Solid State Guitar Amp Forum | DIY Guitar Amplifiers

Please login or register.

Login with username, password and session length
Advanced search  

News:

collapse
Pages: [1]   Go Down

Author Topic: Tube emulation and temperature  (Read 709 times)

aquin43

  • SSGuitar Regular
  • ***
  • Chip Points: 5
  • Posts: 11
    • View Profile
Tube emulation and temperature
« on: November 13, 2016, 10:10:58 PM »

There are many interesting and reasonably accurate tube emulation circuits but none that I have seen seem to address the variation of semiconductor parameters with temperature. Real tubes work so far from ambient temperature that they are largely unaffected by changes within the normal human range of endurance.

I would guess at 0C to 30C as a useful working range (32F to 86F) for an amp that is to sound the same in world wide ambient temperatures. Is this worrying unnecessarily?
Logged

phatt

  • Legendary
  • ******
  • Chip Points: 191
  • Posts: 1673
    • View Profile
Re: Tube emulation and temperature
« Reply #1 on: November 15, 2016, 11:05:57 AM »

Not quite sure what you are trying to explain or understand but  Transistors are Not Thermal devices,, Valves ARE.

The filament (heater) in a Valve raises the temperature of the cathode which then liberates electrons.
In effect the heater *Biases* the Valve by thermal energy.
whereas a Solid State device runs *Cold* and does not use heat to bias into conduction,,it only needs a bias voltage.

Although there are some differences they do share a lot of similarities but you don't just copy Valve concepts into SS circuits and expect instant Valve mojo ,, you have to use  different approaches to achieve same/similar results.

BTY it's not just the Valves the make a difference it's also due to Output transformers.
If you want more technical terms then there are some here who can give you a much more in-depth explanation.

So yes,,,your Concern is unnecessary 8)

Phil.
Logged

aquin43

  • SSGuitar Regular
  • ***
  • Chip Points: 5
  • Posts: 11
    • View Profile
Re: Tube emulation and temperature
« Reply #2 on: November 15, 2016, 02:33:35 PM »

The point I was trying to make is that, since tubes operate internally so far from ambient temperature, the transfer function of a tube stage will not change whether the ambient temperature is 0C  or 30C. This is not true of any of the semiconductor based emulators which will show extreme variations with temperature unless this is compensated for in some way. It may not matter much with hard limiters or clippers but anything more subtle will certainly be affected.

I can't recall ever having seen a design where this has been taken into account.

Arthur

Logged

g1

  • Legendary
  • ******
  • Chip Points: 53
  • Posts: 418
    • View Profile
Re: Tube emulation and temperature
« Reply #3 on: November 15, 2016, 05:31:18 PM »

  I think that at least for the output transistors, they are also operating far enough from ambient temps that we can also ignore the effect of the ambient temp.
Just a guess though.  :)
Logged

J M Fahey

  • SSGuitar Global Mod
  • Legendary
  • ****
  • Chip Points: 413
  • Posts: 4034
    • View Profile
Re: Tube emulation and temperature
« Reply #4 on: November 15, 2016, 10:23:48 PM »

Quote
There are many interesting and reasonably accurate tube emulation circuits but none that I have seen seem to address the variation of semiconductor parameters with temperature.

Oh they do.
Unless itīs the crudest of designs, parameter stability versus temperature change is taken care of.

The basic variation in a Silicon junction, which normally drops "around" 650 mV  , is that it drops 2mV per Degree Centigrade rise, so if you bias a single transistor stage , with its emitter grounded, from a fixed voltage supply, you only need 2 resistors, one in base, one in collector.

In that case, yes, base voltage will change with temperature , and so will collector voltage, a lot more, but practically no circuits are that simple: classic  transistor gain stage uses 4 resistors in a way that operating parameters are quite stable:


A more complex circuit, with 2 or more transistors, in general will have "too much gain"  so negative feedback is used to control gain and reduce any variation to a minimum, both DC and AC.

An Operational Amplifier goes way beyond that, no surprise as they were originally designed for precision lab and Aerospace measurements and Analog Computers.

The one area where temperature variation must be tightly controlled is in power transistor biasing; since currents and power in general are very important, but much worse, drift "the wrong way" and can easily run wild , special care must be taken for compensation.
« Last Edit: November 15, 2016, 10:26:24 PM by J M Fahey »
Logged

aquin43

  • SSGuitar Regular
  • ***
  • Chip Points: 5
  • Posts: 11
    • View Profile
Re: Tube emulation and temperature
« Reply #5 on: November 16, 2016, 07:36:01 PM »

Yes, of course the default design procedures for the linear parts of circuits produce reasonably temperature compensated results, or results in which only the gain varies significantly.

The non linear parts, which form the essence of any tube emulator, will vary with temperature and it will change the clipping points and the low level distortions that are being emulated. That is the part that I was referring to.

Arthur
Logged
Pages: [1]   Go Up
 

* User Info
 
 
Welcome, Guest. Please login or register.
Did you miss your activation email?
* Recent Posts
Jordan Professional 440 Solid State IC Amp Head by teemuk
[April 27, 2017, 07:58:53 AM]


Strange amp in a suitcase by J M Fahey
[April 27, 2017, 05:42:47 AM]


DIY Layout Creator by DrGonz78
[April 26, 2017, 08:40:13 AM]


Marshall Lead 100 MOSFET 3210 Head by J M Fahey
[April 26, 2017, 12:59:30 AM]


More differential amplifier questions by phatt
[April 24, 2017, 09:05:25 AM]

* Sponsors