Shruthi XT not receiving MIDI
  • I’ve had a DIY Shruthi XT sitting in non-operational status for a while and I’m back to trying to get it to work. It powers up and all looks good but I can’t get anything out of the output. One thing I noticed that looks strange is that when I play the controller’s keys I get a variety of responses on the LCD. There is one G# that will display the musical note on the LCD but it doesn’t generate audio. Most other keys don’t show anything on the display. Some display the ~ symbol, which according to the manual signifies control changes. Any ideas what could cause this? My controller does fine with other synths so I’m sure its output is fine.

  • Has it worked fine before? Does it have firmware version 1.01 or newer?

    You might want to make sure all the ICs are still properly seated in their sockets. Give them an extra push.

  • It has never worked. Firmware version is 1.02.

    Of course! Didn’t think about reseating the ICs but it sounds like a good place to start. Thanks!

  • In the past, I’ve been stupid enough to bent IC pins inwards when inserting the IC, and not even noticing.

    Start with the simple things first :-)

    Then post pictures :-)

  • The opto-isolator can sometimes be the culprit. There’s two different chip numbers, 6N137 and 6N138. People have often had to change one for the other.

    Edit: I said often, maybe not that often.

  • I reseated all the ICs. Still having the same result. I also put a new 6N137 in. Same thing. I guess hi-res pics of the PCB is up next? I’ll try to get some when I have some good daylight. Thanks for the suggestions so far!

  • I will also see if I can get my hands on a 6N138.

  • nosamiam: the circuit is designed for a 6N137, I don’t recommend using a 6N138.

  • Ok, I won’t try that.

  • Here are some pics of the top and bottom of the PCB. I hope they are ok. I used a nice camera but I’m a terrible photographer and I know there’s an art to this. Also, the dust and lint is there because it’s been sitting for months and months on a shelf. There was also a residue on the PCB that I tried like hell to clean off (not just flux; it came with a film in the mail that I didn’t try to clean off till it was already soldered. Big mistake). I didn’t get it all off

    Let me know if you see anything obvious or know what I should be looking for.

    1.jpg
    2000 x 1124 - 701K
    2.jpg
    2000 x 1124 - 847K
    3.jpg
    2000 x 1124 - 873K
    4.jpg
    1800 x 1012 - 871K
    6.jpg
    2000 x 1124 - 751K
    5.jpg
    2000 x 1124 - 790K
    7.jpg
    2000 x 1124 - 810K
    8.jpg
    2000 x 1124 - 839K
    12.jpg
    2000 x 1124 - 786K
    10.jpg
    2000 x 1124 - 647K
    9.jpg
    2000 x 1124 - 841K
    11.jpg
    2000 x 1124 - 805K
  • Doesn’t look bad to my eyes. I would recommend to go through each of the components on the MIDI part of the schematic and take a close look at the solder joints. Sometimes solder creates super thin (=width of a hair) bridges between pins.

  • I would go for another 6N137, i had 3 dying in projects so far.

  • I see that your display is applied crooked. Make shure that the bottom (underside) of the display doesn’t touch the solder joints of your MIDI sockets.

  • Well, cleaning of the dust and lint might help as well, certainly when looking for tiny soldering bridges between pins.

  • Another point, I would solder all the mechanical leads of the pots, just makes the design more robust.

    I had some problems back in the days with the optocouplers, did switch them a lot, basically you cannot destroy those ICs, so you could swap them live, but I wouldnt recommend it if you do not exactly know what you doing.

    Ah ok but you say your Dipsplay responds, so I guess some signal arrives at the shruthi, so maybe your tone generater is crooked, you do know that you have to place some filter board beneath the xt to generate sound :)
    So your sound geenration does work in demo mode I assume, then the next question would be if you can get a hold of an oscilloscope to probe some waveform generations..
    and so on.

    PS: I can spot at least one cold solder spot in picture 10 probably the LCD connector gnd, but I dont think that matters for your midi.

  • TheSlowGrowth: I desoldered all the parts in the MIDI circuit and resoldered them. Same deal.

    nightworxx: This is my 3rd 6N137 and it’s brand new. I think I have a DIY project with a 6N137 and it’s operational. I’ll try popping this 6N in that unit and see if it still works correctly.

    cj55: I actually mounted it that way on purpose to make sure there was plenty of clearance. It’s definitely not shorting on anything underneath.

    eelco: I’ll give it a good cleaning. The dust is mostly on the component side of the board since that’s how it sat for those months. Can’t hurt though.

    mex: I do see that joint. I’ll touch it up. I did have a difficult time with this board in general. Some of the pads had some contaminant or something that made the solder act really strange. The solder acted like it had a really high melting point. When I was finished with a joint and would try to pull the iron away, the iron stuck to the solder. I will always, always clean my boards before I solder from now on.

    I’m wondering…
    Could the MCU be bad? It’s the only component having to do with MIDI transmission that I haven’t checked out. They aren’t cheap so I don’t want to just buy another one unless it’s likely the problem. Should I suspect it?

  • Does the sound work when you use the “JAM” button? If yes: Then your processor should be alright.

    So, we’ve checked all the obviousl stuff now… Some oscilloscope images for different parts of the midi circuit would be helpful. Do you have access to a scope somewhere?

  • I forgot about the JAM button! I was so worried about MIDI input. But unfortunately after attaching the filter board I still get no sound. So I may have multiple issues. I do have a scope. I can do some poking around. But if the MCU is bad I probably won’t get far. Easy way to test it definitively?

    I can post some pics of MIDI signals.

  • Yes, some pics of the midi signals would be helpful. Then, start a note with the JAM button and make scope shots of different points in the audio circuit (e.g. a scope shot from the raw oscillator signal, then through the signal path from there)

  • easy, with a scope and the forums help you should fix this in max a week.

    I asusme your midi is working perfectly right, since you get the little note and ~ thing shown in the display which only shows up when the code recognizes some midi signals, but one never knows :)

    So probably there is an error in the filter board, tell us what filter you are using, then start to probe around along the signal path.

    If i recollect right there should be a 39 kHz PWM carrier on one line going into the first filter opamp or soemthing which holds the modulated audio signal, and then crawl all your way into the signal chain until you cannot probe a signal any more.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

In this Discussion