FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 

Dual TM3030: all working ! ( was : debugging )
MUFF WIGGLER Forum Index -> Oakley Sound Systems  
Author Dual TM3030: all working ! ( was : debugging )
EDIT: Now all working and sounding awesome, I'll leave the original post below as it may be of interest to other people who mess up like I did smile

So I finally found some time to finish my dual TM3030 build.
One seems to work well ( still need to calibrate the VCF and VCA but looking good appart from some bleed on accents at max settings - guessing this goes away with VCA calibration ), the other one doesn't respond to MIDI.

Having a working one to compare to should help but does anyone have a suggestion on where to start ? Maybe the optocoupler ?

Note that I was stupid enough to initially put the 5V regulator upside down on both PCBs which resulted in R14 going up in smoke on both units when I first turned the power on. very frustrating
However looking at the schematics it doesn't look like anything else should have received a high current.
Both units'LEDs behave identically when powering on so I'm guessing that the PIC is fine.
Forgot to say that the slide LED is on on both units after powering up.
On the one that works it goes away when I play midi notes above C1 ( or around there, not 100% sure), and goes back on if I play notes below C1.
So I'm guessing this is just a quirk of the firmware ?
Still pretty stuck on this one.
So the opto is fine.
In fact the PIC is actually sending midi out which would confirm that it's fine too.

However it's not sending Gate out.
Tried resetting the channel to 1 by sysex just in case but that made no difference.
I even tried replacing the PIC with the one from the working unit and that made no difference.

The only suspicious thing I can see is that there is some bleed ( around 300mv spikes ) from the midi signal to the PIC +5V feeds( pins 1 & 20 ) on the non working one but not on the working unit.
However I can't see this bleed if I probe the +5V3 hole on the PCB ( probably decoupled out by the caps ? ).
Thus I'm wondering if it's something around the chip or the chip itself that is giving the bleed, and whether it has anything to do with the lack of gate seriously, i just don't get it
So while probing around it became clear that I am missing a resistance somewhere, on the working unit I can measure 22.9K between the probe point next to the frequency pot ( which seems to be a +5V trace that goes to all the LEDs and pin 1 of the PIC ) and the +5V3 probe point whereas on the non working one I can measure some low static voltages ( half a volt ) between these two points when everything is off which clearly comes from caps missing a discharge path.

Problem is I can't even see where that resistance is supposed to be on the schematic ? seriously, i just don't get it
I checked all the 22k on the PCB to start with and they seem fine.
Not quite sure what to suggest at the moment.

Do watch out for resistances across bits of directly unconnected circuitry. The resistance reading will often be via multiple paths including the power rails. It's not a great fault finding aid unless you have a dead short.

The +5V and the +5V3 are not really connected. The +5V3 is the analogue supply that comes from an op-amp output. The +5V comes from a 78L05 regulator chip.

What is interesting is this spiking you are seeing on the +5V line. This could be that the gate output is actually shorted somewhere. It's perhaps trying to go high but instead current is being dumped into a short to ground.

With the power off, what's the resistance between pin 16 of the PIC and GND? It should be quite high, 10K or above.

Thanks a lot for your help Tony, appreciate it !
Just before your message I realised that the 78L05 on the malfunctioning PCB had a partial short between Vin and Vout ( around 2.5K resistance ), so no doubt it's been toasted by the initial reversal.
Luckily I've just found a spare one.
I measured the resistance between pin 16 and Gnd after removing it and it's very high ( 10M range ) on both pcbs.

I take your point and agree regarding measuring resistances, I'm only using that as a very rough comparison between the circuits.
I think there's something else that's not right / toast in the circuit so I'll keep looking into it before putting the regulator back.

Thanks a lot for the info regarding the +5v3 I had completely missed that, makes a lot more sense now.
Darn, so I realised that the 15K I was measuring between the digital +5V and the ground on the working PCB was the output resistance of the 78L05 and installed the replacement regulator on the other one.

I was sure that was the end of it and...the ripple on the +5v is now gone but still no gate out :(
OK finally all sorted !
Turned out that the opto-coupler was actually toast - even though it *looked* like it was transmitting data, it was actually spewing out garbage ( note to self: do not try interpreting MIDI data on a scope ).

This synth sounds absolutely awesome and having two in a rack is great, thank you Tony for this brilliant design.

Altogether this took a loooong time to build, in large part due to too many mistakes I made.
I did learn a few things though, like double check the orientation of parts when the ones I bought are not in the same package as the silk screen indicates oops
Also I'm now quite familiar with the schematics and the PCB layout lol

A couple of pictures to finish off.
The build was quite a bit neater before I started taking things apart, so I'm not super proud but at least it works !
Don't mind the regulators one of which was given new legs and the other I ended up cowboy-SMT mounting as I was sick of soldering it on and off.
Notice the big soviet union polystyrene caps 8_)
Paradigm X
congrats man!

Some sounds are always nice so here is a quick demo sequence :
MUFF WIGGLER Forum Index -> Oakley Sound Systems  
Page 1 of 1
Powered by phpBB © phpBB Group