Mini Synth demo (that one on 'GGG' originally from Funway Into Electronics 3)

[Mark Hammer]

The challenge with this, as I am finding out (and almost done, thank goodness) is that the original illustration and legending obscures the connections between pads.  The redrawer has to keep referring back to the original and the schematic, in order to discern what truly IS a connection/trace, and what is merely a visual artifact that LOOKS like a trace.  The source of the problem lies in the scan of the original, which seems to have introduced some artifacts. I know it may look like black legending on top of a red PCB lattern, but the legending itself produces a kind of red "halo".

I certainly wish it was easier, but as Moe Szyslak  says: "If wishes were horses I'd be eating wish meat every night".

[iainpunk]

Not to go off on a tangent, but doesn't Illustrator allow you to import an image, and then add a 'layer' over it where you can do the tracing, without messing up the original?

Pretty much any graphics package beyond MS Paint allows you to do that, so yes.

some guys on Discord started doing a MS Paint clone that can toggle the ''transparent colour'' to be transparent and show another image through the one you work on. you can then change the transparent colour to show different parts of the image 'behind' your work piece.
i haven't seen any updates on that project since 2017, so i guess they abandoned it.

cheers

[Mark Hammer]

Okay, here we go.  It is perhaps not as smooth as it might be at a higher resolution, but then the original was not exactly hi-res itself.  I cannot vouch for the pin spacing.  If you use it to make your own PCB, you'll likely have to do a few dry runs at different scales on paper, until you find the right size, before committing to a toner transfer.  Feel free to edit whatever strikes you as too ragged.  Because the original was shown  from the component side, looking "through" the board, there is no need to flip the image for use with toner transfer.

(Addendum: I found that if I printed it out at 19% size, it came out just right)

Enjoy.

[Radical CJ]

Just out of curiosity, have you tried it with a mic? It may be fun.

I might be wrong in my interpretation, but given that the input gets amplified and then sent straight into a schmitt trigger, it wouldn't retain any of the qualities of a human voice (vowels, consonants, etc). And the dynamics of the human voice would make triggering unreliable. I would think that guitar should work better than voice, even better if boosted first, and a keyboard should work better than a guitar.

This reminds me a lot of Freppo's CMOS inventions. I can get consecutive hours of enjoyment making noises with something like this.

[TFJames]

Originally I wondered if one of PCB layout programs might let you 'draw' over an image using it as a guide

I know there's one that do this, but didn't bothered too much to learn how to use it. But anyway, people who are used to work with some image editing software (corel, photoshop) may prefer to just use the one they already know how to work with.

Very nice demo, it shows lots of possibilities this one can do. I really liked how it sounds, and may be building one in the future. Thanks for the demo.

Just out of curiosity, have you tried it with a mic? It may be fun.

Yeah, I've been using Photoshop for years, but always found it's vector and/or curved line capabilities a bit limited. However, how embarrassing, I've just checked and worked out you can actually use 'paths' in ways I hadn't realized before, which will be very helpful in future! Thanks for inadvertently making me discover that haha!

Anyway my reason for maybe wanting to use a proper PCB layout tool was just that they would allow better editing of groups of traces, keeping everything connected if you wanted to shift the layout around, and also highlighting everything that's connected to each other, so one might better understand the circuit. And then the big thing, I thought, would be outputting in the right format for any of the ways you might make a board or have it made commercially. But yes, learning the program probably wouldn't be worth it for a one off like this.

And thank you! Glad you like it, and found it inspiring! Hope you have a go, it's a pretty fun effect.

Yes, I did try it with a mic when I first turned it on, but didn't really play with it much because I was bounding away to grab my guitar! I vaguely remember it being okay-ish, but I think Radical CJ is probably more or less correct:

Just out of curiosity, have you tried it with a mic? It may be fun.

I might be wrong in my interpretation, but given that the input gets amplified and then sent straight into a schmitt trigger, it wouldn't retain any of the qualities of a human voice (vowels, consonants, etc). And the dynamics of the human voice would make triggering unreliable. I would think that guitar should work better than voice, even better if boosted first, and a keyboard should work better than a guitar.

This reminds me a lot of Freppo's CMOS inventions. I can get consecutive hours of enjoyment making noises with something like this.

Also I haven't come across Freppo's CMOS inventions, will have to check them out!

[TFJames]

Okay, here we go.  It is perhaps not as smooth as it might be at a higher resolution, but then the original was not exactly hi-res itself.  I cannot vouch for the pin spacing.  If you use it to make your own PCB, you'll likely have to do a few dry runs at different scales on paper, until you find the right size, before committing to a toner transfer.  Feel free to edit whatever strikes you as too ragged.  Because the original was shown  from the component side, looking "through" the board, there is no need to flip the image for use with toner transfer.

(Addendum: I found that if I printed it out at 19% size, it came out just right)

Enjoy.

That's awesome Mark! Well done! I just remembered I did some arbitrarily colour coded versions I made to help get my head around the connections required (and to work out the issue you mentioned with the obscured connections) which may have been helpful. Sorry I didn't think of it before!!

[Mark Hammer]

Etched.  Drilling and stuffing to follow tomorrow.  The holes in the middle of the pads may or may not be a little too big, but I'll find out tomorrow.

[11-90-an]

Etched.  Drilling and stuffing to follow tomorrow.  The holes in the middle of the pads may or may not be a little too big, but I'll find out tomorrow.

Woah MH, that was very fast...  most of the time I just procrastinate my builds... should clearly stop doing that... lol

This reminds me a lot of Freppo's CMOS inventions. I can get consecutive hours of enjoyment making noises with something like this.

Also I haven't come across Freppo's CMOS inventions, will have to check them out!

Yup, Freppo's website is https://www.parasitstudio.se, lots of cool stuff in there..

[Mark Hammer]

Woah MH, that was very fast...  most of the time I just procrastinate my builds... should clearly stop doing that... lol

We'll see how fast, tomorrow.  Temperature's supposed to drop, and the garage (where the drill press is) may be uncomfortable to work in.  Fingers crossed.

[ElectricDruid]

the garage (where the drill press is) may be uncomfortable to work in.

That's "Canadian uncomfortable" he's talking about, which is probably about -15C in terms the rest of us understand!

[Mark Hammer]

Well, it was that low earlier in the day, but with the midday sun we're up to a balmy -8C, so the drilling got done, and I'm stuffing the board now.

Actually, that color-coded image you posted, Tobias, is coming in handy.  I appears that whoever scanned the Dick Smith book had built the Minisynth, and put a stroke through each component in the parts layout as they were installed, making the information hard to read.  So thanks!

What I'm looking forward to is using it with my EHX Attack Decay pedal, which will impose an amplitude envelope on the resulting sound, to make it more like a guitar synth, and make the glitchiness less evident.

[TFJames]

Well, it was that low earlier in the day, but with the midday sun we're up to a balmy -8C, so the drilling got done, and I'm stuffing the board now.

Actually, that color-coded image you posted, Tobias, is coming in handy.  I appears that whoever scanned the Dick Smith book had built the Minisynth, and put a stroke through each component in the parts layout as they were installed, making the information hard to read.  So thanks!

What I'm looking forward to is using it with my EHX Attack Decay pedal, which will impose an amplitude envelope on the resulting sound, to make it more like a guitar synth, and make the glitchiness less evident.

Gosh, it doesn't half get brisk up there!

Oh great, my pleasure, I'm glad it came in handy!

That would be interesting! I was wanting to build/procure some kind of ADSR pedal a while back, maybe I need to look at that again. Look forward to hearing how you get on!

[Mark Hammer]

The Attack Decay is a woefully overlooked pedal, in my view.  It's digital, so it does a much better job of tracking note onset, and also allows one to insert other effects in a loop and process them, using the envelope of the clean input signal.  It's not marketed as a guitar synth pedal, because it doesn't do the fancy stuff of the explicitly synth pedals, but if you feed it an interesting signal it can serve up sounds indistinguishable from synths.  Perhaps just as important - although that won't apply with the monophonic Minisynth - it is able to map modified envelopes onto notes polyphonically.  That is, feed it a cascade of notes, and each will have its own separate rise and fall.

[11-90-an]

[ot]

Well, it was that low earlier in the day, but with the midday sun we're up to a balmy -8C, so the drilling got done, and I'm stuffing the board now.

We once recieved a visitor from Canada once, nice chap, he said that the summer temperatures there are around 6-7C! Which is very, very different than our Philippine Summer 36-40C...

Amazingly he wasn't a complainer and took the heat really well...

[/ot]

[Mark Hammer]

So, planning out some mods.

1) An added mixer stage, for combining the two channels into one.
2) An envelope follower, driving a bandpass filter which is fed by the mixer output.  I'm thinking fed from the output of of IC1A.
3) Fast/slow range toggle for the tremolo.
4) Instead of two switches for individual tremolo, a single master 3-way switch for none, channel 1 only, or both.
5) A normal/mild switch for tremolo, which I *think* will involves changing the value of R35/R36.
6)  I was thinking of putting the various octave up/down options on a 6-position rotary switch instead of two DPDT toggles, but I have to work out what such a switch could give me before committing.

The whole thing seems to be amenable to the 1590XXLG enclosure I had sitting around.  This could be fun.

6-7C in the summer?  Pish posh.  We get plenty of days in the mid-30s.  I like to tell people that the city of Calgary is "the future of weather".  Absolutely no respect for the seasons there.  People can be snowboarding in shorts and bikini tops in February, shovelling their driveways in August, and praying for a white Christmas as they mow the grass in November.

[Mark Hammer]

Finally fired it up.  Interesting how the comparator aspect means you don't hear anything until you exceed the threshold.  The sustain is worthy of Robert Fripp.  But I don't seem to hear the PLLs.  The pads on the board layout suggest the 3-position slide switches are not especially standard.  What did you use on your build?

By the way, I have to compliment you on your build.  I didn't quite realize it at first, thinking that it was a kit, but clearly you did the thing from scratch, so hats off.

For my part, I'm going to have to go over the board tomorrow to see where I've gone wrong.

[TFJames]

Finally fired it up.  Interesting how the comparator aspect means you don't hear anything until you exceed the threshold.  The sustain is worthy of Robert Fripp.  But I don't seem to hear the PLLs.  The pads on the board layout suggest the 3-position slide switches are not especially standard.  What did you use on your build?

By the way, I have to compliment you on your build.  I didn't quite realize it at first, thinking that it was a kit, but clearly you did the thing from scratch, so hats off.

For my part, I'm going to have to go over the board tomorrow to see where I've gone wrong.

Yes, I thought that was an interesting/good feature! And the sustain is surprisingly good too!

But oh no, problem! Yeah I originally bought the wrong sort and ended up paying a stupid amount to quickly get the unusual ones. The ones I used are Devlin SS-23F02-L7 from RS: https://americas.rsdelivers.com/product/devlin/ss-23f02-l7/panel-mount-slide-switch-dp3t-500-ma-50-v-slide/8290611

I'm not sure now whether the single pins connect to the three opposite them, or the three in line with them, but I'm leaning toward the first.

Oh thanks so much! Yes, I came across scans of the project and thought it sounded interesting, so slowly collected the parts and put it together. Thanks again, I appreciate it!

Good luck!!

[anotherjim]

I think they are 3 throw switches and the only alternatives are the similar toggle lever action (same slide mechanism pinout under) which will mean panel legends needs reversing or a 2 pole rotary switch set for 3 throws which would be easier to make wired off-board.
The intended style are rarely stocked, but still made and can be found from Asian vendors like Ali-express. Some vintage instrument spares suppliers carry them as Roland once used them a lot but these sources are expensive.

I don't remember what the subject was, but we have talked about this kind of switches before (may have been a Hens Teeth/DeadendFX subject).
I've a suspicion the pinout was not the expected with the common pin out on its own.
Note that on the schematic, the 2 switch poles are wired opposite ways so as you look at it, if the top left pin is made to its common pin, then the lower right pin is made to it's common pin.
Should be possible to confirm with DMM resistance/continuity beep which contacts are which.

If the project doc doesn't explain fully...

4046 pin 4 connected to pin 3, output from pin 4 = original pitch
4046 pin 4 connected to pin 3, output from 4013 Q pin = octave down
4046 pin 3 connected to 4013 Q pin, output from 4046 pin 4 = octave up.

[duck_arse]

I have one of the original DSE bought switches here [they were real cheep], it is stamped stylised PIC. contact pattern is

Code Select Expand


A1 A2 A3        Bcom
   |             |
Acom        B1 B2 B3


pin spacing is a lttle over 0.1", such that the 6 pins cover 7 x 0.1 grid.

[Mark Hammer]

MANY thanks for that!  Looking at the PCB layout, I now realize that the principle reason my unit was not working properly is because I was treating the pinouts on the board as corresponding to a more typical  toggle.  So far, I've now got the tremolo working, although the monophonic part still eludes me.  Hopefully later today.