Dead MIDI In / Remove LCD...?


So while playing on my Shruthi-1 (SMR4-mkII), having my Minibrute connected to its MIDI In port, all of a sudden the note last played would not stop, even though I stopped pressing the key on my Minibrute. I am guessing something with the MIDI failed then, and the Shruthi could not receive any stop note message, since when I restarted the synth I could not get any note to play via MIDI any longer. However, triggering a test note internally is working fine.

I have tried a couple of different MIDI cables, tested them on other synths, and they all work fine. I have also checked that the receiving channel on the Shruthi did not magically change when playing with it. My guess is that there is one or two bad solder joints somewhere on the board, likely around the MIDI connector or the 6N137. Do you agree? If yes, is there a way to remove the LCD screen without damaging the LCD (or the board) in order to fix these joints?

> My guess is that there is one or two bad solder joints somewhere on the board, likely around the MIDI connector or the 6N137

You could use a multimeter in continuity testing mode to check for that each pin of the 6N137 has a sound contact with the right circuit point. Then once you identify the problem, you might be able to solve it from the bottom side of the board without having to remove the LCD on the other side.

Thank you for your quick answer!

Since I 1) have been taking care of my 1.5 yr old son, and 2) in the process of 1) have hurt my back, I have not been able to check the optocoupler yet. I’m just posting this comment to let you know I have seen it, and am not ignoring it :slight_smile: I’ll let you know how things have developed, as soon as they start, err, develop.

What brand opto was it? I just had an XT in for repair with a failed optocoupler (It was the one that came with the kit, a Vishay i believe). Symptoms same as yours.

@altitude: Seems to be a Fairchild.

However, I just opened up my Shruthi in order to see if I could spot any obvious bad solder joints and measure some voltages. Then the MIDI In was suddenly working again! (I used my laptop to have Ableton Live send signals to the synthesizer, which I also had tried earlier, but in vain.) Something is fishy here. I have been applying some mechanical stresses to the 6N137 and the MIDI ports (i.e. been pushing them back and forth relatively non-gently with my finger) so I am not any longer so sure that there is a loose connection somewhere.

Unfortunately I have to update this thread: The MIDI port have been working excellent since my last update - until now. Suddenly, it just died again. I think the problem is located at some solder point of the MIDI In connector, since if I push the MIDI cable plug from beneath upward, the Shruthi receives notes again. I have tried to resolder the connector pins from the bottom side of the board, without any success.

Any ideas? Is it possible to probe the pins in some way in order to figure out which one is glitchy?

Midi cable into the midi port, other end unconnected, use a multimeter in short detection mode(the one that beeps), touch each pin of the unconnected end with one probe, touch the other probe to the side of the corresponding midi pin’s solder blob/ball(make sure not to hit the pin itself, since that could cause a false positive).

The one that doesn’t beep = the one that’s the problem. :slight_smile:
Edits - You need between two and three beeps out of five(The ones in the middle), since not all midi connections are soldered in by default, check your cable to figure out if all five pins are actually internally connected first !

@V`cent: Worth a try, but for some pins I have to squeeze to probe between the board and the backside of the LCD then…

I guess so. You could start with just the back of the midi connector, to make sure it’s not something mechanical inside of the connector/port itself that is broken, and then move on from there…

And you can of course trace the schematics(can be found on the main page somewhere, I think it’s DIY resources), and see where each pin connects, then probe there instead, so you don’t have to squeeze under the LCD, possibly making something wonky there…

@V`cent: Thanks!

Ok, so I’ve looked through the schematics ( and if I understand things correctly only pins 4 and 5 are coupled to the board? In that case both go through all my continuity tests: Continuity is tracable until legs 2 and 3 on the optocoupler. (The multimeter beeps at leg 3, and shows 220-ish ohms at leg 2, as I guess it should.) I have tried to push the MIDI plug a bit back and forth, but nothing (bad) happens.

So, I guess I have an error somewhere which is hard to reproduce? Perhaps the humidity in the air has been exceptionally high lately, or the stars aligned in some dreadful way. Or have I missed something?

Well, there’s a 220 ohm resistor in the path of leg 2, so that does make sense.

How about the connection(s) from the optocoupler to the atmega? (You can measure on the leg of the atmega, so no need to push under the LCD), although the 10k resistor on leg 8 of the optocoupler won’t beep, and probably shows no connection, try using the resistance mode for that instead.
Also, check if the optocoupler is connected to ground and VCC (should be accessible on the pin header).

If those are correct, try measuring directly on the leg of the optocoupler(not the socket), and directly to the atmega/GND/VCC to see if the socket is the culprit.

@V`cent: Thanks, I think I’ll have time to check those things tomorrow.

In the meantime I have to ponder why I dreamt this night that I had to chop my son into little pieces and make him into - wavetables.