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

Paia 2720 vco & sine/pwm calibration nightmare
MUFF WIGGLER Forum Index -> Fractional Rack Modules  
Author Paia 2720 vco & sine/pwm calibration nightmare
I've been struggling to calibrate the vco and sine/pwm modules for some time now, particularly with the pwm range and tracking. I managed to get a reasonably stable pulsewidth across the keyboard but still gets thinner at the higher notes. Replacing R8 with a 10k trimmer had no.effect. I'm also struggling to understand what the two trimmers for the triangle waveforms do for I don't notice any change in the tone/waveform. It's even worse when I try to calibrate the sine/pwm module. I can't get the pulse range near halfway across the 0-5v.bias without trimming the sinewave to barely audible, and when I get sine sounding righr, the pulse range is off again. This is getting frustrating real fast very frustrating what am I missing?
This is the 2720-2 (or 2720-2A) VCO, correct? There should only be tri/pulse/saw outputs, no sine.

Here's a doc with calibration info, on pages 4 & 5, in case you haven't seen it:

Looks like there should be 4 trimmers total...
RANGE & ZERO are calibrated together to set the tracking & range of the module.
TRIANGLE trims the triangle shape.
PULSE DURATION is used to set the narrowest possible pulse when PW knob is set to min.
Also, some info about calibration, etc. here, courtesy of Dave Brown:
Yes, the vco doesn't have sine but it's companion, the sine/pwm module does produce a sine wave and cv controllable pulse. On the vco, the trimmers for the triangle wave seem to do nothing, and the pwm doesn't track right. Switching R8 with a 10k trimmer had no effect.
My biggest problem right now is the sine/pwm module. When I get the sine wave sounding right, the pwm level is extremely narrow. When I get the pwm to a somewhat usable level then the sine wave is too quiet. Also, like with the vco, I can't set the pulsewith too narrow or it will fizz out halfway across the keyboard. Thanks for the links by the way, I'll check them out after work.
Ok, so I finally got around trying to calibrate the vco and sine/pwm modules, following exactly the instructions for calibrating by ear (I don't own an oscilloscope), and repeatedly ran into the same problems:

On the vco: when trying to set the triabgle waveform to the mellowest sounding tone, the assiciated trimmers have no effect what so ever and no change in the triangle's sound can be detected. When trying to set the pwn rate, when adjusted according to the instructions, at the lowest setting (narrowest pulse wave) I'm left wirh only two keys left to play. The higher (wider) i set the pulse wave, the more keys I have to play with. The pitch seems to affect the pulse width, and the same effect can be observed with the vco connected to bias.

On the sine/pwm module, calibrated to instructions, produces the same issues as in the vco pulse wave, and the sine wave becomea almost inaudible with the pulse range set as per instructed.

Did anyone else have to fight with this? What did you do to fix both modules? Or are my pwm issues a fault in the design? And can someone point out what could cause both trimmers for the triangle wave to fail?

PS: the vco is s 2720-3a, hand drawn circuit and parts layout.
diophantine wrote:
This is the 2720-2 (or 2720-2A) VCO, correct? There should only be tri/pulse/saw outputs, no sine. Here's a doc with calibration info, on pages 4 & 5, in case you haven't seen it:

The VCO in that article (1973 article in Radio-Electronics about the construction and use of the 2700 series synthesizer) is the 2720-2. You should NOT be using any calibration information from that article unless you are sure your VCO is a 2720-2. PAIA released the improved 2720-2A oscillator to replace the 2720-2. The circuit board is different, the 2A tracks better. The manual for the 2A is here: -> 2720A Oscillator

bloodyhell wrote:
PS: the vco is s 2720-3a, hand drawn circuit and parts layout.

Did you mean 2A, or is this someones home brew improved circuit?

If you want to scattershot it, you could try replacing some or all of the transistors in the VCO (same value) and see if that makes a difference. There are only twelve, and two of those are 2n2712 noise transistors. Those transistors are likely over 40 years old now, perhaps one got damaged by a former owner.

And if you haven't already done so, you could contact Scott Lee at PAIA for technical advice. PAIA still responds to questions about vintage equipment, and there is no charge.
Thanks for the reply, I'll try that when I get back home. It's a 2a, my bad. If the 2a tracks better then there must be something very wrong with mine. Were there any other versions for the sine/pwm module as well? Also, any ideas why the trimmers for adjusting the triangle waves don't work? I already replaced a bunch of caps to no avail. I guess the transistors are next.

PS: I took a peek at the 2720-2a manual and I'm positive that I used those very calibration instructions.
John Simonton designed these modules in the most economical way possible. From what I have read, the sine and voltage controlled pwm were left out of the 2700 series VCO for cost reasons. I have not heard of any other versions of the sine/pwm module. When the 4700 series oscillator was released, the sine wave and pwm (manual and voltage controlled) were built in and didn't require a separate module.

There are some corrections by PAIA to the 2720-2A manual that may or may not have made it into your VCO. These are:

- Change resistor R-29 to 150k (brown green yellow) Was 170k (yellow violet yellow)

- In this 2720-2A Voltage Controlled Oscillator kit the 2n4871 Uni-Junction transistor and its selected resistor have been chosen to operate properly with a 5.6v zener diode in the negative supply regulator. Accordingly, make the following changes in this instruction manual: Page 3 - last step Change D4 from 6.8v to 5.6v. Page 12 - schematic Change D4 from 6.8v to 5.v

The trim pots in the PAIA circuits probably had a looser tolerance of 20% or more. Might be worth checking their value (I've had trim pots go bad), and/or temporarily replace with a modern pot. Also check resistor values. These aren't precision circuits but it is possible there is enough looseness with the values that it throws the triangle wave circuit out of whack. Another thing to check is that the transistors are installed correctly. A 2n3904 can substitute for a 2n5129 but it is installed differently than the silk screen.
Thanks a lot! That'll give me a starting point on troubleshooting without doctoring around too much.
What is the watts and tolerance supposed to be? I understand that the brown resistors are metal oxide but the only ones I have I got from Radioshack ages ago, and they appear to be carbon? (beige body) - can I use those kind or should I stick with the kind used on the vco board?
The important thing is that the module was assembled with the correct value resistor. It is easy enough to check the color code of the resistor against the color code in the parts list.

All the resistors, if original, should be carbon film. The tolerance could be as much as 20 percent from that time period. Manufacturing is much improved, and now days, you will probably find carbon film resistors with a tolerance of five percent. The resistors could be 1/8 watt or 1/4 watt, probably 1/4 watt. You generally eyeball it for the size to determine the wattage, if you have no other resources like a specification in a parts list. You can purchase carbon film or metal film resistors for any replacements, it won't make a difference in these circuits.

As for resistor body colors (not the bands) see link-> Component of the Month: Resistors. My understanding is that there are some conventions of body color but it can vary by manufacturer.
MUFF WIGGLER Forum Index -> Fractional Rack Modules  
Page 1 of 1
Powered by phpBB © phpBB Group