Jump to content

Wilba

Frequent Writer
  • Posts

    3,310
  • Joined

  • Last visited

  • Days Won

    2

Everything posted by Wilba

  1. If it's doing things just when your hand is near, sounds like some pins on the 74HC165 are floating (not connected) so your hand is acting like an antenna, channeling interference into that pin. It's a bit like the buzz you hear when touching an audio plug that's plugged into an amp. Check there's no shorts between encoder/switch pins and the heatsink... check all the solder joints there, resistor network orientation, parts in the right place, etc.
  2. NO! You need REGULATED 12V DC for 6581 SIDs. Certainly I've made this clear in the build manual. You should also not have inserted 6581 SIDs unless you get exactly 12V on the points marked 12V. Seems unlikely to have happened with unregulated power supply. Most likely your other issues are due to too much current load, causing the internal voltage regulator to fail, causing the PIC to sometimes not boot and/or other stuff not working. Remove SIDs immediately. Get everything going again WITHOUT SIDs. Get a regulated 12V DC power supply, preferably more than 500mA (to be sure) but a 500mA one will work. Also turn down the LCD backlight brightness to half-way or less. This reduces current load, so when SIDs are eventually put in, you won't go over 500mA. Note SIDs each take 100mA on the 5V supply! So to balance it, the LCD should be drawing ~100mA to keep it all around 300mA or so.
  3. I haven't looked at the Eagle brd file (I can't even load it), but I can confirm both YAC512 should have VSS connected to ground, and on the OPL3 module itself, they are both connected to the ground plane. Maybe because those chips are mounted on the bottom makes it a bit confusing? i.e. pin 2 (VSS) is one pin up from bottom left corner.
  4. Yeah, but has anyone really been far even as decided to use even go want to do look more like?
  5. Wilba

    Sammich LCD

    The negative green LCDs I sell are high-power. Set JBL to 12V and put shunt in JR4, set brightness pot to half-way and no more. That will keep the total current load low enough for a 12V AC 500mA power supply. Any higher and the input voltage will sag, the voltage regulators won't have high enough input voltage to regulate 9V and 5V, and the PIC won't boot.
  6. You could cut off the DIN/DOUT section from the bottom and it would just fit :wink: But whether there would be clearance between the PCBs is another issue. You have to mount the PCB pretty high due to the extrusion at the back.
  7. Also, sometimes a C64 PSU has the 5V pin on the other side of ground. It is safe to connect both those pins together on the socket side (this is what I did on MB-6582 PCBs).
  8. Well, it is in the UK... they tax the crap out of everything going in, and then he probably had to pay £100 for the wind-generated electricity he used to solder it together. :whistle:
  9. Cool! Where did you get the LCD?
  10. Wilba uses and recommends LEDs from Mouser, parts 638-204GD and 638-204ID Excellent diffused and bright light at low currents. 20000 sold, never a complaint. :wink:
  11. Thanks for the tip! Note that discontinued doesn't mean they're not good quality... LCD manufacturers will often discontinue a model and replace it with an exactly identical model with perhaps a newer controller or different backlight.
  12. My 2 cents: Washers aren't absolutely required, but cheap enough to get if you want. There's plenty of clearance around the holes for a standard M3 screw head (well, for Philips head at least), so I don't know what Hawkeye means by "make room for solder pads". Spring washers are like lock washers, they look like a ring that's been cut and bent. I think either will work. I have no idea if aluminium spacers will stick as well as plated brass. Regarding LEDs: It's easy to set the resistors to suit whatever LEDs you use. Ignore the lead standoffs. You can set the height of the LED to whatever you want, but it's easiest to set it to either flush with the panel or exactly the same height as the switch tops. For flat-top LEDs, flush with the panel looks best and is easy to do, just solder the LEDs before the switches, so the panel can be put flat against something so the LEDs don't protrude past that while soldering.
  13. I must admit, from the photos this looks quite badly made. I certainly don't like the idea of mains power wires insulated by a few bits of electrical tape! For your own safety, I recommend not powering this unit again with mains power. If it was mine, I'd start by ripping out the transformer and rebuilding it using power from a plugpack. The schematics are all available at www.ucapps.de you just need to look at the MIDIbox SID section and the Core and SID module sections. But I'm suspecting the problem is power related and the connections between boards are probably still OK... if someone can supply it with the required power, it will probably fire up again and be back in business.
  14. The designer of the SID chip thought it might be cool to "chain" multiple SIDs by connecting the audio out to audio in of another SID. But it doesn't quite work that nicely since they're a little bit too noisy. Also, the audio in can optionally go through the filter (see "ExtIn" parameter in MIDIbox SID). This is useful if you want to filter external audio with the SID filter, or add a "feedback" pot so you get more filter resonance.
  15. When I do the same test with a green LED, I get voltage at B- of 2.77V with trimpot fully anticlockwise and 2.60V with trimpot fully clockwise. The difference in brightness is very small. Maybe you're using a red LED for testing, and the voltage drop is 1.8V so these results are "normal" for a working backlight circuit. Or something else could be wrong, I'm not sure. Last time I did this test, the LED didn't work, and I wasn't getting those kind of voltages and replacing the BC337 fixed it. But I don't think you need to because you are getting some current sinking from the BC337. So now I think that it's most likely the LCD backlight is faulty and not your PCB... so the easiest way to prove this is just send you a new LCD and header.
  16. Possibly, but equally possible is a bad solder joint in the brightness pot, the BC337, or a dead BC337. Note I've run my LCD backlight with a similar setup and not killed it, but I don't set the brightness pot past halfway. Was the brightness pot set to minimum (fully anti-clockwise)? or did you have it at maximum? It might be relevant to what could have gone wrong. It also could be too much current load, but that seems unlikely with 600mA rating and no SIDs installed. What kind of power supply is it? If it's switchmode, they can sometimes do crazy things that I can't explain. I recommend testing the backlight pins on the base PCB with a LED. As above: If you want to test the backlight circuit, you can do this by setting JR4/JBL to "low-power" backlight settings (no JR4, JBL="5V") and inserting an LED into the top two pins of J16 header (cathode in pin 16 "B-", anode in pin 15 "B+"). Test the voltages of the B+ and B- pins relative to ground and report them here... it will help work out what is wrong. Try turning the brightness pot, check that the backlight circuit is actually limiting current. If the LED lights and you can change its brightness, the problem is probably with the LCD backlight. If the LED doesn't light at all, then the LCD backlight is probably OK and there's something wrong with the current limiting circuit (the BC337), though the LCD backlight could also be dead too. I recently fixed a problem where the LCD backlight, the LCD logic AND the BC337 were dead. Don't ask me how that happened, but it can. Do a few tests and get back to me via email... it might be quicker to just assume the LCD is dead and I send you a new one... but you could try to replace the BC337 first. JBP, not JPB, but yes. Those shunts are essential to getting the "orange points" (also the SID voltage) exactly 12V. But this is unrelated to your LCD issue.... putting shunts in JBP will only increase the voltage to the backlight to 12V, the problem isn't lack of voltage there (it did work without JBP shunts, it should again).
  17. Those displays look OK. You need to compare the dimensions with the "standard" one, see here: http://www.midibox.org/dokuwiki/doku.php?id=wilba_mb_6582_control_surface_parts_list You should look at the spacing of the corner mount holes and the size of the bezel (check it will fit in the hole). At first glance, it looks OK. The red/black one I used is here: http://character-lcd-lcds.shopeio.com/inventory/details.asp?id=1287&cat=Lcds&sub=Character%20Lcd
  18. Snoozr: Where is the LCD window?
  19. What a coincidence... yesterday, my wife painted her nails with that stuff. :thumbsup:
  20. Kickflip: if you have a finished MB-6582 that's only missing "Waldorf" knobs, I will sell you 15 of mine, black/transparent or black/red. POIDH! Offer expires in 1440 minutes.
  21. Hi! This forum pets lovers? I destitution to buy such a cat Who has this already exist? What is the kind and what his character?
  22. I suppose... but it is offset by unpainted engraving and no protective paper to assist in painting it. Should the lucky buyer want painted engraving, I have a spare top and rear panel, which I can post cheaply worldwide, I think 6 AUD will cover it.
×
×
  • Create New...