Sauraen

I tried the baking technique and it worked great. The board is not exactly flat, but it's much better and definitely good enough. The board material was FR-4; I had it custom-made, and it was about the same quality as the boards available in the MIDIbox Shop, though maybe slightly inferior.

I tried a hair dryer; it did absolutely nothing. I don't think it was hot enough. I'll keep looking for options.

I don't know, maybe because they wanted us to have to make two and then sell you one? :tongue: Thanks! Once we finish soldering the SID boards, we'll put up a video of the eight SIDs. Hopefully that will happen in a couple weeks. Until the control surface is finished, though, we won't be able to do anything really cool with them all--any suggestions for what I should make the test program in the video? How about simple 24-voice polyphony, and play some MIDI files into it? Or would you prefer a monophonic voice made of 24 saw waves? ;)

Thanks! I'll try the same thing with a hair dryer. :)

They sound fine to me! You might want to put this in Songs & Sounds, though. If you're trying to compare filters, it might be helpful to have a less busy song so you can hear them by themselves. Also, slight differences between the filter capacitors might make a bigger difference than differences in SIDs from the same batch.

For the first set, the part about the ground plane is correct, but they weren't that big, only about 4" x 5". The other has no ground planes but is huge, about 7" x 13". We're not planning to order any more of them, but if we did, what could we change in the design (other than adding a second ground plane to the first one)? Edit: How about this? http://www.circuitrework.com/guides/3-2.shtml

Both sets of custom boards I have ordered for my ASIDITY project (from different companies) have arrived somewhat warped. I left the first set under a stack of books for a few days but that did absolutely nothing, and I eventually soldered everything to them, at which point they are stuck in their warped shape and cannot be bent back. Now I just got another set, and they're warped too. Does anyone have any suggestions as to how I should flatten them out?

Our front panel board arrived today--but for some reason they sent two! Front: Back: Two: :)

That bus is a little confusing, but as best I can tell that is almost correct. If you have eight of these button/LED things with their switch wires connected to the eight inputs of a DIN shift register, just connect the resistor packages' common line to ground instead of +5, as you did. You can still use the DINx4 board, just bend that pin on the resistor packages out and solder an extra wire from there to ground. The one problem is that the 10K resistor connected to the SER input of the 74HC165 still has to be connected to +5, not ground. Usually it is part of the resistor packs, but here because it needs to be different on the other side, you have to use a discrete resistor for that one (and not connect the resistor pack to that pin). I might advise you use discrete resistors for all them on that side of that shift register.

Yeah, I can't wait to hear my MIDI collection played on an OPL3 either! One minor problem with this: the drums take up 3 two-op channels, and you get five separate sounds: bass drum is two-op, while snare, tom, hi-hat, and cymbal are one. So there's only 15 two-op channels left; each MIDI channel would get one. The thing is that MIDIs often have polyphony within one channel, and the OPL3 doesn't natively support this; its driver would have to distribute the voices appropriately. Someday I'd like to build a synth with 2 or 3 OPL3s (or maybe 5--you can get them cheap in bulk on eBay) that just uses the 4-op channels and gives lots of polyphony.

I used Front Panel Designer to draft a panel that I then built by hand, and I noticed that when you copy-and-paste objects, they are not pasted in exactly on the grid, but a little off in both directions. I always have to remember to align every object to the grid after pasting it. That might have been the problem. Check your file--zoom in all the way and see if the little + corresponds exactly to the grid point.

Yes, I have. In my build, every parameter of the OPL3--except CSW and NoteSel--is modulatable. You can even modulate whether a voice is 2-operator or 4-operator, and whether the OPL3 is in percussion mode or not, in addition to obvious things like operator volume and pitch. I've written the drivers for it (LPC17 core) but not built the OPL3 module yet, because I've been busy with other parts of the synth. So I really don't know what it's going to sound like yet. Why do you ask? By the way, it's not possible to make the OPL3 have one 36-operator voice, at least not with all the operators modulating each other. Only 4-op, 2-op, and 1-op (some of the percussion) are supported.

Nice! Are the non-SID parts softsynths in Ableton, or else what hardware?

This is the best OPL3 track I can find anywhere on the Internet. Edit: I especially like the drum fill at 3:22!

That looks like it will work, but what I suggested won't make the LEDs go out when you press the button. The only connection between the LEDs and the button is that one side of all of them are connected to +5. Just take the pin connected to the switch (SW), connect it directly to the shift register (don't connect the resistor package to that pin, though leave it connected everywhere else), and also connect a 10K resistor from that pin to ground. If you mean that the button state will be inverted as far as the software sees it, that is correct, you'll have to switch it back in the software. One advantage of your design is that this button is not inverted, it behaves like all the others as far as the software is concerned. But I don't know how many of these you have, so you might have to buy a lot of transistors for your design.

On the pin of the 74HC165 shift register to which you will be connecting the switch pin of the encoder, don't connect the resistor package to that pin. Instead, run an equal-value resistor from that pin to ground. You may then need to invert the state of the switch in the software. As is, the resistors are pull-ups: if the input to the pin is disconnected (open switch) the IC will see +5 through the resistor, but if the pin is connected to ground (closed switch connected to ground), that will "override" the resistor and the IC will see ground. If you want your switch connected to +5 instead, then switch it so that the resistor is a pull-down, connected to ground. Edit: If that was too confusing I'll draw a schematic.

I also don't have any experience with bad SIDs, but I did build a filter capacitor switching circuit (digitally choose 1 of 4 pairs of capacitiors) for my "MIDIbox ASIDITY" project. At one point I accidentally disconnected one of the filter capacitors entirely (what would happen if you had a bad solder joint) while it was playing, and the SID made some noise but sweeping cutoff did nothing. If this is what you mean when you say "failed", then check the capacitors. My untested guess is that in this situation, changing filter types would change the sound, but sweeping the filter would not work in any of the filter modes, as both capacitors are necessary for all the modes--it's not that one is just for low pass and the other just for high pass. The converse of what @NorthernLightX said is also valid: if you get one SID to work entirely in a certain socket, test all the sockets with that SID.

The boards arrived today!