[music-dsp] Transformers and electricity
citizenchunk at gmail.com
Mon Jul 11 14:17:03 EDT 2005
hi Steve. thanks for this very useful information. i'm glad i decided
to check the list today.
On 7/11/05, Steve Conner <steve at scopeboy.com> wrote:
> Hi all,
> I have built a couple of tube guitar amps so I have an
> opinion on this stuff ;)
> Tubes distort in two ways. If they are not overdriven,
> they produce the low order even harmonics (2nd, 4th,
> etc) which sound subjectively "sweet". This is
> allegedly why tube hi-fis sound "better" than
> transistor ones.
> If they are overdriven they go into hard clipping (the
> output is basically a square wave) and produce lots of
> both odd and even harmonics right up to very high
> During overdrive, an envelope detection effect comes
> into play, that makes the duty cycle of the resulting
> square wave shift according to the amount of
> overdrive. Hence the harmonic balance changes in a
> complex way with the drive level. This effect has
> attack and decay times that depend on the time
> constants of the grid circuit. It's like a synth with
> the amp envelope patched to oscillator PWM, and is
> very important in modelling distorted guitar amp
> That is for a single tube preamp with resistive load.
> A power amp tends to use two tubes driving a
> transformer in push-pull. All of the above effects
> still happen but symmetry of the push-pull circuit
> makes the even harmonics mostly cancel out. Plus there
> is transformer saturation to take into account.
> The way saturation works is: Once the time integral of
> the signal voltage reaches a certain value, the
> transformer gives up. It basically shorts the signal
> out and takes a "bite" out of the waveform as viewed
> on an oscilloscope. The result is lots of low-order
> odd harmonic distortion on loud bass notes.
> The suddenness with which the transformer saturates
> depends on the core material- usually it's a fairly
> soft knee. It comes out of saturation and returns to
> normal pretty much as soon as the input voltage
> Power amps normally run with negative feedback around
> the transformer that adds another layer of complexity
> to it all.
> I don't have any quantitative data on any of this, I
> just fiddle till it sounds good.
> Steve Conner
> dupswapdrop -- the music-dsp mailing list and website:
> subscription info, FAQ, source code archive, list archive, book reviews, dsp links
More information about the music-dsp