Any love for Synthesizers?

There is a lot of discussion going on here, I like that!

I do prefer doing it the old way. It’s like sort of part of the thing. Like recreating an ancient artifact…
Maybe we can go down both roads at once without to much effort?
The other thing is that by using SMD the boards wont be smaller as all the connectors/plugs/potis… need to line up with the front panel.

Here is an idea:

Everybody interested meets at 7pm on Tuesday the 31st and we have a chat and make a plan on how to progress further?

-Sven

I think a meeting is a good idea.

However why do we need to put the PCB all the way along behind the front panel? If you look at lots of the old AMPs ect the switches and potentiometers used to control the device are psychically mounted to the front panel and cabled back to the PCB. As far as I know the only reason its not common to do now is simply to lower costs for mass production, mounting to the front panel requred another assemble process step and a few more parts (wire, nuts ect). We probably won’t benefit from avoiding the need to hand tighten a few things given the hobbyist and one off nature of such a build.

1 Like

I’ve created a Trello Board for this project, will upload files etc. as I can. I’ve added whomever I can find on this thread who is on Trello, apologies if I didn’t add, that can be fixed. Sven, I’ve added you as an admin on the board so you can make whatever changes etc. you like.

I’m half way through converting the first PCB to a dxf, and am getting a bit faster at it, no promises this weekend though, I’ve got work all weekend, along with Easter stuff.

1 Like

I see that HSBNE has a Trello Organization, so if anyone wanted to move it over there that’s cool, didn’t realize HSBNE was into Trello when I first started messing around with this project as a list/board in my private Trello set. Now I’ve made it public, since it’s a HSBNE project, it’d make more sense for it to be over there.

We use our trello for HSBNE stuff rather than member projects, and we’re in the process of cleaning it out and assessing it currently, so we probably wont move it across.

If theres demand for a trello of members projects we could make a HSBNE Projects org or something.

Hmmm, that is an interesting problem. The synth will live at HSBNE and be owned by HSBNE.
The idea is to pull more artist minded people in. Therefore one could argument that it is a space project as it improves the space.
On the other hand, I do it because I always wanted to build one of these…
Could’nt we just make a new board under HSBNE called collaboration projects?

2 Likes

Hmmm, in doing some research on the schematics to try figure out some of the parts used (I’m building the Schematic in Eagle and using that to generate the PCB, fingers crossed, but I’m making slow but good progress, and it’s helping me understand the schematics better) I came across this guy who made a Formant Synth, and he lists some of the hiccups he came across, some of the parts he had problems sourcing etc. Anyway, just thought I’d share it so we can try pre-empt some difficulties.

1 Like

I just drooled a little there…
Look at all dem knobs and dem cablez and … oh god!

See its funny, I agree with you on all the knobs and indicator lights, but I looked at the same cables and immediately worried about how you’d prevent electrical noise pickup as 1/4" TS cables are neither twisted nor shielded… and while some audio quirks are fine, ain’t nobody likes a 50hz hum from main power.

Are you absolutely certain that synth patch cables aren’t shielded, or are you extrapolating from a random cable you’ve encountered?

Also

From what I’ve been told, unwanted interference and large PCB’s is the goal.

For people that will TL:DR what I’m about to say, skip to the last paragraph and just take my word that there is sustained reason for what I say there. Otherwise I will start my more in depth explanation of why with the assumptions, and reason for those assumptions, that where made to reach this conclusion.

First off I am basing this on the sets of standards for audio gear, so I its not applicable if they are using custom parts that do not align with these standards in any way, but seeing the simplistic nature of the plugs\cables anyway and the time\expense to do so I do not feel this is unreasonable to rule out.

I also assume they are not using the TinyTelephone standard (bantam plugs), which is possible, but I deem unlikely as this was a comparatively modern standard developed for patch panels and portable gear without the space for XLR, rather than the sorts of bulky gear being shown. They also use a 4.4mm connector, typically with minimally higher width of the entire plug, so normally look different than the cables in the video, however without seeing actual pins of the connector I can’t tell. Seeing as a standard for patch panels seems fiiting to the role I will also note that bantam plugs tended to be fairly poorly regarded by professionals of the time due to their limited power carrying ability, lower rated insertion cycles and decreased connection reliability vs the bigger 6.3mm plugs that where basically the only option in patch setups before the TinyTelephone standard was created.

Finally I assume there is no separate sub category for synth cables that I have not heard of as I cannot see the need for one, and after a quick poke around the internet I can find no mention of such a thing. But with this list of assumptions noted,then I will proceed by stating that they are using standard 1/4" TS cables.

These cables where originally designed for telephone exchanges, and are one of the oldest sorts of connector still in use today. They operate on a very simple naming system, firstly the size of 6.35 mm 3.5mm or 2.5mm which is the diameter of the plug, followed by a code that describes the conductors (T=tip, R=ring, S=sleeve).

From the video they appear to by using the screw together type of older style plugs, which where only common in TS, although there where definitely molded version made in TRS that came about slightly later as stereo headphone became a thing. TS cables by definition carry an unbalanced mono signal, and doesn’t provide a contact to allow grounding of external shielding. TRS cables generally carry unbalanced stereo, but some mix desks have used them in “balanced” (actully just differential signalling rather than truly balanced) mode as a slightly smaller alternative to XLR buy using both Tip and Ring to carry the monaural signal, one the inverse of the other, while passing ground down the Sleeve. However this never became common as the Sleeve contacts last so it generally creates a large amount of noise on insertion, and if a TS cable was used instead of a TRS would either bridge the Regular and inverse signal amplifiers output, which early solid state gear didn’t like, especially with line level and other “loud” outputs. Worse, if the other end of the TRS cable carrying a balanced signal was put into a device with a TS socket then the Ring that carried the inverse signal would be left powered but unconnected, which while tolerated just fine by modern amps it typically destroys valve based gear. XLR cables never had these sorts of problems and could also provide cable locking, so even with the slight size disadvantage they continue to dominate the analog side of the pro audio environment even to this day.

So am I certain its Unshielded? No. But on the other hand not only is it highly likely, it would create just as many problems if it is in fact operating on one of the less common setups that actually allowed shielded balanced lines.

Okay, you seem to be extrapolating from old telephone patch panels which used similar connectors and also pro audio gear to flat out guess how synthesizer patch cables are constructed, instead of actually looking up how synthesizer patch cables are constructed.

First, the type of the connector has zero correlation with what wire is used with it. Basic shielded cable is so common these days that you would almost certainly have to explicitly specify that you wanted it unshielded to actually get one. Second, shielding does not require a third conductor. Two conductor patch cables use the ground wire as a shield. Nobody uses TRS connectors for this application. Finally, much synthesizer patch wiring does NOT carry audio data, but control voltages. I even found a reference that warns against using pro audio cable for patch wiring because it causes issues when carrying control voltages.

All of these things are irrelevant, anyway. Companies that sell actual synthesizer patch cables explicitly mention that they are shielded, because it’s a marketing point. Two seconds with google was all I needed to find multiple references to the fact that proper synth patch cables are shielded.

TLDR: don’t just argue from first principles when basic research will get you the right answer.

In that case there is a separate set of standards for patch cables? Fine, it was listed as an assumption. Assuming that everything you say is right however, that’d give us no ability to patch signals that are “balanced” (again here I’m also counting systems that output a 180 degree rotated differential signal, even though these are not technically considered balanced) via standard patched connections. I realize that this probably isn’t the end of the world for most applications, but given the size of these things anyway is there actually an advantage in using all 1/4" TS connectors rather than XLR for Audio?

Also BTW to prove I have not gone completely mad yet, many syths still use banana cables, of which i have never seen a shielded example nor have I been able to find one searching online, and every main sore of patch cables seems to sell both shielded and unshielded as options (I can only guess that that’s for cable capacitance or flexibility reasons).

The connections between synth modules aren’t balanced. These are patch cables, designed for building up a synthesizer configuration, not for connecting the synth to other audio gear. You just convert the synth output to balanced form when you want to do that, and then you use a standard XLR cable.

As for synths that use banana cables, there’s this:

“But modular synths which use banana plugs only use one plug and jack, and the patch cord only has one conductor; they rely on the power supply providing internal interconnections between modules for the signal return path. This causes a problem when interconnecting two modules that are in different cases, or when it is necessary to connect to some equipment that uses a different connector standard. Modular synth manufacturers who make equipment that uses banana jacks usually offer an interconnection module that makes it possible to establish a ground connection between cases, so that a signal return path exists.”

Just because other people are videos of posting expensive, large format modular synth gear, meet the SX-150 Mark 2, a ridiculously cheap, tiny prebuilt unit that’s available on amazon. I’m not even suggesting that this is anything like a replacment for what we are hoping to build, but its a true analog unit and worth less than $30aud delivered, so it has a lot of potential for hacking as an input for projects, can be used as a tone generator or as a way of getting a user friendly physical interface for your DAW software, which will in turn provide all the modifier options like EQ or compressors that such a basic unit lacks (just make sure you disable the speaker so it doesn’t drive you mad.

http://incompetech.com/wordpress/2015/04/stuff-review-sx-150/?utm_source=feedburner&utm_medium=email&utm_campaign=Feed%3A+incompetech%2FajSa+(incompetech)

So I was making good headway with the board, but I’ve hit a snag with the schematic and the PCB Traces shown (For the Keyboard Interface unit, page 18).

The Pot [P7], shows a connection between terminal 2,3, #3 on IC6, C6 cap, and the Anode of D2, with terminal 1 being connected to the Cathode of D2, and #6 on the IC5.

The PCB however, shows a connection between 1,2 (or 2,3), and the Cathode of D2, with the Anode being connected to Terminal 1 or 3. The IC’s don’t seem to connect up correctly as well.

This may well be a simple case of I don’t know what the hell I’m doing, with a simple solution. If someone could be kind enough to have a look and see if I’m crazy or not I’d appreciate it, otherwise I might just bring my laptop along on Tuesday night and nut it out then.

1 Like

After a lot of swearing, I’ve finally gotten something out of Eagle (Besides the decision to learn Altium, because Eagle is a piece of crap). I’ve posted a .dxf and a .png of the trace for the Keyboard Interface Circuit onto the Trello Board, so you can mess around with that, if it’s not quite right, you can always shift things around with the .dxf and make another image to print, or you can generate your CAM files if you’re happy. All the holes should be good, since it was generate with a PCB design software, but just be careful of the Transistors and Pots, as it’ll depend on what you get as to whether the holes will work, you might need to tweak that.

It doesn’t line up perfectly with the original image, but at this point it’s taken this long to get this far, so I’m just putting it out there now, rather then getting hung up on making it perfect. Hopefully it’s of use.

As a side note, are you guys thinking of making your own keyboard from scratch, or just butchering up an existing keyboard with your own circuitry?