MidiDAC Slide

[welhamc]

Evening wigglers

Just a quick question about my the MidiDAC's slide function. It seems that there may be some issues with the Auto setting of the slide circuity on my module

On auto mode, pressing a key say, two octaves higher whilst holding down the first causes a slide as anticipated (the resulting note is not in tune to the first note, and releasing the second causes the initial note's pitch to be shifted up a few semitones but this is probably just because the module could do with a recalibration). However at higher slide times there is still an audible pitch envelope effect with discrete key presses, which is more noticeable the further the two notes are apart, and the higher the slide time. The slide circuitry with the switch set to "on" acts as it should albeit with a similar shift in pitch to the compared to the auto circuit (again I imagine a calibration issue)

I've had this module for nearly 5 years and have noticed this since I built it, but assumed this may just be how the circuitry functions. It seems about time to double check with the experts!

[Synthbuilder]

... the resulting note is not in tune to the first note, and releasing the second causes the initial note's pitch to be shifted up a few semitones but this is probably just because the module could do with a recalibration.

There is no calibration that affects the pitch differently of notes under slide compared with notes without slide. The new note should be nearly exactly the same as one played without glide. There is a small difference, usually less than a cent, because of offset current of the op-amp. If you do have a significant difference then there is something wrong.

Firstly, if you have an issue 4 board, make sure that D1 is not fitted. There is a place on the board for D1 but it should be empty. The next thing to do is to test for unwanted offset current in the op-amp U4. Play any note without glide and the glide pot set to its minimum. Listen to the note and monitor, if you can, the pitch of the note with a tuner of some sort. Turn the glide pot up and see if the pitch changes. If the change is significant, ie. over 5 cents, then there the op-amp is probably faulty and needs replacing.

The other thing to do is replace the 4066 chip. Remember that this is a 4066, ie. CD4066 or MC14066, and not 74HC4066.

Tony

[welhamc]

Hi Tony

Thanks for getting back to me so quickly

Unfortunately I do not have a spare AD712 or a CD4066 to hand, but I swapped out U4 for an LF412 instead. However this had no effect on the unit, and the slide time pot acts much the same as a "coarse" pot control would on an oscillator.

I'll have to try replacing the CD4066 and seeing if this works, otherwise it will probably be a tear down and troubleshoot type job

[welhamc]

For reference, the pitch shift of a note at min slide to max slide time is nearly two octaves

[Synthbuilder]

For reference, the pitch shift of a note at min slide to max slide time is nearly two octaves

Whoa... that's horrible. That suggests something wrong with the 4066. I think.

A 4016 can used in place of the 4066 too - at least just to check to see what the problem is.

Tony

[welhamc]

Thanks for the help Tony, I've ordered some 4066s, will keep you updated!

[welhamc]

Success! Swapping out the CD4066 has solved the issue, and now have fully functional slide circuitry. Thanks for your help Tony!

[welhamc]

It looks like a similar issue has come back... I'm not having the issue with the slide pot acting like a tune pot anymore, but playing a note (note A), playing another simultaneous legato note (note B) and returning to note A still with legato results in a pitch of note A +40 cents give or take

Is this a problem with U4?

[Synthbuilder]

I'm not sure what's wrong to be honest. It sounds like the 4066 has gone bad but that shouldn't happen so quickly after the last one died. Just to make sure I understand the problem, is the unwanted pitch shift only when playing legato and the slide switch to 'auto'? What happens when the slide switch is set to 'on', is it still out of tune when playing legato, or is it now always out of tune?

One other thing to check is to remove U5 completely and try using the midiDAC as before. You should not get any appreciable slide but altering the slide time should not produce an obvious pitch shift. If it does then U4 is at fault.

Tony

[welhamc]

I reshuffled my modules and reseated all cabling as part of that and now its functioning as normal... modular can be elusive sometimes

[Synthbuilder]

I reshuffled my modules and reseated all cabling as part of that and now its functioning as normal... modular can be elusive sometimes

I think I know what caused the problem. I think it's the midiDAC's LEDs and your power cabling.

The earlier midiDACs drew additional current to light the LEDs. When an LED is on, it takes an extra 4mA and that current is pushed back down the 0V connections. In issues 1, 2, and 3, the LED current went down the digital 0V on pin 3 of the power connector, and the analogue 0V on pin 2 was undisturbed. However, this occasionally caused the DAC and 4066 to fail if the user connected the power cable while the power was on, or there was a fault on the cable or distribution. So in issue 4 the two grounds (0V) were connected together. Now the analogue ground and digital ground were using the same pin 2 connection. Thus any current taken by the LEDs when lit had to go back down the common 0V. This causes a small detuning of the analogue CVs when either or both of the LEDs are on.

The amount of detuning should be minimal, certainly less than 1 cent. But if the power cable to the midiDAC are too long or too thin then you will get a small but not insignificant voltage drop across them when the LEDs are both on. So I think your problem was caused by the power connector either not making a good connection with its header, or that the cable used to power the midiDAC was too long.

Later issue 5 midiDACs resolve this problem by using a different LED driver circuit similar to that used on most of my other modules with LEDs. However, the use of shortest and thickest power cables is always recommended.

Tony

[welhamc]

This makes perfect sense - thanks Tony

I've had nothing but headaches using the MTA push tool to make power cables. I always continuity test them before use, but it seems that after time/moving modules around sometimes they can become a bit temperamental

Perhaps I will go down the molex route for my power cables from now on

Chris

[Synthbuilder]

I've had nothing but headaches using the MTA push tool to make power cables. I always continuity test them before use, but it seems that after time/moving modules around sometimes they can become a bit temperamental

The key is having the right cable size for the chosen plug. But, yeah, I'm not keen on them either if you only have the smaller hand tool. I find the 0.1" ones even worse though, which is why I now always use Molex or equivalent for the 0.1" interconnects. It's cheaper too.

For the power cables I often use bigger Molex sockets, but with the MTA headers on the board to keep modules compatible with MOTM's MTA specification. Molex cables work well into MTA headers, but I don't think you can do the opposite.

Tony