My first DIY modules - starting a Kosmo format build

[juggle]

Update on the clock divider coming up…it works…but there was a small error in the stripboard layout. Bonus points if anyone can find it before I make the post and get it up

I also did a few other general updates today. The top rail in my case was bugging me. It was pulling away from the case and when I’d screw modules in they’d just kind of wedge between the rail and the case instead of actually screwing into it. Plus…on one side it was just over 12mm thick so some LMNK modules just wouldn’t fit over there…which was a bit of an issue since that’s where I planned on putting them.

So - I took the rail out and sanded one side down about 2mm tapering off towards the middle where it was the right thickness. Took me 3 or 4 passes with the sander but finally got it where I was happy. Then put glue on it this time before screwing it back in. And broke out some good clamps to hold it in. Now modules actually screw in properly.

While I had them out I did a little cleanup on the VCA’s since the waxed lacing I ordered came in on Thursday:

It’s quite a bit neater. Not quite telco quality but better than the rats nest it was. I still need to improve my lacing technique a bit but I figure that will come with practice. I’m not huge on the lacing I got though. It was listed as telco quality and many of the reviews on Amazon for it said it was exactly what people were used to using in telco…but…it’s rounder than lacing I’ve seen in electronics before. So it doesn’t sit quite like I’d expect it to. It’s also a bit big for this size of wire.

On this module I used 30awg silicone insulated wire which I really enjoy working with - I also have 22awg and 18awg but the 18 is way too heavy for this kind of work, and the 22 is almost too heavy as well - even though the 30 is almost definitely too fine for this, it’s really hard to get pins into housings with it because of how fine the wire is…but my 22awg is sometimes hard to get into housings because it’s too thick. Going to pick up some 24 and 16 since I need more wire anyway and am still looking for that sweet spot

No photos of the case mods…it doesn’t really look any different…just the screws are a bit more secure and the LMNK modules now fit anywhere in the case.

I’ll share details about the divider shortly - may get a little long winded…

[juggle]

Alright -Short version: I’ve got a fully functioning clock divider!

All of my files (including DIYLC source, and f3d/step/stl/3mf for the panel) for this are in my github here: Modular-Synth-Build/Clock Divider at main · jhitesma/Modular-Synth-Build · GitHub

Long version - buckle up.

So I decided to go with the clock divider circuit shared by Niklas Ronnberg here based on an original design by Matthias Herrmann for his Fonitrokik synth but modified to be non-inverting and added a startup pulse. Sounded exactly like what I wanted so I breadboarded up part of it to test that it would work with 2N2904 transistors and with that test successful started drawing up a strip board layout.

The resulting layout is:

It’s bigger than I wanted, had way more trace cuts than I wanted. And just feels a bit clunky…but…it was fairly easy to build and it works. At least…it does as pictured. The version I built (pictured in an earlier post) almost worked.

Here’s the board without components to make trace cuts easier to see since a few are rather hidden:

It fits on a piece of Tayda “large” stripboard with 4 strips to spare. But I left them on since I designed the panel to mount a full length board (or…so I thought.)

I’m going to go ahead and document the full build here since I haven’t seen that done too much and figure it could help someone just getting started (not that this is a very useful module when first getting started…but…it’s what I documented.)

First thing I do is mark out the cuts to be made on the component side, this is easier than trying to figure out where they go on the strip side and lets me make sure I’ve got them all placed correctly:

Once I’m sure they’re marked correctly I use a single pin header pin to transfer them to the strip side. I just poke the pin in where marked - and then stab it with my marker on the strip side. For multiple cuts in one column - I just count them out instead of transferring them individually:

Oh, that marker I use…I love it. I saw it suggested by Adam Savage on his tested channel and was intrigued enough to order 3 of them off Amazon. They’re great for transferring marks in long skinny holes…and just all kinds of general marking as well - kind of like a long skinny fine tipped sharpie:

Then I double and triple check that everything is transferred and marked on the strip side correctly:

And once I’m confident I get out a small drill and just twirl it in my fingers 2 or 3 times to cut the traces:

After cutting one last sanity check:

Then I start adding jumpers next. I use my 30awg silicone wire for most of these. For shorter jumpers I’ll just use a piece of cut of component lead sometimes (There’s one I did that way on here.)

I strip one end of the wire and tin it:

Then lay it in place and mark it at the hole it will go into (sorry it’s hard to see my mark since this was a black wire…didn’t think that through too well for the photo!)

Then I strip to the mark - tin and solder it in place.

Once all the jumpers are in I add any lower profile sockets and the power connector (I also added D1 here before I remember to take a photo.)

For there it’s like building any other board. I start with the smallest/lowest components and work up. Diodes first, then resistors, small capacitors, transistors, and finally the bigger caps.

Now I discovered my first mistake. I used the board measurements from DIYLC to plan the panel and it’s mounts. But…that only works if you’re using cut off pieces of board - these boards are actually a bit bigger than DIYLC expects resulting in:

The need to drill a new mounting hole that matches my panel.

Also - I positioned the board so the mounts on the panel would clear the jacks…but didn’t take into account where things attached on the board. So this is less than ideal:

Thankfully there’s room I can move those wires behind the reset jack up a few columns to get them the clearance they need in this case.

I also forgot to get a photo of how I did the grounds but you can kind of see it in that photo. Basically I put the jacks in with the ground lugs facing the center and each LED into the panel with the ground to the center. Then bent one leg up and one down so all the grounds tied together. I didn’t really have room to use connectors on the side of the board against the panel so decided to not use them on either side in this case even though I generally prefer having connectors between the panel and board:

But man…that’s a lot of wire to manage!

I still need to lace it up a bit better…but twisting the pairs and soldering it up helped a little bit:

From there it was a simple matter of giving it power and making sure nothing smoked or blew up and the IC socket had correct voltage on the correct pins. Since I did I popped in the IC, put it in my case and gave it a clock signal.

And it worked! Until it got to the 1/64 division and both the 1/64 and the 1/128 came on. Hmmm…oh well, I added those two myself anyway as the original circuit stopped at 1/32. Let’s do a little patch to see how it works anyway:

Clock divider works

I may adjust the resistors on the outputs since they do seem a little low. I haven’t measured them yet to see what they’re actually at. But you’ll notice in the video that the LED’s on my noise drum don’t really show up when triggered from the divider and the output from them is quite a bit lower. I do seem to remember Kristian saying he noticed the output from this was a bit low and he changed the resistors in his circuit as a result. (though his circuit uses a different layout on the outputs with a voltage divider since it runs the outputs off 12v instead of 5v…so I may need to make bigger changes to boost that…or…just live with it.)

And what was causing the 1/64 and 1/128 divisions to tie together? If you look at the original stripboard I posted yesterday and the one I just shared you should be able to notice an extra trace cut.
E23 to be exact. The 1/128 output was accidentally bridged to the 1/64 trigger. Added that cut between them and now it’s working as expected!

So - few small mechanical issues that weren’t quite perfect with the mounting hole position and the location of some of the outputs. But I’m very happy with it and the options it opens up.

Kind of want to build a logic utility module now…but need to finish a few others that I’ve already started first - and really can’t wait for those boards to get my seq v4 connected so I can really start to have some fun.

[juggle]

So with the line driver boards for my sequencer interface scheduled to arrive today I wanted to try and get the panel ready…or at least test it since there were a few aspects of it I wasn’t sure about. Good thing I did…there are a few things I need to address:

First glance it looks pretty good. The shelf for the line driver board seems sturdy enough to withstand the forces of plugging/unplugging an RS-232 cable. And the mounting bosses for some boards to mount the DPDT pushbuttons to seem to put them at a proper height.

But…I need to raise the analog output board a bit higher to clear the jacks. It’s a square board so I have 4 mounting options but the board is so tightly packed there just isn’t any space along any of the edges. This is the closest fit:

I could remove that header…it’s not exactly necessary. But still things are a bit too close for comfort. So need to do some adjustments here.

The biggest issue that’s going to be the trickiest to address is the alignment of the DPDT buttons:

As I feared once mounted to perfboard they don’t quite line up. I’d debating the best way to address this. Ideally I should choose a spacing based on the perfboard/switches and then adjust the holes accordingly. But I like the spacing of the holes as they are and doing that would make the board a bit more cramped while still leaving more empty space which is a really lousy compromise. I could just make the holes a bit bigger…but then the switches would just look sloppy.

So…I need to think about how to address this a bit better still.

And I still need to find a way to mount a USB port for +5v power out to the sequencer. And before I reprint I want to confirm the line driver fits and the RS-232 port lines up correctly anyway.

Not sure I have enough white PLA to print another…starting to get low. Thankfully we’re finally cooling down and temps are only in the low 100f range and no longer hitting 110f regularly so I’m probably safe to order more PLA again

[willow2x]

Nice to see another 3D printing synth guy here =)
Could you not change out the headers for right angle ones?

Also do you actually juggle? I’ve been into circus based skills for as long as i can remember =D
Poi, Juggling, CJ, Clubs, Staff, Devil/Flower Sticks, Yoyo… all kinds.

[juggle]

@willow2x

Nice to see another 3D printing synth guy here =)
Could you not change out the headers for right angle ones?

I could, but it wouldn’t give enough clearance. And since I need to reprint the panel anyway to adjust the switch positions and fine tune the line driver mounting I may as well adjust the position of the analog output board as well.

As for my name…I can…but not well enough to justify the name As many nicknames do it came from college. My freshman year I was bored one evening and decided to teach myself to juggle with some tennis balls that were laying around. And I quickly found that while I could almost juggle them normally - I had no problem “reverse juggling” then by bouncing them off the floor. And while the dorm floors were poured concrete…apparently tennis balls were like mallets against a drum head as far as the girls on the floor beneath us could tell. When they came up to see what the noise was about I was branded with the name - I fought it for a few years but finally gave in and accepted I was stuck with it

The line driver boards just arrived moments ago and as I suspected - even having the .brd file to open in Eagle and import into Fusion the lack of models for the RS-232 connector means a few of my assumptions about just how things were fit were a touch off. I expected the connector to stick past the edge of the board for example - but it’s actually flush. I had planned on mounting things so the connector was flush with the front of my panel so the shape of the attaching cable wouldn’t matter. But this means I either have to leave room in the panel cutout to accommodate the cable - or I have to shift the whole board forward a few mm and open the opening a bit so the board fits along with the connector. And I think that’s the way I’ll go since I don’t want to get into a situation where switching to a different cable means I have to redesign the whole panel. So I need to shift the driver platform down about 2.5mm and shift the holes forward and enlarge the opening for the port to allow the board to fit as well.

Rethinking my plan to add an opening for a USB port though due to the current draw from the Sequencer. I use a 2a USB plug to power it as even a 1.5a can have issues. And since I’m running the modular off a 1.5a 12vac wall wart currently…seems like I should just power the sequencer off it’s own supply instead of building another supply into the modular case just for the sequencer.

[juggle]

Some good progress tonight, but the sequencer isn’t putting out CV just yet

I made some modifications to the panel and used the last of my white PLA to print it out…and I really do mean last. There’s maybe 6’ of filament left on the spool. Will have more Wednesday but looks like I’m not adding any new modules tomorrow or fixing any issues on this one!

Thankfully, my fixes were sufficient that this will be usable - even if I do still want to do some fine tuning.

I lifted the analog output board away from the front of the panel another 11mm. And beefed up the mounting legs so it doesn’t just snap them off (I hope…they may need some additional reinforcement since this is a pretty heavy board with the 16 trim pots on it!) So the jacks now clear…just barely but they clear.

And the RS232 fits now - though I did forget to shift the receiver board mounts forward a few mm so it’s slightly recessed instead of flush:

It’s minor…but I REALLY don’t like the way the board is visible at the bottom. So…if I can get a cable on here tomorrow (Need to dig one out…REALLY hope I’m right that I still have a few stashed away!) then I may just add a bit more plastic to the design to cover up the board instead of shifting this to be flush like I was planning.

A little peek in the back with both boards installed:

Ok…so how about the buttons. What I did was before printing the panel I tested my changes by using a drawing from Fusion 360. I exported the front face of the panel as a drawing and printed it out - then folded it halfway across the button holes and held it up to them to see how well they lined up:

Much quicker than printing a new panel to confirm The photo is misleading since I couldn’t hold everything still wtih one hand and snap the photo with the other. The buttons did appear to line up just about perfectly.

So - I modeled up the perfboard in Fusion and added holes that lined up with the mounting bosses I had modeled between switches. I then exported those as drawings and cut them out so I could glue them to the board as a template:

I held it up to a light to make sure the holes were aligned down the center of a row of holes since the buttons span one hole - this way it should all line up:

Next I cut it out. I really like to use my bandsaw with a thin blade for cutting boards…but…it snapped on me the other day and I haven’t dug out my jig, flux, and silver solder to repair it yet. Honestly haven’t had to do that in so long I’m really not sure where the jig is at this point. And I’m TERRIBLE at cutting boards with a utility knife. Or at least I was until I came up with this trick a few days ago. I use another board as a guide and hold it in place with a few pin headers just pushed into the holes as guide pins:

Probably hard to tell from the photo…but those two strips of pin headers aren’t soldered. They’re just there to keep the two boards aligned while I make some cuts along the edge to cut the lower board.

Then a could quick holes with my drill a few 3mm screws and I was able to solder the buttons in:

The two strips of buttons are reversed. So this has to be rotated 180 to fit on the other side. Oddly enough it fits better there than on this side. Which I guess makes sense looking at my patterns again since I used the pattern for the other side. They should be the same but I guess they’re very slightly different. It lines up usably on this side…but not perfectly:

Kind of hard to tell I know…but the buttons are slightly tilted. From the front it’s even harder to tell:

And they all work. Though two of them rub ever so slightly. And I do agree with the guy who’s design I copied that these buttons really could benefit from a bit more range of motion. It’s just a bit hard to tell which position they’re in. I suppose I could add some LED’s and transistors to indicate bi-polar/normal…but that seems like more work than it’s worth.

Overall I’m really happy with how this is coming out.

Tomorrow I need to make up the 2nd button board, wire up the jacks and the buttons (no one has chimed in on if I got the wiring right in my previous post…I’ll test with the continuity meter but sure would appreciate confirmation that I’m not way off base!) connect the two boards…make up a power cable. And make up some more 10 pin cables (the interconnect between the transmitter board and sequencer uses the same 10 pin cables eurorack does…and so will the connection to the DIN board that will be on the next panel I make up with the gate outputs.)

I also need to add labels on the jacks/buttons. I’m really torn here. My plan to to number them:

1 2
3 4
5 6
7 8

But…I’m thinking I may change it to be:
1 5
2 6
3 7
4 8

Which…doesn’t feel as “right” to me…but would make the wiring much cleaner and more logical since each button board could be on it’s own connector then. If it keep it staggered then each button will have to be on it’s own connector which will use up a lot more connectors. Actually…it would have to be 2 connectors for each button because I have lots of 2 pin connectors but no 2x2 connectors. So I either have to do 2 connectors for each button. Or 2 connectors with wires from both boards if I keep the staggered numbering plan. Or if I switch to the vertical numbering I can do each button board on it’s own connector which seems “right”. It will also let me put each row of jacks on their own connector as well - again making for cleaner wiring.

So…lots of cables/wiring to do tomorrow…but if I do that then assuming I do still have a cable…I should finally have my sequencer connected! No gates just yet…but…at least I’ll be able to get CV’s out of it which will be a huge jump forward.

[ChristianBloch]

Next level techniques

[juggle]

Well, progress. But…not CV outs yet

And I’m not quite sure why.

First stumbling block…I got one end of one of the cables between the sequencer and the line driver module backwards. So…there’s a risk I may have fried something (like the line driver IC’s)

I fixed that…though still unsure if the IC’s are any good now or not but still wasn’t getting the LED on the analog output board to light up when I connected the RS-232 cable between the two line drivers. Well, found 4 cables when I dug one out…so let’s try one of the others.

2 of them are out of the running. They’re the M-F and I need M-M for this. Oh well. The one I was using was a parallel cable from the old laplink software package from…oh…about 30 years ago. It’s about 2m long and nice and flexible…would have been perfect. But…apparently it’s not a straight through cable. Which left me with one last cable.\

The worst of the 4. From an old parallel port zip drive. About 1.5’ long and very stiff. I remember I hated it back when it was used for an actual zip drive because the cable was so stiff and the drive so light that the drive would never sit right. Oh well…it’s better than nothing.

And…the light came on! Alright…we’re getting 5v from the sequencer to the analog out board across the line drivers. All is not completely lost!

Ok…so now I need to wire up some of the jacks and wire up the +/-12v to the analog board.

I just wired up the first 4 jacks, and just used jumpers to set them into normal mode for now so I don’t have to deal with the buttons tonight.

The power wire was kind of annoying…I swapped out the pin headers for a JST-XH so it would be keyed and I’d be less likely to get things reversed. But I had to make a cable that’s 10 pin modular power on one end and 3 pin JST-XT on the other which was a bit tedious. But…it tested out ok so should be good to go there.

Let’s go into the menus on the sequencer and get the CV’s into test mode so I can adjust the trim pots. And…I can barely see the screens. That’s…not good:

Hmm, disconnect the cable between the line drivers…and the screens are back to normal. So…apparently the analog board is pulling too much current or my power supply just isn’t up to the task it’s being asked to perform.

But it’s almost 11PM and I’ve got a lot to do tomorrow so I need to call it a night. Tossed in a quick order for new line drivers IC’s so they’ll be here by Friday in case I did kill them (spent more on shipping as I did for the chips since they’re not that expensive but shipping was $8 to have them here this week since I really want them by the weekend). Maybe tomorrow I’ll try a different power source for the sequencer or think of something else to do to debug…I could try just bypassing the line drivers…but the cable for that is really confusing to make so I was hoping to avoid it. And it seems to be more of a power issue anyway.

Oh well…no sequencer for me tonight

On the upside…I did get the 7 segment display and 3mm led’s for my oscillator today so that’s now complete:

Yeah…I had to go with blue…but it’s a bit brighter than I expected. May have to swap some resistors to tone it down a bit.

And part of why I didn’t get more done on the sequencer is I spent too much time trying to tune the oscillator. I just can’t get it to stay on the same note between octaves. Been fiddling with it two (or is it three) nights now and can almost get it close…but one octave is always at least a semitone off. Can’t seem to find any info about how to use the HF trimpot either so wonder if I’m missing something there. I dialed in the reference voltage…so I’ve just been messing with TRK and CNTR - mostly with TRK and when I get it close CNTR to pull it in. I’ve been using an external tuner and the built in is agreeing with it now that it’s working as well.

So, rather frustrating evening Hopefully tomorrow sees better progress.

[Caustic]

The blue looks good! Good progress!

[ChristianBloch]

The VCO tuning is indeed fiddly. You should make sure your power supply provides enough and stable juice first of all. Then turn HF track all the way down, then just keep finding the center frequency and adjusting tracking. You may also have to match your precision resistors better if you really can’t get it close enough. At the end, adjust HF tracking if your upper range isn’t quite getting there. Also, you may be trying to tune it an octave too high or too low.

[eric]

Have you tried this :

1222 Tuner VCO Module

I finally got around to powering up my #1222 VCOs, I was waiting for the mica caps to arrive from AliExpress but they’re still at least a couple weeks away, so I gave up. While poking around trying to figure out why I can’t seem to be able to calibrate the frequencies (more on that later), I noticed a resistor getting quite hot. It is R42, the resistors in series with the 4.1 voltage reference. [image] I did a bit of math : The current through R42 is I=V/R= (12-4.1)/220 = 35.9mA which seem…

This just makes a lot of sense and I will build mine with a R42 of 1K from the start.

[ChristianBloch]

Also make sure to read @antoine.pasde2’s follow up further down.

[juggle]

I did use the 1k for R42. And was fairly careful to match my 10k precision resistors…got them closer than Sam’s in the build video…there was only 0.01 difference between all of them according to my budget fluke. (It’s a fluke…but it’s one of the cheaper 115’ s for the overseas market you can get off Amazon.)

I had missed that other discussion about matching the 100k resistors I may have to look into that. And I keep meaning to measure the voltages off the switch but forgetting to.

Two other things I’m not clear on:

1. I keep seeing references to 5 octaves, and it gets wired to 5 pins on the switch…but Sam limits the switch to 4 positions and there are only 4 precision resistors. Should I have limited it to 5 positions instead of 4?

2. As per Sam’s video I’m first just trying to get all 4 octaves to tune to the same point. But does it matter what my fine tune is set to while doing that? And the tune knob below it seems to be more of a CV offset than an actual tune (I haven’t looked closely at the schematic yet.) as it only appears to have any effect when I have a CV source. I’ve been tuning with no CV source but maybe that’s the wrong approach?

Also - it seems the comment about being an octave too high or too low may be right - as my external tuner is going out of range on the two lowest octaves and not recognizing a note. Should I use the center trimmer to bring everything up an octave or two? This is with no CV source just the VCO running free.

Honestly I haven’t put much time/thought into the VCO yet since I’m still trying to get this CV interface from the sequencer working. Until then I’m not overly concerned since I don’t have a source that’s “in tune” So I haven’t done my own research to try and answer these questions yet like I usually would.

Thanks for the help and input everyone!

[fredrik]

I keep seeing references to 5 octaves, and it gets wired to 5 pins on the switch…but Sam limits the switch to 4 positions and there are only 4 precision resistors. Should I have limited it to 5 positions instead of 4?

ref is calibrated 4.0 V, so there are 5 positions (0.0, 1.0, 2.0, 3.0, and 4.0 V).

[heckseven]

That’s a really good question. I might try that later tonight just to see.

[juggle]

Did a tiny bit of debugging on the sequencer over lunch.

Remembered that I have a USB power meter so plugged it in between the power supply and the main board on the sequencer.

Looks like without the line driver modules it pulls 300ma. With the line drivers it draws 450ma. The power adapter is rated for 2a so should have no problem with that.

But…since I’m powering over USB I’m limited by the voltage regulator on the STM32 Discovery board…and I’m only seeing 4.5v on the 5v bus. And I know from past experiences with STM32 Disco boards that their on-board regulators and power circuitry are…questionable at best. I forget the details but I know a flight controller I had built out of one had all kinds of problems because of the way their power is setup…and the 3v output is unusably low (which is causing issues with a ethernet port I was trying to add to the sequencer.

I did try a different power adapter…but same result which didn’t surprise me. So…I’m probably going to switch to powering this thing off 5v directly instead off off the USB. I ordered a little meanwell 5v 3a power supply but will rig something up off my lab supply this evening to test with. I strongly suspect the issue is just that the built in regulator on the Disco can’t handle 45ma reliably.

Of course there’s also a possibility that I fried the DAC on the analog output board with that bad cable last night. Which would suck since that’s $20 from mouser or digikey (only $12 from Newark but I don’t order from them as often.)

[juggle]

So…bit of good news. Looks like the regulator on the discovery is a big part of the problem. Hooked it up to the lab supply instead and the backlight on the LCD’s is much more stable and when I hook up the line drivers it doesn’t dim out anymore. In fact it now pulls more current than it did before - so the regulator was the problem:

Or at least…it was certainly a problem.

Bad news is I’m still not getting any CV out

I twirled the trimmers all the way from one extreme to the other and only saw 0.002v of change or so.

I don’t think it’s the line drivers either. I found that there’s a debug routine built into the sequencer firmware that lets you individually trigger the lines between the receiver board and the analog output board. And using that I was able to confirm that signals are getting from the sequencer to the analog output board as expected. The voltages I was seeing were just a hair lower than I expected (i.e. 3.8v instead of 4v) but should be close enough. And do seem to indicate that at least the line drivers are working as expected as signals are getting to the analog out board.

I do see VERY slight voltage changes as I change the calibration settings on the sequencer. But no matter what I do I’m never seeing even 1v out.

I did confirm +/-12v on the proper pins on the analog board. So that’s correct.

There are only 3 IC’s on the analog output board. 2 TL074’s which I doubt could be the problem If my reversed cable yesterday fried something it would be something powered off 5v and the opamps only get +/-12v so it seems unlikely that sending 5v down the wrong pins would damage them. But…I’ve got plenty of TL074’s on hand so let’s try the old auto mechanic trick and just swap in a new part to see if it fixes things.

As usual…nope not a lick of difference.

Which leaves just the DAC as a likely problem The $12 chip that’s $20 (all before shipping) if I order it from anywhere that ships quickly. Oh…wait…it’s up to $16 now Quest components has them…but they’re closed on Oct 1st for a holiday? They’re in CA what the heck is the holiday tomorrow, apparently it’s the international day of older persons but who the heck closes for that? It is armed forces day in South Korea…so despite being headquarted in CA maybe they’re owned out of South Korea. Anyway they have a minimum order of $25…so…scratch that option.

So looks like $26 with shipping from Digikey for one chip…and may have it on Monday, probably not until at least Tuesday

It’s also the least fun to replace being SMD.

Since there’s no chance of having one here by the weekend (Well, it’s not impossible but I’m not paying another $23 on top of the $20 for the chip just to get it here quicker.) I think I’ll try and do some more debugging to 100% confirm I’ve killed it before ordering a replacement. Because I’m not convinced that hooking up a cable backwards on the line driver TX could have killed the DAC on the analog out board. For that to happen the 5v would have had to go through the TX driver, over the connecting cable, then out of the RX driver and to the wrong pin on the DAC. I haven’t looked closely at all the schematics…but…that seems highly unlikely to be possible.

So there’s probably something else I’m overlooking.

I did neglect to put jumpers on one set of normal/bi-polar headers the first time I powered it up…but that shoudln’t have killed anything. It would just leave some of the inputs on the op-amps floating. Not ideal…but shouldn’t have killed anything.

Oh well, need to get some sleep.

I did seem to make some progress with tuning the 1222 at least. I checked voltages at the switch and they’re all looking good. One was .001v higher than the others. So…doesn’t seem like my resistors are likely to be the issue. I cranked the CNTR up to raise things a few octaves. I also moved the ring on the switch to give me one more position so I have 5 positions I can select. Now I get readings on my external tuner of C4, C5, C6, C#7 (though C6 and C#7 cause the tuner on the 1222 to freak out) with each reading a bit higher than the previous. But my 5th position despite being in the highest position gives the lowest note. And it’s a B. according to the internal tuner. The external doesn’t register it (this external tuner isn’t great on low notes. A3 is the lowest it will read) Oddly enough the internal tuner also shows the C4 as a B.

I’m pretty sure I have the switch wired wrong. The labeling on the PCB kind of confuses me:

The silk is backwards…but I read it as center pin on the left. Then going right from there Pin 4,3,2,1, and I assume 5 on the last GND pin.

I have the stopper on the switch here:

Which gives me 4 “clicks” for 5 positions

1 CLICK 2 CLICK 3 CLICK 4 CLICK 5

But with how I have it wired that last “5” is the lowest tone followed by 1 then 2,3,4.

So maybe it should be 5,4,3,2,1 despite the label?

I need to stop dealing with this at the end of the day when my brain is fried. I can build boards just fine in this mental state…it helps me relax. But debugging things…nope…that’s best deal with on weekends when I’m not brain fried from debugging code all day

[juggle]

Ah…instead of going to bed I looked at the 1222 again.

And figured out the switch. Somehow it never hit me that those were V’s on the silk. They look the same forwards and back but still I missed them. Then it made sense. Shifted my wires around and now I get all my octaves in the correct order. Horray!

And I think it may be time to replace my tuner. I think I’ve had this thing since high school (which puts it frighteningly close to 30 years old). And I’ve never been very happy with it (a cheap Seiko SAT500…looks like they still make them!) haven’t used it in years since I mostly just use my phone now. Then again…maybe it just doesn’t like the synth level compared to a guitar level.

Apparently it just doesn’t like the square wave output. Switched to the sawtooth and now it reads lower registers better. Shifted CNTR so I get C1-C5 which seems a bit better. But…still not getting it to stay centered between octaves. But…I think it’s just a matter of some time spent fiddling now and I’m too tired to keep at it tonight. Will try tomorrow and suspect I’ll finally get it - or at least get it closer.

[juggle]

Woke up this morning with another idea to test the line drivers. I can hook up a digital out board for gates. That should let me double confirm that the line drivers are working. I mean…I’m already pretty sure they’re working since I can see the lines changes going to the aout. But being able to get gates/triggers working would help confirm that the problem is with the aout and not the line drivers…

[juggle]

Well, the only built up digital output boards I have are in my SID synth and I soldered to the pins on them so they’d be a pain to wire into anything else. But…I have a few that I ordered back in 2005 and never built up since I never finished the control surface for the SID or for the FM synth I started either. They’re old R2 boards (they’re now up to R6) and lack decoupling caps and use single row pin headers instead of 10 pin shrouded headers like the current R6 boards…but…they’ll still work. So built one up:

Too late for me to figure out the wiring and software to enable the gate outputs tonight though. Will deal with that tomorrow or this weekend.

Did some more debugging on the analog out…and it’s looking pretty much like the DAC must be the issue I confirmed that voltages are changing right at the DAC pins when I activate the test mode so signals are getting from the main board out to the analog board…just the DAC isn’t doing anything.

I posted over on the midibox forum to see if anyone who’s more familiar with it has any ideas: http://midibox.org/forums/topic/21500-aout_ng-on-seqv4-not-getting-any-output/

And the cute little 5v power supply I ordered showed up:

I haven’t wired it up yet. I’m still debating what to do for a case on the sequencer…right now it’s just a chunk of 1/4" MDF I cut on my CNC that the control surface mounts to and the main board, midi board and line driver board just kind of dangle. I’m debating if I want to make an actual case or just mount them to the back of the board and keep it a bit more exposed.

I did print some legs that snap into the panel and hold it up:

I like the simplicity of them…and being able to show off the “guts”

But…if I add this power supply and have 110v running to it…well…that requires a bit more safety.

I have a fused/switched IEC320 jack: https://www.amazon.com/gp/product/B00511QVVK?psc=1&redirect=true&ref_=oh_aui_detailpage_o01_s00&tag=tv-auto-20 I used one on my first 3D printer and have a spare left over. I was actually thinking about using it on my modular but just got a wall wart supply instead of a transformer for now. But…I may print up a little box that holds the power supply and this little guy and can mount to the back of the sequencer…just…the sequencer is so light I’m not sure I really want to run a fully 110v power cable to it.

So maybe I’ll make a little box for this and just run 5v over to the sequencer. Meh…I’ll sleep on it and decide tomorrow or this weekend. It’s definitely running better off straight 5v than it did off USB though so worth the effort to do it right.

Oh - and did a bit more tuning on the 1222 and getting it closer and closer. All 5 octaves now stay on the same note and the shift between them is getting smaller and smaller. I just don’t have the patience to sit there and fiddle until it’s right in one sitting But I am getting closer!