AndrewMartens

Heh, I know how that goes! "That project sounds cool, when is it going to be ready?" ... "It depends, I could have it working in a few weeks, but I would lose my job, the house would look like crap and my wife would divorce me."

Really? Just make up a company name and use your home address, that should work. Something like "HL-SDK Enterprises".

For testing, I powered both my Core and SID modules from an unregulated "12V" DC power supply - in other words, 13.8V (which allows for the voltage drop before a 12V regulator). I could have done a dual-regulator (13.8 -> 9 -> 5) setup, but instead I ran the 7805 directrly from the DC supply voltage. Even with the whole unit (Core, SID and LCD) only drawing about 200 mA, the 7805 gets rather warm. I had a spare TO-220 heatsink around, so I threw it on there to help dissipate some of the heat. But supplying any more than 15V would be a real push...

Wow, toroidal transformer and everything. Very nice! I'm probably going to make do with a crusty old 24V transformer I looted from a spare parts / junk box at my old company...

It should be possible to easily get the necessary components for the older boards. Both of my SID boards were of the old revision (I just finished building one) and the parts should be obtainable from your favourite parts shop - Digikey, Mouser, Jameco, local electronics shops, etc.

Excellent designs! :D I look forward to seeing the finished product. Don't forget that eight SID boards will take up some space (depends on how deep a rack case you're going with, but it shouldn't be a big issue). I had considered doing either two FMs or two SIDs in a 1U rack, but just couldn't be bothered to do a nice design for it. Now I'll probably just keep my spare SIDs, OPL3s and PCBs around as a backup in case something happens to my working set.

Well, it's finally finished! I'm just finishing up my documentation for the Wiki, but in the meantime I'll post up some final construction pictures. I'm not the type to spend a lot on an enclosure or a panel, so I grabbed a sharpie and went to town labelling it by hand. The white coating on the metal is just plain old Tremclad spraypaint, which worked quite well. As I mentioned above, I cut the hole for the LCD way too wide (wide enough to fit the LCD through, actually) - so rather than having to order a bezel that may or may not be the right size, I used that trusty sharpie again on a piece of hard clear plastic, and drew in a crude window. Everything else on there should be pretty straightforward. The knob on the joystick is actually borrowed from my Red Sound Dark Star - I haven't opened the case on that one, but it looks to be exactly the same joystick. Here's a few more "making of" pictures: Drilling/cutting template Joystick wiring Putting it all together Quality workmanship? No, not really. Front panel view Left angle product shot Right glamour shot The ports around back The MIDIbox SID makes some friends

For the 2SC945, try this shop: http://www.mainelectronics.com/bulksemi.htm It's local to me, and I've bought stuff from them in the past. Not sure if they have hundreds, or if $0.17 (CDN) each is a good price or not. Worth a call, though. And if they're crap, I can go by and rough 'em up a little (no, not really, but I will go in and complain if you have problems). As much as I don't want to reveal my sources, they even have CA3080Es (DIY synth freaks take note), though they're a bit pricey at $1.50 each. Still...

Very interesting. It sounds like his approach would work very well for single notes; it's essentially a threshold detector that then does a period -> frequency -> MIDI note conversion. Trying to process all six strings from a guitar would be a different matter, though (unless you wired a custom pickup so that the element for each string went to a different output)...

... and you'll be met with blank stares and confusion ;D

I considered the cost of designing kits for some other projects as well - and it's hard to justify. Generally it makes far more sense to just put in some overtime at work than to throw together a bunch of kits; not like it's intended as a money-making operation, but spending hours every week packaging up kits to ship out isn't exactly a great use of free time. PCB + hard-to-find components is one thing, but complete component kits is a different beast altogether. And what if your sales estimates are off, or people defraud you, etc.? Then you either get stuck with a bunch of extra boards/components, or lose money, or both...

Quick answer: probably not. While the PIC may be powerful enough (if you're lucky, but I doubt it would be to do this with a quick response rate in realtime), unless you've studied signal processing and assembly language programming, this is probably too ambitious. Longer answer: However, I seem to recall someone once making a PIC frequency analyzer for a contest a few years back. The PIC hardware has improved since this point, and if you were to selectively filter around the guitar string frequencies, it may be possible. However, you'll probably have to code this in assembly for speed, and have enough signal processing knowledge to figure out how to do the analysis. That part could be done on a PIC, given enough time and knowlege. The real challenge would be doing this fast enough to trigger MIDI note events, and that is where I'm 99 44/100% sure the PIC will not be fast enough. You'd probably want something more along the lines of a 56303 DSP. Check this out for a start, though...: http://www.circuitcellar.com/library/print/0998/Lacoste98/index.htm

Just rereading your original post here, doing an assignable joystick would be a lot more effort on the software-support side. You can also probably find a cheaper joystick at surplus somewhere else (I think I paid about $1 for mine). That said, the previously-mentioned links worked fine for doing a fixed cutoff/resonance setup.

If it's the socket itself that's broken, just use some sharp pointed cutters to hack the legs off just above the PCB (on the socket side), then desolder them one-by-one.

I heard that Futurlec takes about 2 weeks before they ship... I'm unfortunately at 20 days already, and it's still sitting at "Order Entered". ARGH. It was either $80 at Futurlec, or $220 at Digikey, foolish me thought that waiting a couple of weeks wouldn't be so bad.

In case you didn't find them, these two threads were all I needed to get my joystick working: http://www.midibox.org/forum/index.php?topic=5708.0 http://www.midibox.org/forum/index.php?topic=3375.0

Sorry, I've ended up shelving these projects for now. Probably for good. It was a tough decision, but there just isn't enough time in the day for work, renovations, running the house, and relaxing to fit many projects in there. I'm working on a few different projects these days (as well as a MIDIbox FM) with better musical capabilities than these other chips. That's not to say that the AY-3-891x, YM2149, and POKEY aren't great chips in their own right - it's just that the SID more than satisfies my lo-fi needs.

Heh... mine is technically "PCB mount", except that it's a series of small PCBs that are then each mounted to the front. Including the LCD and joystick, I had 18 screws holding everything in place - not too bad, except for some drilling issues on the buttons. Ideally I would do a custom DIN + encoders + buttons all on one PCB, but that's going to have to wait for my "Green Bastard" synth coming late next year.

Well, it's been a month, but I finally have had time to get back to the SID again. After reading up on various types of capacitors, I tried swapping in some 470 pF ceramic caps (good 5% COG/NP0 ones) for the polyester film caps that I had put in there. The sound was a bit smoother, but the filtering was very heavy, and at the higher levels of cutoff only just started to open up a bit. Adding any resonance just brought in a load of distortion. Based on this thread - http://www.midibox.org/forum/index.php?topic=3589.0 - it sounds like I need some smaller caps. I'll have to wait until my order arrives from Futurlec, because it has some variety packs of ceramic and mylar caps in it. On the plus side, I managed to get my front panel painted, labelled, and mounted all of the parts tonight. Fixed the improper drilling of the joystick screws, had to fix a few other drill spacing issues with some washers - the PCB holding the four buttons under the display is actually mounted on a slight angle so that the rightmost button doesn't bind on the edge of its hole. I also managed to fix my "cut the LCD hole way too large for some unknown reason" issue by drawing on a piece of clear plastic with a sharpie, and mounting it in front of the LCD. Pictures will have to wait for a day or two until I get a chance to buy some white model paint and go over all the shiny screw heads that are messing up the look of the panel. Then I'll tweak the customizations a bit more, and then post them to the wiki... same goes with some before/after samples of that darned filter. Getting close now! :D

Clock dividers would probably work well if you wanted a set-and-forget solution - even a 4-bit counter would work well enough (div by 16) though you could go as high as a 12-bit one (ie 74HC4040) if you really wanted to. Alternately you could just set a high-priority interrupt on the PIC at the appropriate rate, and just toggle an output pin. There might be a slight amount of jitter with that approach, though it wouldn't require any extra ICs.

Too busy... building... no time to make patches right now! ;D

That's my general understanding of the concept, yes. Depending on the sample rate you may be able to get away with just using a generic opamp instead of the LM398 (which I believe can hold a voltage for quite a long time). Also, you may not need the buffer after each S&H amp - but I haven't checked the LM398 datasheet to verify that. Looking at the Matrix1000 schematics now, it appears that if you have a fast enough sample rate, just a simple R/C network may be enough to maintain the control voltage for the CEM: http://wolzow.mindworks.ee/analog/files/OberheimMatrix1000Schematics.pdf Check out page 4, they have the 4051 outputs going to a cluster of R/C networks, each of which is connected directly to the CEM. UPDATE: Heck, going to the CEM3396 datasheet - which I should have done in the first place - they have the output of the 4051 into a single cap, and then into the CEM. Even less components! Hmm, those CEMs are looking like a lot of fun. Maybe I'll have to buy one somewhere so that it can sit in my parts cabinet for years until my other MIDIbox projects are done... ;D [note to self: this bodes extremely well for my unrelated project which is also planning on using a fast-sampled S&H after the output mux]

Heh, I know how THAT goes! I made a nice cardboard template, drilled/dremelled out the holes, and I still got stuff wrong. For some reason I cut out the LCD to the size of the metal edging, instead of just the display portion (now I can see inside my MB SID, greaaaaat), and completely messed up the positioning of the corner screwholes for the joystick. At least the pushbuttons were okay... once I used the stepper bit to make them a bit larger so that they wouldn't rub the edges of the holes due to my imprecise "measuring". ;)

Well done getting everything back online!

Just downloading it for a listen now. Sorry I missed your post about your live PA at Shine. I haven't been in a clubbing mood much the past couple of years, but if I could combine electronic geekery with it, then that would probably give me a good excuse to take $WIFE out dancing for the night. Make sure to post again the next time you're playing somewhere - I'll try to keep an eye out for it. And now, just another few minutes and I'll be listening to your set... [update] I really liked the first 6+ minutes. Around the 8 minute mark when it gets really percussion-heavy, that's not really my thing. 12~13 minutes I got back into it again. It may not be album-quality-polished, but that's not really necessary for a live PA. Bet it sounded really good live. Nice work!