Full plate voltage 12ax7 preamp to Class D?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
So I have seen this trend a few places and don’t know how to design a power supply that works.
Henriksen’s new Forte guitar amp has a 12ax7 running at “full plate voltage” into the preamp and then into a class AB power amp I suspect.
Milkman’s Them Amp pedal has a 12ax7 preamp running at “high voltage” then into a class D board.
I don’t think these are just 12-24v buffers.

What kind of power supply could run a “high voltage” 12ax7 preamp and a class D board?
 
This thread should probably be in Instruments and Amps.

High voltage for a 12AX7 probably means 200-300V supply rail i.e. the voltage you need if you want low distortion but don't want to use a CCS anode load. Used properly, a 12AX7 will simply amplify the signal - just like a properly-used opamp! Hence for a guitar amp no point in doing it except for marketing reasons?
 

PRR

Member
Joined 2003
Paid Member
> What kind of power supply could run a “high voltage” 12ax7 preamp and a class D board?

Write down your estimated power needs.

One 12AX7 is likely 240 Volts 0.002 Amps.

One transistor POWER amplifier is likely nearer 24V 2A, 48V 1A, much variation.

Note the vast difference in V and A on each supply.

It makes no sense to build 240V 2.002A, then waste-off 90% of the voltage (over 400 Watts of heat!) to get 24V 2A.

There are ways to kick-up 24V to 240V. Until recently this was non-trivial. Today you can buy pre-engineered up-converters, and that may be one path.

The other path is, clearly, TWO supplies. With incredible luck you might find the power transformer from one of the several hybrid tube/transistor amplifiers. More likely you buy two power transformers, and go on from there.

Note that a LOT of commercial boxes run 12AX7 or 12AU7 as low as 24V to the plate, and people buy them.
 
Hi

"High-voltage" is relative. Higher than what?

Recom has a DC-to-DC converter that produces around 200V out from various possible input voltages depending on the specific module (R15-150B). You can stack two to have up to 400V. These are about $50-60 so not likely to be what is used in any of the products mentioned above.

Considering that a 12AX7 only needs 2mA or less for both sections combined, which even at 300V is only 600mW of power, it is simple to use a voltage multiplier (caps and diodes) driven by a CMOS oscillator. Rolls does this for phantom power for microphones. The alternative is to use a coil and mosfet to pulse it, driven by a controller chip - all standard stuff for hobby switch-mode projects.

When a tube is used for actual gain and not as a buffer, it imparts a bit of character to the sound. So, tubes are added for more than just marketing.

All the digital mixing and recording guys have software plug-ins that emulate vintage tube gear and other analog stuff to help warm up the sound. Digital sound has been a big boom for analog and especially tubes. At first it was thought that everything before digital was obsolete, now you can't get enough of it back - haha!

Class-D amps have lots of problems. A friend of mine was telling me about these great class-D amps he got from Europe. The distortion was listed as kinda low but at wierd frequencies. It was suggested to me that these freqs were possible null points of the supply and the amp where distortion bottomed out into a resistive load. There's some marketing!
 
Used properly, a 12AX7 will simply amplify the signal - just like a properly-used opamp! Hence for a guitar amp no point in doing it except for marketing reasons?
I think a LOT of people will totally disagree with that. There is nothing better than a tube for a guitar preamp; the 12AX7 being one of the best---all the modeling amps in the world are just trying to get THAT sound!
 
The key word was "used properly". And in a forum other than instruments and amps (it was moved here after that post) that implies linear operation with no clipping/distortion.

Now, the whole point of them in guitar amps is to not use them properly. Rather to to run them at weird and wonderful operating points where they become rather non-linear and distort in positively euphoric ways.
 
Use a 12AX7 properly and you won't hear it. Part of using it properly is to provide enough voltage.
Agreed, and this is exactly what you want for Hi-Fi. Of course, if the tube's distortion is inaudibly low, there is also no point to using it for Hi-Fi in the first place, when a transistor does the same job far better, and with equally inaudible distortion.
Of course, you could provide plenty of voltage and still abuse it - but if so, what is the point of claiming the voltage as a feature?
About seventy years ago, Leo Fender looked at the example 12AX7 circuits in one of the tube-manufacturer catalogs, and borrowed it for his guitar preamp designs. That circuit used a 1.5k cathode resistor, 100k anode resistor, and about 250 - 300 volts B+.

If you draw the load-line and graphically estimate the second-harmonic distortion of this circuit, you will find it generates several percent distortion before it starts to clip either signal peak. This is enough to hear with a guitar as the input signal, if you have reasonably good ears and there is adequate SPL.

This is particularly true when the signal has gone through several half-12AX7 stages, each contributing, say, 5% THD, so you may very well have 10% - 20% THD by the time the signal hits the speaker. Absolutely horrific for Hi-Fi, but this amount of distortion produces glorious, lush-sounding "clean" guitar tones, of the sort the Fender Twin Reverb is famous for. It's called "clean" because you don't hear it as actual distortion, only as a richer sounding guitar.

So why is the high anode voltage a feature? Because there have also been many guitar preamps and pedals that use a 12AU7 or even 12AX7, but run it on a piddly 12 volts or so.

Run on 12 volts at the anode, these valve-based pedals distort heavily, like a diode-clipper based distortion pedal. This makes the buzzy distortion beloved of those who grew up listening to metal, and sounds nothing at all like a Fender Twin. We're talking something close to 100% THD here - what you get if you clip a sine wave into a square wave, doubling its RMS power (hence, 100% THD.)
In the context, probably best to regard the claim as meaningless marketing puffery.
Perhaps. But perhaps it is an honest attempt to tell the buyer she is getting a subtle "valve clean" sound, rather than buzzy distortion.

It's really hard to know what's what in the world of valve audio, where superstition and myth are everywhere.


-Gnobuddy
 
Do you mean, from a low-distortion solid state amp? Like the Fender Frontman 25?

For me, this doesn't do it. To my ears, this sort of clean sounds thin and uninspiring. Other words to describe it might be cold, or sterile. Compare with the "clean tone" from any of the famous classic Fender tube amps, and there is a lot missing.

I have been playing my semi-hollow electric guitar into a perfectly clean SS amp for some time now, and I have been trying for months to find some way to warm up those thin, sterile cleans. Adding a graphic EQ to shape the frequency response helped a bit.

Adding a filter that simulates a guitar speaker/ cab frequency response helped a little too.

Adding a little reverb and delay helps to hide the thin cold sound another wee bit.

Adding an EHX Overdrive Glove set for an almost inaudibly tiny amount of overdrive helps a little bit too. (It probably squishes the spiky attacks as well as adding a tiny amount of low-order harmonic distortion.)

But to me, after all this, it still sounds less attractive than just running the guitar through a single carefully biased and EQ'd 6AG5 (a tiny RF "pentode" that's really a beam tetrode.)


-Gnobuddy
 
I would have to try the Frontman with another speaker, I am not impressed with the stock one. I am not sure about it and it's clean IC path. I did play through a two transistor preamp with bass and treble into a SS amp and it sounded fine. Maybe the circuit was non-linear. I will have to get back to you on that. You did get me to question the Frontman. I have a Peavey Raptor and I find it real hard to EQ through the amp. The guitar I used and liked through the other SS amp was a G&L ASAT Classic (slightly better guitar). Still the Frontman is lacking so does not seem guitar related. But also the speaker is in stock condition and not beat to hell out of from severe use. The speaker with the other amp was, in one word, loose. That is an interesting idea. I did put a bell transformer on a different Frontman speaker (bought before getting the Frontman and I didn't know the make of the speaker) and with a fan as cooling had it humming way out in the garage for a few months. One day I'll have to try a swap.
 
Last edited:
Are you saying that ALL Fender amps have 10-20% THD? I never measured one, so I'm just asking.

Maybe. Depends on what power output you are producing. To drive the outputs more you need a higher voltage from your gain stages before it. I looked up some distortion measurements for the 12AX7, I have not found the chart I was looking for, maybe later. But for now we can use this one.

Vacuum Tube Evaluation_full_dual_triode

It lists distortion at high gain, 23 dBu which for many of us does not mean much so a quick conversion is in order (I could get out my old slide rule but this is easier).

dB dBu dBFS dBV to volts audio conversion digital - calculator volt to dBu and dBV dB mW SPL dB decibels 0 dBFS - convert dB volt normal decibels relatioship relation explanation analog audio absolute level true rms convertor converter decibel to dbf

So 23 dBu is 11V rms or 31V p-p.

Seems all the tubes run under 2% distortion at this point, let's check out the low level, 16.7 dBu which is (6.8V or 19V p-p). You get a smattering of 0.25, 0.5 and 0.5% ones.

So to make things easier let's use a a Champ type amp for an example since it is about as simple an amp you can get. Using the datasheet values we need 13V p-p to drive the 6V6 to full output. But full output is 8-12% distortion (depending on using 4.5 or 5.5W output) and fairly loud if going through an efficient 12" speaker. The 12AX7's in the chart can easily drive the 6V6 to full output with probably under 0.5% distortion if you are going for maximum clean, turning up the amp volume and the guitar volume down. Drop the signal so that the 6V6 is putting out 1W and its distortion should drop in the 2% region. The 6V6 will be making most of the distortion, the 12AX7 not so much.

Now if you have an amp with a tone stack things get a little different. You are loosing voltage across the stack which the triode has to make up. Using the example of a Silverface Champ with a bass and treble control you could be getting more distortion out of the triodes. But these amps are known for their clean sound rather than the earlier Tweed models. They reduce the midrange content and if we still need to drive the 6V6 to a 1W can we do it without running the triodes in the 5% distortion range?

Love to spend more time on this but I got things to do.
 
I think a LOT of people will totally disagree with that. There is nothing better than a tube for a guitar preamp; the 12AX7 being one of the best---all the modeling amps in the world are just trying to get THAT sound!

In around 1982 I found a circuit in Wireless World for a soft limiter. It was described as giving an over driven valve type distortion. So I built one for guitar and with a little extra gain and some top end lift it sounded very good. You could change the sound with a 2m2 pot. It was basically a op-amp with two transistors in the feedback loop controlled by a pot which gave soft clipping.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.