ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

SS tube power amp emulation

9/18/1997 10:49 AM
M Helin	SS tube power amp emulation As I have lately seen so many tube amp schematics and discussion on subject, I've been wondering if it's possible to emulate tube power amp and tube power supply circuits in solid state designs. I mean such ideas as: 1) using current limiters (chokes, resistors) between PSU and power amp to emulate tube compression effect, 2) trimming bypass cap values affecting attack time response, 3) using output transformers (1:1) 4) other ? MH.

9/18/1997 10:49 AM

M Helin

SS tube power amp emulation

As I have lately seen so many tube amp
schematics and discussion on subject,
I've been wondering if it's possible
to emulate tube power amp and tube power
supply circuits in solid state designs.

I mean such ideas as:

1) using current limiters (chokes, resistors)
between PSU and power amp to emulate
tube compression effect,

2) trimming bypass cap values affecting attack
time response,

3) using output transformers (1:1)

4) other ?

MH.

9/18/1997 12:53 PM
R.G.	There have been lots of smart people thinking about this for quite a while (maybe decades). The best shot seems to be with some variant of a power FET and some auxilliary circuits, possibly a low power pentode. A power FET actually has a "triode region" where it does act like a triode. It might be possible to use a small-die power FET to do this emulation.

9/27/1997 6:21 PM
Jack Orman	The common MOSFETs, IRF511 etc., are enhancement mode devices. The key to getting the MOSFETs to sound like tubes is to use depletion mode devices. I have several on my work bench that I plan to drive with a high V+, 250-300v, and see if I can get some tube sound. Let's see, if I remember correctly, the higher the gate bias voltage required for turn on, the longer the linear part of the channel will be, which I think will give you a longer region to make the turn-on (ie: soft clip). Also, I'm designing an FET Big Muff... should be interesting. -JACK http://members.aol.com/jorman/

9/27/1997 6:21 PM

Jack Orman

The common MOSFETs, IRF511 etc., are enhancement mode devices. The key to getting the MOSFETs to sound like tubes is to use depletion mode devices. I have several on my work bench that I plan to drive with a high V+, 250-300v, and see if I can get some tube sound.

Let's see, if I remember correctly, the higher the gate bias voltage required for turn on, the longer the linear part of the channel will be, which I think will give you a longer region to make the turn-on (ie: soft clip).

Also, I'm designing an FET Big Muff... should be interesting.

-JACK
http://members.aol.com/jorman/

9/30/1997 2:01 AM
M. Helin	That's interesting, what type number MOSFETs exactly then are the depletion mode devices? Generally, I feel that there's been too much emphasis on emulating harmonic quality of tube amplifiers, the dynamical qualities are not considered to be so important. I think that dynamical characteristics are prior to harmonic, human ear first detects the envelope of the sound (like the initial attack of the guitar string). In real world they interact - compression tends to event out harmonics, clipping generates more harmonics etc. I just thought that if we make a solid state amplifier work like tube amplifier in it's environment (mainly power supply, chokes, caps), it would behave - compress, sag, clip and distort - like a real tube amp. Not quite, but close enough to be musically enjoyable to play with. I have played with some very softly clipping (and expensive) solid state amps (didn't try then at full volume), it was funny, it was actually more funny than trying to play a tube amp at low volume. Mikko

9/30/1997 2:01 AM

M. Helin

That's interesting, what type number MOSFETs exactly then are the depletion mode devices?

Generally, I feel that there's been too much emphasis on emulating harmonic quality of tube amplifiers, the dynamical qualities are not considered to be so important. I think that dynamical characteristics are prior to harmonic, human ear first detects the envelope of the sound (like the initial attack of the guitar string). In real world they interact - compression tends to event out harmonics, clipping generates more harmonics etc. I just thought that if we make a solid state amplifier work like tube amplifier in it's environment (mainly power supply, chokes, caps), it would behave - compress, sag, clip and distort - like a real tube amp. Not quite, but close enough to be musically enjoyable to play with. I have played with some very softly clipping (and expensive) solid state amps (didn't try then at full volume), it was funny, it was actually more funny than trying to play a tube amp at low volume.

Mikko

10/5/1997 6:05 PM
Jack	Envelope is very important. In making analog snare drums, a popular trick is to add in a fast attack pulse to give a click to simulate a stick hit. Obviously the amplitude and duration of the pick hit will be a part of the overall sound in a guitar signal. A fast attack SS circuit doesn't duplicate the tube envelope very well. However the harmonics that are present contribute a lot to the sound. There is an article on tube simulation on the net that describes tube distortion in terms of how it affects the width (duration) of the pos/neg parts of the signal. It is claimed that tube distortion causes one half of the signal to have its width increased, and the other half decreased. This, of course, adds in even harmonics. I intend to investigate signal width modulation in the near future. -Jack

10/5/1997 6:05 PM

Jack

Envelope is very important. In making analog snare drums, a popular trick is to add in a fast attack pulse to give a click to simulate a stick hit. Obviously the amplitude and duration of the pick hit will be a part of the overall sound in a guitar signal.

A fast attack SS circuit doesn't duplicate the tube envelope very well. However the harmonics that are present contribute a lot to the sound. There is an article on tube simulation on the net that describes tube distortion in terms of how it affects the width (duration) of the pos/neg parts of the signal. It is claimed that tube distortion causes one half of the signal to have its width increased, and the other half decreased. This, of course, adds in even harmonics. I intend to investigate signal width modulation in the near future.

-Jack

10/6/1997 5:38 AM
Mikko H.	There are indeed many previous studies on tube sound emulation with solid state circuits (I'll write more about these later, maybe). It has been claimed, that the "tube sound", or the "tube distortion", is the result of the following phenomena (among others): - Tube grid bias shift, this will cause asymmetric clipping and even order harmonics (mainly second, some fourth and sixth). - Tube saturation, in push-pull power amp this will give odd-order harmonics (mainly third and some fifth order). - Interaction of output transformer and speaker load, this is mainly harmonic function, it will boost some low frequency range (moderate Q at about 200 Hz or less). - Class B power amp crossover distortion, this results in even order harmonics. - Tube rectifier 'sag', this is mainly dynamic effect like compression. - Cathode-biasing in power circuit, this is like rectifier sag, compression is the result. - Slow attack time of power tubes, I think this works like a low pass filter, but only on high voltage level changes, it's more like a slew rate limiter. - Miller effect on preamp tubes, between gate and anode, works like low-pass filter again. Some of these features can be/ have been emulated with SS circuits. Roland's TubeLogic emulates tube rectifiers (there's a TUBE/DIODE switch on Bluescube series amps). Peavey people have been quite busy trying to develope decent SS emulation of tube preamps. Tube bias shift is quite easy to emulate with SS, I've been thinking a simple mod for TS-9 (haven't tried it yet) like this: R R D +---/////---+---/////---+--

10/6/1997 5:38 AM

Mikko H.

There are indeed many previous studies on tube sound emulation with solid state circuits (I'll write more about these later, maybe). It has been claimed, that the "tube sound", or the "tube distortion", is the result of the following phenomena (among others):

- Tube grid bias shift, this will cause asymmetric clipping and even order harmonics (mainly second,
some fourth and sixth).

- Tube saturation, in push-pull power amp this will give odd-order harmonics (mainly third and some fifth order).

- Interaction of output transformer and speaker load, this is mainly harmonic function, it will boost some low frequency range (moderate Q at about 200 Hz or less).

- Class B power amp crossover distortion, this results in even order harmonics.

- Tube rectifier 'sag', this is mainly dynamic effect like compression.

- Cathode-biasing in power circuit, this is like rectifier sag, compression is the result.

- Slow attack time of power tubes, I think this works like a low pass filter, but only on high voltage level changes, it's more like a slew rate limiter.

- Miller effect on preamp tubes, between gate and anode, works like low-pass filter again.

Some of these features can be/ have been emulated with SS circuits. Roland's TubeLogic emulates tube rectifiers (there's a TUBE/DIODE switch on Bluescube series amps). Peavey people have been quite busy trying to develope decent SS emulation of tube preamps.

Tube bias shift is quite easy to emulate with SS, I've been thinking a simple mod for TS-9 (haven't tried it yet) like this:

 
         R           R         D
 
   +---/////---+---/////---+--

10/6/1997 5:48 AM
Mikko H.	TS-9 mod, another try with pre-tags. R R D +---/////---+---/////---+--

Page 1 of 2

Next>

Last Page>>