[music-dsp] tube non-linearities

[Kursad Kurt]

----- Original Message -----
From: "James Chandler Jr" <jchandjr at bellsouth.net>
To: <music-dsp at shoko.calarts.edu>
Sent: Wednesday, February 14, 2001 3:04 AM
Subject: [music-dsp] tube non-linearities


> Hi, Shifty, Sergio and others--
>
> Thanks for the good ideas and references.
>
> On non-symmetrical leading and trailing edges in clipped half-waves, ya'll
> have come up with at least three good possible explanations--
>
> 1. Driver clipping.
>
> I need to find time to dissect some amplifiers. Fixed a lot of tube amps
20
> years ago, but haven't disassembled a guitar amp in a dog's age.
>
> On typical solid-state power amps, the driver transistors provide the
> voltage gain, and the output transistors are just voltage followers that
> provide current gain. The drivers typically clip at about the same level
as
> the output transistors.
>
> Tube output stages don't look like voltage followers.

In pspice simulations pentode output stages look like current sources, when
there isn't any negative feedback, that is.

>Anyone know what
> peak-to-peak deviation a tube driver must generate, in order to fully-clip
a
> tube output stage?

The driver stage in your Peavey is a kind that is never overdriven, because
its tone sucks if you overdrive it:-) (as far as I know from folks who have
done so:-) )
The one that sounds pleasing when overdriven is the that marshall uses, and
called "long tail PI". There was a thread about it, some time ago. As far as
I know voltage level is adjusted such that that driver stage is overdriven
immediately after power tubes get overdriven. (It doesn't make much sense to
get too much gain from power tubes, so, just after the hint of power tube
clipping, driver stage clipping begins). But you should check the
correctness of this info:-)


>Often the driver tube is operated at a lower voltage than
> the power tubes, so maybe the driver stage is more susceptible to clipping
> than the output stage? OTOH, maybe the output tubes provide enough gain th
at
> the output tubes clip much sooner than the driver output level can reach
the
> power supply rails?
>
> 2. Transformer saturation
>
> Transformer hysteresis seems plausible. One would think a high-quality
> transformer would be resistant to saturation, but OTOH it is doubtful that
> mass market guitar amp companies go to unnecessary expense on their
products
> (G).

There is a "boutique" amp manufacturer that copies the old Marshall Plexi
transformers exactly (not just impedance and power rating).Have a look at
http://www.mandgamplification.com


> 3. When the plate voltage drops at high current flow, it may temporarily
> under-bias the output tube (since the control grid voltage probably
remains
> at about the same DC point)?

In a pentode, plate current should be reasonably independent of the plate
voltage, because electrons are already accelerated by the "screen". That's
why they behave like a current source anyway, I guess.

-Kursad

>
> In my ignorance, this seems plausible too.
>
> James Chandler Jr.
>
>
> dupswapdrop -- the music-dsp mailing list and website: subscription info,
> FAQ, source code archive, list archive, book reviews, dsp links
> http://shoko.calarts.edu/musicdsp/
>


dupswapdrop -- the music-dsp mailing list and website: subscription info,
FAQ, source code archive, list archive, book reviews, dsp links
http://shoko.calarts.edu/musicdsp/