High Gain Guitar Preamp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm working on the design for a high gain 50W head for modern metal. I've laid out my ideas for the preamp:

Preamp1.png


All tubes are 12AX7s. This is not really an original design, just a collection that I've put together. The first two sections are from Brian S.'s (AX84 forum) L3 amp, the 3rd (cold clipper) and some other tidbits are from the Peavey 5150, the final gain stage, the cathode follower, and the tone stack are from AX84's UberGain.

The schematic is preliminary, and will need tweaking. The coupling caps (C4, C5, C8) are probably too large, and the first couple of stages will probably need some caps across the plate resistors to decrease the gain at high frequencies.

This pre will be feeding into a P-P 6L6GC power amp targeting about 50W in class AB1.

The power supply will based on an Antek toroid with a solid state rectifier, possibly with a cap multiplier for the preamp section.

Any suggestions are welcome!
 
If you want to reduce level to the following stage, then it is better to unbypass some part of the cathode follower, then lowering stage gain AND causes NFB that increases linearity. R17 is too high and is loading nothing. In fact, considering Miller capacitance, it is adding a low pass filter.

I´m not musician, I´m talking only based in electronic theory and practice.
 
If you want to reduce level to the following stage, then it is better to unbypass some part of the cathode follower, then lowering stage gain AND causes NFB that increases linearity. R17 is too high and is loading nothing. In fact, considering Miller capacitance, it is adding a low pass filter.

I´m not musician, I´m talking only based in electronic theory and practice.

I understand where you are coming from, but increasing the linear operation of a stage is the exact opposite of what a preamp like this is supposed to accomplish. The values and circuit blocks I'm using here are pretty standard in all guitar amps.
 
OK, but explain me some questions I cant undestand. In the 4th stage, you gain say, 50 times. But the 2:1 attenuator consisting in the 2 resistors R21 and R22, then the signal to the 5th stage is 50 * 1/2, or 25 times. Which is the advantage of such circuit, versus adding a small cathode resistor unbypassed to obtain a 25 times gain in the 4th stage?? This way the 4th stage allows similar gain, but less distortion thanks to local NFB. I know that musicians like you use distortion deliberately, but help me understand you.
 
OK, but explain me some questions I cant undestand. In the 4th stage, you gain say, 50 times. But the 2:1 attenuator consisting in the 2 resistors R21 and R22, then the signal to the 5th stage is 50 * 1/2, or 25 times. Which is the advantage of such circuit, versus adding a small cathode resistor unbypassed to obtain a 25 times gain in the 4th stage?? This way the 4th stage allows similar gain, but less distortion thanks to local NFB. I know that musicians like you use distortion deliberately, but help me understand you.

The entire goal is more distortion, not less. So, one stage amplifies the signal until grossly distorted, and then the attenuator drops the signal level down, just so the next stage can distort it some more - all in the interest of creating more complicated tones.
 
I know that musicians like you use distortion deliberately, but help me understand you.

It's all about "touch sensitivity" or whatever term is in vogue today. The trick is to get multiple stages to be at the verge of distortion at the same time. This way the amp can be relatively clean and with a twist of the volume knob on the guitar, or even just playing the notes a little harder the amp transitions into distortion, with each stage contributing its own characteristic sound. The guitar player can then "play the amp" by altering his picking style.

Multiple volume or gain controls (3 in this case counting the one on the guitar) allow more flexibility. Crank up the guitar and the input gain and back down the master gain to get lots of distortion at low volume, or crank the master and back down the other two to clean up the act.

The high valued grid stoppers are common in high gain amps. They affect the frequency response, stability, and yes throw away some excess gain. Find the schematic for a Soldano SLO-100 for a similar design.

I have built a 6 stage high gain preamp for screaming lead guitar. The issue with 6 stages of gain can be hum, noise and microphonics. You may have to try a bunch of 12AX7's for the first tube to find a quiet one. Even the second tube can contribute some microphonics. I have several hundred 12AX7's and I wound up using a military spec 5751 for the first stage. It has a slightly lower gain than a 12AX7 and is built like a tank for less microphonics. I used DC heaters to kill hum. Some of my grid stoppers were 270K carbon comps.
 
OK, boys. I have a friend, Carlitos who once explain the question of deliverately distort the signal, and the theme of acoustically couple the speaker to the guitar to create Larsen effect as a musical resource, but I have no idea how do you use it, and how to help you get it. Then, this thread isn´t for me. Thanks for the time and patience dispensed to me.

Good luck.
 
My goal has always been touch sensitivity, and I do a lot of palm muted chugging for rhythms. This takes a metric ton of gain to sound good.

I have built a 6 stage high gain preamp for screaming lead guitar. The issue with 6 stages of gain can be hum, noise and microphonics. You may have to try a bunch of 12AX7's for the first tube to find a quiet one. Even the second tube can contribute some microphonics. I have several hundred 12AX7's and I wound up using a military spec 5751 for the first stage. It has a slightly lower gain than a 12AX7 and is built like a tank for less microphonics. I used DC heaters to kill hum. Some of my grid stoppers were 270K carbon comps.

My current amp (L3 designed by Brian S. at AX84) has four stages feeding into a cathodyne EL84 output stage. Works very well for me, but I want more :devilr:

And, I need more power than the EL84's will give me.

Thanks for the tip on the 5751. I know one of Sovtek's 12AX7's is built to the old 5751 specs; I'll be sure to try them. I'm planning on DC regulated heaters for the preamp, but I really haven't had any problems with heater induced hum in my L3 with it's elevated AC heaters and tight wiring.
 
I would add some on-board screw-pots for those resistors, maybe even some front-panel inter-stage volume controls. You know how it works, so in-use you'll have no problem intuitively making adjustments.

Perhaps my favorite preamp is two Fender channels in series, with a volume between and a volume after them both. I like your input almost directly to the first grid, and you may find playing with the resistors in that front-end can change how it works (and the amount of treble) with the inductance of your pickups (unless you're using powered pickups). Those first resistors in series and parallel also affect what happens when you turn down the volume on a Fender versus Gibson guitar and whether the hum comes up when you turn down a Fender a bit; those resistors determine the input impedance that affects whether a strat's higher-resistance volume control lets the hot line get too far from ground. You might consider a trm pot to ground useful to adjust to match your guitar better if the gain is really high. Really good move not using some two-input-jack resistor network that throws away half your gain. Also, I really like to boost treble early, then roll off the treble later; so I would put a very fender-ish bright switch and cap across that first volume pot (or a bright pot in series with a cap both in parallel with the volume). Its effectiveness obvioulsy changes with the volume setting, but that's OK in-use (at full volume a fender bright switch does nothing). Main use for distortion modes is that the guitar signal gets treble boosted early and then reduced later; but the distortion generated later only gets its treble reduced, making for a really great creamy distortion that emphasizes the beats and interactions of multiple-string signals rather than the 'transistory buzz-box' fuzz distortion high harmonics that's effective for single-notes but very irritating. Plus if you turn down or bypass the later stages you can get great chucka-chucka strat stacatto with that treble boost when set for 'clean'.

You may have to decide whether you want what I call "auto-correct" where distortion and limiting make every note sound about the same, which is really great for machine-gun playing with lots of pull-offs and hammer-ons and assorted single-note tricks, but the distortion is alway on full, and notes are all the same volume, infinite sustain, until silence like a noise gate. It covers up the mistakes very effectively and hides the method. Or do you want a bluesy more expressive touch type of distortion that's more emotive and really has 'feel' and changes timbre radically (witout much volume change) when you pick harder, and changes from cleaner to really beatty when you play two notes simultaneously and stretch one and distorts like a sax or human voice does when pushed harder (but like them doesn't really get much louder just more urgent). The two styles dictate different topologies. With lots of interstage volume controls, a few stage bypass switches, and an early bright switch or second set of early tone controls you can end up with a much more versatile amp. Personally I have little use for lower-voltage devices like the Chandler or any stages that emulate one, but that's just personal; but you might consider a bypass switch. Beating a dead horse, I like EQ before distortion and EQ after distortion, as together that allows me to alter the tone of the distortion independently from the signal. Turning up the "early treble" and down the "late treble" turns down the treble on the distortion that originates between the two.

One other thing to consider, if you ever use any envelope-following devices, like a radical compressor or a trigger-wah auto-filter, lovetone meatball or that sort of thing, you'll want to be able to tap into your circuit right after the first stage to rectify and smooth an envelope-following sense line for the voltge control on your VCF or VCA EARLY while there's still plenty of dynamic range before it's all compressed out, then have the actual filter (or amplifier) section applied after the final output with all it's rich distortion. That's just a nice jack to have for the future (maybe an isolation cap too). You usually don't need an entire insert send/return but at least a send. That way a trigger-wah voltage control can be much more expressive, yet its filter applied to the harmonic-rich output. A wah sounds very different inserted early versus late. A noise gate is best sensing there at a very early stage too, and its VCA can be there or later; ideally again you'd want the noise gate sensing the signal before it's compressed, but you'd have to decide whether it should then turn down that early input or turn down the later master volume.

Also, a LOT of devices sound better in between stages instead of before them all or after them all. So you might consider some send/receive insert points. I usually like spacial effects like reverb and echo late after everything, but sometimes the reverb sounds better driven earlier before the dynamics are all compressed out.

If you want some shimmery highs with that high gain, consider an Aphex aural exciter late in the stages (like maybe after the preamp output). It adds treble harmonics without turning up the treble noise.

But it sounds like you know what you want...and perhaps you don't need so much versatility.
 
Last edited:
Found it... It's the Sovtek 12AX7LPS. It has the spiral wound filament and larger plates like the 5751

The LPS is optimized for low hum, much like the 7025. I have found that in many cases the long plate tubes are more prone to microphonics. It's not because of the long plates, it's the long grid. The longer the grid, the more it can vibrate, and the lower its resonant frequency. The 5751's that I have all have short stubby plates and 3 mica supports to control vibration.

I have made amps so microphonic that feedback resulted when the amp was placed on top of the speaker cabinet.....without even plugging in a guitar.
 
Of course an amp really shouldn't be placed on top of a speaker cabinet, despite the tradition and convenience the only thing worse is the 'combo' format that's so practical but even more hostile to the electronics. Some of the VOX amps had external steel frames, and of course there's racks. But the world is probably ready for a better alternative. Perhaps someday some kind of phantom power thru the guitar cord will facilitate a preamp (with a "standard" interchangeable module format) in the guitar; if you don't need to "work" the knobs fewer people would care where the power amp sat. I wonder whether a tube preamp in the guitar would have similar microphonic problems. But I typically digress, so I return you to your original thread subject.
 
You might want to take a good look at the TAD 7025WA select for this application.
I have been playing with these a lot lately and in all honesty, I find it to be about the lowest noise tube produced. They have no shortage of richness even in some of the fussy high gainers.

Since I started using these in guitar amps all of the common 12AX7 gibberish has disappeared.

Anyway, looks like a fun project, just thought I would throw in my .02.
 
Perhaps someday some kind of phantom power thru the guitar cord will facilitate a preamp (with a "standard" interchangeable module format) in the guitar;

I have been working on that one for a while. Phantom power as it exists today will run a solid state preamp, and barely power one speciallized tube. I have discovered the LIPO batteries used by model airplane and hellicopter builders. These are light and have enough energy to power a 3 tube preamp for several hours.

I wonder whether a tube preamp in the guitar would have similar microphonic problems.

Yes there are microphonics, but in small doses they seem to be a good thing. Instead of picking up energy from the speaker and making feedback, the tubes pick up energy from the strings adding fatness to the sound.

I have been tinkering with subminiature tubes. They come in two flavors. There are directly heated types intended for battery operation that draw very low heater power, but have pretty bad microphonics and low gain. There are also indirectly heated subminiature tubes that work much like their larger brethren, and eat as much power. Many of them are very quiet tubes.

Now putting a vacuum tube power amp and speaker inside a guitar is a different story. I'm still experimenting with that one.
 
You might want to take a good look at the TAD 7025WA select for this application.
I have been playing with these a lot lately and in all honesty, I find it to be about the lowest noise tube produced. They have no shortage of richness even in some of the fussy high gainers.

Since I started using these in guitar amps all of the common 12AX7 gibberish has disappeared.

Anyway, looks like a fun project, just thought I would throw in my .02.

Cool! Kind of pricy, but just for V1, V2 it would be worth it if it lowered the noise common with these high gain amps.
 
Cool! Kind of pricy, but just for V1, V2 it would be worth it if it lowered the noise common with these high gain amps.

The circuit looks okay to me, but a note about the "bootstrapping" on the cathode follower. I tried this in my amp and found it to be excessive and unnecessary. It added a fuzzy/scratchiness that I didn't care for. Obviously I'd encourage experimentation for your own tastes. For reference, my pre-amp is essentially a cross between a Bogner Shiva and a Soldano SLO. Gain ranging from blues/classic rock to thrash metal.

I've got "Chinese", Svetlana and E-H tubes in my high gain pre-amp and they're nice and quiet. They're cheap from thetubestore.com. I've got the E-H in the DC coupled cathode follower position and it sounds great. I tried 6 different types in that position (in all positions actually, but focused on that one) and found it to be the richest, most widely usable at different gains. FWIW, I found the Sovtek sounded like complete crap compared to Svetlana and E-H. Cold, no character, and too compressed.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.