VCO problem

[sator]

Hello,

First, thank you Yves for creating and sharing such magnificent modules! I'm building a synth to use on stage. The first module to go through testing was the VCOs (I've built 3 of them). Here are the scope views of waves:




As you can see saw is somehow inverted and square is, well it isn't square . No problem with sine or tri. Checked all VCO modules, they all have the same problem. I went through parts list and checked all of them to make sure, couln't find anything wrong (obviously there is). Here is one of the modules:



Power inlet of the board has -14.92 / +14.94. Output of voltage regulators are -13.51 / +13.56. Tried bypassing the regulators, nothing changed. There is no subtitutions for parts (only my silver mica cap is 5%). Any help is appreciated.

[blue hell]

You're most likely just looking at sound card artifacts (that is when you are using a software scope, which is what it looks like in the pics). If so, due to AC coupling the square looks like it does and likely the input inverts (not uncommon for sound cards) which would flip the saw upside down.

[sator]

Thanks for the fast reply. I have built MFOS VCOs before and calibrated the same way I'm trying now. I didn't have this problem then. But I'll definitely look into buying a scope.

[yusynth]

[sator]

Thanks Yves for the reply. I've mingled with the sound card connection and voila! Everything is working!




Made a reverse connection for SAW shape (i.e. connected ground to signal input and viseversa). SQR still seems a little odd but it sounds right I guess. (By the way all scope images are taken at 112 Hz)

Cheers..

[ark]

The square wave doesn't seem odd to me -- it's what you get when you run a true square wave through a high-pass filter with a very low-frequency cutoff. So probably your sound card has some kind of DC blocking capacitor that acts as a high-pass filter with a cutoff of 10 Hz or so.

[tazer]

Hi

may i say thanks for all your amazing info! Have built a few of your modules with great success. Building a vco at the moment, when i fired it up there was a short circuit somewhere and i got smoke off one of the 10 ohm resistors. Cleared the short circuit and tried to power on and nothing. Tried to change the chips and the smoked resistor and nothing, unfortunatly my skills with a multi meter are very basic and it was a home etch so the problems could be plently i dont really know if it worked before the big fry up or not.
But while looking at the 2 component layouts for bc547 vco, one black and white the other green there is a discrepancy with the diodes at the bottom right corner. There is 2 diodes side by side above the 47k trim pot, on the black and white comp layout they sit one way and on the green layout they face the other way. I tried both ways with no luck but it would help if i knew which way was the right way.
keep up the good work!

[yusynth]

[tazer]

thanks for your speedy reply, will try and follow your instructions about finding the voltages and let you know how i get on and if any discrepancies-cheers

[tazer]

right so slight progress, there is some noise coming out of it-from the triangle out it looks kinda like a ramp with the top lopped off-the is also low in volume signals coming from sine and pulse, though weak. All the wave forms look abit weird, there is nothing coming out of the saw that i can tell.

i took the measurements on the meter-it seems my music from outer space bi polar supply only puts out 12/-12v
lower leg upper res +11.87
upper leg lower res -11.81
upper leg upper res +11.7
lower leg lower res -11.64
pin 8 opa 2134 (is this ok?, have also tried tl072) +10.41
pin 4 opa -10.41
jumper at 220pf +3.41

any idea's would be more than appreciated, i've been over the solder points but maybe i need to do that again.
many thanks

[yusynth]

The VCO is designed to run with a +15/-15V power supply, it cannot work properly if powered with a +12/-12V PSU.

[tazer]

i thought that may be the case when testing it last night, at least i know it hopefully works. I have a 12-18.5v ac supply that i'm using but the mfos boards only seem to output 12/-12, dunno if they can be adapted. What are others using to gain 15/-15 volts? i've built 11 modules (inc 4 of yours and this is the first that needs the full 15/-15 think i need to pump up the voltage! any ideas about the best way to achieve 15/-15 would be gratefully rec'd though i think mfos do a 1.5-2amp adjustable pcb that goes from 9-15 bi polar.

thanks anyway-your a great help to us novices

[tazer]

Hi Yves

I got my last vco sorted-thanks also have got a +/- 15 power supply sorted. I've built another and having a few problems with it, its a home etch so it could stem from that, but i have checked it over and it seems ok.

Basically there is waveforms coming from all 4 outs. The saw and square are the only ones that really look like they are meant to, the triangle and sine both have sections that seem squared off, also the pulse width seems only to work in a very small portion of the pot-maybe 5% of the pot very strange. As soon as you move the pot clockwise it sounds like there is a small amount of pulse width change but then it cuts outs when it hits 5%, as far as i can tell all the components are ok. I took these voltage readings which seem close to the ones you asked me to take the last time i had a problem.
on the lower leg of the upper 10 ohm resistor : you should get +15V( i get +14.71)
on the upper leg of the lower 10 ohm resistor : you should get -15V(i get -14.75)
on the upper leg of the upper 10 ohm resistor : you should get +14.9V (roughly) (i get +14.5)
on the lower leg of the lower 10 ohm resistor : you should get -14.9V (roughly) (i get -14.53)
on pin 8 of the OPA2137 : you should get something like +13.8 to +14.2V (I get +13.24)
on pin 4 of the OPA2137 : you should get something like -13.8 to -14.2V (I get -13.31)
on the long strap beside the 220pF cap : you should get something like +4.8V (I get +4.24)

they seem close any ideas would be appreciated i'm abit stumped.

[yusynth]

Hi the measurements are OK, the slight discrepancies are not impacting.
Now if you have the right sawtooth shape (swinging between for -4.5V to 4.5V) then the other waveshapes should be otained easily provided you follow the trimming instructions at my site.
The fact that the pulse range is so compressed means that your sawtooth is not balanced with respect to 0V, as said above the sawtooth must be swinging between -4.5 (+/-5%) to +4.5V (+/-5%). This balanced is obtained by measuring the ouput signal (with a scope) and trimming T3 until the sawtooth signal is well balanced, once this is obtained then the other waveshapes should be obtained easily by fine trimming T3 (triangle/sinewave) and T4 (sinewave).

If you cannot obtain a balanced sawtooth by adjusting T3, then there may be a faulty or a misplaced component.

[tazer]

thanks yves will check it out and let you know how i get on. At least there is activity in there thats always a good sign. I've tried changing the IC's with no luck.