5thumbs wrote:First of all, good analysis and thank you for it. I'm not real solid on the science of why the R6 changes produces the perceived change in bass response. Part of me wants to be lazy and fall into the "higher input impedance causing more pickup loading, hence more bass and presence" stuff...but I'm not convinced that's the right explanation to just blindly accept.) I'm going to read over your post again later and think about it (provided I get some time in the Tech Cave tonight.)
Truthfully, I'm not sure how much of an effect the filter created by C3 and R6 has on the overall tone. Looking at the graph for a high-pass filter, you have a decade in frequencies between the -3dB corner frequency of the filter and the point where the curve ends and you have no attenuation. So with a corner frequency of 33.9Hz, you have a frequency range of 33.9Hz to 339Hz where you have at least some attenuation between 0dB and 3dB. That's certainly a wide enough range to encompass most of the fundamental frequencies of the guitar's open strings, pretty much ending on the high E (E4=329.6Hz). But in that decade, there's only 3dB of attenuation at most. I was of the understanding that 1dB was about the minimum change the human ear can hear. I wonder if the gain provided by the op-amp is effecting this somehow, by somehow magnifying the difference between the highs and lows of the guitar signal, not just the overall signal level.
5thumbs wrote:OTOH, transistors are notorious for being non-linear in their amplification of frequencies in the human-audible range. (This is one of the reasons I usually think of opamps for boosting gain, rather than transistors...and yes, I am lazy.
) The re-biasing of Q2 appears to eliminate a lot of treble boost (a.k.a, frequency-dependent gain), so it might be that we are getting a treble cut and calling it a bass boost.
I really wish I understood transistors better, as then I could dig into data sheets and draw up graphs for how the Q2 stage is operating, like I can with a vacuum tube triode. To be fair, a triode amplifier can be non-linear in it's amplification of frequencies in the human audible range to, depending on how it's set up In a cathode biased triode, gain is much lower than it could be, due to cathode current feedback introduced by the voltage drop across the cathode resistor. That's why you'll sometimes see a capacitor in parallel with the cathode resistor; the charging/discharging action of the capacitor as the cathode voltage fluctuates helps to stabilize the cathode voltage, getting some of the stage's gain back. But this ends up creating a filter which causes the tube to have more gain at higher frequencies than lower ones. The only way around that is to a) either use a large enough capacitor, that the knee of the filter is below the lowest frequency of interest, or b) use a diode or diodes to get the voltage drop needed to bias the tube, since the diode's voltage won't fluctuate with current.
I hesitate to try to figure out the Q2 gain stage with tube knowledge more than I already have, as a) I'll eventually start Doing It Wrong as tubes and transistors have different behavior, and b) it'd be veering the thread away from the topic of tuning a DS-1 for maximum awesomeness of tone.
But honestly, it's becouse it's my friand request. I have his unit to fix bad switch, he also asked me to do some mods to it, like keeley. I have go ahead and added this midrange control I was tocking about + Synth mod (two switchable mods + Synth bypass). He also ask me to mod it to use regular power supply, an this is why. I can't resist my friand's requests 
Your friend's MIJ DS-1 can already use PSA and other recent power adapters 
I use a switch with 3 resistors for three different gain settings I've found that I liked most.
)