[ ? ] Continuously variable phase LFO?

[R.G.]

Yeah, that's what I was saying.

See http://geofex.com/FX_images/ramp-lfo.pdf, first circuit. The pots, switch, and first opamp are for the ramp up/down stuff. From the output pin of the first opamp on is a triangle/square voltage controlled oscillator with vastly more range than an LFO needs. Since the triangle has fixed positive and negative peaks, implementing a sine shaper with under 5% distortion is only a matter of scaling so that a pair of clipping diodes rounds off the peaks.

It doesn't have anything like the super performance, accuracy, and other features of the waveform chips, but it does give you a square for locking and a triangle for making the variable-phase marker, as well as a shaped-sine output which is fixed in phase with the triangle. Two of those (two dual opamps) a comparator, and a CD4046 should do it.

[puretube]

R.G.: tried to find info on a VCLFO on geofex like you mentioned yesterday,
but couldn`t find the: "search GEO" on the left hand side in the "regular features, anymore...


btw, my 2-OTA "sine-sine_90" or "simpquad" offers no tri, square, or other
"harsher" signals at any point (except maybe the "trough zero" points, which actually are "through half supply-voltage"),
than a soft-clipped (=rounded) sine at pin3...  - seems to be some kind of analog equilibrian pendulum... 

[moosapotamus]

This thread is making my head spin.

I've only had snippets of time to try to keep up since recovering from the massive ice storm that knocked out our power for about six days (some folks in our area still don't have power restored - going on two weeks for them). Actually, we are still kind of recovering. Picked up some of our pets today that my brother in-law was keeping safe for us. We've had below freezing temperatures almost every night. Pile on all the holiday stuff and I barely have time to go poop.

Anyway... you guys have posted so much cool stuff to try out. I need a three week vacation.

gez - the stepped lfo thing sounds really cool!
puretube - really want to try your ideas out, too. thanks for posting the schemes.
jc - your stuff always blows me away... even when I can't completely get my entire head around it. anxious to see what you've been describing.

Thanks!
~ Charlie

[R.G.]

There's nothing new under the sun. I found an article from Elektor 1997 showing a CD4046 used for its phase detector and an XR2206 used for its sine/tri/square output to generate a sine or triangle from an arbitrary rectangle wave.

[TELEFUNKON]

Note however, that unlike the more "usual"  3stage hi-pass or lo-pass phaseshiftoscillators, which need a gain of ~27 in the loop
(can`t remember the gain needed for 4 stages...), the "45er`s" loop gain is merely slightly higher than 2 

Yes, that's par for circuits which buffer between the RC phase shift stages. I remember seeing a circuit in one of TI's analog app notes using four opamp buffered RC stages and the necessary gain was microscopic.

It seems that each stage is operating at a "gain" of about one over the fourth root of two, about  0.84.

BTW.: in the other "usual" 3- or 4-stage phaseshiftoscillators and phaseshifters per se,
I prefer the lo-pass configuration because of its inherit smoothing capability in combination with AGC,
noise rejection, and forgivingness cc. being overdriven at its hottests spots...
(if it weren`t for the neccessity of floating (series-) variable resistors... 

Yep that is nice.

Here is what a certain Mr. Bode had to say about cascaded allpassfilter versus cascaded lowpassfilter (quadrature) phaseshiftoscillators:
M.S.O.

[TELEFUNKON]

Isn't there some way to manipulate two triangle waves to give a phase varying triangle?
If so, this could then go through a triangle to sine converter & there you have it.

Here is a sawtooth version: B. Hutchins
also see Elektor 10/1982
and ETI 1/1983

[Paul Perry (Frostwave)]

Thanks Telefunkon! "Une belle affaire" indeed

[gez]

Hope your Xmas wasn't ruined, Charlie!

Found a little time today to do a bit more work on this.  The OTA tracking filter I cobbled together worked beautifully and I got a really nice sine over a wide range of frequencies.  Don't have the chips to do the phase shift, though, and won't until next month...but it's feasible.  I'll mess around with some switched cap filters I have, as that will probably take the parts count down...will have to do some data-sheet reading.  As with all 'simple' ideas, they end up more complicated than anticipated as you work through to a final design.  It will be low current consumption, though - IIR those wave shaper chips are real hogs, although RG's idea does sound simpler than mine.  All swings and roundabouts...

[gez]

IIR those wave shaper chips are real hogs

Just checked the data sheet of the MAX038: 80mA - 100mA per chip! 

Plus split-supply required?

[gez]

Got the parts through and, as you can see, it works.  Only the first 8 settings of the rotary pot are shown above, but the phase shift continues for the full 360 degrees (16 positions in total).  No filtering shown, hence the stepped nature of the waves.  No points for the camera work, either...

So far the parts count is 1 cap, a handful of resistors, 5 eight-pin CMOS chips, one pot and a specialist rotary switch.  I did a little trick to make 4051 chips 16 step.  They're cheaper than 1-of-16 multiplexers and I saved a little board space in the process.

I have some switched-cap chips in my parts bin somewhere.  Not sure if they'll fit the bill.  Filtering could be done otherwise, using an LM13700 (at the expense of final amplitude).  The 4060 would have to be voltage-controlled, though.  Should be easy enough: wire the inverters of the 4060 as a Schmitt and wire it to a dual op-amp.

Is it worth it, though? (simple on paper, but the end result is quite a complex circuit).  All depends on whatever everyone else had in mind, I suppose...

[R.G.]

Nice work Gez.

I've always liked stepped-approximation sine techniques.

[gez]

The story so far:



Apologies for the quality of the sketch.

Outputs would most likely need to be buffered (suitable op-amp followers) to be of any use.

4V peak-to-peak with a 9V supply.  Parts count would be higher still to end up with smooth sine-waves.  In its favour, current consumption is low.

[gez]

MF10 did the trick.  The frequency is too slow to be dual-scoped, hence the weird second image, but hopfully you can see the phase-shift.

Theoretically, gain can be increased by reducing in value the 20Ks to pins 4 and 17 of the MF10: anything down to 10K.  Might cause clipping with a 9V supply, though, so I kept things to 2V peak-to-peak.

Data sheet didn't have any info on the input bias current of the MF10's input amp.  Might be possible to sidestep the op-amp followers and stick in high value resistors (IM input resistors).  Shouldn't cause too much distortion, and would do away with a chip.  Also isn't clear whether the 10Ks from 2-4 and 17-19 are necessary.  When I could be bothered I'll pull them and see what happens.

Waveform is smooth, although it has a 'thickened'/'serrated' look to it.  The data sheet said this would happen.  IIR, Penfold uses a small value cap to smooth this out.  Probably not necessary (good enough for our purposes).

Clock bleedthrough might be a bit of a problem...a PCB designer's nightmare?

Although a relatively simple circuit the parts count is mainly chips, which is a pain.  Phased sines are not something that can be done easily in the analogue domain (it seems).  The outputs are amplitude stable, though, and the circuit has reasonably low current consumption (unfortunately the MF10 can hog around 8mA or so, IIR).

The schematic should be checked against data sheets/CMOS cookbook.  Can't guarantee I didn't make a mistake when drawing it.

Well that's my entry.  Do I win a cream cake?

[R.G.]

Yep, the serrations will be with you from any digital/switched cap filter. They would be there in a uC version too, to a degree depending on the output strategy. However, for an LFO, it is likely that they would be completely inaudible.

Good work. It looks nice. Yes, cream cake for you! 

[slacker]

That's really clever Gez, I like the way you flip the supply to the voltage dividers to get 16 steps, very cool.

[gez]

Why thank you Sirs!

I'm wondering if it would be easy to rig up a decoder chip and display to the binary switch, so that you can see which position is selected.  I've got some recent articles from EPE that go into using low-current displays.  Might have to do some digging.  As for a decoder chip...ideas? (off the top of your heads).

[gez]

Update

Did some empirical testing of the MF10.  With a 9V supply it can deliver an output signal of around 6.5V peak-to-peak before clipping occurs, so setting gain for unity presents no problems.  The resistors to pins 4 & 17 form the input resistors of two separate inverting op-amps.  The other, 10K resistors provide feedback to these inputs.  By making all resistors equal value, unity gain (4V peak-to-peak in this case) is achieved. 

No info is given in the data sheet as to how much bias current the inputs of the op-amps require, but I made all resistors 1M and it functioned fine.  This meant that I was able to pull the op-amp buffers feeding the MF10.  Although there may have been some distortion of the stepped signal due to loading (I didn't check), there was no discernable distortion of the final sine waves.  In short, if you up the value of all resistors associated with the MF10 to 1M, you can pull the ICL7621 op-amp, thereby simplifying the circuit a little. 

[puretube]

Gez:
here`s a nice one (from 64 years ago...  )
to get rid of the steps in a non-filter way 
(US pat.#2488297).
(can be adapted to those newfangled semiconductors, though...)

[gez]

 

I think I'll stick with the filter!

[puretube]

Gez:
here`s a nice one (from 64 years ago...  )
to get rid of the steps in a non-filter way 
(US pat.#2488297).
(can be adapted to those newfangled semiconductors, though...)

Of course, those newfangled adaptions would only work for triangle-waves (like in the vintage circuit...),
but not for sine-wave construction  .