Are pins 13 and 14 of U5 joined together?
You mean bridged with soldering?
I mean they should be joined on the PCB, 13 and 14 are joined to form a voltage follower, if they are not connected then I don’t know what the results will be, just meter the pins on the chip when its in the socket and see if they are shorted together 0 ohms.
Yes, they are indeed , 0 ohms on the multimeter
Then I’m at a bit of a loss, next thing up the chain is the PIC chip itself. I guess unless you have a scope then looking at the PWM output is not easy.
Hi Craigyb, you were right!!! one of my LM324N is defective, new chips arrive monday…
Have a good day
Thanks for your help
Mike
2 Likes
Glad to hear it, I couldn’t really get my head around the fault other than being the lm324 or possibly the level converter. I built a similar design as a test shortly afterwards with 28 channels and it works perfectly with 4504 and lm324.
1 Like
I really appreciate all the help
Have a great day Craigyb ( Greg is my brother 
)
See you
Mike
1 Like
the level converter is the 4504? Right?
Correct, converts 5v to 10.6v that allows headroom for 0-10v PWM to send to the RC filter and then to the follower input of the lm324.
1 Like
Ok thanks! I’ll wait for the new chips and I’ll come back with the results…
Kind regards
Mike
WOW! 28? that is quite a big number of channels indeed!!! 
Well it is 8 channel polyphonic, so that’s 8 pitch CV and 8 velocity CV channels, plus 4 dedicated to pitch bend, modulation, aftertouch and breath control. Then there are 8 gate outputs, the polyphony can be reduced and if you do that each note frees up two outputs and a gate output to be used for other CV controls and drum triggers etc. it’s still in development, but I’m happy so far.
Well, I just love what I just heard so far. Quite an impressive beast. Keep on the good work:wink: