Jump to content

latigid on

Frequent Writer
  • Posts

    2,516
  • Joined

  • Last visited

  • Days Won

    147

Everything posted by latigid on

  1. Ah, one more thing: make sure that you use a decent PSU for the USB/power. If the PSU is weak or noisy you might get unstably power for the display. Good luck!
  2. Seems mostly good but I would check the soldering on the waveshare pins again Is the 10uF capacitor correctly oriented? Probably it is but I can't tell from your picture. What is the voltage at the reset pin (should be +3v3)? Below is the J15C pinout. You can check for interconnections at the top of the waveshare pins. I.e. make sure that you get contact all the way through. Net N$9 is the chip select signal that goes to pin 15 of IC2 (also check the soldering there). And here is the same for the display/plate with the connections labelled.
  3. Hi, Maybe the jumper powering the display is not installed? Could you post a photo of both sides of your Core PCB? Best, Andy
  4. Your proposed part substitution should work. Check the interboard spacing: 16mm M2 spacer 3M FF header socket 8.26mm TE thru-board header 6.5mm 14.76mm, you have >1mm to spare. FWIW I've used large strips of FF sockets for years, just cut down with a dremel-like tool and you lose that position as the tool "melts through". But it works fine and is cheaper than individual sockets of the right size.
  5. Right, there are no F buttons on the midiphy SEQ v4+. Here are a couple of approaches: Press 'Exit' until you reach the menu then the 10th button Assign a bookmark to the page; see seq_ui_pages.h for a list of pages to jump to Freely assign any other button that you don't use (e.g. metronome) in the hardware configuration Hack the firmware/code yourself Request such a feature
  6. Hi, sure is! Top row is equivalent to Wilba GP buttons. They action items on the OLEDs or e.g. in track mode toggle gates/triggers on and off. Bottom selection row changes function depending on the selection LEDs (around the datawheel/dial). So instead of Wilba's 4x track groups and 4x tracks, you have all 16 tracks available on individual buttons, (up to) 16 step views or whatever the track supports. Likewise you have up to 16 layers available, whereas you only had layer A/B/(C: multipurpose) on Wilba's. Layers are either parameters or triggers, check the amazing user guide here. Other than that it's quite similar, so most of the smaller dedicated buttons will be the same. Shift = Select!
  7. Wow, what a saga! I'm really sorry that you had to go through all of that. I'll discuss with Peter as to how we can compensate you for the trouble. It's the first time we've seen something like this and of course it will be reported to the manufacturer. Seems like fab errors crop up from time to time but it always sucks to get stumped by them. Well done again for sleuthing it out. Hope the rest of the build goes better. Best, Andy
  8. Getting a bit OT here... Peter and I have been working on this and will have a solution ready for light bleed that could work for existing SEQs too. If you bought the midiphy L4 caps, they don't really like to come off again after insertion. The fully clear ones are safer in this regard. So for L4 caps you might want to wait a bit... Or, there are other light-blocking ideas. Heatshrink doesn't work consistently enough as the bottom will pull away from the switch body. Something like LiquidTape could be a good bet but it is very stinky and you should work outside with it at least or even wear an air filter mask. For keycaps, you could use a nail file (or borrow the better half's) or sandpaper to gently scrape around the stem of the cap, at least at the very tip to bevel the edge somewhat.
  9. Would have to try it some time I still don't have my SEQ mounted in its case ; I think the UX idea with illuminating the respective track/layer SB LED seems logical but I haven't played with drum tracks enough. Another one that was intended to be developed was the mute/solo display rather than doubling up on redundancy. An obvious question is how forks in the software are managed if they diverge from the main branch and only Thorsten has "full" commit privileges. But that is a discussion for another thread probably.
  10. Here is the "minimum system" with +5V in red, 0V in blue. Also PA11/PA12 as before. Try pulling off the diode too!
  11. There is no explanation other than a weird TVS diode pulling something up/down. I would try with that removed. You could even only resolder the waveshare with pa11/12 and the +5v pins. Can provide info when I'm back home or you can check the waveshare schem.
  12. No it wasn't. Peter is into the code now and would know if it can be implemented /reimplemented.
  13. Also that the waveshare is off, could we see that too? E.g. was the crystal desoldered, the jumpers removed, but resoldered correctly for instance?
  14. Sure, already suggested: R30 to PA11 (yellow) R31 to PA12 (pink) Other side of the resistors: R30 to USB 3 R31 to USB 4
  15. Soldering looks very clean! Was the v407 breakout actually soldered in? If so, impressive that you managed to get it out so well! If it wasn't soldered in, well then the pins probably never had good contact with the LoopA-Core PCB. Other things: try desoldering the TVS diode and powering without it. Also the (soldered -- brave!) optocouplers don't seem to have enough on them, would recommend a touch up.
  16. I think I notice this behaviour too and it could be improved! What I would expect: From SB Track or layers, press and hold SB Mute, the Sel row controls mutes, also displays on the OLEDs. Releasing SB Mute would revert to the previous Sel. Does that fit with your idea?
  17. I think what I would expect is the Selection (Sel) row to always indicate the function indicated on the 8x selection buttons (SB). So if I am in Track, the Sel row should choose tracks 1-16/GxTy. Switching (click and release) to the mute mode will show the mutes. From Track, click and hold on Mute should allow temporary switching to mutes, then when released back to Track. For me, it seems somehow to get "stuck" so the mute SB stays on and the OLEDs stay stuck in mute. I wonder if there is a conflict with option #9 or similar?
  18. Hi Bruno, My understanding of the selection buttons (around the datawheel) is: click and release: change the selection row to that function click and hold: temporarily switch to this function, when released it will revert back to the previous selection. Does it work that way on your side? Best, Andy
  19. Post some pics of the PCB to let another set of eyes go over it for something obvious? There's not a lot to the USB but I would check/scope the line between the 22Rs and PA11-R30/PA12-R31. Do the header pins contact the MCU chip pins? The bootloader is the normal MIOS one. When you "reflashed" was it with boot hold or over JTAG? It would be worth trying a) to upload any MIOS software e.g. _NG to see if that sticks or b) to wipe the chip and start again from bare metal. Is an SD card inserted during any of this? Does the 3v3 rail remain stable during startup? Do you see any LED or OLED activity?
  20. Check the 22R resistors R30/31, TVS diode. Check the USB pins PA11 PA12 make it to those resistors and the MCU chip (2x2x25 pinheader soldering). Check for shorts to 0V. Check PA10 (USB ID) is floating and no jumper is in J20. Check the USB socket. You might also try reflashing the bootloader with hold mode (JPA0, needs at least R102 mounted) or even via JTAG if it got corrupted somehow.
  21. Hehe, Peter does indeed have a soothing voice :) Maybe check that your Waveshare 407v is set to Boot Config: Flash? Best, Andy
  22. Would have to leave Peter to answer specifics about the LoopA firmware but I will see what I can answer. I think I suggested that one too :) Might be tricky as the LoopA would have to track when things were last armed. But if there is a sensible way to do it and it's possible to code it I'm for it! It's a real-time MIDI recorder/looper so a different interface than a step sequencer. You still have to enter the notes by playing them but you can quantise after. No "song mode" at the moment as far as I know. But maybe it will be added at some point. No MIDI export at the moment but it's on the list. I don't know the naming structure. On the list! What is your specific workflow? If you want to edit individual steps then a knob-per step is best, hence the SEQ. If you want to "noodle" a jam and tidy it after, then LoopA. No, the UI requires both 40x2 OLED displays and a knob/"GP button" per step. Thanks! All was based on Peter's initial concept with some minor additions. Best, Andy
  23. We're making good progress! It's a fancy new app with a fancy new display and fancy new other things. :)
  24. Hi Steve, Please refer to the BOM e.g. https://www.midiphy.com/en/shop-details/137/47/midibox-seq-v4-lh-full-essential-kit- This is the whole "kit BOM" but each PCB also has its own BOM page for easier reading, plus you won't get confused with the coded "shop names" of the PCBs. There you can cross-reference all parts with the silkscreen names on the PCBs. You could in effect use the slightly longer pins for everything. I think the "standard" pins are only used for jumper headers etc. Where it's important is interconnecting the lemec and plate PCBs with through-PCB headers/sockets. A "standard" header length works but might not give quite the right contact, therefore slightly longer ensures a better connection. Have fun building! Best, Andy
  25. Hi Robert, great! I was thinking of suggesting that you trying another card, but you said you tried two and formatted already so I didn't want to patronise you. I can only think that certain card models don't work properly and we also notice that in the SEQ. I don't think there is an upper limit and I typically use 2 GiB SanDisk ones myself. You could try the 8GiB one in the LoopA for a true test of intercompatibility :) Have fun! Best, Andy
×
×
  • Create New...