MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Dual TM3030 considerations
MUFF WIGGLER Forum Index -> Oakley Sound Systems  
Author Dual TM3030 considerations
terjewinther
I am going to build two TM 3030s inside one rack, as others have.
What kind of considerations do I need to address?
I will have only one on/off switch, and split the power to both units.
As for MIDI: can I just split the MIDI in to both units, and then use MIDI through out from one of the units, or is it better to make a chain as MIDI in - unit1 - MIDI through to - unit2 - MIDI through to - MIDI out?
I will also make a simple mixer/adder and make a stereo output, so that if only one output is used, you have a mono signal from both units. As long as a buffer the signals in and out, nothing fancy needs to be done?
Anything else?
Synthbuilder
The easiest way to do the midi thing is simply to connect the midi in of the second unit to the midi out of the first.

But you could try connecting the midi data line (pin 18 of the PIC) of the first TM3030 directly to the midi data line of the second. You'd need to make sure that both U2 and R4 are not fitted to the second TM3030.

You can't simply split the midi input signal to both TM3030s. Although it may work the midi spec doesn't allow for this.

As for mixing - the TM3030 already has the same output mixer circuit as the original 303. It's just that I haven't fitted the extra 100k & 1uF cap that the original has. If you cludge these two components onto the first board. You could use this to sum the second TM3030's output into the first by using the NC lug of the second's output socket.

Tony
terjewinther
Great - thanks!
I will use the MIDI chain then.
Thank you for the mixing tip!
terjewinther
I have finally been able to complete the dual TM3030. Wow, does things take time. Had to scroll several pages down to find this thread again.
Both of the TM3030s works and sounds great, and I will start calibration tomorrow. There are two peculiar things that I need to debug:
- One of them has PIC version 1.1, while the other one has version 1.2, and somehow version 1.2 seems to not catch up on all gate signals. Maybe 1 out of 10 keystrokes does not trigger it, and it sometimes laches up the glide as well. What were the version changes of the programmed chip?
- One of the TM3030 (the one with PIC version 1.1) has a highpitched fluctuating sound at the output, sort of a grainy highpitched noise. Rather soft (downed out whenever there is a sound from the synth), but still clearly audible, and it is present at all times, also when there is no gate signal present. It sounds like a digital signal bleeding through, but I might be mistaken. Any idea where to start debugging?
Even though it is not calibrated yet, it sounds great in stereo! And the mono/stereo auto switching/mix works like a charm.
Synthbuilder
terjewinther wrote:
One of them has PIC version 1.1, while the other one has version 1.2, and somehow version 1.2 seems to not catch up on all gate signals. Maybe 1 out of 10 keystrokes does not trigger it, and it sometimes laches up the glide as well. What were the version changes of the programmed chip?

As far as I know there were no changes to the midi handling firmware.

I think it's the opto coupler or maybe the way the midi is wired up. Is it the second one in line that has the note drop out problems? Maybe swap the opto-couplers around and see if that changes things. If so it maybe best to replace the slower one with a new one from a different make. I have always had good results from Fairchild parts.

terjewinther wrote:
One of the TM3030 (the one with PIC version 1.1) has a highpitched fluctuating sound at the output, sort of a grainy highpitched noise.

OK, this may be the cause of the other problem because this sounds like midi is leaking into the audio. Does it stop when you disconnect the midi lead on the unit? If so, this would suggest a problem with the grounding of the units.

The input to the second TM3030's opto coupler should not have any ground connection. Both wires should just come from the output of the first TM3030's midi output.

Tony
terjewinther
I tried swapping the opto-couplers. No change.

Yes, the digital leak disappear when MIDI is disconnected.
MIDI out from the first is connected to MIDI in on the second, using only the two active leads. Ground is not connected.

I tried connecting the middle ground pin to ground (chassis), but still no change. What else must be done to solve grounding problems?


Synthbuilder wrote:
terjewinther wrote:
One of them has PIC version 1.1, while the other one has version 1.2, and somehow version 1.2 seems to not catch up on all gate signals. Maybe 1 out of 10 keystrokes does not trigger it, and it sometimes laches up the glide as well. What were the version changes of the programmed chip?

As far as I know there were no changes to the midi handling firmware.

I think it's the opto coupler or maybe the way the midi is wired up. Is it the second one in line that has the note drop out problems? Maybe swap the opto-couplers around and see if that changes things. If so it maybe best to replace the slower one with a new one from a different make. I have always had good results from Fairchild parts.

terjewinther wrote:
One of the TM3030 (the one with PIC version 1.1) has a highpitched fluctuating sound at the output, sort of a grainy highpitched noise.

OK, this may be the cause of the other problem because this sounds like midi is leaking into the audio. Does it stop when you disconnect the midi lead on the unit? If so, this would suggest a problem with the grounding of the units.

The input to the second TM3030's opto coupler should not have any ground connection. Both wires should just come from the output of the first TM3030's midi output.

Tony
terjewinther
Ah... Found the error:
I removed the special two-to-one mixer line, with the extra 100k and 1uF cap and all problems vanished. Of course now the mono/stereo options vanished as well.
terjewinther
And here is the final solution:
I carefully inspected the original TB-303 output mixer, and found that the additional extra "mix in" input needs to be grounded when not used. So I tried that, and of course it works!

But: I only have one NC lug on my switchcraft jack. How do I make this extra mix input grounded when the input is used?

Hmmmmmm.........
Synthbuilder
Are the extra resistor and cap mounted onto the TM3030 board? If so this is the best for them. You need to minimise the distance between those components and the summing node.

Is the wiring to the extra components done with screened cable with the screen connected to ground at one or both ends? If not it may be worth a try doing this.

The other thing is perhaps to add a 10K resistor from the NC lug of the output socket to ground. This will gently tie the floating input to ground and may reduce the pick up of unwanted signals.

Is your midi wiring as short as it can be? You may want to consider using screened cable for your midi connections too if they run close to the output sockets.

Are you still getting dropped midi notes? That does sound like an opto problem.

Tony
terjewinther
Synthbuilder wrote:
Are the extra resistor and cap mounted onto the TM3030 board? If so this is the best for them. You need to minimise the distance between those components and the summing node.


Yes. Originally I mounted them on the output sockets, but moved them onto the TM3030 board. They are now very close to the node between R101 and R102.

Synthbuilder wrote:
Is the wiring to the extra components done with screened cable with the screen connected to ground at one or both ends? If not it may be worth a try doing this.


I can try that. I am using standard cable now (7/0,2mm if I remember correctly).

Synthbuilder wrote:
The other thing is perhaps to add a 10K resistor from the NC lug of the output socket to ground. This will gently tie the floating input to ground and may reduce the pick up of unwanted signals.


Good point. I was thinking about that, but considered using a larger value, like 100k, or 470k or even 1 Meg. I could try with various values with clips before I solder it on.

Synthbuilder wrote:
Is your midi wiring as short as it can be? You may want to consider using screened cable for your midi connections too if they run close to the output sockets.

Are you still getting dropped midi notes? That does sound like an opto problem.

Tony


The MIDI wiring is as short as it possibly can. The board-to-board wiring follow the edge of the boards. The Midi wiring is not that close to the output sockets, but there is of course some proximity. I am still getting dropped notes on the second unit. I swapped the opto-coupler, but still the same situation. I believe I have some additional opto-couplers I could try. I could even swap the midi order, so that Midi input goes to unit 2 first, and see if that changes anything.

Thanks for the tips - I will give them a try and see what happens.
terjewinther
Ah - hahaha.... hyper
Tony Allgood - you are just the greatest synth mind around!

I changed the cable from the NC lug of the second TM3030 to a screened cable, connected one end of the screen to ground, and again placed the two extra components (1uF cap and 100k resistor) on top of the first TM3030´s mixer junction. I also took a 12k resistor I had lying on my desk from the NC lug to the ground tag, as you suggested. And guess what: now it works perfectly! No hum, no noise, no highpitched interference. Just quiet when there is no sound, and full dual TM3030s in stereo or mono whenever I want to.

I also swapped the opto coupler on the second TM3030 to a similar type as on the first, and then changed the internal MIDI connection from the first to the second to a screened cable (because it did cross the output cables, as you had suspected), using the ground tag of the MIDI out on the first TM3030 for the screening of the cable. This drastically improved the performance of the second TM3030. There is still the occasional difference, but now it is more like the second TM3030 is missing, say, one of every 50 to 100 keystrokes. More than good enough for my use. And analog music could use some variations now and then anyway. hihi

Again: thanks a lot for great products, and for the brilliant support.

Terje Winther
MUFF WIGGLER Forum Index -> Oakley Sound Systems  
Page 1 of 1
Powered by phpBB © phpBB Group