Jump to content

latigid on

Frequent Writer
  • Posts

    2,516
  • Joined

  • Last visited

  • Days Won

    147

Everything posted by latigid on

  1. Maybe a dedicated 595 chip is needed (no matrix), otherwise your gate signal would have a 1/8 duty cycle? Or I guess I misunderstand: you use the DIOMATRIX as a simple DINX2 DOUTX2.
  2. Yep, this is how AOUT modules work! For the switchable solution I developed quad-channel THT boards, should still have some available if you are interested. Otherwise: op amp with gain, two inverting amps in series, second with switchable gain also as a summing node (offset for bipolar CV).
  3. Well done #38! Might be a record :-) Best, Andy
  4. Chapeau! Congrats to #37! Enjoy your new machine!
  5. No idea about your error, I know some people have trouble formatting and the trick is usually to use another computer (Windows) or so. Why not create a verified SD card on your side and ship that to the customer? If it doesn't work, then if must be a hardware issue (e.g. SD socket soldered too hot).
  6. Sure, all parts and ports are compatible. You can also use a wCore if you prefer to breakout your SD card and LEDs to the rear panel, otherwise the DISCO core should work without requiring RES-SD. The wCore J16E header is a combination of J16 plus the LEDs on the DISCO board plus reset. Just as you wrote The I2C header is 1:1 pinned with the corresponding I2C header on the Core. So it is more convenient to wire up. You can of course go for 8x I2C outs with the firmware "hack", but this was not implemented in the midiphy SEQ v4+ on TK.'s recommendation. Still, it was shown to work in at least one case, so if you feel it is worth it, go for it! Basically the same Yes, all fine, AOUT (not AOUT_NG) in the SEQ config. There are schematics on the wiki but all was designed to be 1:1 pinned with existing boards. The I2C modules can theoretically connect analogue inputs and the ADCs are quite decent. So these headers are used to connect the AINs. In practice there is nothing in the software to scan them, so the headers should remain unpopulated unless you plan to write I2C drivers/modify the app. Greetings, Andy
  7. It doesn't! Maybe you should put all of your questions in one topic rather than making one topic per question?
  8. Probably MABPM 5S. DIN sockets are quite standard -- no surprise :) To be extra careful, measure the distance between pins at the edge of the PCB. Should be 10mm but there are two standards, hence both can be used on the PCB (two pairs of holes).
  9. I have the same "IDC crimping tool" but favour simply using a bench vise. For micromatch MOW cables I leave the polarising tab outside and it works well for me. If you have different tools that work for you, all good! As far as testing other cable types go, there is a plan to build a "backpack"/"shield" that spans across the IDC16s. Then in theory any connector could be wired in. It's a really fun project to build and you can see it working straight away. :)
  10. Hi Alistair, Generally I would only populate parts where there is a silkscreen indicator. This is certainly a double-sided board and the Core is quite a basic circuit, so you won't need any jumpers I am 99% sure. All pin headers seem to be indicated with a J prefix. Any plated holes that don't have a corresponding silkscreen mark are probably vias that join the top and bottom copper planes. You will probably be able to see the copper tracks leading to the vias from the difference in colour. Or if there is no track, the via connects to a 0V/ground plane But yes, it is harder to make out with black solder mask! If you are unsure, put your multimeter in continuity test mode (or resistance) and check the electrical connections. Best, Andy
  11. I can't access your google drive. You can upload images here but the size should be less than 2MB and certain extensions (tiff I think) don't work. PNG/JPEG should be fine. If it has green soldermask and visible traces/copper on both layers I am pretty sure you don't need bridges.
  12. Is it a two-layer board with green soldermask? Then no bridges are required. If it is a single-layer board, find the "quickview" or .brd file on uCapps. Probably the red wires represent the top bridges, with blue as the bottom layer already patterned on the copper side. Feel free to post a picture of your board here to check.
  13. Just post here! I think you would make great friends with Mr. @ssp :)
  14. Here you go! http://ucapps.de/midibox_ng_manual_fs.html Have fun!
  15. You could write an NGR script to address LEDs in sequence: https://github.com/midibox/mios32/tree/master/apps/controllers/midibox_ng_v1/cfg/midiphy It is a very hacky script and could be much nicer with a few for loops. You can use the MIOS terminal command set dout all 1 (or 0 if the cathodes sink into 595 inputs). It really depends on your setup, whether you have transistors, whether you drive LEDs in a matrix etc. You will need a better +5V PSU though as whatever MCU you are using might drop out of regulation with too much draw.
  16. Wow, shiny forum! Nice!

    1. flyweight

      flyweight

      Yeah very nice! well done

  17. Not sure if it is supported as I think it expects a Standard Control Surface (SCS). To change the "LCD configuration" you first must upload the bootloader app and choose the appropriate LCD type (including OLEDS :) )
  18. Hi! @k2z3k0 does a lot of that though I don't know the exact workflow used http://midibox.org/forums/search/?q=k2z3k0&type=forums_topic&nodes=18 SEQ uses the concept of loopback tracks, so you can control a track with another one. Best, Andy
  19. @TK. makes sense! I tried to make the wCore etc. compatible with the DISCO so no surprises!
  20. Some people noticed that the LoopA can get stuck in booloader hold mode. As a potential remedy, install resistors R101 and R102. If you don't have the exact values it doesn't really matter. JPA0 should be left unjumpered except if you want to enforce bootloader hold. With the resistors installed the state of PA0 is properly defined rather than floating. If you don't notice the issue, no need to change it! The BOM is updated to reflect this.
  21. Congrats and thanks for the kind testimonial! Totally right about SMT not being the devil, well as long as it's conceived for hand soldering in the design. :) #36 is your serial number
  22. I'm very glad; that's a great resolution! :)
×
×
  • Create New...