Wilba

You can fix that problem by using bits of that circuit I posted earlier. You can get a gate/trigger output from the ribbon using an opamp that sense touched vs. not touched. Look at the circuit between the ribbon and the "Gate Out" output. Again, I have to say I haven't actually built this circuit yet... and pay attention to the power symbols, which aren't too common. My guess is that the + input of opamp IC3C is tied to VEE (negative rail) via a 10M resistor. Any touch of the ribbon is going to bring the + input above ground.

Count me in for XY Pads and any of their softpots... I've only got one sample of their softpot... eventually I'm going to make a MIDIbox ribbon controller and eventually was going to arrange a bulk order if enough people wanted to make their own. But that's a long way off...

I love the modular concept of MIDIbox, and I've already done a "modular" MB-SID with four SIDs. For MB-SID v2, I was motivated to make a single board solution because I really love laying out PCBs, and it would be much smaller than a modular solution (i.e. removing redundant shift registers from SID modules, components from Core modules, etc.). So I'm making a single MIDIbox PCB so I can squeeze four PICs and eight SIDs into a case that's about two-thirds the size of a C64 "breadbox". But since an MB-SID can have a lot of combinations of display sizes and control surface designs, there can never be one single PCB that suits all people, for some, it will have too many "modules", for others, not enough, and since most of the reason for doing a single PCB is to optimize the space and make it a certain size to fit a case, then there's never going to be one size that suits everyone. So my advice is make your own single PCB if that's what you really want, but if you've got heaps of room in your intended case just buy the ones from SmashTV since once the case is closed, you can't tell the difference ;D

Can the opamp be replaced with one that doesn't require a negative supply, like an LM324? It would be nice to use the same supply that's in the MB-SID already. I don't know enough about audio signals... what is the voltage output of the SID module, including the protection buffer? If it swings either side of ground, then forget I said anything :-[

TK, is there some way we can test for this EUART bug with a Core using MIOS? i.e. some MIOS app that echos MIDI In to MIDI Out, and send a lot of messages to it from a PC and watch for a zero byte sent back?

I am glad to hear the SID you got from me works! ;D

Great work, I like it. I noticed this on eBay - cheap PCB manufacturing service: http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=250010384030&ssPageName=ADME:B:EF:US:2 Maybe you might dare give them a go and let the forum know if they're any good?

I recently realised while designing a new PCB for MBSID v2, that since each SID module in a stereo pair will have common data shifted in from the one Core, and only the separate CS pins used to set SID registers on each SID differently, then you can use a common pair of shift registers for two SIDs. This reduces the amount of PCB space a lot... in my current design I have the two shift registers underneath the two SIDs (in perpendicular orientation) and a sort of common bus between the two SIDs. Looks pretty cool ;D

I've been waiting over a week for offe to get back to me, and still no reply from downix. Sorry guys, I'm not waiting anymore. Current orders: DrBunsen: 1 analogbrotha: 1 Analog-X: 4 Wild_Weasel: 2 th.bass: 5 th.bass: I can only sell you 5 of the new-old stock chips, I do have another two 8580 chips but they cost me about AU$30 each (salvaged from C64s) and I'm not selling them for a loss. If you really want 7 then I can sell you those two as well for AU$30 each, as I don't really need them anymore. OK, everyone PM me with: * email address (for PayPal) * postal address (inc. real name) * whether you want to pay postal insurance or not (extra AU$7.50) Due to recently losing one package of two SIDs in the mail, I will now offer postal insurance where available, which I think is about AU$7.50 per parcel. If you do not choose to pay the extra for postal insurance, I cannot refund and certainly can't replace any SIDs lost in the mail. I am not a business able to cover the occasional loss of things in the post. That said, I leave postal insurance optional but highly recommended. I would feel bad if I had your money and you didn't have your SIDs. There are no more to sell, I will send out PayPal invoices after I get everyone's details and post them all on Monday morning. I will also test all the SIDs again, to be sure ;D

I have a few more to sell now, and I am still waiting for offe and DrBunsen to reply to messages, and now I can sell two to downix if he wants them. I have 13 left to sell. Current orders (still unconfirmed, unpaid): offe: 2 DrBunsen: 2 analogbrotha: 1 or 2 downix: 2 Analog-X: 4 Due to recently losing one package of two SIDs in the mail, I will now offer postal insurance where available, which I think is about AU$7.50 per parcel. If you do not choose to pay the extra for postal insurance, I cannot refund and certainly can't replace any SIDs lost in the mail. I am not a business able to cover the occasional loss of things in the post. I am also giving fair warning to people with current orders: you must reply to your personal messages promptly or your order is cancelled and I offer the SIDs to others.

Still waiting for offe to get back to me regarding his two.

I guessed it would be wrong PIC connections... ;D good to hear you got it going!

Always test voltages by putting negative (black) on ground, positive (red) on pin to test. Test 74HC595 is not in the wrong way! Test 74HC595 pin 8 is joined to ground. Test 74HC595 pin 16 is joined to +5v. Test 74HC595 pin 10 is +5v Test 74HC595 pin 13 is 0v. Then test connections, using multimeter (test 0 ohms resistance!) and test no short between pins/wires next to these connections: PIC pin RC4 -> IC2 pin 12 PIC pin RC4 -> IC3 pin 12 PIC pin RD5 -> IC2 pin 11 PIC pin RD5 -> IC3 pin 11 PIC pin RD6 -> IC2 pin 14 IC2 pin 9 -> IC3 pin 14 PIC pin RC5 -> SID pin 8 PIC pin RC2 -> SID pin 6

Maybe you can make something to test easier. Solder a 1k resistor to an LED. First test this between 5v and ground. Then insert into SID's IC socket between pin to test (like A0, D0, etc.) and ground. Then you can watch for the LED to light from your PC ;D I believe the pin being tested will go high until a new modulation event, all the others will go low. You should also not have the SID in the socket when testing.

I just thought if enough people were getting these buttons, then maybe a group order of some boards to go with them would be a good idea... not just for a monome clone. It doesn't have to be 8x8... it could be 4x4 or 4x1 even...

It looks like the monome just uses plated tracks - looks shiny to me. I wish I could see a closeup shot of the "intertwined star pattern" they use, perhaps spark fun will do the same on their demo board. Why be so concerned about oxidation of plated tracks though? I've got PCBs that have been sitting in a damp shed for years and their tinned tracks are still perfectly shiny. Just out of interest: are there enough people who want to make a monome clone (using MIDIbox) that we could get a bulk order happening for an 8x8 button PCB?

Replacements have arrived, I can sell 2 each to offe and DrBunsen now. Guys please read your messages!

I can also confirm successful burning of PIC18F4620 with winpic800 and my trusty old JDM burner. Bugger. Now my PICkit2 just dropped in usefulness. ::)

All SIDs get hot. 6581 gets very hot, 8580 not so hot. A very very HOT 8580 might mean it is dead.... fused internally or something... or you are supplying with more than 9v. In your case you should upload setup_8580.hex, it is correctly configured to 8580 and will work whether you have a control surface or not.

First, measuring voltage of cheap variable PSU is not a good test. With no load, this will be much higher than it is rated. It will be at the rated voltage at the rated load, like 500mA or 1000mA. Where are you measuring the 11.5? The picture looks like one of your regulators is removable. Is this the regulator for the SID? Which one, 7809 or 7812? Try taking it out, and measure voltage going into pins 1 and 2. This should be around 14v DC above ground. Measuring pins 2 and 3 of the 7809 in PSU should be 9v DC.

I think you can just use an opamp to detect whether it's being touched or not. This circuit here: http://www.paia.com/LabNotes/index.htm does a lot of fancy things like detecting two contacts, but if you look at IC2C, that's what is used to trigger sampling of the ribbon, i.e. if there is a touch on the ribbon, it will turn on the switch at IC1C and get sampled at IC3B, when there's no touch, IC3B holds that voltage. You could simplify this because you've got a PIC doing ADC so just use a logic level input (DIN pin) to detect when there is touch and only do new ADC when it is high.

It does program PIC18F4560 though, you just have to connect it up to a 40 pin IC socket with a header and a few wires.

I emailed the PICPgm programmer and he says he'll put the ID value feature in the next release and email me. Bit late for me, cos I've got a PICkit2 on order.... Impatience is the mother of impulsive spending.

I've tested with a PIC18F4620 with a JDM burner that works with PIC18F452. The results: IC-Prog doesn't support PIC18F4620, I tried using it with PIC18F452 setting and it doesn't work. PICPgm supports PIC18F4620, and seems to burn and verify fine, but has the same problem as PIC18F452, it doesn't write out the ID Value so the bootloader is probably all screwed up with a COM port baud rate (i.e. garbage on MIDI In instead of boot request). Another program I tried called WinPic (http://people.freenet.de/dl4yhf/winpicpr.html) also supports PIC18F4620 but has some wierd issues writing the config bits. It also seems to get confused with the ID Value location (0x200000 to 0x200007). I'm not bothering anymore trying to make this old JDM programmer do new tricks. I think I will get one of those PIC burners from Microchip, it's USB, powered from the USB, compact, relatively cheap... bill might be interested too, since he's got no parallel port ;D

I've tried PICPgm with my PIC18F452 and a JDM programmer. It does burn the PIC, very fast too, and verifies OK, but when i put it in a Core it would repeatedly send Pitch Bend messages instead of the usual SysEx upload request. I reburned with IC-Prog (as I have done before, a long time ago) and it worked. What I noticed is that PICPgm has no way of setting the "ID Value" like IC-Prog does. If I burn with IC-Prog, a known-good PIC (works in Core), then dump PIC to hex file with PICPgm and then view that hex file with IC-Prog, the ID Value is FFFFFFFFFFFFFFFF So my guess is PICPgm doesn't know about the ID Value, it does an erase before burning and thus the bootloader thinks it's using a COM port not a MIDI port hence the garbage Pitch Bend messages. I really wish I could get it working though... I might have to work out if there's some magic I can do to the hex file to set the ID Value...