Do it right or don´t do it 
Here´s Garnet Herzog, the secret weapon behind Canadian band Bachman Turner Overdrive (hey, if they added the effect´s name to band´s name means they were serious about it )
12AX7 driving a *power tube* , a pentode; a triode simply doesn´t cut it, driving an output transformer into a resistor for convenience but even better a real speaker.
And then attenuating that signal to drive a real LOUD amplifier.
The "pedal" looks like a tiny head ... I wonder why
Here´s Garnet Herzog, the secret weapon behind Canadian band Bachman Turner Overdrive (hey, if they added the effect´s name to band´s name means they were serious about it
)12AX7 driving a *power tube* , a pentode; a triode simply doesn´t cut it, driving an output transformer into a resistor for convenience but even better a real speaker.
And then attenuating that signal to drive a real LOUD amplifier.
An externally hosted image should be here but it was not working when we last tested it.
The "pedal" looks like a tiny head ... I wonder why
As you may have noticed, there are / were lots of 12AX7-based guitar distortion / overdrive pedals, but none that I have come across used an output transformer.
The reason for this is that the 12AX7 is only comfortable driving really high impedance anode loads - like the 100k anode resistor used in thousands of preamp stages. Unfortunately, it's really difficult to make an audio transformer with a 100k primary impedance - that would require an enormous number of coil windings, and an enormous primary inductance.
The Herzog that J. M. Fahey mentioned started out as an experiment by guitarist Randy Bachman. He used an entire Fender Champ between his guitar and his main guitar amp, with a volume pot wired where the speaker should go.
Valve guitar amps don't like working without a speaker (or dummy load) connected, so Bachman kept frying his Champs. Eventually Bachman went to guitar amp tech / inventor Gar Gillies to see if he could fix the thing and make it work, which he did by basically adding a dummy load, and then building him a big amp to go after the Herzog: "Gar" Gillies' Herzog® - All Tube Guitar Effect - garnetamps.com - Home of the Garnet™ Amplifier Company
Champs tend to be harsh-sounding when overdriven, so I'm curious how "Gar" Gillies got Bachman's Herzog + main guitar amp to sound so smooth. Listen to the song "American Woman" by Bachman-Turner Overdrive to hear what I mean.
Personally, I have my doubts as to whether the output transformer is necessary go get good sound out of a valve preamp pedal. But if you are determined to go that route, you need a valve that can drive much lower loads than a 12AX7. I don't think you need to go quite as big as a 6V6, though, if you are willing to experiment, and know enough to tinker safely with valve gain stage design.
I think one possibility is a 12AU7, which, from the datasheet, looks like it could drive something like a 10k - 25k output transformer with careful design.
Another possibility is a small pentode (as Fahey said, we have all come to expect the sound of overdriven pentodes from a guitar amp.) There are lots of little NOS pentodes from the era of valve radios and valve TV sets out there, unloved and unwanted, and some can be had quite cheap. For some reason, triode-pentode valves tend to be even cheaper, and they give you a free triode to use in your preamp as well as the output pentode. There are even some that contain a triode that appears to be identical to the ones in a 12AX7.
A while ago, I worked for a while on a project very much like the one you've outlined. I used a 6JW8 triode/pentode valve, eventually preceded by a 6AQ6 triode as the design progressed.
I did get it to work on a breadboard, but I fell out of love with it along the way, and didn't follow through to a finished design. I did post a fair amount of detail to a thread started by Printer2, starting around post # 19: Mini-amp for Output Tube Distortion
-Gnobuddy
R.I.P. Fred Nachbaur
The single best example I've seen is the small amp emulator in Fred Nachbaur's Dogzilla
Unfortunately, the domain has expired but the Wayback Machine appears to have captured it here.
The second (and much easier & closer to your original question) is the Real McTube. Also by Fred, and also only via the wayback machine.
The single best example I've seen is the small amp emulator in Fred Nachbaur's Dogzilla
Unfortunately, the domain has expired but the Wayback Machine appears to have captured it here.
The second (and much easier & closer to your original question) is the Real McTube. Also by Fred, and also only via the wayback machine.
I used one time a full EL84 push pull amp as an effect pedal to dial in the distortion I needed and then amplify by a transistor amp into a guitar cabinet to get the distortion at any volume. It was great and I still use it sometimes at home.
But currently I am using a custom built high gain preamp at home. It has an ECC82 power amp, SE parallel, if I remember correctly.
There was a series of half rack preamps from Hughes and Kettner - i think it was Blues, Cream and Metal Machine, the schematics can be found on the net - at least one of them has a small tube power amp inside - that was the inspiration.
I am more of a metal guy, so I use most of the distortion from preamp, but without a tube power amp, the sound is not perfect. Even the small triode one works pretty well.
But currently I am using a custom built high gain preamp at home. It has an ECC82 power amp, SE parallel, if I remember correctly.
There was a series of half rack preamps from Hughes and Kettner - i think it was Blues, Cream and Metal Machine, the schematics can be found on the net - at least one of them has a small tube power amp inside - that was the inspiration.
I am more of a metal guy, so I use most of the distortion from preamp, but without a tube power amp, the sound is not perfect. Even the small triode one works pretty well.
Years ago I was looking for something easy (pedal/preamp) to get that tube overdrive sound.. I tried many designs but never been 100% satisfied.
Then I found a nice little schematic called 'firezog' which is a kind of 'line out' that can be used on the Firefly amplifier. So I thought it would be nice to built that circuit inside a box so I can connect the output of any small tube amplifier to the input of any other (bigger) (tube) amplifier, I mean convert 'on the fly' any small tube amp into an Herzog. That little box is really useful to me.
Then I found a nice little schematic called 'firezog' which is a kind of 'line out' that can be used on the Firefly amplifier. So I thought it would be nice to built that circuit inside a box so I can connect the output of any small tube amplifier to the input of any other (bigger) (tube) amplifier, I mean convert 'on the fly' any small tube amp into an Herzog. That little box is really useful to me.
There are a couple of things about that Firezog that concern me:
1) If the 4 ohm / 8 ohm switch ever fails (and it will fail), there will be no load at all on the amp acting as the Herzog. This is likely to quickly destroy the Herzog-amp. This is the classic problem of a $2 part failing, and causing hundreds of dollars worth of damage.
2) Even a small 15-watt Herzog will put out about 11 volts RMS, or over 31 volts peak to peak, at the speaker output, at full volume. The big amp that follows the Herzog adaptor, on the other hand, probably wants 20 mV at its input, but we can expect it will handle up to maybe 6 V peak to peak (about the maximum a standard guitar FX pedal can put out.)
So it would be a good idea to insert an additional resistor in series with the volume pot, to cut down the maximum signal level there, and allow you to use more of the pots range without worrying about blowing up the input stage of the post-Firezog amp.
3) Using a 1M pot for the Firezog is just silly. All it does is create an enormously high source impedance, which increases thermal noise (hiss), and increases hum and noise pick up.
I've modified the Firezog to eliminate these problems, as the attached schematic shows. Each of the three problems has been eliminated.
In particular, if the 4 ohm / 8 ohm switch fails open, there will still be an 8 ohm load across the Herzog-amp, instead of the intended 4 ohms. Any reasonable valve amp can handle this much load mismatch without failing.
If the 4 ohm / 8 ohm switch fails shorted, there will be a 4 ohm load across the amp, instead of the intended 8 ohm load. Once again, any reasonable amp can handle this much load mismatch.
The other changes are trivial, lowering the output level pot to 10k, and inserting a series resistor to cut the maximum output level to a safer value.
Note that there is a huge omission in both the original, and this improved, Firezog: there is no speaker emulation filter, not even a capacitor to roll off the fizzy harsh highs coming out of the overdriven input amp. While the original Herzog had the same omission, "Gar" Gillies designed a special new main amp for Randy Bachman to eliminate the fizz and harshness, and give him that smooth, sustaining lead sound he wanted.
-Gnobuddy
Attachments
We do! Nowadays we call them distortion pedals, overdrive pedals, fuzz-boxes, dirt pedals, et cetera, and they are everywhere - just about everyone who plays electric guitar has at least one, and many have several. Their function is exactly what Randy Bachman intended - to go between guitar and amp, and add distortion and sustain to the guitar sound, beyond what the guitar amp itself can create.
Today's Herzog's are usually solid-state, which is much cheaper and more convenient for the manufacturer, but there have been quite a few valve-based ones too, from the Chandler Tube Driver to the Effectrode Blackbird.
None of the hundreds and hundreds of solid-state Herzogs - at least none that I've ever heard - have produced the long, smooth, sustain of the original, which is remarkable for its complete lack of harshness. Instead, many of us have trained our ears to tolerate, maybe even enjoy, the harsh buzz of clipping semiconductor diodes.
-Gnobuddy
I am very sure that the solid state of pedals are only that way due to costs of production/traditionalism. With modern day mini transformers, I can’t see how it should be too difficult to modify a well known tube mic pre and fit inside a guitar pedal. There are even those subminiature tubes, but my knowledge on these is very limited. Anyone tried using those?
You can certainly make a very simple tube preamp using a single triode. Run it off a high enough voltage, and you can get a gain of roughly 50x with a single triode from a 12AX7; divide that down with a pair of resistors, and you will have a tube guitar preamp.
The thing is, a single stage like this adds very little "tubey" sound. If you have good ears, and a guitar with reasonably beefy pickups, you may be able to hear just a tiny touch of "tubeyness". It will be very clean, though. Much cleaner than, say, a blackface Fender amp. In my experience, most people, including many musicians, will not even notice the 5% or so THD that the single triode adds to the signal.
Since you have two triodes in the 12AX7, you could extend the circuit by adding one more gain stage, with more attenuation before and after to keep the signal down to line level. This will increase the amount of "tubeyness" just a little, and allow a little fizzy-sounding overdrive. It usually doesn't sound much like an entire tube guitar amp, though. But it might be what you want, I don't know.
You can lower the power supply voltage, and try to make a tube run on as little as 9 V. Slight variations on this theme have already been done dozens of times. The "Real McTube" is one popular DIY example. The "Matsumin Valvecaster" is another DIY-friendly option. From what I've heard, these low-voltage versions tend to have no clean tone at all - you can choose from distorted, or more distorted. They sound less fizzy than a preamp tube running on 350 volts, but instead tend to sound darker, growlier, and less "tubey".
Google will turn up many, many commercial tube guitar preamp pedals that were, at one time or another, available for sale. You can find schematics online for some of the older ones (Chandler Tube Driver, for instance), but you will not for many newer ones - they are proprietary devices after all.
So there really isn't much of a question whether it's possible to build a tube preamp for guitar. Of course you can! But virtually none of these use an output transformer, for the reasons I mentioned in my first post: firstly, it is not at all clear that an OT is necessary, or will make the preamp sound better. Secondly, an OT needs a relatively beefy tube to drive it, a 12AX7 won't do the job. Thirdly, an OT is quite expensive.
Looking back at your first post, I think I (and some other respondents) may have mistaken your intention. If all you want is some sort of tube overdrive pedal (and not something that sounds like an entire guitar amp), my suggestion is to forget about the output transformer, and try out something like a Valvecaster.
Google will turn up vast numbers of excited Internet posts about the Valvecaster, and some sound clips here and there. It is easy to build, the circuit is simple enough to cram into an FX-pedal enclosure, and it operates on low voltage, which is much safer.
My own goal is something else - a tube guitar preamp that can be plugged into a clean, flat-response P.A. system or powered loudspeaker, and which produces a sound similar to a good tube guitar amp.
-Gnobuddy
But transformers do make a difference in sound. Like at the different mic pre’s. transformer vs transformer-less is a big difference. Some even go far enough to say that the transformers has as big impact on the sound as the tube Itself. I guess what I mean is a actual pre amp, similar to a mic pre, that has a much lower headroom, and probably several more gain stages to induce the more ‘tube sound’. I wouldn’t want a clean boost, and the distortion from the starved plate valve designs don’t even sound the same.
Yes, they certainly do.
Anything to do with electric guitar (and audio in general) is a minefield these days, full of marketing, superstition, unsupported opinions and beliefs, outright B.S., and very, very little actual trustworthy information. It can be very hard to tell one from the other sometimes.
Personally, I am deeply skeptical of the belief that the transformer has an impact anywhere comparable to the tubes. That doesn't mean you have to share my opinion, of course!
-Gnobuddy
Yes, they certainly do.
Anything to do with electric guitar (and audio in general) is a minefield these days, full of marketing, superstition, unsupported opinions and beliefs, outright B.S., and very, very little actual trustworthy information. It can be very hard to tell one from the other sometimes.
Personally, I am deeply skeptical of the belief that the transformer has an impact anywhere comparable to the tubes. That doesn't mean you have to share my opinion, of course!
-Gnobuddy
I know. This is why it’s hard to know what to believe. Music electronics isn’t like anything else, it’s like the least scientific science there is. Almost like a art/science hybrid!
This is very easy to prove: the person making the claim should simply measure frequency response and distortion spectrum of the transformer. From that, we can immediately, and objectively, tell if the transformer makes an audible change in sound.
But I bet you won't see actual objective engineering data like this...not even for extremely expensive aftermarket guitar amp transformers that come with all sorts of unbelievable advertising claims (you probably know the brand I'm thinking of.) No proof at all, though, just outrageous claims.
A good audio transformer in a mic preamp makes no change in sound at all - it has a bandwidth that covers the entire audio spectrum, and inaudibly low levels of distortion.
A transformer can lower the amount of hiss from the preamp, and this was why they were originally used.
Audio signal transformers deal with microscopically small currents and voltages. A good one has such little distortion that the human ear cannot detect it.
If someone claims there really is a big difference in sound in microphone transformers, they can simply show us the data to prove their point: show us the frequency responses, distortion spectra, maybe intermodulation distortion as well. These days, you can easily measure all these things with an old laptop and some free software.
Without measurements, these claims are worthless. In my opinion, these claims are most probably designed to make customers spend more, by buying the more expensive preamps (which contain transformers.)
Guitar amps are different. They drive a lot more current and power through their output transformers. Under these conditions, there is no doubt that transformers generate some harmonic distortion. But I suspect it is much less than the distortion generated by the output tubes under the same conditions, and therefore, not a significant factor in the sound.
When you pick up your dog, you don't notice the extra weight of her collar...all our senses are limited in this way, and smaller stimulii are overpowered by bigger ones. In audio, it's sometimes called masking. 50% THD from the output tubes will mask 5% THD from the transformer, so we can't hear the smaller contribution.
But I could be wrong, of course. Maybe guitar amp output transformers really do change the sound. When someone shows me hard data proving this (or I generate such data myself), I'll believe it.
Until then, I remain skeptical. Especially when it comes to unverified sources of uncertain quality on the 'Net, without proof, it doesn't exist!
-Gnobuddy
The thing is, audio electronics wasn't always like this!
From the late 1800s (starting with Bell Telephone) through maybe the 1970's or early 1980s, audio was a true engineering field. Very smart, very well educated scientists and engineers worked in the field. They used sophisticated lab equipment and carefully conducted listening tests to draw conclusions, both about the limits of human hearing, and also about the technical requirements that audio electronics had to meet in order to be audibly perfect. They published peer-reviewed technical papers that led to further research by others. Important conclusions were carefully checked and tested repeatedly by different research groups working independently.
By the time the Compact Disc came along, almost all those requirements for perfect sound had been met. Distortion was inaudibly low. Frequency response of the electronics was wider than the human ear's. Wow, flutter, background hiss, all had been pushed far below audible levels. The one remaining weak spot was the electromechanical devices in the audio chain - microphones and loudspeakers.
So the field of audio was mature, and all the biggest hurdles of the past were gone. We had perfect sound, except for our speakers and mics.
With no big challenge remaining, the scientists and engineers left the field. And the marketers and audiophools moved in, replacing science and engineering (which work) with subjective opinion (which led to such wonderful things as slavery, astrology, belief in witchcraft, the Dark Ages, et cetera.)
That's where audio stands today, especially on the 'Net. Most of what you hear comes from people who have never studied science or engineering, who discard the principles of science completely, and simply believe whatever they want to believe. "Red wire sounds better!" "Mpingo disks made my stereo have more separation!"
But I agree with you about guitar electronics in particular. That is the only place I know of where "bad" amplifiers sound good to us, and "good" amplifiers sound bad. As many people have pointed out, we shouldn't really think of electric guitar amplifiers as amplifiers in the Hi-Fi sense; rather, we should think of them as part of the instrument itself.
Considering the huge progress made by, for example, the AX84 participants a couple of decades ago, we should know more than we do. Certainly the people developing the best digital tube amp simulations know a lot more about tube amp sound than they used to. But all that information is locked up, top secret stuff inside big corporations like Yamaha and Roland.
So we're still stuck with our soldering irons, our oscilloscopes, our little stash of tubes, and our ears!
-Gnobuddy
Re-inventing the wheel again
And then the industry decided to dumb it down to THD and watts and declared everything else inaudible and irrelevant.
Which is where the mainstream has been since the '40s. (see Crowhurst's writings from the '50s for a voice in the wilderness)
Unfortunately (as Gnobuddy pointed out) the renaissance in the '90s (founded on "if we can hear it but can't measure it we're measuring the wrong things") got subverted by "measurements are irrelevant". The supposedly rationalist "main stream" of engineering (e.g. Sony) kept inventing newer forms of "perfect sound forever", most compromised to be commercially viable and each of which took a decade or so to work what now needed measuring and to what degree (e.g. clock jitter, SMPS).
At the same time, actual recording and mastering techniques went backwards.
There are some corners of the web (including around hereabouts. And vale Siegfried Linkwitz) where listening and measurement are taken seriously. But just not where the money is.
In the end, most music listeners picked convenience (ipods and MP3s) over fidelity, anyway.
</diversion>
Seymour Duncan. Twin tube range (e.g. the classic)
<diversion>
And then the industry decided to dumb it down to THD and watts and declared everything else inaudible and irrelevant.
Which is where the mainstream has been since the '40s. (see Crowhurst's writings from the '50s for a voice in the wilderness)
Unfortunately (as Gnobuddy pointed out) the renaissance in the '90s (founded on "if we can hear it but can't measure it we're measuring the wrong things") got subverted by "measurements are irrelevant". The supposedly rationalist "main stream" of engineering (e.g. Sony) kept inventing newer forms of "perfect sound forever", most compromised to be commercially viable and each of which took a decade or so to work what now needed measuring and to what degree (e.g. clock jitter, SMPS).
At the same time, actual recording and mastering techniques went backwards.
There are some corners of the web (including around hereabouts. And vale Siegfried Linkwitz) where listening and measurement are taken seriously. But just not where the money is.
In the end, most music listeners picked convenience (ipods and MP3s) over fidelity, anyway.
</diversion>
Last edited:
You need to talk to Grant Wills. Or Lynn Olsen. Both who've spent a fair bit of time measuring misbehaviour in transformers recently. Or Rupert Neve.
Relevant to the guitar bit, the interesting stuff seems to be around how and when the core saturates and whether the output tube/tubes can get there.
My observation is that "traditional' amps have undersized (cheaper) OPTs, high gain and jazz/acoustic amps tend to have correctly sized (for guitar) transformers. These are still smaller than a similar wattage HiFi transformer.
But I take your point on the general lack of correlation between measured performance and price point. Even if you take volume into account.
I have a Fender Super Champ XD, and a Fender Princeton Reverb reissue. Both have almost identical push-pull 6V6 power sections. The SCXD has a smaller output transformer.
There is an audible difference, but it is not what I expected from all the claims I read: when you turn up the power on the SCXD, the bass starts to diminish, with no audible change in timbre. The PRRI doesn't exhibit this effect.
My hypothesis is that the beginnings of core saturation in the smaller SCXD lower primary inductance to the point where it starts to reduce bass response. An EQ change, without any audibly perceptible accompanying change in THD.
That unexpected result bears repeating: there is no change in timbre that I can hear, much less a glorious improvement one way or the other. The smaller OT doesn't contribute better (or worse) tone; the effect it creates is subtle, and can be pretty well duplicated with a tiny twist of the bass control on the PRRI. Musically, I find it pretty much irrelevant.
At least for the sort of guitar tones I use (nearly clean, which should be more revealing of small amounts of distortion generated by the transformer), it's hard to avoid coming to the conclusion that the OT seems to be an irrelevant red-herring as far as contributing any sort of audible distortion to a guitar amp goes. (An undersized OT apparently does contribute some mild EQ effects, easily duplicated with a few resistors and capacitors in a Herzog, without the need for an expensive, heavy, bulky, lump of iron.)
...so the amps that are designed to produce more bass (and/or frequency response to lower frequencies) use bigger transformers - exactly what you'd need to maintain (more or less) the same core saturation level!
That seems like straightforward engineering to me, dictated by the entirely routine textbook characteristics of audio transformers. No hint of mystical benefits from a smaller OT, other than the one Leo Fender spotted - lower price!
Sure! Six-string guitars in standard tuning don't go below 82 Hz, while Hi-Fi was expected to go down to at least 30 Hz, and ideally 20 Hz. That's two octaves lower, and any audio engineering textbook of the era would explain why you'd need a much bigger transformer to get the same amount of core saturation when you drop the frequency by a factor of four.
It would actually be engineering idiocy to use a Hi-Fi sized OT in a same-power guitar amp!
(The same effect has enormously shrunk the power transformers inside our power supplies; for the same power output, the huge lump of iron needed at 60 Hz has given way to a comparatively tiny blob of Ferrite in an SMPS switching at 100 kHz or more.)
I think the audible contribution of the guitar amp OT may have been quite considerably overestimated; I'm led to the conclusion that if the guitar amp OT can produce a full-power bandwidth from roughly 100 Hz to roughly 5 kHz or better, that transformer will almost certainly be more or less completely inaudible. Whatever little distortion it produces appears to be completely buried underneath the much bigger distortion from the valves.
I have seen a small guitar amp OT generate enormous amounts of distortion on my 'scope - but only at very low frequencies around 20 Hz, far below the guitar bandwidth. Not only does the guitar not generate these frequencies, the guitar speaker doesn't reproduce them, either, so this has no audible consequence at all in a guitar amp.
-Gnobuddy
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.