Category making things

MIDIbox SID-LC v1.0-PROTO PCB Layout Draft 1
(Click for 300dpi version…)

Wow, I didn’t think I would get this finished tonight, but… Well… I have. This is the first draft of the PCB layout for the MIDIbox SID-LC v1.0-PROTO. That is, the first draft of the design for the prototype for the MIDIbox SID kit I’m hoping to produce and sell.

While it is true that I’ve already built a working SID-based synthesizer (photo gallery retired), I want something a bit more. So, I’ve come up with this… It’s a bunch of the MIDIbox modules all squished on to one board. Basically, this is everything one needs to build a MIDIbox SID Step A (Basic Control Surface) on one easy to assemble board.

I’m feeling pretty tired now, so I think I’ll head to bed. Well, as soon as I make a backup of all that work.

Suffice to say, I’m pretty happy that things have come this far… Now I just need to order a complete set of parts for it, confirm that they all fit the on-paper layout, then order a set of prototype boards.

If that all goes well, then I’ll start seriously looking into what it will take to assemble 50 or 100 kits and sell them.

Well, the SIDbox / whatever MIDI problems are sorted. I’m also able to load apps and things like that… The problem now is that it’s not making any sound. I’m not sure why yet, but that’s okay. At least one half of the device appears to be functioning as designed.

Well, power works.
(Click for more SIDbox photos…)

Well, the SIDbox PCBs came in today, and thusly, I assembled one. Well, as much of one as possible… I’m still waiting for some more sockets to arrive tomorrow, and then I need to finish building a PIC programmer. But after that I will hopefully have a working prototype of a single board SID-based sound module.

For background, I’m using what are basically a bunch of different designs from the MIDIbox projects, implemented by me on a single PCB. It will also run the MIOS software which Thorsten Klose has so kindly shared with the world. The main project has a bunch of small ‘modules’ which can be wired together and fitted together in a case to make a synth of one type or another. Well, I wanted a specific type, build portions of a number of the different modules, so I made just that.

I’ve actually been thinking seriously about selling these as kits, and after asking Thorsten (TK) about it in the MIDIbox Forum, I’m seriously thinking about doing it. If I do this, I’ll be merging 5 or 6 of the MIDIbox modules (I can’t remember right now — but it’s a good bit more hardware than seen above) on a single PCB and sticking it in a single case. With my design it’ll be possible (just with the main PCB) to have 8 buttons, an LCD, and all the stuffs I mentioned before. I’m still not sure if I want to do this work, but… I may… We’ll see. :)

Anyway, this post was supposed to be about SIDbox (photo gallery retired) photos, so I’ll link to some specific ones here:

· Mmm. SIDs. A MOS 6581, MOS 6581R3, and CSG 8580R5.
· The nifty note sent by along with some ridiculously generous SIDs.
· 600dpi scans of the finished PCBs. (photo gallery retired)
· Ahh, a nicely laid out array of six SIDbox PCBs.
· Ooh, and the SIDbox PCBs even fit in the enclosure. They fit a little too well, but that’s all right.
· Test fitting parts to be sure there won’t be any dimensional surprises when actually soldering. Everything looks good from here…
· Well, the power supply works. I was getting a solid 5VDC and 12VDC, which should be ideal for a MOS 6580 SID.
· The PCB with all parts fitted except for two sockets and all ICs.
· The bottom side of the SIDbox PCB, where one can see a hardware bugfix. After I ordered the boards I found out about a potential problem, and the blue wire should resolve that.
· The SIDbox PCB fitted with a PIC18F4620 to test on-board ICSP. It works.
· A photo which may look familiar to those of you who liked the laser printed board layout plus parts photo from before…

So, yeah. If I’m lucky I will have a working SID-based sound module either tomorrow or Saturday. If not, I’ll probably have a lot to learn… And I’ll be really upset with myself… We’ll see. For now, I’m pretty excited.

The SIDbox PCBs are here! Granted, these are considered prototype boards for the MIDIbox SID-LC I’ve been working on. Or… Maybe I’ll stop here. We’ll see later. :)

Tonight I made a parallel port PIC programmer from the MIDIbox PIC Burner schematic. Except I left off the power supply (I’ll use my benchtop one (photo gallery retired)) and added a DIP18 socket for the PIC16F88.

I’m able to switch V_PP, V_DD, Data, and Clock on and off with the proper voltages coming out of them. Unfortunately the software isn’t recognizing the programmer nor is it able to identify which PIC has been inserted in it. Ah well, I think it’s a matter of parallel port settings, which I’ll mess with another night.

Oh, and I had to etch two PCBs for it, because I screwed up the first one. I also learned what was most wrong with the PCB for the Casio SK-1 MIDI Modification which I screwed up. I think I didn’t let the photoresist ‘develop’ for long enough. I’ve found that with a fairly warm lye solution (the bottle had been soaking in hot tap water for half an hour) it takes about 1:30. With a room temp / cold solution after 12 minutes things still aren’t right. And when a contrast-y image appears, there is still some photoresist left in some points, which makes for an unclean etch.

At least now I know. Now I can go to bed and stuff. Oh, but only after I wash the lead off of my hands.

Well, it looks like I’ve identified a problem with the SIDbox layout. Per this thread I shouldn’t be using a stand-alone oscillator for the SID. Whoops. That means any finally assembled boards will have a jumper running from where the oscillator goes over to pin 18 of the PIC. Not a big deal. Won’t break anything, it’ll just work around a potential problem.

Second, I asked the guy who writes and has designed the core of the MIDIbox stuff if he’d be all right with my selling these boards as kits. He had some suggestions, but he sounds pretty okay with it. I’m just not sure I wish to make the investment. The thread about that can be found here if you are interested.

So, I’m laying the board out again, and I’m going to be getting six prototype boards which should work well, but aren’t usable as kit boards.

I guess the question is whether or not I should attempt to put this together as a kit. I’m thinking an initial (or single?) run of 50 would work well… I just wonder if I can afford it. And I still can’t answer the question of why I want to do it. It just seems like something fun to do.