MIDI in not working [solved]

Hi,

I just finished assembling my Shruthi kit. I don’t have a MIDI keyboard so I’m trying to trigger it with a x0xb0x. I’ve verified the x0x is sending MIDI out by recording a MIDI track in Protools.

-I can hear the test sound fine.
-When I had the LCD installed I didn’t see a note indicated on the screen. I’ve since removed it for troubleshooting.
-I’ve reflowed all my solder points on the ICs and connectors.
-As suggested on other threads, I put a scope on pin 6 of the 6N137. I can see pulses.
-I don’t know if this is strange, but the chip I received in my kit for IC4 is marked as ATMLU012. I searched for a datasheet for this and couldn’t find one. Is this the right chip compatible with AT24C64C?

Any help would be appreciated.
luther

Holy never mind!

This is my first experience with MIDI and I knew nothing about channels. I changed the channel on my x0x and everything is working fine.

“Welcome to the ShruthiVersum™” :wink:

@ luther_r: as I still have no clue on my problem with midi in, do you mind try to use a led instead of the scope (as suggested here) and see if the led lights up? I don’t have a scope and I don’t want to buy one and use it anly for 5 minutes in my entire life… led is not working for me but I cannot say it’s because of 6N137 failure or something else. Thank you very much.

Ok I just tested this for you. I sent a repeating sequence into the Shruthi via MIDI. I put the LED between pins 6 and 8 of the 6N137. It lights but I can’t detect any flicker from the MIDI data. I also tried a 1k resistor in series (smallest I had lying around). That dimmed it, but it still looked steady to me. Sorry, you’re probably have to get a logic probe ($25) or find somebody with a scope.

@muldee: maybe it’s the same problem I had - bad solder joint on the GND pin of the IC socket?

I spoted a few GND Problems on some boards (the LCD Trimmers GND Pad and obviously (other Thread…) the Opto’s PIN 5…), maybe a Hint to change the Thermals on a future Revision?

I was thinking of switching to a hatched/grid ground plane instead of a fully filled one to minimize this problem, but couldn’t find any information as to how this would solve the problem. Maybe this gain by a factor of 2 in copper area won’t change much…

I think its not the copper Plane itself, maybe the small lines that connect the pad to the GNDplane are to big and therefore suck to much heat from the Solder when using a weak soldering Iron (like me: I usually use a 8W Needle (intended for SMT-use) , but for some GND connections i cant get the solder flowing so i use for these a big ugly 20W thingy). So I suspect the diameter of the lines is to big, as you cant change the thickness of the copper layer (you chose “thickest” did you?) just try to make the lines thinner in the next revision of a board to test…

@luther_r: first of all thank you for your willingness. What you wrote is generating in me a lot of questions: why bridging the pin 6 and 8? I supposed it had to be something like the led anode on Ground (pin 5) and the led cathode on the Output of 6N137 (pin 6). I tried every possible bridging with a led but the one you tried… a led connected to a 220 ohm resistor in these configurations:

  • anode on pin 5, cathode on pin 6: the led isn’t lit and nothing happens on midi activity
  • anode on pin 5, cathode on pin 7: the led emits a very small quantity of light
  • anode on pin 5, cathode on pin 8: the led lits with a very bright light

Have these info any meaning?
By the way I ignored the existence of the logic probe, thanx for the hint.

@Pichenette: the ground pins on the IC are the 11 and 31, right? (looking to the schematics…)
edit: never mind, I saw your post on midi not working, you meant the pin 5 of the opto…

Since the MIDI opto out is at 5V in absence of signal, and drops at 0V in presence of signal, putting the LED between pin 5 and 6 will give a lit LED in absence of MIDI data and a dimmed LED in present of MIDI data. Putting the LED between pin 6 and 8 should give the opposite - by default, no difference of potential across the LED, but a 5V potential difference only when there is some MIDI messages.

Anyway, MIDI messages are so bursty that I don’t believe they can be seen with a simple LED.

So, as I experienced no light with a led on pin 5 and 6 in absence of midi I suppose something is not normal… I mean, the midi opto out is not at 5V…

Anyway, as soon as my finances will let it, I’ll go and buy a scope.

Same here. If you get one with x+y in you can also use it as Korrelationsgradmesser so if you Girl asks whats this for you always may claim its for checking important things like the monocompatibilty of your first #1 Hit :wink:

well, as my first scope I was thinking about the DSO nano V2, wich is quite unexpensive compared to the others and I think it is sufficient to a noob like I am. I don’t know if I can use it as a “Correlation meter” but I don’t mind… mine will be a great #1 hit, even if it’s not monocompatible :D.

@luther_r: would you mind put [solved] in the title?

Wait a minute!
Since the MIDI opto out is at 5V in absence of signal, and drops at 0V in presence of signal putting the LED between pin 5 and 6 will give a l LED in absence of MIDI data and a dimmed LED n present of MIDI data. Putting the LED betwee pin 6 and 8 should give the opposite – by defaul no difference of potential across the LED, but a 5V potential difference only when there is some MIDI messages

If I check voltages on 6N137 in absence of midi I got this:

  • com probe on pin 5, the other on pin 6 = 0,23V
  • com probe on pin 5, the other on pin 7 = 4,96V
  • com probe on pin 5, the other on pin 8 = 5V

Any idea about this?

The voltage on pin 6 is not normal. You should read 5V there, and a small drop (say 4.8V) when there is intense MIDI traffic.

Ok, it’s time to desolder the lcd. Wish me good luck and crossfingers everyone.

Yo Muldee!

Cut the pins with a DREMEL or something in hole to remove the LCD and then cut the plastic between the single pins and desolder them one by one. Theres next to no chance to desolder the 16 Pins in one piece…