My MFOS modular expansion box project (work in progress)

Discussion & Documentation for Ray Wilson's synths, modules and noise boxes.

Moderators: Kent, dslocum, synthcube

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

My MFOS modular expansion box project (work in progress)

Post by Bodo1967 » Mon Jul 02, 2018 7:23 am

... or whatever you might call it.

Anyway, I am currently building 7 MFOS modules that'll go into a 5U case about 20 inches (50 cm) wide.

Modules will include (one each):

1) Wave Freaker
2) Quad Timbre Bank
3) Noise Cornucopia
4) CV & Gate Expander
5) Delayed Modulation
6) Echo Rockit
7) 24 dB/oct VCF from the 'Oldies but Goodies'

I made (Echo Rockit) or will make (24 dB/oct VCF) the PCBs for the last two ones myself. The latter one isn't available anyway, and my decision to build two Echo Rockits* was before I decided to order the other five PCBs. If I had had that idea two weeks sooner or so, I'd have added them to my order.

* one in a conventional case as a birthday present for a friend, and one for myself as a module

The case is half built (frame is ready). It'll get two wooden "rails" underneath so I can put it on top of my Ultimate (the blue one in the build thread).

The 1) - 5) PCBs are almost finished except for very few parts which are still on their way. Most of the front plate holes are drilled, and I've begun inserting pots and switches so I can start with the panel wiring soon.

Panel layouts are my own design, because I decided to make all connections with both 1/4" and 3.5 mm jacks - makes it more flexible, for instance for using it with my TTSH as well. I'll of course share the designs in Scribus or PDF format if anyone's interested.

The 24 dB/oct VCF wasn't originally planned, but when I bought sheet aluminum (500 mm wide) for the front plates they cut it for me according to my requirements, and that left a rest about 2.7" wide. That's when I thought "why throw it away, just make another module and add it". I decided on that VCF because it's relatively small (don't have space for 15 pots or jacks on that panel), and I figured a 24 dB/VCF might be a nice addition to the two 12 dB/oct VCFs in the Ultimate & Expander.

However, I'll include a 4-input mixer stage with the VCF module, using the signal mixer circuit of the Ultimate's audio mixer.

I'll start adding photos soon and turn this into a build thread.

Oh, and speaking of photos:
@synthcube - if you wish to add pictures of my Ultimate to the gallery on the MFOS website, I'd be happy with and honored by that!

And I'd like an MFOS T-shirt, but they seem to be out of stock :(.

Edit: corrected a few typos and added two missing words
Last edited by Bodo1967 on Fri Jul 06, 2018 1:15 am, edited 1 time in total.
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Re: My MFOS modular expansion box project (work in progress)

Post by Bodo1967 » Thu Jul 05, 2018 2:23 am

Bodo1967 wrote:I made (Echo Rockit) [...] the PCBs for the last two ones myself.

The case is half built (frame is ready). It'll get two wooden "rails" underneath so I can put it on top of my Ultimate (the blue one in the build thread).

[...] I'll start adding photos soon and turn this into a build thread.
PCB layout printed on paper and attached to one side of the PCB

ER 01
Image

A few pilot holes drilled (marked with felt pen) so I can use those to align the backside print properly.

ER 02
Image

I printed the front and back side layouts (each one twice) on transparent inkjet printer foil, stacked those prints atop each other for better optical density, exposed them to UV light, developed them and etched them using sodium peroxo disulfate - 2 stages:

ER 03
Image

ER 04
Image
... why buy it for $100 when you can build it yourself for $150?

User avatar
hermbot
Ultra Wiggler
Posts: 765
Joined: Sat Jul 16, 2016 7:42 pm
Location: Wisconsin

Post by hermbot » Thu Jul 05, 2018 10:47 am

Sounds like a great project. I've considered doing something similar in a spare Pelican case I have. I was considering doing several MFOS modules in a custom panel or several NLC. Or maybe a mixture of both...

Looking forward to see how this goes!

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Fri Jul 06, 2018 3:33 pm

Here's the 1) - 5) module PCBs, more or less finished except for the ICs:

Wave Freaker
WF 01
Image

Noise Cornucopia
NC 01
Image

CV & Gate Expander
CVG 01
Image

Quad Timbre Bank
QTB 01
Image

Delayed Modulation
DM 01
Image

And this is the case with the panels placed loosely in no particular order*. The 'rails' at the bottom are bespoke to fit on the MFOS Ultimate.

* from left to right it happens to be
Wave Freaker - Quad Timbre Bank - Delayed Modulation - CV & Gate Expander - Echo Rockit - Noise Cornucopia - 24db/oct VCF

Gen 01
Image
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Re: My MFOS modular expansion box project (work in progress)

Post by Bodo1967 » Fri Jul 13, 2018 1:54 am

Bodo1967 wrote:7) 24 dB/oct VCF from the 'Oldies but Goodies'

However, I'll include a 4-input mixer stage with the VCF module, using the signal mixer circuit of the Ultimate's audio mixer.
I etched the PCB on Wednesday, drilled it yesterday, and added all resistors and wire jumpers. There are very few parts I had to order, but that's no big deal.

The PSU for the case is also almost finished; I etched and drilled the PCB on Wednesday, there's just two electrolytic caps missing - hope I get them today or tomorrow.

More pics to follow soon (probably on the weekend).
Bodo1967 wrote:And I'd like an MFOS T-shirt, but they seem to be out of stock :(.
Perhaps the print layouts are still available so I could print my own (or have it printed)...?
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Fri Jul 13, 2018 3:40 pm

The 24 dB/oct VCF so far:

VCF 01
Image

There's only two 100 pF caps and a 100R trimmer pot missing on the PCB now. and of course the ICs.

The wires are due to the audio mixer I kludged on the board (bottom left). (+) and (-) voltage connections, plus a wire from the mixer out to the VCF in. The rest of the kludging is on the bottom side of the PCB.

It won't win a beauty contest, but so what.

And this is the 'conventional' Echo Rockit, the one that's gonna be a birthday present for a friend:

ER-CONV 01
Image

PCB is fully equipped (except I haven't inserted the ICs yet), panel wiring is finished and I've begun connecting the panel and the PCB.
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Tue Aug 21, 2018 2:58 am

Just a small update:

Noise Cornucopia and the 24 dB/oct filter are ready and working.

Delayed Modulation - extremely strange behavior in all (!) of the submodules. Neither the LFO nor the envelope generator nor the VCA seem to work properly. I have no idea why.

Wave Freaker - some problem. Something pulls the -12V to ground (as in, short circuits it), and I haven't yet found the culprit.

Quad Timbre Bank - The input signal doesn't get past the 100k input resistor. Voltages on the ICs are OK, and again I haven't found the reason yet.

Echo Rockit (the one in the blue box; haven't even drilled nor populated the PCB of the module version yet) - that thing drives me crazy. LFO and the oscillator are working nicely, but I only get 0.5 V at the 78L05 that is supposed to feed the PT2399. Hence no delay whatsoever, and again I don't find the reason.

2 out of 6 are working. Not a very good rate. Wouldn't have thought a module could be way more troublesome to build than a MFOS Ultimate, a TTSH, a Jasper, an ARP 1601 or an Oakley SRE-330 :bang: .

:deadbanana: :deadbanana: :deadbanana:
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Sun Aug 26, 2018 11:59 am

Bodo1967 wrote:Delayed Modulation - extremely strange behavior in all (!) of the submodules. Neither the LFO nor the envelope generator nor the VCA seem to work properly. I have no idea why.
Some progress here. The LFO now behaves properly. I had to insert another 1M resistor between the LFO rate indocator LED and its connection point. Apparently the LED drew too much current which affected the LFO signal.
Also, I now get more or less reliable signals at the output. However, the VCA bias still needs looking into: the trimmer pot for the VCA that should be set so you get as close to 0V when no gate is applied doesn't get be lower voltages than approximately 3 V.
Bodo1967 wrote:Quad Timbre Bank - The input signal doesn't get past the 100k input resistor. Voltages on the ICs are OK, and again I haven't found the reason yet.
Bad solder joint. I don't make these often, but occasionally happens... solved. Works basically, haven't tested the CV modulation yet.
... why buy it for $100 when you can build it yourself for $150?

User avatar
wiperactive
Wiggling with Experience
Posts: 446
Joined: Sat Apr 12, 2014 12:17 pm
Location: Gloucestershire, UK

Post by wiperactive » Sun Aug 26, 2018 4:51 pm

Watching with interest. Have several MFOS original issue PCBs waiting their turn in a seriously over-long pile of pending projects including a couple of Quad Timbre Banks and Wave Freakers. Not much out there on these as finished projects, especially the QTB which should be an even more potent filtering set-up when optionally multiplied as Ray had suggested, so I look forward to how this one shapes up for you.

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Tue Aug 28, 2018 9:12 am

Bodo1967 wrote:Echo Rockit (the one in the blue box; haven't even drilled nor populated the PCB of the module version yet) - that thing drives me crazy. LFO and the oscillator are working nicely, but I only get 0.5 V at the 78L05 that is supposed to feed the PT2399. Hence no delay whatsoever, and again I don't find the reason.
Update: I gave up on my self-etched PCB yesterday :deadbanana: , and ordered 2 original MFOS PCBs from the UK instead.
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Mon Sep 03, 2018 5:02 am

A few more pics:

QTB 02
Image

I used self-cut-and-drilled aluminum profile pieces as holder brackets.

QTB 03
Image

One gets an idea of how the finished module looks now - by now, I replaced the paper drilling template with transparent adhesive foil, and the buttons and missing 3.5 mm jacks are in place.

WF 02
Image

... same procedure for the Wave Freaker.

WF 03
Image

Concerning the above mentioned short circuit between -12V and GND: I'll disconnect all wires except power supply from the PCB. If the problem then still exists, it's on the PCB. If not, it's a panel wiring problem. Then (provided it's a wiring problem) I'll start resoldering one after the other and check for an arising connection between -12V and GND after every singe wire so as to locate the problem source. We'll see...

And I'll do the same for the Delayed Modulation.

And, last pic for today, this is the 24 dB/oct VCF.

VCF 02
Image

The additional IC in the upper right does the audio input mixing (I used the 4 input mixer circuitry from the MFOS Ultimate - not a big deal).
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Wed Sep 05, 2018 6:32 am

Bodo1967 wrote:Update: I gave up on my self-etched PCB yesterday :deadbanana: , and ordered 2 original MFOS PCBs from the UK instead.
Update: They arrived yesterday, and I'll (re-)start the build soon, perhaps tonight.
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Sat Sep 08, 2018 9:59 am

Bodo1967 wrote:
Bodo1967 wrote:Delayed Modulation - extremely strange behavior in all (!) of the submodules. Neither the LFO nor the envelope generator nor the VCA seem to work properly. I have no idea why.
Some progress here. The LFO now behaves properly. I had to insert another 1M resistor between the LFO rate indocator LED and its connection point. Apparently the LED drew too much current which affected the LFO signal.
Also, I now get more or less reliable signals at the output. However, the VCA bias still needs looking into: the trimmer pot for the VCA that should be set so you get as close to 0V when no gate is applied doesn't get be lower voltages than approximately 3 V.
Update, and a positive one at that: I got the module working today. Sometimes mistakes are so obvious you don't see them although (or because?) they are so simple. We call that "not seeing the forest for all the trees" in German:

Ray apparently forgot a connecting wire between the panel elements (jacks, pots and such) and the PCB in the panel wiring diagram. The GND connections are only intra-panel. Added that wire, and *bingo*. I could even remove the 1M resistor I had added in the LED subcircuit, and I could bias the VCA to 0.00 V - as it should.
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Mon Sep 10, 2018 2:01 am

Bodo1967 wrote:Wave Freaker - some problem. Something pulls the -12V to ground (as in, short circuits it), and I haven't yet found the culprit.
Well... removing the decoupling caps didn't improve anything, and neither did disconneting all wires between the panel and the PCB. But when I took the PCB off its holder bracket I finally found it: The PCB had a tiny scratch right underneath the (aluminum) holder bracket, and of course this scratch had to be on a trace. Yes, the -12V trace. So I put the caps back in and placed a little bit of foam rubber between the PCB and the bracket - and yep, no continuity anymore. I'll reconnect the remaining wires soon, but that should've solved the problem.
Bodo1967 wrote:2 out of 6 are working.
4 out of 6, considering the QTB and the DMM, and hopefully 5 out of 6 if and when I get the Wave Freaker running sometime this week.
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Mon Sep 17, 2018 2:45 am

Wave Freaker
Bodo1967 wrote:The PCB had a tiny scratch right underneath the (aluminum) holder bracket, and of course this scratch had to be on a trace. Yes, the -12V trace. So I put the caps back in and placed a little bit of foam rubber between the PCB and the bracket - and yep, no continuity anymore. I'll reconnect the remaining wires soon, but that should've solved the problem.
It did for the -12V continuity. Solved so far. But when I put the ICs back in, I suddenly hat continuity between +12V and GND. WTF? Immediate assumption: The previous -12V continuity may have damaged an IC.

So I removed all ICs again and re-inserted them one by one, measuring for continuity after each single IC. And it indeed turned out to be the CD4066 that caused continuity. So I replaced it, and yes: no continuity anymore. However, the Wave Freaker still isn't working.

By now, I am suspecting my LM339 to either be faulty or to be something else but not real LM339. The signal (I used a triangle wave from my Minibrute for testing) is present perfectly on the inputs (proving the buffer stages are working as they should), but there is absolutely nothing coming out of the outputs - flat line on the oscilloscope, also after replacing the "LM339N".

I'll try and replace one with an LM2901 (I have some, and they should be sufficiently compatible at least for testing) and see whether that provides any useful signals.

At least, I finally got the first Echo Rockit (the standalone unit in the blue case) up and running yesterday. The supply voltage for the PT2399 was OK all the time :bang: , I had measured against GND and not against battery negative :doh: .

Replacing the self-etched PCB probably wouldn't have been necessary - by now I assume that the (first) PT2399 was destroyed when I put on supply voltage when at first a few contatcs beteween front and back side of the PCB were still missing, and that it probably would have been sufficient to just replace the chip... :roll:

The second Echo Rockit (the module) is also on its way, the components are on the PCB, and I'll do the panel wiring soon.

There's good progress with the CV & Gate Expander; about half of the panel wiring (or even a little more, since the GND connections are complete) is done.
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Mon Sep 24, 2018 2:11 am

Good progress on the weekend.

Continued and finished the panel wiring on the CV & Gate Expander...

CVG 02
Image

... tested it using the Arturia Keystep, and it works nicely. So it went into the case:

Gen 02
Image

The modules in here are, from left to right:

(1) Delayed Modulation - (2) a gap for the Wave Freaker - (3) 24 dB/oct VCF - (4) Noise Cornucopia - (5) Quad Timbre Bank - (6) a closed-by-now gap for the Echo Rockit - (7) CV & Gate Expander.

And I finished (including successful test) the Echo Rockit (module version).

ER-MOD 01
Image

It already sits in the right gap on the above picture.

That makes 6 down, 1 to go. I'll try and troubleshoot the Wave Freaker probably this week - try LM2901 instead of my "LM339N", and replace the CD4013 which may also be faulty.
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Tue Sep 25, 2018 1:14 am

Bodo1967 wrote:I'll try and troubleshoot the Wave Freaker probably this week - try LM2901 instead of my "LM339N", and replace the CD4013 which may also be faulty.
I replaced the ICs yesterday, but this led to no improvement whatsoever. I'm at my wit's end for the moment. I don't even have an idea where to look next. :deadbanana: .
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Sat Feb 02, 2019 10:13 am

OK, to wrap up this thread: 6 Modules (all except the Wave Freaker) are working fine.

And I have, as of today, officially given up on the Wave Freaker. I am sick and tired of wasting yet more time without being able to find the mistake. I have proper signals at the inputs of the LM339 (tried several from different sources, and LM2901 as well), but nothing reasonable is coming out of them.
... why buy it for $100 when you can build it yourself for $150?

User avatar
wiperactive
Wiggling with Experience
Posts: 446
Joined: Sat Apr 12, 2014 12:17 pm
Location: Gloucestershire, UK

Post by wiperactive » Sat Feb 02, 2019 11:47 am

Thanks for sharing your work.

I have several MFOS kits which are still way back in a huge pending list after years of SDIY inactivity. Currently, I've just tackled a long overdue Paia 9720 dual VCO kit as a practical 'refresher' and confidence booster, and also to get familiar with the masses of flying wires typical of MFOS projects! Happily it worked first time. Still some oscillator calibration to do as a finaliser.

If and when I get onto my Wave Freaker I'll report back on my build and outcome to see if any light can be shed, though your result as an experienced builder is a little unsettling at this juncture with little or no other builds, as far as I'm aware, being posted elsewhere.

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Sun Jul 14, 2019 12:47 pm

There's a second module cabinet on the way... 8-)

It will contain mostly, but not entirely, MFOS Modules:

1) VC LFO (two of them as one double module)
2) Dual lin/log VCA
3) 8 Stage Phaser
4) (Triple) Saw to Triangle Converter - I am currently designing a PCB for this
5) CV distributor (schematic will be taken from the SL Ultimate Expander, and I'll design a separate PCB for 3 or 4 of them in one module)

... and, non-MFOS ...

5) Oakley Sound discrete ladder filter
6) YuSynth ADSR (two of them as one double module)
7) "Bouncing Balls", LFO based envelope generator with acceleration during the repeats (just like a bouncing ball, hence the name). I'm currently working on the PCB layout.

... and we'll see if and how much space will be left.

And I hope to get the WaveFreaker working somewhen.
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Wed Aug 28, 2019 5:12 am

Some update:

Positions (1), (2), (3), (4) and (6) are working now (I still need to make the panels though).

PCBs for (5) are on the way. (6) is mostly finished but I need to make the panels first (I need to fit in the pots in their corresponding holes before soldering them).

And there's more to come...

8) Flight of Harmony Quad Plague Bearer
9) CGS Serge Triple Wave Multiplier
10) CGS Modulo Magic
11) CGS Infinite Melody
12) CGS Diatonic Converter
13) CGS Bi-N-Tic Filter
14) CGS Burst Generator
... why buy it for $100 when you can build it yourself for $150?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Sat Oct 19, 2019 4:24 pm

I have now started making the panels, and the modules will go in the second box (which is also ready and equipped with two PSUs (one each with +/- 12 and +/- 15 V, and I may even add aonther +/- 12 V PSU if required) by and by as they get ready.

The f(h) Quad Plague Bearer (8) and the Saw to Triangle Converter (4) are now complete with their panels:

Image

Image


Image

Image
... why buy it for $100 when you can build it yourself for $150?

johnstilton
Common Wiggler
Posts: 126
Joined: Tue Oct 09, 2018 10:31 am
Location: West Chicago

Post by johnstilton » Wed Oct 23, 2019 4:14 pm

Thanks again for sharing your work. I hope to find more time to make progress on some of these this winter.

I am very curious to know the results of interfacing the MFOS and CGS circuits.

Did you buy the transistor or the diode suprladder PCB from Oakley?

User avatar
Bodo1967
Veteran Wiggler
Posts: 659
Joined: Wed Aug 10, 2016 2:27 pm
Location: Germany

Post by Bodo1967 » Thu Oct 24, 2019 3:00 am

johnstilton wrote:I am very curious to know the results of interfacing the MFOS and CGS circuits.
Oh yes, so am I :lol: !
johnstilton wrote:Did you buy the transistor or the diode suprladder PCB from Oakley?
The transistor ladder filter. Needs troubleshooting though since it doesn't work (yet) :confused: :waah: .

BTW, did you receive the PCB in the meantime?
... why buy it for $100 when you can build it yourself for $150?

johnstilton
Common Wiggler
Posts: 126
Joined: Tue Oct 09, 2018 10:31 am
Location: West Chicago

Post by johnstilton » Thu Oct 24, 2019 11:55 am

BTW, did you receive the PCB in the meantime?
not yet. maybe it got held up in customs. this happens sometimes (esp. with merch from England for some reason).

Post Reply

Return to “Music From Outer Space”