Posted: Sun Jun 22, 2014 8:31 pm Post subject:
More voltage controlled 40106 fun
Haven't breadboard this yet, but the idea struck me when looking at the MFOS Weird Sound Generator schematic and wondering about how the gate control works and how to put it in my next build.
I had wanted to adapt the WSG core to wire the transistor cv in and the gate straight to the front panel for patching. Part of me thought two cv ins per oscillator was a little over kill too (though I might still go for it!)
Anyway, it struck me that adding a pot between the cv in and the diode would affect the pulse width of the gate and could even end up in cv control. Turn the resistance all the way down for classic WSG gating, and all the way up for pure FM, with things in between possible!
I plan on testing this tomorrow or at the weekend, hopefully have some scope traces and sound clips too.
NOTE: I didn't set the internal impedance of the signal source so the results are not going to be very accurate, though they do hint towards what is possible. I tried setting an impedance of 10k when I used the triangle wave test, but I am skeptical it will be accurate either.
40106 CV TEST.JPG
Description:
Filesize:
10.12 KB
Viewed:
19155 Time(s)
40106 cv test (tri - impedence adjusted).JPG
Description:
Filesize:
118.93 KB
Viewed:
751 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
40106 cv test (square).JPG
Description:
Filesize:
80.72 KB
Viewed:
818 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
40106 cv test (sine).JPG
Description:
Filesize:
105.35 KB
Viewed:
809 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
Also, for those of you who are like me and enjoy taking a fake/easy triangle from the cap out put: the output there looks very interesting.
Not sure how to tame the gain differences though. I've been using inverters to amplify the tri out from there, but the much bigger dips down to 0v will need much more headroom...
Some more analysis of this: It's not exactly FM, but it's altering adjusting the pulse width of the sync/gate effect. It seems to be doing this by adjusting how fast the cap discharges when the diode is forward biased - which also means it will have less "recovery time" needed to charge up enough to start oscillation again.
The pot also effectively adjusts the knee of the diode, which causes larger and larger ranges of cv input to make the cap charge slower rather than fully discharge.
Unfortunately the saw-like output is not possible. It's an error in the simulation caused by me setting the cv source as having an output impedance of 0 ohms! If you check the scale, the simulation calculated it as less than 0.01Vpp.
I'm going to breadboard this Friday because I also want to try the diode the other way for some more CV experimentation but I think I am most likely to go for the WSG-style NPN transistor across the pot for modulation.
I'll post back with results when I get them. I'll try to take sound clips of all of them
This is going to be part of an article I plan on doing about 40106 oscillators and adding cv to them along with explanations of the hows, whats and and whys.
Overall, if you only want to use square LFOs, then I recommend a 100k pot for the series resistance in front/behind the diode. Good range of sounds available there. If you want to use sine/triangle then a maybe 20k max, but non-digital values have limited use. Not as cool as I hoped but definitely still useful.
There is a threshold point of resistance, over which the output stops being gated and becomes "true" FM. This threshold area is really tiny but could yield some very cool results. I might go back and try pin pointing it and then using a fixed series resistance. Results with analog waveform lfos might be better then.
You can also flip the diode round but it basically sounds the same. Instead, the discharging is slow and the voltage it hits when it gates is high instead of low.
40106 wsg diode (cv sweep, 25k resistance).mp3
Description:
Sweeping a buffered cv source with about 25k resistance inserted. Kind of disappointing. Has made me write this off for non-square LFOs. Very small range of voltages have an effect. Increased resistance makes the range become even smaller
Filesize:
95.48 KB
Viewed:
1514 Time(s)
40106 wsg diode (cv sweep, no resistance).mp3
Description:
Buffered CV source sweep with no series resistance. You'll notice that it is either off or the pitch set by the pot. Kind of useless
The most interesting in my opinion. CV and carrier frequency quite close. All that's being adjusted is the series resistance. Really interesting clashing, dissonany sounds here.
Should have mentioned. In all oscilloscope shots - top is square out from the 40106. 2V/div. Bottom is from the cap. 1V/div.
40106 back bias diode cv (audio rate cv, higher freq carrier, no resistance, increasing carier freq).mp3
Description:
This one is a little different. This is when the carrier frequency starts off too low to charge the cap when the modulation is high. Interesting pwm effects
Filesize:
287.66 KB
Viewed:
1732 Time(s)
slow charge fast discharge.jpeg
Description:
This is with the resistance set just around the threshold that pushes it into true FM instead of gating.
Filesize:
17.47 KB
Viewed:
18847 Time(s)
resistance increase 2.jpeg
Description:
Resistance close to reaching the threshold. Notice how the cap charges up to a certain point when the cv is low but can't get high enough to flip the schmitt trigger
Filesize:
17.49 KB
Viewed:
18848 Time(s)
carrier too slow.jpeg
Description:
No sound because the cap can't charge in time.
Filesize:
21.54 KB
Viewed:
18848 Time(s)
carrier just fast enough.jpeg
Description:
This is when the carrier is *just* fast enough to get one transition in. Very nasal pwm sound
Filesize:
101.62 KB
Viewed:
740 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.
Interestingly, having the diode the other way round actually has the exact same effect. It took me a little bit of time to wrap my head around that myself!
Right, you can switch the oscillator on and off in both cases. But the diode in reverse direction switches it ON with a 1 (Vdd) and OFF with a 0 (GND). I find that more logical.
Joined: Apr 14, 2013 Posts: 21 Location: Arkansas, USA
Posted: Thu Jul 31, 2014 5:23 pm Post subject:
Thanks commanthe for the PDF and the great explanations. It really shows me some new uses for the 40106. _________________ Vine username: "billsship"
Modulargrid
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum You cannot attach files in this forum You can download files in this forum
Forum index » DIY Hardware and Software » Lunettas - circuits inspired by Stanley Lunetta
FAQ
Calendar
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Chat Room
