开云体育

Hi all,

This has been discussed before, and I have since been in contact with the person who did this (Keith Sterling) to their Matrix-12, so I wanted to share the details here. This was done over 15 years ago, so there may be better parts than specified below now, but these parts have worked fine for Keith since then.

I have put some photos that I will refer to here in the Photos section. AD = Analog Devices.

-Replace the TL074 at C213-713 with AD713JN (essentially a quad AD OP275) (Pic "AD713JN"). Keith: The OP275 has a much higher slew rate than the TL071/072/074. I did my mod a long time ago and there are probably better chips out there now, for example the Burr Brown OPA4132, but the Analog Devices I used yields noticeable improvement.

-Replace TL072/82 at C215-715 with AD OP275 (Pic "OP275 OP176 and caps"). Keith: Also you must solder small (50pf) caps across the feedback resistors if you don't want your Matrix-12 to be a 12 voice theremin (check photos for exact placement). These caps are small enough to not affect the sound in any way.

-Replace TL081 at C249-749 with AD OP176 (pic "OP275 OP176 and caps"). Matt: These are obsolete now but I managed to find some at utsource on ebay.

-Replace the voice summer op-amp (TL072) with AD OP275 (Pic "OP275").

That's it!

In addition to replacing the op-amps (which I know some people disagree about in terms of possible improvement), I was wondering about audio cap replacement. This is another contentious issue, but I have heard at least one re-capped xpander in person, and it did seem to sound better (sharper?). Unfortunately, I couldn't A/B with an uncapped one. Nevertheless, on a 25+ year old machine, I can't see it hurting and if there was a possibility of improving sound quality, it seems work doing.

Somewhat related, I wanted to ask people about the effect of oscillator volumes on audible distortion. I almost always kept osc. volumes at 63, thinking that there is no difference between 63 or less in terms of gain or saturation. However, testing this yesterday (lowering the volumes to ~30, with high resonance and no filter modulation) I did seem to notice that, at lower volumes, there was a mild decrease in some mid-upper resonant frequencies.

Of course it's not uncommon to have a bit of extra distortion/saturation with max volume on some analog synths, but it's not something I associate with Oberheims (most don't even have osc volumes!). And it surprises me that I would have to turn the volumes down so low to hear any difference. I realize it's a bit subjective, but can anyone else comment on their experience with this, i.e. am I really hearing any differences?
I'm curious how changing op-amps/capacitors might affect this, or if it's just inherent to the design.

Somewhat related, I wanted to ask people about the effect of oscillator volumes on audible distortion. I almost always kept osc. volumes at 63, thinking that there is no difference between 63 or less in terms of gain or saturation. However, testing this yesterday (lowering the volumes to ~30, with high resonance and no filter modulation) I did seem to notice that, at lower volumes, there was a mild decrease in some mid-upper resonant frequencies.

Of course it's not uncommon to have a bit of extra distortion/saturation with max volume on some analog synths, but it's not something I associate with Oberheims (most don't even have osc volumes!). And it surprises me that I would have to turn the volumes down so low to hear any difference. I realize it's a bit subjective, but can anyone else comment on their experience with this, i.e. am I really hearing any differences?
I'm curious how changing op-amps/capacitors might affect this, or if it's just inherent to the design.

Having done some more testing, I can address my own comment a bit better. This may not be news to some people, but distortion is definitely present with both oscillator volumes 63 and it's most noticeable when using high resonance. The seems to appear more as a kind of squashing of harmonics rather than pure overdrive, and leads to some of that grunge that I find annoying on the Xpander. Things clear up a little at 45, more at 30, and the best at around 20. At 20, and with high resonance, the resonant peaks stand more (not surprising, since you're not turning the resonance "volume" down) whereas at 63, the peaks seem to get compressed within the overall signal and other subtle inharmonic distortion is introduced.

Without resonance, it's harder to hear any difference between the different volumes unless you start boosting the high end with external EQ (as I did), but the distortion is still there. All of this is pretty subtle to be sure, but in case you, like me, are looking for a way to tame some of the grungey overtone, knocking the osc volumes down will help somewhat. Wish I'd tried this some time ago!