Compensating multi-stage feedback (was: RE: all tranny vca+ )

Magnus Danielson cfmd at swipnet.se
Sun Jul 2 22:28:24 CEST 2000


From: Don Tillman <don at till.com>
Subject: Re: Compensating multi-stage feedback (was: RE: all tranny vca+ )
Date: Sun, 2 Jul 2000 06:59:39 -0700 (PDT)

>    Date: Sun, 02 Jul 2000 15:14:40 +0200
>    From: Magnus Danielson <cfmd at swipnet.se>
> 
>    Certainly, but now comes my key question:
>    Are you sure that what you perceive is due to types of distorsion?
>    Are you _really_ sure?
> 
> Pretty sure.  Anything's possible, but in the case of the mixers
> the signal isn't going through a very complicated circuit and that
> circuit isn't supposed to be doing anything exotic to the signal.

I recall the issue about having capacitors in the signal path. Just a quick
recap:

In order to have to trim for DC balance one just use DC blocking cap. Now, for 
this cap not to have a too great influence of signal as froming a highpass
filter with the resistive load you better have a pretty large cap in the
1 to 10 uF range normally. Now, this usually means an electrolytic cap, and
those are not really that linear, so... also they age pretty bad if you do not
handle them correctly.

>    My point is, I do not object to people hearing something sounding
>    good or bad, but what I have learned is that very few people are
>    able to pinpoint the correct process of destruction or enhancement
>    (as below). 
> 
> Ahh, I know what you mean.  I'm pretty good at picking these out
> though.  This particular "distortion" has a sound that's very
> distinctive, very constistant and is fairly obvious.

Well, you must understand that I can not know this by just see a few statements
over email. Now that we are really are getting into specifics more than a
quick comment it becomes easier to accept things.

>    > Other classic examples are the Pioneer and Technics hifi
>    > receivers from the mid-70's distortion spec race.  
> 
>    Well, was the measurement applicable to the process of destruction?
> 
> Certainly not, and that's the point.  Pioneeer and the others changed
> the goal from making a great sounding amp to making an amp that tested
> great on this one standardized measurement.  And this pointed out in
> very clear terms the uselessness of this particular standarized
> measurement.

Right.

>    Now, to summerize what I have heard and what I know of which may
>    destroy or enhance sound:
> 
>    Linear filtering
>    Impulse responce (_should_ be the result of linear filtering)
>    Cross-over distorsion (small signal distorsion)
>    Linearity (medium signal distorsion)
>    Clipp and clipp behaviour (large signal distorsion)
>    Feedback loop behaviour on distorsion
>    Amplifier load impedance behaviour
>    RF intermodulation (any non-linear section will act as a mixer)
>    Noise and noise modulation
>    Cross-talk
> 
> Good list!

Thanks! Note, the above list has very broad covering points and you must pin
down each specific mechanism under them. I am sure we can come up with many,
many, many of them. Actually, I think we _should_ come up with a lot of them
and discuss each of them.

> Here are my contributions:
> 
> Effects of junction capacitance varying with signal voltage

Good point, covered by the linearity point

> Power supply interaction

Uh, _what_ power supply interactions do you intend?

Yeah, power supplys can surely mess things up.

> Effects of bias changes changes with waveform transients

Good point, covered by the linearity point.

> Hey this is fun!

Yes, and it is my intention to keep it fun!

Oh, my point of amplifier load impedance behaviour may not be obvious to
everyone. Depending on the phase relation between current and voltage will the
output stage behave quite diffrently from stage to stage. Running into a almost
perfect resistor is really just toooo nice of load, you need to run into a
capacitor and into an inductor and see what happends. Unless you can handle
these loads you will get nasty clipp distorsion, often only occuring at high
amplitudes and greatly input signal frequency dependent. You can have an amp
preforming 0.001 % THD but providing 20% when put into a realistic load and at
a high level. Also, these cases are nasty since this is also where you pop
your trannies even if you beleived you did a sound dimensioning. Think! Learn!

Cheers,
Magnus



More information about the Synth-diy mailing list