perplx

So my sammichSID had been working perfectly for years. Yesterday, it suddenly started to show messages like it was updating the firmware, even though I hadn't sent it any. I got scared, unplugged the power and the MIDI, and powered it back on. The right-side SID chip's filter no-longer worked. :pinch: So I start to think, "It may not be a good idea to put the Sammich in a MIDI-thru chain right after two Shruthi-1's connected in poly-chain mode; maybe all the SysEx sent with a generic manufacturer ID got the Sammich confused. Maybe it received something meant for the second Shruthi and interpreted it as a firmware upload." So I fired up MIOS Studio, established communication, and sent a real firmware update. The update process looked like it worked, but now, after showing the MIOS 1.9 screen, it's stuck on "Launching CS", won't respond to MIOS studio, and makes no sound. :frantics: I tried uploading the firmware blind, but it gets no reaction from the Sammich. Now, I'm very deeply attached to my Sammich, it's been very good to me. I figured I'd ask about the proper course of action before I make things even worse. Is there a way to yell at the PIC loud enough that I can re-upload the firmware? Should I / Could I just get a new PIC to make the problem go away?

Thanks! The LCD wasn't to blame, putting an LED between B+ and B-, as you suggested, didn't show any light. I followed the traces and found one of the legs of the voltage-select jumper had a dodgy-looking joint; re-heated that and I got a soothing blue light from the LCD. So, my bad. :whistle: Thanks for the support! You rock so hard, it's kind of crazy.

So I'm almost done with the sammichSID. It works, and sounds even better than I expected! Thanks Wilba, nILS, TK et al.! I just have a small problem: the LCD works, but the backlight doesn't turn on. I can only use it with a lamp - not ideal. I got the low-power blue LCD, and the jumper is in the appropriate 5V position. I tried the backlight pot, it doesn't change anything. I might've mishandled the LCD somehow and damaged it, but since the display part works, I figure it's worth trying to fix. So, do the excellent folks here have a clue how to test what works? where to test the LCD for what backlight current? Thanks!

Ah, that what caused the confusion. I meant the rows on the panel, which are then the columns in the matrix. (right?) My idea was 16 red LEDs, 16 green, and 16 blue, using only primary colors to reminisce about the lo-res old computer graphics. That was why I was thinking about more than one color of LEDs running on the same resistor. Wow, those are low. I guess that's fine, since they're diffuse, but I have trouble finding a 3mm blue LED that matches the red and green. Everlight doesn't make a diffuse blue like the red and green you linked. Mouser part 604-W7104MBC/K looks close electrically, but it's a clear LED with a low viewing angle, and it's the closest thing that's in stock. I guess I'll try DigiKey, or use all-clear LEDs... :unsure: Good point, I have a few spare 595's from some things I tried out with an Arduino. I'll make sure things look good before soldering. Thanks for your help!

I'm currently waiting for the next batch of kits to ship, so I can join the fun. This got me so excited, I had the idea to use different colors of LEDs on the different rows of the panel. More blang. :drool: Thing is, if I understand correctly, the resistors that limit the current to the LEDs are on the columns of the matrix, not the rows. That means resistors of different colors need to run on the same resistors. I'm a bit worried about relative current specs vs. relative brightness of different colors, since I don't actually know how to calculate the max forward current, etc... :ermm: :huh: Does anyone have specs for the LEDs to use? The proper method for calculating LEDs' specs? Or should I just keep it simple and go monochrome? Thanks!