yet another core midi problem

[/dev/null]

Hello everybody,

I started to built myself a midibox sid with one core so far. The core seems to be up and running wirth MIOS installed.

When I power it up I get the single upload request which makes me think that midi out is working properly. :)

Where my problems start is when it comes to the midi-in:

I already spent quite some time here: http://www.ucapps.de/howto_debug_midi.html but I don't make any progress:

my approach to find out what's wrong was:

1. connect a midi cable between midi-in and midi-out of my midi interface, which is connected to my linux laptop: works fine: I was "uploading" the SID app and the in-monitor showed the same as the out-monitor. So I guess it is not a pc-hardware/software issue

2. Check the resistor values at the MIDI In Port: meassured them and seem to be fine

3. TEST IN3: Instead of soldering I just took an led and connected it between the first two pins of J11: It as lit all the time and when I started the "upload" it was flickering with the sysex messages.

4. TEST IN6: standalone optocoupler-test: shows the expected results. so i skipped test 7 since i didn't want to solder something together and the optocoupler should work after test 6 anyways, right?

TEST INOUT1: Yeah thats where things stop working. I just unplugged the PIC, put a jumper between pin 2,3 of J11 but the in-mon stays emty while "uploading".

I hope the information is enough. I appreciate your help and of course the whole project!

[stryd_one]

Welcome aboard! :)

Please read the link in my signature - it is very important to COPY and PASTE your logs.

For example, one time someone said that they did the loopback test and they recieved the same sysex that they sent. In fact, they recieved a different sysex and didn't notice it. It's good to have a fresh pair of eyes double-check these things.

Please do ALL of the tests and post results from each.... We'll see how we go when we have all the info :)

[/dev/null]

OK here I go again:

TEST PC1 and TEST PC2:

As I already mentioned I am using MIOS Studio on a linux laptop:

pc-midi-out -> pc-midi-in

first block of the 6581-sid-hex-file:

midi-out-monitor:

timestamp [unknown] | Sysex message: F0 00 00 7E 40 00 02 0C 00 00 20 15 7B 63 3F 04 3B 5E 1B 78 28 1D 71 5F 42 51 6F 0D 7C 18 6E 78 6F 61 74 77 46 7E 04 4F 3C 39 70 27 3B 63 4F 01 5B 5E 1C 78 0E 7D 71 67 41 1F 6F 0E 3C 00 1E 78 73 61 7F 77 46 7E 0F 67 3C 37 70 7F 3B 63 3F 04 03 5E 1A 78 3A 3D 71 57 43 6F 6F 0D 3C 1F 3E 78 6B 60 0E 77 47 1E 00 6F 3C 39 70 54 3B 63 2F 05 4F 5E 1A 78 3F 1D 71 57 40 24 00 00 00 02 20 00 00 00 12 00 00 00 01 10 00 00 00 09 00 00 00 00 48 00 00 00 04 40 00 00 00 24 00 00 00 1F 7F 7A 09 08 00 00 15 48 40 08 01 54 44 01 00 1B 24 20 1C 01 56 22 01 40 0D 12 10 0A 00 67 11 00 40 06 29 08 03 00 31 48 4F 40 03 04 44 7C 00 17 64 27 60 01 3A 22 3E 00 0B 32 13 70 00 78 11 01 00 07 51 08 09 00 3E 08 40 50 03 78 44 05 40 13 24 28 30 00 50 24 03 20 05 42 20 1C 00 0F 11 43 20 06 51 0C 1B 00 23 08 61 60 02 60 46 0E 40 03 44 39 08 00 3A 23 48 60 05 62 1C 64 00 5C 11 66 30 06 29 0E 34 00 38 48 73 28 01 58 48 1B 00 0E 04 40 1F F7

midi-in-monitor:

00000000074542 ms | Sysex message: F0 00 00 7E 40 00 02 0C 00 00 20 15 7B 63 3F 04 3B 5E 1B 78 28 1D 71 5F 42 51 6F 0D 7C 18 6E 78 6F 61 74 77 46 7E 04 4F 3C 39 70 27 3B 63 4F 01 5B 5E 1C 78 0E 7D 71 67 41 1F 6F 0E 3C 00 1E 78 73 61 7F 77 46 7E 0F 67 3C 37 70 7F 3B 63 3F 04 03 5E 1A 78 3A 3D 71 57 43 6F 6F 0D 3C 1F 3E 78 6B 60 0E 77 47 1E 00 6F 3C 39 70 54 3B 63 2F 05 4F 5E 1A 78 3F 1D 71 57 40 24 00 00 00 02 20 00 00 00 12 00 00 00 01 10 00 00 00 09 00 00 00 00 48 00 00 00 04 40 00 00 00 24 00 00 00 1F 7F 7A 09 08 00 00 15 48 40 08 01 54 44 01 00 1B 24 20 1C 01 56 22 01 40 0D 12 10 0A 00 67 11 00 40 06 29 08 03 00 31 48 4F 40 03 04 44 7C 00 17 64 27 60 01 3A 22 3E 00 0B 32 13 70 00 78 11 01 00 07 51 08 09 00 3E 08 40 50 03 78 44 05 40 13 24 28 30 00 50 24 03 20 05 42 20 1C 00 0F 11 43 20 06 51 0C 1B 00 23 08 61 60 02 60 46 0E 40 03 44 39 08 00 3A 23 48 60 05 62 1C 64 00 5C 11 66 30 06 29 0E 34 00 38 48 73 28 01 58 48 1B 00 0E 04 40 1F F7

TEST PROG1: The display already shows the copyright-text and then "Ready." So I guess this is redundant. I had the PIC preprogrammed with MIOS from doc.

TEST CORE1: yes the crystal is oscillating.

TEST CORE3: I couldn't find any bad solderings.

TEST CORE4: Vdd is between 4.98 and 5.01 on all check-points.

TEST CORE5: ground is alright, too.

TEST CORE6: midi plugs should be connected properly (I think the following tests are also suggesting this)

TEST CORE7: polarity of the diode and the diode itself is working.

TEST CORE8: resistor values are ok.

TEST CORE9: of course I have it cross-linked.

TEST CORE10: I don't notice a lot of upload requests.

TEST CORE11: The cap etc should be fine. How could I test this in more detail? The alive-message I reveive when powering up is correct.

TEST OUT1 and TEST OUT2: I get an "alive-message" when powering up the Core. So this should be redundant.

TEST IN1: all the resistors seem to be fine.

TEST IN2: the midi plugs should be ok since the following tests work.

TEST IN3: the led is lit and flickers when sending a series of sysex-messages.

TEST IN4 and TEST IN5: I haven't tried it but it has to work since the previous test works, right?

TEST IN6: the optocpupler-test works, too.

TEST IN7: I haven't tried it, assuming that the previous test makes sure the optocoupler works. Tell me plese if I am wrong.

TEST INOUT1: I was sending the beginning of the 8581-hex-file and the midi-in is empty so no log availible.

I hope I got it all together now and thanks for the fast reply stryd_one

[TK.]

Which MIDI Interface are you using exactly (manufacturer)?

There is an upcoming TEST_INOUT2, which isn't documented yet:

Could you please check, if the transfer is working when R6 (1.2k resistor) is replaced by a smaller one, e.g. 470 or 220 Ohm. Note: if you don't want to replace the resistor for this check, you could also connect a 1k in parallel (resulting resistance is ca. 500 Ohm... thats ok for this check)

Best Regards, Thorsten.

[stryd_one]

Yeh TK's probably on the money there (no surprise!) :D

Something else you could check, is the traces between the midi I/O and opto/core - a short in the right place will make it go a bit pear shaped...

TEST IN7: I haven't tried it, assuming that the previous test makes sure the optocoupler works. Tell me plese if I am wrong

Yeh man that's what I thought too.... but a couple months back in the chat I spent about 2 hours troubleshooting because I made that assumption, turned out there was a short under the opto that wasn't seen because I told the guy that test wasn't needed . oops! The test isn't meant to check for such things, but this was a lucky break ;) He had a couple of loose pins in the opto socket and had spilled solder on there while fixing it up. Sometimes it's better to do the test and "know" for sure.

Regardless, that's just theory now because TK is gonna be right... we seem to be seeing a few of these lately. If you're using a USB midi interface from a hub, as an alternative you might like to try it direct from the port on the PC, or with a powered hub if you have one. But I think that trick only helped with one person, the others had to do the mod TK suggested.

Good luck!

[/dev/null]

Thanks a lot both of you for the help. I think I really could narrow it down now to a broken 6n138. I ordered myself some parts today so I will report once I tried it out with the new opto.

Btw. my midi interface is an Edirol UM-3EX. The 220Ohm resistor could unforunately not solve my problem but I guess I will keep it in there and hopefully this and the new IC will make it work for me.

[stryd_one]

Btw. my midi interface is an Edirol UM-3EX. The 220Ohm resistor could unforunately not solve my problem but I guess I will keep it in there and hopefully this and the new IC will make it work for me.

If you have a chance to try out someone else's midi interface that might be worthwhile too... As for the resistor if it didn't help then you probably should change back to the original.

Good luck, keep us posted!

[/dev/null]

ok i finally got the new opto and time to try it out. all i can say is that it's working now :)

thanks for your help. these little ICs obviously are not as robust as they could be.

[stryd_one]

Yay!!

Sometimes it's better to do the test and "know" for sure.

See, there's method in my madness ;)