Author |
Message |
Harry
Joined: May 09, 2017 Posts: 20 Location: Texas
|
Posted: Tue May 09, 2017 1:27 pm Post subject:
Graphic Oscillator Subject description: looking for help on this old circuit |
|
|
Hi, first time poster here looking for help on this:
At this point I have the circuit built up exactly as shown in the schematic, 9v supply and using a second cd4046's vco to feed the input.
So, it seems to be working for the most part, but with a couple issues. First, the output of the circuit shows that sliders 2 and 3 are reversed, meaning as you move the 2nd slider it alters the 3rd bit of the waveform instead and vice versa, same goes for sliders 6 and 7. This affects the inverted and non-inverted phase of the waveform.
I'm still new to digital logic, but going through the schematic it seems the concept should work, pull voltage off the wipers sequentially and then reverse and invert the phase. I have double checked my wiring and plan on triple checking when I get home.
Second issue I'm getting is very low audio output. The schematic description suggests feeding this into a vcf so this can't be intended. I think I was reading about a half volt on the scope with the volume at max and just barely audible over the noise with my amplifier cranked. I've yet to verify this, but the article mentions the voltage at top and bottom of the pots should alternate between .53v and .47v. Wouldn't this leave only at max .06v range coming off the wipers?
Anyway, any helpful troubleshooting tips or insight would be appreciated! I know often these old magazine articles are prone to typos or printed out as untested ideas. I haven't found anything on the net to verify the schematic so it'd be nice to know if I got something wrong or if there is a flaw with the design. |
|
Back to top
|
|
|
PHOBoS
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
|
Posted: Tue May 09, 2017 2:27 pm Post subject:
|
|
|
first of all
interesting circuit. when I read graphic oscillator I first thought it would be a basic sequencer but the inverting and reverting
is a neat idea to get more steps and a symmetrical waveform. The sliders seem to be connected in the right order so maybe
you made a small mistake somewhere. As for the voltage, my calculations tell me that with a 9V supply there is a voltage of
0.53V over the pots, ignoring any voltage drop caused by the XOR gate and tolerances in the resistors and pots. That's not a lot.
I first thought the opamp would double it because it has a gain of x2 but with the 47K resistors attached to the wipers of the pots
it brings this back down to x1. I am not sure why they added these resistors as the voltage and current is already low to begin with.
So if you want a higher output voltage lower the value of these (or in my opinion discard them alltogether) and lower the
value of the 47K resistor attached to pin 2 of the opamp as well. Say you want an amplitude of 5V then (after discarding the
other resistors) change it to 10K. _________________ "My perf, it's full of holes!"
http://phobos.000space.com/
SoundCloud BandCamp MixCloud Stickney Synthyards Captain Collider Twitch YouTube |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue May 09, 2017 2:55 pm Post subject:
|
|
|
Re. the slider swap you have exchanged the low and the middle bit for the 4051 eiter at the 4051 side (pin 10 and 11) or at the 4024 side (pin 11 and 12).
Re. the low output level .. there should be about 0.5 Volt over the pots, reversing direction each half cycle. The output amp amps about two times .. so there should be about 2V of output signal .. or so it seems to me .. a bit low for a synth maybe .. but should not drown in noise at least. When not working I'd expect that to be in the final stage ...
Assuming you do not have an oscilloscope ... you could measure with a meter set to AC between ground and the output of the 4051 (pin 3) to check if its about 0.5 V there (when a sine is set up with the sliders as in the example in the article).
Edit: maybe PHOBoS is a bit more right re. the ouput levels _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
Harry
Joined: May 09, 2017 Posts: 20 Location: Texas
|
Posted: Tue May 09, 2017 3:22 pm Post subject:
|
|
|
Thank you!
Yeah I thought it was a pretty novel circuit too. From a cost/space perspective it also offers more resolution than an audio rate sequencer.
Yes, that does definitely confirm some of my suspicions. That helps with my confidence in bringing the gain up.
Hopefully I do have a misplaced wire somewhere screwing with the sequence, I'll see in a bit. If not I can always just rearrange the pots
Now, once I have it running properly I'd like to convert the pll to a standalone vco so an external oscillator wouldn't be needed. Would this be feasible? Would pin 3 remain connected as is? |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Tue May 09, 2017 3:34 pm Post subject:
|
|
|
Actually .. talking about sequencers .. the reversal and put inversion .. when separated out as individual functions .. are cool for sequencers too ... two things the MANIAC has (had - does it still exist?.)
You'd want a bit more voltage for a sequencer .. but can amp it up .. oh and DC coupling .. but this could be a nice starting point. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
PHOBoS
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
|
Posted: Tue May 09, 2017 4:07 pm Post subject:
|
|
|
I think you are the MANIAC these days jan
The 4046 makes a great simple standalone VCO, unless you want V/Oct tracking then it get's a bit more complex but that
can be done too. Pin 3 wouldn't be used anymore and if you want lower frequencies you would have to add an extra divider
between the output of the 4046 VCO (pin 4) and the CLK input of the 4024 Counter (pin 1). Right now it actually uses the
PLL function in combination with the counter to multiply the incoming frequency. _________________ "My perf, it's full of holes!"
http://phobos.000space.com/
SoundCloud BandCamp MixCloud Stickney Synthyards Captain Collider Twitch YouTube |
|
Back to top
|
|
|
Harry
Joined: May 09, 2017 Posts: 20 Location: Texas
|
Posted: Tue May 09, 2017 7:37 pm Post subject:
|
|
|
Right on, traced my board again had the 4051's A & B select reversed. Jumpered the 47k resistors and replaced the 741s with 6k8. Have a much healthier 4.3v now, will probably bump the gain up just a bit more. Totally happy with this oscillator now that I can finally hear it, definitely met my expectations. Appreciate the help guys!
Phobos, you use perfboard too? Thought I would be in the minority as much of a pain ic's and jumpers can be I still much perfer it to anything else. |
|
Back to top
|
|
|
PHOBoS
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
|
Posted: Wed May 10, 2017 2:53 am Post subject:
|
|
|
yep, I am a perfert
I know there are different methods to it and the way I do it is that I first design a layout on a 0.1" grid using an
old (free) version of Sprint-Layout. Then I copy that layout by bending the component legs or use some stripped cat
cable for longer 'traces'.
What I like about perf is that you can get from circuit design to a working circuit pretty quick and if something is not working,
you made an error in the design or want to add something it is relatively easy to do. (compared to an etched PCBs that is)
Good to hear you have a higher voltage now. At first I thought the 4.3V seemed a bit low with a 6K8 resistor
although the 4051 might have something to do with that. However, I don't know how you are measuring it. If you
measure the output with a voltmeter on the AC setting then you are measuring the RMS voltage so the amplitude
will actually be higher than that. If you don't let the oscillator run you could measure the DC voltage straight from
pin 6 of the 741.
But there is something else that might be causing a lower voltage than I expected, the 741 itself. Although I have a
bit of a soft spot for it (it was my first opamp ) a TL072 might be better here because it has a much higher input
impedance. As long as you don't boost the voltage up too much so it gets close to the rails it should work fine in this
application.
btw when I woke up this morning I actually realized how it does the reversal. I hadn't given it much thought yet
but although I knew you can reverse the direction of a binary counter by inverting the ouputs and I also knew XORs
are great for this since you can turn the inverting on/off, it never occured to me to do this by using another output of
a counter. I was also thinking that instead of using the XOR gate to reverse the polarity and the current limiting resistors
you could do it with 2 comparators and get a higher voltage/current to begin with. Although I do like how they used that
last XOR gate for it. _________________ "My perf, it's full of holes!"
http://phobos.000space.com/
SoundCloud BandCamp MixCloud Stickney Synthyards Captain Collider Twitch YouTube |
|
Back to top
|
|
|
Grumble
Joined: Nov 23, 2015 Posts: 1294 Location: Netherlands
Audio files: 30
|
Posted: Wed May 10, 2017 4:47 am Post subject:
|
|
|
The 741 is not your typical rail to rail output opamp..
Looking at the specs I see that when powered from +/- 15 volt, the output swing is typical +/- 14 volt, so it looses about 1 volt per rail typical (worst case is even 3 volt per rail!)
I can't find what the losses are when powering from 9 volts, but I'm afraid it will not be less than 1 volt per rail... and the TL072 isn't better in this area.
I guess thats why the 10k pots are in between the 10 k resistors: to keep the voltage coming from the 4051 away from the rails. |
|
Back to top
|
|
|
|