many of what are considered to be modern high gain tube preamps really only add one more triode beyond what the JCM800 had. And since many of them also use a cathode follower which only has a gain of one, it just looks like there is more going on than there actually is.

So in the case of the JCM800, there was the somewhat cold biased first gain stage with the cathode bypass cap that boosted right at the midrange frequencies, followed by a very cold biased second gain stage, followed by a hot biased third gain stage. And - only the first stage got the cathode bypass cap.

Between the cold bias and the pair of treble peaking circuits, it adds a lot more buzz and distortion than you might otherwise have.

Something like a SLO or Recto adds a fourth gain stage, but much of what is happening with the JCM800 serves as a sort of framework. The biasing on two of the triode stages is typically not as cold as the first stage in a JCM800, but it's just a touch colder (than the center bias value found in many Fenders and other similar amps). And the cathode bypass caps are typically 1uF combined with the 1k8 resistors, which gives the same mid-emphasized corner frequency as the JCM800's 680nF and 2k7 pairing.

The higher gain amps then put the super cold bias on the third gain stage by means of the very large 39k cathode resistor. The fourth gain stage then is something you don't have in the JCM800, and it's usually biased and cathode-bypassed like the first two gain stages.

I remember catching a brief glance of a schematic for either Marshall's DSL or TSL amp, and it looked like there were four triode stages with the tone stack coming right off the plate of what would have been V2B.

The treble peaking circuits also will typically serve as sort of hardwired volume controls in between stages when combined with a grid leak resistor that forms a voltage divider. You also sacrifice some distortion definition when the resistance to the triode's grid is very large, because of the Miller effect - those triodes have capacitance built into them.

And something like a Trainwreck Liverpool or Express only had three triode stages. They usually have a very distorted sound because there's the combination of colder and hotter biased stages, and the third gain stage is voiced very brightly, not unlike the bright input of an older Marshall amp. There's no real circuitry to drop down the distortion between stages in those Trainwreck amps, and I think there wasn't even a grid resistor on the input. So they represent just how crazy you can go with just three triodes.

All of this discussion leaves out the phase inverter and how much distortion that it adds. But I think if you look at this like a "Trainwreck plus one" preamp, that you can get a pretty decent amount of distortion out of four 12AX7 triode stages. I think the hardest part of this design would be getting the four cathode resistor values to where they would compliment each other, and not be too cold. It would also be easy to contribute harsh sounds in the signal path if too many cathode bypass caps were used. This is where switching would come in handy with figuring out when there was too much.

I think that buzzy and aggressive would be easily done with four triode stages. More than enough gain for that.

DrNomis · Post by **DrNomis** » 15 Apr 2015, 01:38

okgb wrote:Right! there is only so much gain you can make a particular tube produce , and if it were easy likely someone would have done it, but
the pentodes can have a higher gain then the 12ax7 , you may have to investigate , the 6DJ8 I believe is a pentode & triode in a 9 pin package
pentode with gain of 500? but them you have problems with microphonics, so broaden your search and keep looking!

The 6DJ8 is actually a Twin-Triode tube, I've seen it used in some HiFi preamp designs.....

Note: The 6DJ8 has the same pin configuration as the 12AU7/12AT7/12AX7 except that pin 9 is connected to an internal shield rather than the centre-tap of the heater......

http://en.wikipedia.org/wiki/6DJ8

okgb · Post by **okgb** » 15 Apr 2015, 03:40

Right you are , there's something about the 6dj8 thought the two halfs are different [ 1 12ax7 , 1 12au7 ? ]
and won't help the O.P. who wanted all the gain out of one tube ? [ the point was that at 2-3 tubes the jcm 800 wasn't as much as people
wanted if the O.P. was asking for the gain in one tube ]

Intripped · Post by **Intripped** » 15 Apr 2015, 05:17

On my opinion the OP is looking for a tube version of the fuzzface

DrNomis · Post by **DrNomis** » 15 Apr 2015, 08:52

If you set up a 12AX7 so that you have two Triode gain-stages cascaded, with careful biasing you can get the circuit to clip similarly to a Fuzz Face, connecting the two Triode sections up the same way as in a Fuzz Face is not going to work simply because a Triode works differently to a Bipolar Transistor, firstly a Triode is a Voltage Amplifier and a Bipolar Transistor is a Current Amplifier, that is to say a Triode is controlled by applying a Negative Voltage to the control grid to reduce the flow of electrons from the cathode to the anode, without the Negative Voltage on the control grid the Triode fully conducts, on the other hand to get the electrons to flow from the Emitter to the Collector you need to Bias the Transistor so that it conducts and this is done by applying a small current to the base of the Transistor, basically the opposite to a Triode....

I have actually tried connecting a 12AX7 up in a Fuzz Face circuit and know from experience that it doesn't work....

drewl · Post by **drewl** » 15 Apr 2015, 13:45

I think the OP was wondering why two "stages" of a Fuzz Face obtain so much distortion compared to two tube stages.

I've built a lot of different types of amps and you can get quite a bit of distortion from just two 12AX7's.

Also tweaks to the power amp can enhance the overall gain, making it sound more like one of the high gain variants.

Some of Mesa Boogie's designs also get a lot of gain without using many more stages like the Rectifiers do.

FiveseveN · Post by **FiveseveN** » 15 Apr 2015, 14:16

This is one of those cases when using the proper terms is important. You can get various amounts of saturation with the same amount of gain. Headroom is of course essential, linearity of the transfer function as well (though something we usually have little control of) but biasing is also important: abruptly cut off half of your signal and it will sound quite "dirty". Go even further, with "folding" and "coring" and the non-harmonic products will fill up your spectrum to whatever levels of fuzz you want.

But people usually just want to ape classic designs. Then it should come to no surprise that you need three triodes to build a three-diode preamp. One can come very close by replacing the first with an equivalent SS gain stage (like Blackstar et al), since it provides little of that elusive "tube distortion" everyone seems to worship, and it's masked by further distortion anyway.

caburron1ero · Post by **caburron1ero** » 15 Apr 2015, 23:16

resistor ronker ,maybe someone will come with an idea ,i ve read that effectrode mercury fuzz first stage has much more plate volts than second stage.the idea is to use two triodes and some tone stack(faustone klipper has a unique circuit in the distortion like a valve schmitt trigger .

Rocket Roll · Post by **Rocket Roll** » 17 Apr 2015, 19:47

Soldano:

More information: http://www.ampbooks.com/mobile/classic- ... -preamp-6/

As far as "tube Fuzz Face" goes, here's something: https://www.diystompboxes.com/smfforum/ ... ic=49696.0

okgb · Post by **okgb** » 18 Apr 2015, 14:08

And I can add from experience owning a few different tube preamps, that although you can get a fair amount of distortion
out of two tubes, the quality and sound is not as complex or satisfying as ones with more tubes.
example the Mesa formula preamp , a great clean channel but the overdrive ch [ 2 tubes ] has nothing exciting or compelling that makes
you want to use it, especially compared to something like the CAE 3+ which never fails to satisfy!
Side note Carvin made a tube [ hybrid ] gtr preamp called quad X in the 80's which used something 7 tubes one into the other on the overdrive ch
9 tubes in total , I believe , just so the unit had more tubes than other preamps of the time , it seemed to make it more noisy than add good qualities!

MoonWatcher · Post by **MoonWatcher** » 19 Apr 2015, 15:02

drewl wrote:I think the OP was wondering why two "stages" of a Fuzz Face obtain so much distortion compared to two tube stages.

Something like a Fuzz Face will give the impression of more distortion because there is nothing to really attenuate the treble frequencies.

A triode like a 12AX7 has ~150pF capacitance at its input. When an amp design has grid resistance, this forms a lowpass filter.

And that is only one source of a lowpass filter in the typical tube preamp circuit - there is usually more attenuation to keep unpleasing treble frequencies in check.

So to approximate the non-treble-attenuated aspects of a Fuzz Face, a tube preamp would need treble enhancing circuitry, primarily in the form of highpass filters. Along with those, the stages could be cold biased to also increase the emphasis of clipping in the treble ranges.

Again - amps like Marshalls tend to have the treble-peaking circuits that consist of a 500pF/470k pairing. The issue becomes that the 470k into a triode's grid creates a lowpass filter that removes trebles above 2k2 Hz. So you really don't want to go with a grid resistor that is much larger than 200k - this will give you full-range up to about 5k3 Hz. Obviously 180k or smaller will ensure that nothing in the audible range is removed.

The second gain stage in many Trainwreck amps would typically have a 2n2 decoupling cap that was simply followed by either a 56k or something like a 150k resistor shunted to ground. That would yield a highpass filter of 500 Hz to 1k3 Hz. There was NO grid resistor into the third stage. Then the third stage was cold biased with a 10k cathode resistor.

So - to get the impression of the same overall amount of distortion as a Fuzz Face, the triode stages simply need to be configured as described above, which is probably the opposite of what you'll find used in guitar amplifier preamps, especially ones tailored for large amounts of distortion. The Soldano link was provided, and it makes it very clear what was done to heavily attenuate both the bass and treble content at the stages that actually are capable of clipping.

The truth be told, something like a Big Muff is probably more similar to a high-gain guitar preamplifier than one might realize. The three gain stages all will typically have trebles above 700 Hz rolled off. That is an extremely low frequency for a lowpass filter. The two stages that actually employ diode clipping then also have their frequency bands tailored to narrow the range of what is clipped. The Fuzz Face is absent of these design elements altogether. And comparatively-speaking, so is the typical high-gain tube/valve guitar preamplifier.