STM32->DIN Sync ?

[Nirokesan]

I thought this was going to be simple, but I haven't got it to work. Is anyone using a DIN Sync out with an STM32?  I searched the forum and most of the posts involve the older core and boosting the signals from 3.3V to 5V.

During my build, I just assumed that I could use a DOUT SR set to clocks and start/stop, wire a pair to a DIN Socket, and be good to go.  That didn't work.  I thought perhaps I had the pin-out of the socket wrong (top two google images for DIN Sync out contradict one another), but swapping them didn't work either.  In fact, one way made the Run light of my 303 come on, and the other way made the selected pattern light flash.  I don't get that.  I traced the signal from the 303 back thru the cable and believe that the correct configuration of the socket (face side) is:

clock                        S/S

    NC.                 NC

               Gnd 

 

I started looking into DIN Sync standards and read that the clock should be a square wave and should have a delay of 9ms for the first step.  Could this be the issue?  I also read in the other threads mentioned above that some people got the Sync to work from the old core using clock lengths of 3-6ms (this must depend on the chosen tempo, not optimal). 

What are my options here?  I'd prefer to use the Clock SR from my DOUT as the Sync sockets are already built into that box, but if I'm forced to connect directly to the STM32, I will.  

Thanks for any advice/shared experience.  Surely someone must have DIN Sync working from the STM32!

[gerald.wert]

Are you using Seq V4?

A8 should be Din clock output

A9 should be Din Sync start stop

It looks like things changed a bit from F1 to the stm32F4 there is not a J5C. The pins that were A8 and A9 went to PC4 and PC5 on the STM chip. On the F4 core these are output on J5B as A4 and A5 are you able to test to see if you have din sync there? I would think you would still want to add a 741 if there is clock there to protect against shorts and put you at 5v. I am guessing it is not there and the software has changed though as they are no longer labeled A8 and A9 anymore.

You should be able to use the SR on your DO too. I am not exactly sure what you are using for DO though. You just need to figure out what SR # it is.  If you are using a quad IIc it does not have a shift register on it you have to supply it. Also if you are using the line driver it does not have a SR connected to the din Jack you could connect one though.  NG can be super helpful to figure out what number a SR is. You can test for it in the console.

Once you find your SR # I believe you will need to set it up on the CV configuration page  and in the hardware configuration file before it will work. Definitely interested in what you find out. I have the din sync jack on my Seq on the midi IIc but have not hooked it up to anything yet. I was in the same situation as you where it is a bit more to it and just do not have the time at home right now to sort it out. I do have an Xoxbox with din sync so am definitely interested!

 

I think if you have the line driver if you go here:

http://www.ucapps.de/midibox_seq_manual_hw.html

and go to the bottom of the page it gives an example configuration that is working.

Edited December 15, 2017 by gerald.wert

[Nirokesan]

44 minutes ago, gerald.wert said:

Are you using Seq V4?

Thanks for your reply, Gerald.  Yes, I'm using SeqV4 with latest software. 

 

51 minutes ago, gerald.wert said:

On the F4 core these are output on J5B as A4 and A5 are you able to test to see if you have din sync there?

I cannot test J5B without disassembling my sequencer enclosure, and I'd prefer not to if there is another solution, especially as there is no room on the rear panel to add any more ports. 

 

53 minutes ago, gerald.wert said:

if you have the line driver if you go here:

http://www.ucapps.de/midibox_seq_manual_hw.html

and go to the bottom of the page it gives an example configuration that is working.

my hardware and config file are all set up as you can see in this GIF : https://i.imgur.com/BFUeDRB.gifv    The right-most column of sockets and LEDs is showing the SR assigned to clock and start/stop.  You can see that I have different clock divisions set for the separate clock outs, and all Start/Stop outputs are held "high".   I'd prefer to use this DOUT SR for DIN Sync to my Roland gear since I can supply different clock divisions to each machine, and because the case is already built with the provision for two DIN ports.  

The questions are....why is this DOUT SR failing to drive my 303, has anyone else got this working with STM32 and DOUT, and how?

[Nirokesan]

https://i.imgur.com/7CDCKkM.gifv

DIN Sync out via the DOUT Clock+S/S SR# works fine with my 606, however, I needed to use a trigger length of 2ms at least otherwise the 606 played in some broken sort of half-time.  The above GIF shows a trigger length of 3ms (maybe 4ms) and a tempo of 128 bpm.

So the failure I experienced before must be the fault of my 303.   I've tried with two different cables that have both been tested for continuity.  I just did a test between my 606 and 303 alone and discovered some irregularities.  Both units play fine by themselves.  When I connect the two together, the clock rate on the 606 also flashes on the 303 showing that the clock is getting through the socket and interrupting the internal (this is different from the connection to the DOUT as the MIDIbox clock is not running until you press play). When I press Run on the 606, the clock on the 303 stops flashing but the Run light is on, no sound from the 303 (just like with the MIDIbox). Obviously I need to look at the schematics and take this issue elsewhere.

(edit: looked at schematics, a bad pin 3 solder joint on the 303 socket could be the problem. Will examine)

 

Bottom line is.....DIN Sync from the DOUT works just fine if the trigger length is set larger than 1ms (increase if necessary), I can keep my breakout box design, and I can use different timing divisions.

Edited December 17, 2017 by Nirokesan

[Nirokesan]

I opened the 303 and tested the socket connections which all seemed fine, but there was no solder on Pin 2.  Maybe it broke free from the pad and pin? I don't know.  Added solder there and reflowed the other pins.  Works fine now, trigger length is 3-4ms here, can't remember