Midi Thru port not working

Xopek · 2022-08-10T16:28:33.461Z

Hi!
Looks like my midi true port stopped working.
What happened: OT mkI 1.40A OS.
Midi in and midi out work great.
I changed cables, midi message sources, psu. Nothing. Nothing comes from midi thru. I checked the soldering of the connector with a tester - everything is ok. As far as I understand, there is only one active part between midi in and midi thru - an opto-isolator (I forgot shoot a photo). Any fresh ideas on how to test it? Elektron support is on summer vacation, so looking forward to trying your advice…

guga · 2022-08-10T21:33:51.780Z

you could try with a MIDI monitor, see if the computer receives any message from the MIDI Thru

edit: are the MIDI settings on the device connected to the MIDI Thru port correct?

trackerjack · 2022-08-10T21:55:29.435Z

Don’t quote me on this, but I’m assuming there should be another active part like an inverter / schmitt trigger of sorts, for buffering, driving the (thru) output current loop and with hysteresis to clean up the optocouplers output… (could all be one integrated part though I guess)

Xopek · 2022-08-11T05:43:41.509Z

guga:

you could try with a MIDI monitor, see if the computer receives any message from the MIDI Thru

edit: are the MIDI settings on the device connected to the MIDI Thru port correct?

I’m used hardware monitor from my midi merger or
MD with clock and ch receiving settings. I’ll try computer midi interface and software midi monitoring. Come back.

Xopek · 2022-08-11T05:52:43.054Z

trackerjack:

Don’t quote me on this, but I’m assuming there should be another active part like an inverter / schmitt trigger of sorts, for buffering, driving the (thru) output current loop and with hysteresis to clean up the optocouplers output… (could all be one integrated part though I guess)

I saw something like 6n138 there. I need to open ot again and take photo…

trackerjack · 2022-08-11T07:52:13.797Z

That’s the opto-coupler, which by itself shouldn’t be able to drive the output afaik…

Tbh, I would just wait until support is back in town.

Supercolor_T-120 · 2022-08-13T10:44:17.808Z

Optocouplers can definitely fail for weird reasons, they’re pretty fragile. No idea if that’s what happened to yours, but it has happened to me a couple of times. Once all it took was putting a piece of gear with an acrylic enclosure into a new case that apparently had a lot of static charge in the foam lining from the factory. MIDI input still sort of worked, but it depended on what the source was - with some controllers it wouldn’t work at all, with others it was inconsistent and had things like wrong o skipped notes, completely random velovities, etc. Passing any MIDI through an original MIDIpal first made it work nrmally. Eventually figured out the optocoupler had been damaged just enough that it became extremely sensitive to the leading slope of incoming pulses, so it needed an extremely clean MIDI signal to work. Swapping in a new one fixed it and it’s still working find 6 years later. The moral is if you’re repairing, designing or building something, always socket the optocoupler! I’ve only had to replace two or three but that’s still more than any other component except power supply capacitors (and patch memory batteries but that’s not a repair, that’s routine maintenance).

Something like that seems really likely in this case. Is the OT optocoupler socketed? If it is there’s no harm in swapping it if you can get one easily (I still have about half a dozens so I haven’t had to buy one since the supply chain got bad).

avantronica · 2022-08-13T13:14:56.052Z

Supercolor_T-120:

Something like that seems really likely in this case.

Just curious if it can be found on many devices other than at the IN given the op said IN working great - so i’d not have bargained on an optocoupler between the in and thru - i’d have thought the responsibility to protect would fall on teh receiving IN

Just curious, not challenging especially as @Xopek mentions there being one in between so i guess it’s perhaps possible to short it ? or use a diode if it’s not typical to have an opto-isolator in that part of circuit (if that assumption is even valid)

I’ve looked a fair bit at both OT boards and there’s nothing of that nature socketed

Xopek · 2022-08-13T13:15:12.467Z

In this thread I’m founded good resolution photo.
In this we have H11J3 optoisolator and HC4050 high-Speed CMOS Logic as mentored trackerjack.

093566×2365 1.97 MB

avantronica · 2022-08-13T13:18:33.303Z

Xopek:

In this we have H11J3 optoisolator

If only the one then … so it’s indeed only used at welcoming the IN signals - has no(if IN is solid) bearing if i’m not mistaken on the Thru which is presumably not a passive ‘clone’ of IN - so the focus might be elsewhere on the board

Xopek · 2022-08-13T13:20:48.293Z

avantronica:

Xopek:

In this we have H11J3 optoisolator

If only the one then … so it’s only used at the IN - has no bearing if i’m not mistaken on the Thru which is presumably not a passive ‘clone’ of in - so the focus might be elsewhere on the board

I think you are right. so it could be hc4050

Supercolor_T-120 · 2022-08-13T13:43:29.663Z

I think you’re right, I reread and I thought the OP had actually found a separate optocoupler on the thru port.

Still recovering from being sick all week and my focus isn’t the best right now.

Xopek · 2022-08-29T21:46:05.940Z

Xopek:

hc4050

If anyone is interested: it was just that. Replaced the 4050 and everything became ok.
Thank you!

avantronica · 2022-08-29T22:12:30.496Z

Xopek:

Replaced the 4050

Impressive skills, looks fiddly !

aircoholic · 2022-11-17T20:08:10.427Z

I wanted to say a big thank you for everyone posting their valuable knowledge in this thread!
This Info has helped me tons more than the Elektron customer support.
My Midi Thru and Midi In didn’t work properly with the Torso T-1 I bought, but worked fine with my other gear.
Elektron Support told me it doesn’t sound like a problem with the OT and tried to blame Torso Electronics. I did refer to the info in this thread, but apparently a technician said it doesn’t help to replace the optocoupler.
I did it anyway. First the HC4050, which didn’t really change much. After replacing the H11L3 (not H11J3!!) everything worked like a charm again, Midi In and Thru now work with the Torso T-1, like they’ve known each other for years.

Thank you all who contributed to this valuable thread again!!

Xopek · 2022-11-17T21:27:41.786Z

aircoholic:

H11L3 (not H11J3!!

You are right!

Supercolor_T-120 · 2022-11-22T22:30:52.551Z

aircoholic:

a technician said it doesn’t help to replace the optocoupler.

On paper they’re supposed to either work or not work, so I can’t blame them really. I only know about it because someone with a lot more experience suggested it to me years ago and was right.

It’s interesting that the 4050 was an issue for someone, too. I wonder if it’s unlucky static discharge into the MIDI jack making parts fail - that’s the only cause I could work out for mine (I probably already mentioned it but it happened after putting it in a new case for the first time, and the foam was carrying a lot of static so I assume some of discharged into the input jack and messed up the optoisolator in that case). It seems unlikely that something coming from another piece of gear would damage it.

thermionic · 2022-11-22T23:39:20.867Z

Supercolor_T-120:

On paper they’re supposed to either work or not work, so I can’t blame them really. I only know about it because someone with a lot more experience suggested it to me years ago and was right.

It makes sense to prioritize the myriad of configuration failures that are more common, but once you’ve isolated (sigh) all other factors…