Shruthi XT midi in problem

I’ve just finished assembling an XT and everything seems to work fine except for the midi in (haven’t checked midi out yet). When I try to control the XT via midi (e.g., Ableton or Axiom 49 midi keyboard using matching midi channels) it doesn’t recognise any input. However, when I set the XT’s midi channel to 0, it does recognise sporadic and unpredictable Note On and control change messages. The keyboard works with other synths so I’m sure that’s not the problem. Could it be a faulty 6N137, a bad solder joint, or some other issue I’m not seeing? Any help would be much appreciated.

I’d check the orientation of the opto-isolator chip, and all solder joints in that area.

Also, it should be relatively easy to trace back from the opto to the connected pins of the ATMega. Check there too for bad joints.


The 6N137 is oriented with the dot (pin1) facing towards the midi connectors, which I take is correct. I’ve also reflowed all of the pins of the ATMega, the midi in connector, and the 6N137 socket. It’s still behaving weirdly though. For example, moving some of the faders on my Axiom 49 sends Note On messages to the XT (the note icon appears) and it plays random sounds / patches.

Sounds to me like the processor might be running at the wrong speed. Did you set the fusebits accordingly so that it uses the crystal? Did you install the right crystal on the board?

Hi, there are some things i would check:

  • did the MIDI cable work before ?

  • is the setting of the MIDI keyboard working with another synth ?
    I use for hardware checks an old, externally powered keyboard, just to rule out possible MIDI bogus of the computer.

Then next:

  • is the Shruthi correctly configured for your filter type ?

  • what MIDI options are set in the configuration menu ?

  • is the 6N137 sitting really tight in its socket ? Are all pins in the socket ? (this shit happened to me more than once !)

  • same for ATMEGA !


  • did the MIDI cable work before ?
    Yes, no problem there.

  • is the setting of the MIDI keyboard working with another synth ?
    Yes, the Axiom 49 works fine with an Analog Four. Also, when I try to sequence the XT with an Octatrack it doesn’t receive any messages.

  • is the Shruthi correctly configured for your filter type ?
    Yes - lpf for the SMR4 MKII.

  • what MIDI options are set in the configuration menu ?
    Tuning = 0, Channel = 0 (or whatever matches the channel on the Axiom), Midi Out = off

  • Are the ATMEGA and the 6N137 sitting really tight in their sockets ? Are all pins in the sockets ?
    Yes and yes.


-Did you set the fusebits accordingly so that it uses the crystal?
Can you please elaborate what you mean here? I bought the AVR and EEPROM pre-programmed from Modular Addicts. How would I check? The crystal I’m using is rated at 20MHz.

When I get home from work I’ll check the midi and audio out to try to narrow down the problem, and I’ll also post some hi-res pics of the board.

Okay, so with a pre-programmed Atmega you should be fine.

Have you checked the opto-isolator chip is inserted correctly? Sometimes an IC leg can get folded over on itself when the IC is inserted.

It could also just be a bad opto IC. You could try a new one, and see if that helps.


I wasn’t able to upload any pics last night, but will try again tonight. I’m very sure the opto-isolator chip is inserted correctly, but I suspect it could be internally damaged somehow. I’m waiting for some new ones to arrive. I read that the opto-isolator only deals with midi in, not midi out, is this correct? Therefore if it is only the 6n137 that is the problem, midi out should not be affected, right?


Well, I changed the opto-isolator chip and it’s exhibiting the same behaviour. But now I’m confused, the XT is receiving midi messages after all, but they seem garbled.

Assuming the Axiom 49 is communicating with the XT on the appropriate channel, should it be as simple as plugging in and playing? With the Axiom connected to the Analog Four’s midi in for example, I can immediately play notes on the keyboard that correspond to notes on the A4.

What happens with the XT, however, is that if I try any input from the Axiom (e.g., key, fader, mod wheel, etc.) it changes parameters on the XT in weird ways. Where is the problem likely to be? What’s more, I get a strange display symbol which isn’t documented in the user manual, what does that mean?

Thanks for your help so far guys, I appreciate your patience.

There is not too much for the MIDI in . The 6N137 a diode and one or two resistors .
Is the diode in the right direction . If a midi message is recieved and no sound comes out - it can be also the Filterboard Maybe there is the problem .

Thanks for the response, I’m starting to pull my hair out here. What might be the problem with the filterboard? Sound is coming out when I move a key/fader/button on my midi keyboard but it’s not what I expect it to be.

The symbol you see on the display is the pitch bend lever icon. It should only be displayed when you use the pitch bend on your Axiom49.
What exactly happens when you play different notes on your keyboard? Does the display then show the note icon?

Ah ok. When I play different notes on my keyboard, usually I don’t get any sound. Either nothing happens, or, the pitch bend, note, or control change icons randomly appear. If I start the internal sequencer using the jam function and then I play a note or move a fader/mod wheel on the midi keyboard, it immediately causes the xt’s sequencer to stop. It’s really doing my head in trying to troubleshoot this.

Strange. Then it is a problem with the digital side. But there is not much involved. Only the optocoupler circuit with the 6N137 a diode and 2 resistors. And then only one connection to the microcontroller.
I would resolder all connections (also on the microcontroller) with a little bit more fresh solder. Looks like there is a little bit too less solder on some joints.

Thanks, when I have time over the weekend I’ll try to resolder everything and see how it goes.

Looking at the soldering, some of the pins of some of the pots might need resoldering (extra solder and/or reflowing what’s there).
If your pots have bad connections, or are not connected at all (!), you get a ‘floating’ input to the microcontroller, which starts to do random things then … which is what seems to happen.

I resoldered all the connections, but it’s still the same. I might need to get an expert to look at it.

Check the 10K resistor value between the optocoupler pins 6 and 8, if it’s too high, the pulse raise/fall times will be slow, and you’ll get garbled input.