Has anyone made a working version of this module?
I have had on/off success literally !! If theres an R in the month it works ..sometimes…otherwise its driving me bonkers. I suspect the E203 which I’ve found unobtainable ..I’ve tried many replacements but without it seems any luck.(This of course if its the fault)
Maybe there is a mistake in the schemtaics?
Reach out to CZ directly, he likely has it and has a pretty good understanding of what makes it tick, or not, or what to test.
I was just going over the schematics. The old 1967 schematic uses a 2N4303 in place of the E203. The 2N4303 is still out there and can be sourced. The one Moog used was those black/white round button type package. I have them in both the TO-92 and round/button package. Wonder now if there is any difference?
Both the old and newer schematics are almost the same with 2 substitutions I can find. The old lists a CL-1 diode, replaced by a 1N4148 in the newer one. And the 2N4303/E203 sub. I’m no tech, so I can’t help with trouble shooting. There are a few qualified techs that may chime in on this list. Might want to try one of those official Norlin #991-041064-001 types if you haven’t already.
Pinouts can vary between the TO-92 packages and the round ones.
Try moving your transistor around to other positions.
Also, many other FETs will work, perhaps even Roland’s favorite, the 2SK30.
Good luck.
Thanks for the input …I’m pretty sure its the Jfet that has to be precise.
The 2N3189’s I use in my 911’s work fine but not in the Dual Delay.
I’ve noticed a sub for the E203 is a NTE466 which is listed as a high speed trigger .
But I’ll order CZ’s suggestion (thanks C Z) and see …with postage
£20 for two!!
This module has been a pain for the last 10 years ..
It works…then stops working…I hope this time I fix it for good.
I’ll re post results .
a. What’s a 2N3189? Typo for 2N3891?
b. Suspect your own mistakes first, not the components. We humans tend to assume a lot.
c. NTE and ECG transistors are often poor substitutes and very high priced.
d. Here’s some pics and diagrams that might be of some help: http://www.synthfool.com/911a
One additional question, using a scope, does your main cap (C1) charge up when given an S-trig?
It won’t need that transistor (Q9) to do this and can be left out. (and if incorrect, should be left out.)
If no cap charge, the fet isn’t an issue yet. Have you traced the circuit with a scope?
No Kevin..no scope..yet.
Ive ordered the 2N 4303 ..I’ll see what happens when that’s in.
A small Moog circuit , this has given me more trouble than any others …apart from 921b , but even that’s fully working now.
But again many thanks for your help and time. I’ll post results .
Ok…got my 2N4303 …with no pin out..and NOTHING on web regarding GDS Pin out.
Its the round /dome/black-white type
Ive got a socket …somewhere ..but I need to know correct pin out as my circuit may be wrong. Anyone got a link?
The pinout is listed on the Moog 1125 Sample/hold schematic. I always have to look at those twice to make sure the orientation. Has a circle with the three terminals offset lower labled S D G and says top view. Can look it up on that 1125 page. These were used in a few other models too.
Link to a 1125 schematic:
http://machines.hyperreal.org/manufacturers/Moog/schematics/1125-2.jpg
And a link to the older 911A schematic using the 2N4303:
http://moogarchives.com/911as.jpg
Thats good C Z..thanks for the links..this is the best pin out info for the b/w dome head 2 N4303 lve seen…I’ll place it in and test.
Regards…
Thanks to all !! Module working.
But any builders watch out for R10.
My schematics say 27k but CZ’s info states 2.7k!! I guess the dot blended on to the “2”… D’oh !!
Glad that worked out for you!
Good eyes on that resistor error. Can’t believe I missed that, as I was studying both schematics side by side looking for errors/differences. Could have saved you some time and money had I caught that. 
I have seen a few errors like this, that seem to just be copy errors. The old 60’s Moogs would come with a packet of blueprint schematics. Some single page like the 911A and some were large fold outs four times that size like the 901A.
Anyhow, I have noticed at some time they re-drew these blueprints for the Norlin modular service manual. It just looks like they did not proof read these very good as there are tiny and simple errors like that one. They don’t look deliberate. For instance the 901C blueprint has the top rotary square wave positions tied together. Yet this is ommited in the Norlin print. Copy error?
The 984 Four channel mixer had a similar omission. The bottom right cap on the original has a curved and straight line indicating polarity, though it is missing a + sign. On the norlin print it has two straight lines. Small errors like these seem as though they are just copy errors, but can cause confusion or a circuit not to work properly.
Another big one is the transistor matching schematic. Correct on the NPN circuit, but missing the 10K resistor on the PNP circuit.
Thanks for pointing that out. The best way to document these is putting this text up on this forum.
That’s why they started the convention of “2r7k” where “r” denotes the decimal point to eliminate konphusion.
Even worse are schematics that do not have dots where junctions should be (Buchla, argh!!!)