Mini-Modules : Minimoog DIY clone on its way.

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, Joe., luketeaford, lisa

Post Reply
User avatar
jdelgoulet
Wiggling with Experience
Posts: 270
Joined: Tue Jun 05, 2012 4:12 am
Location: Brive , France

Post by jdelgoulet » Thu Nov 15, 2018 5:19 am

For those who will buy a parts kit from SynthCube (https://synthcube.com/cart/guinguin-mme), be sure to buy the PCB+Panel without the rare parts from my shop.

Thanks for the kind words fuzzbass ! One correction thought : pdf of the schematics for this project are available to anyone who buy a kit from me.

aabbcc
Wiggling with Experience
Posts: 313
Joined: Fri Feb 02, 2018 10:47 am
Location: Sweden

Post by aabbcc » Thu Nov 15, 2018 7:09 am

Maybe this has been asked before but has somebody done a comparison of this against the boog?

I want dat moog sound and I want to build but the boogs cheap second hand prices are tempting me :deadbanana:

User avatar
jdelgoulet
Wiggling with Experience
Posts: 270
Joined: Tue Jun 05, 2012 4:12 am
Location: Brive , France

Post by jdelgoulet » Thu Nov 15, 2018 8:32 am

About what ?? a Boog ?? Never heard of that ... :bang: :bang: :hihi:

More seriously, I haven't. Some did. I received one comment saying the boog was more 'greety' than the MME - not sure what that means in real life. I had other comment from people owning a real Mini saying the sound of their MME was the same (as the real old Mini).

Honestly, without having heard the Boog, I'm sure it is as close as a MME compared to a real Mini. If you are after the sound, the boog is a good choice. Is you are into DIY (you like soldering etc) and want to support small manufacturer , then get a MME :mrgreen: or simply get both :party:

aabbcc
Wiggling with Experience
Posts: 313
Joined: Fri Feb 02, 2018 10:47 am
Location: Sweden

Post by aabbcc » Thu Nov 15, 2018 9:19 am

jdelgoulet wrote:About what ?? a Boog ?? Never heard of that ... :bang: :bang: :hihi:

More seriously, I haven't. Some did. I received one comment saying the boog was more 'greety' than the MME - not sure what that means in real life. I had other comment from people owning a real Mini saying the sound of their MME was the same (as the real old Mini).

Honestly, without having heard the Boog, I'm sure it is as close as a MME compared to a real Mini. If you are after the sound, the boog is a good choice. Is you are into DIY (you like soldering etc) and want to support small manufacturer , then get a MME :mrgreen: or simply get both :party:
Thanks!

Haha will probably end up getting both lol...

User avatar
fuzzbass
Super Deluxe Wiggler
Posts: 2087
Joined: Thu Feb 06, 2014 6:08 pm
Location: Florida, USA

Post by fuzzbass » Sat Dec 01, 2018 6:27 pm

I'm into the calibration routine. Even though the first synth I ever played was a Minimoog, around 1976, I don't have one, and have never had my scope inside one. So forgive me if my questions are noob. Also please forgive if it has already been dealt with above.

Is there a sample and hold for pitch CV built into the glide circuit? That would explain the behavior I am seeing. It seems like pitch is read on positive gate transitions. disconnect gate CV, and it stops reading pitch CV changes. Release a key and the pitch CV being "held" droops very slowly. Holding the key holds the pitch. Maybe there is a cap here that is not the best choice? Gate CV seems to bleed into the pitch CV and this is not slow or subtle, it can clearly be heard (my Kenton is putting out 10V gates). Should I be using 5V gates?

Is this normal? The primary reason I noticed this is because I am deviant, not following the calibration routine to the letter, but rather reading the VCO frequency at the expansion Molex header, not end to end through the synth.

Also, the PWM (all three) seem to min/max out ~25% / 75%. What resistors should I look at to refine this?

Thanks all.
Wired for weird

User avatar
fuzzbass
Super Deluxe Wiggler
Posts: 2087
Joined: Thu Feb 06, 2014 6:08 pm
Location: Florida, USA

Post by fuzzbass » Tue Dec 04, 2018 7:43 am

And while I'm asking questions, what is the pitch cv voltage range that the glide/cv section can track? On the unit I built, seems like ~ 0V to +4V. I guess on the Minimoog this section did not need to support a range wider than the keyboard, but I want to be sure I don't have a defect. Thanks.
Wired for weird

User avatar
Kevin Mitchell
Wiggling with Experience
Posts: 279
Joined: Tue Jun 27, 2017 10:55 am
Location: Springfield, Massachusetts

Post by Kevin Mitchell » Tue Dec 04, 2018 1:54 pm

fuzzbass wrote:what is the pitch cv voltage range that the glide/cv section can track?
I used a 5 octave keyboard (0 to 5 volts) with no trouble with CV pitch on the MME module.

User avatar
nosotokarjazzy
Learning to Wiggle
Posts: 29
Joined: Fri Jun 10, 2016 2:48 am

Post by nosotokarjazzy » Sat Dec 29, 2018 7:33 am

what are your tips & tools for crimping micromatch connectors ?
they are so frail, I can't manage to close them without breaking them :bang:

doeppler
1-Post Wiggler
Posts: 1
Joined: Sun Dec 30, 2018 9:51 am
Location: The Netherlands

Post by doeppler » Sun Dec 30, 2018 9:54 am

Do you plan to release EagleCAD board and schematics?

User avatar
Rigo
Chief Spam Killer
Posts: 2572
Joined: Mon Dec 27, 2010 4:48 am
Location: Belgium

Post by Rigo » Sun Dec 30, 2018 10:15 am

doeppler wrote:Do you plan to release EagleCAD board and schematics?
What a friendly first post ...

You could have started reading the last page of the thread, where it says:
jdelgoulet wrote:One correction thought : pdf of the schematics for this project are available to anyone who buy a kit from me.

User avatar
fuzzbass
Super Deluxe Wiggler
Posts: 2087
Joined: Thu Feb 06, 2014 6:08 pm
Location: Florida, USA

Post by fuzzbass » Thu Jan 24, 2019 2:01 pm

Back in December I got a bit stuck calibrating an MME build - and I took a break from it to complete a commission build. Now it's back on my desk.

Caveat: I have never been inside a genuine Model D, or previously built the MME.

Oddness with pitch tracking. As pitch CV rose past +3.7V, pitch tracking flattened out; pitch would not be driven higher. The problem was isolated to the Glide circuit. That part appears to be a sample and hold; that is pitch CV is read and stored in a cap during gate-on events. Looking at gate CV, the input on this unit is biased ~-1.3V, and this was summed with the incoming 5V gate to produce +3.7V during gate-on. I remember that number from somewhere... :hmm:

The CV source I am using is a Kenton Pro Solo MKII. When I started, the gate out was configured in the default way: Gate signalling at +5V. Consulting Kenton's manual I found their recommendation for Minimoog D is setting the gate type to "S-Trig no pull-up". This setting did not open the path to allow the Glide circuit to sample/hold a new pitch CV. Trying all versions, I found that "S-Trig high pull-up" mode (inverted gates, +10V off, 0V on) not only functioned, it solved the pitch tracking issue. :guinness:

I would otherwise stop here and start annoying my kids, pets, etc with the synth. But this is a commission build and I want to be sure...

Is this the expected behavior?

(update) S-Trig can't possibly be the solution here since it drives the EGs in reverse.

Thanks.
Wired for weird

User avatar
jdelgoulet
Wiggling with Experience
Posts: 270
Joined: Tue Jun 05, 2012 4:12 am
Location: Brive , France

Post by jdelgoulet » Fri Jan 25, 2019 10:02 am

That is interesting. So it seems a S-trig works for you (for the glide) but in the MME (like in the real Minimoog) the glide works with a V-Trig signal. On the real Mini, the V-trig is at +10v but a +5v showed that it was enough so I did not modified the original circuitry.

On the MME, the gate signal you put at the gate input is directly sent to the glide circuit but also go to a V-trig to S-trig converter (R388, Q98, D7). Make sure you have a 2N3904, a 10k resistor and D7 is in the right position.

Julien

User avatar
fuzzbass
Super Deluxe Wiggler
Posts: 2087
Joined: Thu Feb 06, 2014 6:08 pm
Location: Florida, USA

Post by fuzzbass » Fri Jan 25, 2019 1:50 pm

jdelgoulet wrote:That is interesting. So it seems a S-trig works for you (for the glide) but in the MME (like in the real Minimoog) the glide works with a V-Trig signal. On the real Mini, the V-trig is at +10v but a +5v showed that it was enough so I did not modified the original circuitry.

On the MME, the gate signal you put at the gate input is directly sent to the glide circuit but also go to a V-trig to S-trig converter (R388, Q98, D7). Make sure you have a 2N3904, a 10k resistor and D7 is in the right position.

Julien
Is there a reason you did not put v-trig to s-trig converter in front of the gate input at the glide circuit? On my MME, the glide circuit works with s-trig, and does not work with v-trig (+5V gate). Also, which version of the Model D did the glide/kbd circuit come from?

Thank you.
Wired for weird

User avatar
jdelgoulet
Wiggling with Experience
Posts: 270
Joined: Tue Jun 05, 2012 4:12 am
Location: Brive , France

Post by jdelgoulet » Sat Jan 26, 2019 4:59 am

Is there a reason you did not put v-trig to s-trig converter in front of the gate input at the glide circuit? : yes. There is no need since the circuit expect a V-Trig signal (+10v)

On my MME, the glide circuit works with s-trig, and does not work with v-trig (+5V gate) : unfortunately this is what you see on yours. But it works with +5v without issue for everyone (at least this has never been reported to me and the +15 MMEs I built did not have the issue you are facing).

Also, which version of the Model D did the glide/kbd circuit come from?
MME is based on REV2 but this the glide circuit is the same for all three versions.

User avatar
fuzzbass
Super Deluxe Wiggler
Posts: 2087
Joined: Thu Feb 06, 2014 6:08 pm
Location: Florida, USA

Post by fuzzbass » Sat Jan 26, 2019 7:43 am

jdelgoulet wrote:Is there a reason you did not put v-trig to s-trig converter in front of the gate input at the glide circuit? : yes. There is no need since the circuit expect a V-Trig signal (+10v)

On my MME, the glide circuit works with s-trig, and does not work with v-trig (+5V gate) : unfortunately this is what you see on yours. But it works with +5v without issue for everyone (at least this has never been reported to me and the +15 MMEs I built did not have the issue you are facing).

Also, which version of the Model D did the glide/kbd circuit come from?
MME is based on REV2 but this the glide circuit is the same for all three versions.
Thank you. There must be some as yet unidentified defect in my build. Back to the bench...
Wired for weird

User avatar
jdelgoulet
Wiggling with Experience
Posts: 270
Joined: Tue Jun 05, 2012 4:12 am
Location: Brive , France

Post by jdelgoulet » Sat Jan 26, 2019 9:59 am

By the way, do you have this behaviour when the glide is engaged (switch 'On') or all the time when you use the main CV input ?

User avatar
fuzzbass
Super Deluxe Wiggler
Posts: 2087
Joined: Thu Feb 06, 2014 6:08 pm
Location: Florida, USA

Post by fuzzbass » Sat Jan 26, 2019 12:49 pm

jdelgoulet wrote:By the way, do you have this behaviour when the glide is engaged (switch 'On') or all the time when you use the main CV input ?
Same behavior with switch engaged or not, and increasing glide time also does not affect problem.

Also, when pitch CV threshold ~+3.7V is passed, Q94 goes into HF oscillation, and it's collector jumps to near +10V.

I built Dr. Bob's S-trig converter on breadboard, with high pull up (since it appears in minimoog schematics that R10, Filter Contour Generator, functioned as such). Pull up was connected TP2. The processed trigger was inserted input side of R356 (5.1K) which was lifted from the board. Result was mixed:
1. Glide Circuit now tracks/holds cvs much higher than +3.7V, but;
2. Q95 (which is a switch) opens on key down when CV changes to lower value, but opens on key up when CV increases.

... so this is not the solution either.
Wired for weird

User avatar
fuzzbass
Super Deluxe Wiggler
Posts: 2087
Joined: Thu Feb 06, 2014 6:08 pm
Location: Florida, USA

Post by fuzzbass » Mon Jan 28, 2019 12:01 pm

Resolved.

The problem turned out to be that the v-trig to s-trig converter was loading down my gate source, a Kenton Pro Solo MKii. jdelgoulet suggested earlier that I try buffering my CVs, and I thought I had tried that, but I guess not carefully enough to find it the first time around. I repeated the test this morning, buffering only gate, and quickly saw a difference. The glide circuit's pitch CV output is now tracking -1V to +5V (six octaves) and with range switches on the panel, this is plenty.

Since I want my MME to be happy with the Kenton, I changed the series resistor R388 in the trig converter from 10K to 100K, and now the Kenton does well without a buffer. This did not affect the operation of the contour generators.

Thank you jdelgoulet for being patient with me on this one.

Summary:
The MME common pitch CV input is processed via the keyboard glide circuit. This section is a sample and hold, with variable glide. Gate is required here to switch open the charge path for sampling.

The contour generators in the MME require old style s-trig (switch to ground) signals to fire. Rather than spend years explaining the difference to everyone, jdelgoulet instantiated R. A. Moog's suggested v-trig to s-trig converter in front of the contour generators, so we all can use typical +5V gates to fire them.

The keyboard glide circuit's gate input seems to work fine with +5V gates not converted to s-triggers.

The v-trig to s-trig converter that R. A Moog devised, and jdelgoulet added (ref: Q98, D7 and R388, shown on schematic with contour generators) has lower input impedance than a typical op amp input. The keyboard glide circuit also takes the gate signal, but before the converter. So gate going into the glide circuit can be affected by insufficient drive for the converter. Some gate CV sources may not drive the converter sufficiently to maintain +5V at gate-on. I found one: The Kenton Pro Solo MKii. Changing the Kenton to produce +10V gates made no difference because the issue is power, not voltage - the Kenton is going to produce a consistent drive power on its outputs, regardless of voltage settings.

A quick test to see if this happening: monitor your +5V gate while connected to MME. With my Kenton, it was sagging down to +3.7V, and this was the max voltage that the glide circuit would track. Another indication is when your common pitch CV input seems to hit a ceiling not encountered when running pitch CV directly into the individual VCO CV inputs.

This glide tracking issue may also arise if a gate signal has been passively multed to many destinations in additions to the MME.

I found that when the gate CV sags, the point to which it sags is proximal to the maximum glided CV obtainable from the kbd glide circuit. One can process the gate CV with a buffer to drive more current and satisfy the trigger converter. I experimented further with it and found I could isolate the gate input to the kbd glide circuit from the trigger converter by bumping up the value of R388 in the converter, from 10K to 100K. This change did not affect the firing for the contour generators.

(update) I found the gate input to the glide circuit was also loading my interface down. I obtained further improvement by changing both R355 and R388 from 10K to 1M.

YMMV.
Wired for weird

User avatar
synthcube
you will be assimilated
Posts: 2143
Joined: Wed Dec 21, 2011 6:42 pm
Location: Boston
Contact:

Post by synthcube » Mon Feb 25, 2019 4:36 pm

credit where credit is due, fuzzbass does amazing work, really.

this is likely going to have a display place of great honor in the new synthCube store outside of Boston...

its an amazing project of you haven't looked into it more closely..

Image
facebook: https://www.facebook.com/2SynthCube

synthcube webstore www.synthcube.com/cart
music from outer space: www.musicfromouterspace.com
blacet research: www.blacet.com
MOTM DIY Analog Synthesizers: www.motmsynthesizers.com
music from outer space euro smt: www.mfoseuro.com

User avatar
fuzzbass
Super Deluxe Wiggler
Posts: 2087
Joined: Thu Feb 06, 2014 6:08 pm
Location: Florida, USA

Post by fuzzbass » Wed Feb 27, 2019 8:08 am

synthcube wrote:credit where credit is due, fuzzbass does amazing work, really.

this is likely going to have a display place of great honor in the new synthCube store outside of Boston...
Thank you!
Wired for weird

Karma
Learning to Wiggle
Posts: 47
Joined: Sun Oct 20, 2013 6:25 am
Location: Naoned (france)

Post by Karma » Fri Mar 08, 2019 9:56 am

Hey julien, i send you a private message to have the schematic!!!

audiocommand
Learning to Wiggle
Posts: 3
Joined: Thu Dec 05, 2013 2:36 pm
Location: South Germany

MME noise filter sustane

Post by audiocommand » Sat Mar 23, 2019 6:24 am

Hello,

Have my MME finisht and it work great but with little problems...!

Have no pink noise! withe works great!

Cutoff pot doesn't work but I have a cutoff reaktion with the ENV, emphasis works!

Have no reaktion on the VCA sustane!

The unit is only calibrated in the voltage section the VCO/VCF/VCA part is not calibrated! In the 4v part i have only 3.85v but i read this is ok!

Pleas can i get a schematic pdf.

Thank's for help greethings Dominic

User avatar
Check Mate
Aquatic Beta Particle
Posts: 190
Joined: Mon Feb 07, 2011 1:18 pm
Location: Red Hills of Lardossia

Post by Check Mate » Wed Apr 03, 2019 7:11 am

Dear Julien,

I hope you can help me (or somebody else) because I still don't know how to properly calibrate the modulation mix part.
Especially steps 6:
Adjust Offset trimpot (R384) to null any offset. (What offset should be nulled/how?)

and Step 11:
While switching OSC-MOD control on and off, adjust Offset trimpot (R384) for no pitch change.

Why do I need to change the Value of R384 again in step 11 if I already nulled the offset at step 6?? If so I'd change the offset again or not?

Also when I turn the OSC mod switch on on my unit the OSC pitch is significantly lower.
No matter what I do I always have a pitch change when switching OSC-Mod on and off.

Do you have an example how it should sound with OSC-mod switch on and full modulation applied or a good example.
Also some scope pictures would be nice to know how to perform the calinration.

Thanks!

User avatar
jdelgoulet
Wiggling with Experience
Posts: 270
Joined: Tue Jun 05, 2012 4:12 am
Location: Brive , France

Post by jdelgoulet » Wed Apr 03, 2019 9:04 am

Basically, steps 4, 5 and 6 are part of 'the same step' : you need to null the offset between signals you have on TP10 and TP11. You could read it :

Connect oscilloscope channel 1 to TP10 - VCA input. You should observe oscillator 3 waveform. Then , connect oscilloscope channel 2 to TP11 - VCA output. Adjust R384 to null any offset between the 2 signals.


It is written to do it again in step 11 because I don't know what scope people might have. In step 4, 5 and 6 you do it visually then in step 11 you do it by hear to be sure you don't hear any change in VCO frequency when you engage the VCO modulation (having the MOD-LEVEL know set to 0).

ie : Imagine a live situation. You want to add modulation but 'not yet' so you have modulation MOD-LEVEL set to 0. You switch OSC_MOD to 'on'. At this time you don't expect the VCO pitch to change .

If you always get a pitch change - even with mod-level set to 0, then it might be a matching issue between Q27 and Q97.

User avatar
Check Mate
Aquatic Beta Particle
Posts: 190
Joined: Mon Feb 07, 2011 1:18 pm
Location: Red Hills of Lardossia

Post by Check Mate » Wed Apr 03, 2019 12:31 pm

Thanks Julien for the quick reply!
There is something weird going on also with the Balance trimmer (maybe it is broken or not soldered correctly).
I'll check over the weekend.
Cheers :guinness:

Post Reply

Return to “Music Tech DIY”