Electro Harmonix Stereo Memory Man Echo/Chorus

[miketbass]

Here's one that you don't see too often - 90s reissue "Stereo Memory Man" from the stranger days of EHX/Sovtek. I picked this up for a good price to repair and after replacing the U1 opamp (JRC4558 gone belly up) I had a decent sounding MN3005 based delay.

This is a single BBD chip delay based on the legendary Panasonic MN3005 that was a reissue of a unit developed in the late 70s. Designed to be a stripped down version of the DMM this unit sports a proper stereo output setup and a switch to toggle between "echo" and "chorus". While the Deluxe version could run both simultaneously, the SMM was designed to select either/or. Many claim that the chorus setting is one of the better analog chorus sounds produced.

After getting this going I began to check calibration and give it a test drive. My unit tops out at 280ms max delay which is right in range for these pedals. The clock frequency is determined by a 240pf cap and can be adjusted by raising or lowering this value. As many of these do, my pedal suffers from prominent clock whine setting in past 2 o'clock or so. Usually this can be mitigated with the clock cancel trim however upon probing my BBD output I noticed something odd, leading to a question for anyone with experience with these....

My dpdt switch to toggle between "echo" or "chorus" mode seems to do the bit just fine, but there seems to be some bleed over between the settings. Meaning that in delay mode I can see the clock frequency bounce up and down a hair (a few hundred hz or so) and in "chorus" mode I get a fairly prominent slapback echo along with the modulation. Ever turn the feedback knob full CCW and strike a staccato note? Sounds like that running in front of a chorus pedal.

Usually I have an example on hand to compare a circuits function to, or a wealth of videos and other info to check against. Theres not a whole lot to read up on these and most mentions of the chorus setting are rather brief. Is this a typical behavior of this pedal? I'm wondering if some leaky caps in the LFO and clock circuit are causing some cross talk here, but it also may just be how the circuit operates. Listening to a few YouTube clips (even with the poor sound quality) leads me to believe otherwise. Any help??

Attached are a schematic of the circuit and a shot of the pedal with a few other EHX delays.

[miketbass]

Well I decided to dust this off and revisit the issue of the chorus setting. I have since acquired 2 more memory man units with chorus, both late 70s - early 80s. One is a blue and black version, the other an original SMM that this pedal is a 90s reissue of.

First things first, they do sound different from this modern edition. Speaking only of the chorus sound they give a beautiful lush chorus. The blue/black is particularly nice and leaves the delay time pot connected when the chorus is engaged to give more tweakability.

More to the point clock frequency measurements show a frequency that wiggles around 60khz (measuring at pin 2 of the MN3005) on both of the older units. This newer one "wiggles" around 30khz which is much too long of a delay time - which causes this weird modulated slapback sound when engaged. I have checked all connections against the schematic and find no discrepancy. Playing with some component values lets me vary the delay time (namely the 10k resistor tied to pin 2 of the 4047) but other parameters change with it, such as the "width" of the chorus.

I am not particularly familiar with using the 4047 to clock the MN3005 or modulated delays in general. Does anyone have any suggestions as to where to begin troubleshooting? A note: the modern SMM and the vintage unit appear to have the same components, values, and wiring for the chorus feature yet perform much differently. Attached is the vintage schematic.

[miketbass]

Well, we have resolution. I began tracing the clock circuit further and had started checking caps. In the process of lifting legs and resoldering, the issue has disappeared. I suspect a bad solder joint on the 1uf cap C34 is to blame. For anyone wondering - this reissue SMM is kind of a lemon in the bunch. The old version runs quieter with less aliasing and clock noise, longer delay time, and a better chorus sound. The 90s version sounds to have less low pass filtering and may work better for short slapback stuff but the overall sound seems lesser.