DS-1 mod+Math Question

[soupbone]

I've been modding the ds-1 for a while and one mod I did was changing r6 from 100k to 150k,and the other was changing r9 from 22ohm to 10ohm.Jack Orman has a very similar mod; http://www.muzique.com/news/fat-mod-for-the-boss-ds-1/ The only difference is he changes r9 to 1k.I like the sound of mine a little better because it still retains some gain.I can't figure math-wise why the 10ohm in my mod seems to work?Thanks on advance.

[Transmogrifox]

Here is a hint:
Miller Effect

Consider the effect of this on C4, and the effect of C4 on the frequency response of this gain stage.

[PRR]

> changing r9 from 22ohm to 10ohm.

The transistor Emitter internal impedance (at ~~0.6mA) is 50 Ohms.

R9 at 22 gives 72 ohms total.

R9 at 10 gives 60 ohms total.

Hard to see any "large" difference, 72 or 60. How much difference do you hear?

1K is a pretty radical change. Almost a different pedal. "Much less harsh" may be the least of the differences.

[soupbone]

> changing r9 from 22ohm to 10ohm.

The transistor Emitter internal impedance (at ~~0.6mA) is 50 Ohms.

R9 at 22 gives 72 ohms total.

R9 at 10 gives 60 ohms total.

Hard to see any "large" difference, 72 or 60. How much difference do you hear?

1K is a pretty radical change. Almost a different pedal. "Much less harsh" may be the least of the differences.

I had tried r9 at 1k but it sounded a bit sterile.I changed it back to 10ohms from my original mod,and it seemed to have more gain and maybe a little more dynamics?As well as cutting down a lot on the fuzz.So,Could i maybe at try r9 at 1omhs with even better results?Thanks!

[Transmogrifox]

I had a chance to simulate this.  Your mod makes what I would consider to be a very subtle difference in the frequency response:
Pass-band gain increase of about 2 dB.
High-pass -3dB cut-off changes from 604 Hz up to 713 Hz.

The added gain serves to lower input impedance through the feedback network while the input impedance looking into the base of the BJT goes down simultaneously (as PRR pointed out).  That is the reason for the cut-off frequency increase instead of what you might have expected to be a decrease.

This would tend to make it sound a little more "tight", but not "fat".

Down at the frequencies where "fat" would come out (<500 kHz) there is only a 0.5 dB difference in the gain.

The miller effect I alluded to earlier has no effect until over 100 kHz so that is irrelevant.

Your mod DOES make a noticeable effect on the bias.  DC bias changes from about 4.5V (stock model) down to about 3.5V.  This means you will be clipping on the negative polarity sooner than the positive polarity.

It is likely that the dominant effect you hear is the asymmetrical clipping caused by the bias shift.