Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: leftu2 on March 03, 2026, 11:50:20 am
-
Amp has has 4 gain stages with a plate driven tone stack and a LTP phase inverter with marshall values.
My goal is to get the max gain out of the amp with plenty of asymmetrical/ even order harmonics .
From my research, the best way to measure the voltage between stages is using peak to peak voltages / Vpp.
I know you build gain between stages to keep things from getting mushy and bloated.
What I can't find out is what is a good range for an acceptable Vpp for each stage. For ex, I've seen no more than 2-3 Vpp at V2b before clipping begins, I've seen no more than 10 Vpp before clipping begins at V2b.
I know the voltage will climb as you go through each stage but again, at what point is it too much? What I don't want to do initially is "just your ears", I need a starting point, a range for each stage, then I'll tweak using my ears and what distortion feel I want.
(i just bought a scope because I'm tired of guessing when tweaking pre amps)
There are 2 voltage dividers already there. I can make a voltage divider off the gain control feeding V2b if needed.
With this configuration, I'm getting a lot of distortion that has a nice feel in between spongy/tight on the open E and A string in standard tuning but there's still some "bad" stuff going on. There's still a bit of bloated sound/feel even with the gain way down.
I suspect I need to lower the Vpp between stages to clean it up a little. Again, I just don't know where to start because I don't know what the acceptable Vpp should be between stages.
Last night I changed the 1st 3 plate resistors to 220k and that got me the compressed distortion I want. If I understand it right, the combination of the plate resistor and cathode resistor sets the clipping point and changing the value of the the voltage divider resistors to increase/decrease the voltage to the grid won't change the clipping point of the tube??
If I'm wrong, then I need to find the right combination of grid input voltage, plate resistor & cathode resistor values?
I've been modding the schematic which shows all component changes, B+ voltages and Vpp voltages. This is where it's at now.
-
the better "starting point" is at the speaker, what does the signal look n "sound like" there?
next, the PA n driver
everything prior is simply the "speacil sauce" that add flavor to the "sound"
-
the better "starting point" is at the speaker, what does the signal look n "sound like" there?
next, the PA n driver
everything prior is simply the "speacil sauce" that add flavor to the "sound"
Thanks for the reply. What is the "PA n driver"?
-
power tubes (Power amp) n whatever dives them, ...PI for fixed bias,... driver tube for self biased.
-
In general, I agree with Shooter's approach. In your case, this is a pre-amp, so we assume you're looking at the output in the absence of a power amp? If you are using a power amp, what is it? It does matter.
From my research, the best way to measure the voltage between stages is using peak to peak voltages / Vpp.
I'd love to know where you got that. Volts are volts! Vrms is the standard, but ok, you're trying to see the effect of each of your stages. I get that.
I think you may be better off using a scope to look at your waveform following each stage, pairing that with listening. Voltage in isolation really doesn't tell the story. And frankly listening is the more important of the two.
This process of dialing in a particular sound is not an easy task. And if you're changing multiple things at once, it is really tough. Best to focus on one stage at a time.
-
If you want to increase harmonic distortion in the preamp, make your signal amplification stages more 'non-linear'. This tends to occur normally in tube preamps anyway, but you could consider things such as:
1) keeping more (rather than less) bypassing in the cathodes (This is because cathode current feedback makes everything 'more-linear' and so reduces harmonic distortion). Note if you want to include a cold clipper stage, which is usually unbypassed, keep this at the end of the preamp chain.
2) increase/steepen AC load line gradients (e.g., through higher AC loading after each gain stage - to get more asymmetry in each output signal swing - which means you'll probably need more stages to get more drive in the power amp)
3) don't use local NFB loops in your preamp (as these reduce harmonic distortion - i.e., same outcome as unbypassed cathodes)
4) Use 12AX7s (rather than 5751s, AT7s, AY7s, AU7s). Note that there are also reasons for breaking this rule - e.g. a 12AT7 in a Long Tail Pair inverter can sound really good and help a PP output stage to function 'better' in overdrive which IMO makes an amp sound nicer too.
5) Use dc-coupled pair normal gain stage and cathode follower like the 5F6A/marshall for the Tone stack driver (as these sound amazing when they're overdriven, but they also provide excellent buffering for a TMB tone stack when the amp is being played 'clean'). Note that this might seem somewhat contradictory to the aim in 2) above, but it in reality these act like one joined-up assymetrical gain stage when overdriven and IMO you gotta have it if you want a great sounding guitar amp.
6) If your aim is to get more even-order harmonics, then you should aim to keep as many 2nd, 4th etc harmonics of the fundamental in the signal as possible - and the 'best' way to achieve this is to have alternate hot and cold biasing in each successive preamp gain stage. You will still get some odd-order harmonics if you do this, but you will get 'more of a mix' of even-order and odd-order harmonics. (Otherwise, if all preamp stages are centre-biased, then when these are driven into 'symmetrical clipping', the result is more odd-order harmonics, which probably will sound 'harsher' - which might or might not suit you). Note that will all the gain that you will get if you have 3 or more preamp stages in the preamp, you'll want some inter-stage attenuation otherwise you'll end up with mush.
YMMV.
-
In general, I agree with Shooter's approach. In your case, this is a pre-amp, so we assume you're looking at the output in the absence of a power amp? If you are using a power amp, what is it? It does matter.
From my research, the best way to measure the voltage between stages is using peak to peak voltages / Vpp.
I'd love to know where you got that.
From the internet. AI spit this out. I'm still learning obviously and sometimes I have no other resource than to ask a question in google.
----------------------------------------------------
Why Use Vpp for Preamp Stages
Identifies Clipping:
Vpp represents the total swing of the audio waveform from its lowest to highest point. To know if a stage is distorting, you need to know if the signal voltage exceeds the supply voltage (Vcc) or biases, which Vpp directly tells you.
Real-Time Dynamics: A guitar signal has a high peak-to-average ratio. An RMS meter might show low voltage, while a peak-to-peak measurement on a scope will show that you are hitting 2-4Vpp on loud strums.
***here's the power amp section. Ignore the phase inverter as it's a cathodyne design. I've converted it to a LTP using standard marshall jcm 800 values.
-
What I can't find out is what is a good range for an acceptable Vpp for each stage.
In general, the max peak input a stage can handle before clipping, is equal to its bias voltage. So if a preamp stage has 2V bias, it can handle about 2Vpeak (4V pp). Beyond that you are overdriving it. Same for the power tubes.
-
I agree that Vp-p can be an important and useful metric, perhaps moreso than Vrms.
But for me, its utility comes when assessing the onset of distortion / the transition from clean to overdriven to clipping.
You're working on mushy distortion in an overdrive preamp, so you're well past that area, the majority of ADSR signal envelope will be full on clipping.
I interpret 'mushy' as too much low end / too much bias shift / insufficient high mids & treble. Obviously that's within the overdrive preamp section, so unless you're also overdriving the power amp (?), it would be unlikely to be beneficial tinkering with that stuff there.
Show you schematic and we could make some suggestions.
Have you got and read Merlin's latest books?
-
I usually solve / tune in both clean sound and distortion sound (tone) with filtering. I find especially pre-shaping the tone (removing low end) before overdrive circuits useful. After all, that`s the way most of the high-gain amps are constructed..
-
Vpp represents the total swing of the audio waveform
and that total in a preamp is seldom >2vpp (see Merlins comment on bias). since it takes ~~~ 1vpp or less to drive a preamp tube, so as long as the signal can overcome the losses between stages, smaller vpp is better
-
I'm still learning obviously and sometimes I have no other resource than to ask a question in google.
You figured it out by coming to this forum and asking real people with real experience. IME, AI answers can be a useful starting point, but AI really isn't "I" -- it's just regurgitating what the model scrapes from everywhere, and is not authoritative.
Merlin Blencowe's books (see https://valvewizard.co.uk (https://valvewizard.co.uk) are very well written and well worth reading. Designing an amp and voicing it are not simple tasks, and it takes time.
I think an inexpensive oscilloscope will help you, once you learn how to use it. Best way to look at Vpp and waveform . . .
-
Show you schematic and we could make some suggestions.
Have you got and read Merlin's latest books?
Here's the full schematic with all my notes made at different points. Plate driven tone stack (for now) I might need to go to a cathode driven tone stack as tubeswell suggested. The cathodyne phase inverter is now a LTP with jcm 800 values.
I'm getting pretty good definition except the E string has a bit of a weird buzz to it. I feel like I need about 3 more #s on the pregain to get where I want to be distortion wise.
I will get Merlin's book for sure.
-
Merlin Blencowe's books (see https://valvewizard.co.uk (https://valvewizard.co.uk) are very well written and well worth reading.
I think an inexpensive oscilloscope will help you, once you learn how to use it. Best way to look at Vpp and waveform . . .
I did say I just bought a scope in my OP, I got a Hantek DS02D15. 2 channels, 150Mz, signal generator.
Now I need to know what to look for as far as what a "ugly/bad" wave looks like while probing the circuit.
Iv'e done a little research but most of the examples I see, don't deal with guitar amps and say "if the wave looks like this, there's light clipping, or moderate clipping, or way too much clipping"
I'll be getting Merlin's book.
-
I usually solve / tune in both clean sound and distortion sound (tone) with filtering. I find especially pre-shaping the tone (removing low end) before overdrive circuits useful. After all, that`s the way most of the high-gain amps are constructed..
Changing coupling caps was one of the 1st things I did after swapping the cathodyne phase inverter to a LTP configuration.
-
Vpp represents the total swing of the audio waveform
and that total in a preamp is seldom >2vpp (see Merlins comment on bias). since it takes ~~~ 1vpp or less to drive a preamp tube, so as long as the signal can overcome the losses between stages, smaller vpp is better
That's what I found out today. Changing the original high pass filter to a 470k/470k voltage divider before V2b definitely helped clean things up even though I only had 7 Vpp at the grid before i put the divider in. I wouldn't have thought 7 Vpp was too much...