[ ? ] Continuously variable phase LFO?

[puretube]

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?

Let`s share the cream, or the cake...

why trying to create a sinewave, and then chase it through an MF10 (or an LTC1060 - pin compatible...) ?

Create a square, divide it by 100, chase it through half of an MF10 (Low-Pass/clocked by the original square): out comes the sine...

(BTW.: filtering those staircase-ripples is frequency-dependant, again - and it does bother me - Rock`n`Roll or not...)

To get rid of the staircase-steps:
create a square, divide it by 1000, chase it through half an MF10 (Low-Pass/clocked by 1/10th of the frequency of the original square),
chase it through another half of an MF10 (Low-Pass/clocked by the original square): voila, off are the freaks!
[this way, the frequency influencing cap becomes a reasonable size for LFO purposes, too...  ]

To get phase-shifted sinewaves:
create a square, divide it by 100, and divide that one by 2 and call it phase one;
take the 1/100th in another "channel", and use a comparator/opamp to (voltage-controlled...)PWM
the pulses;
divide these by 2 to create 50% duty cycle and call it phase two;
chase both phases through the 2 halves of an MF10 (Low-Pass/clocked by the original square) :
out come the phase-shiftable sines (voltage-controllable)...

To get a quadrature (90°) pair:
create a suare, divide it by 100, chase it through the well-known 4013-to quadrature-flipflop-circuit,
chase 2  90° seperated outputs from the 4013 through half an MF10 each (Low-Pass/clocked by 1/4th of the original square)...

Or:
create a square, divide it by 100, chase it through half an MF10 (you guessed it: Low-Pass/clocked by the original square)
call it "phase one",
take it and run it through the other half of the MF10 (again you guessed it: Low-Pass/clocked by the original square),
and call it "phase two" : nicely separated by 90 degrees. but: about 1/3 of the other phase`s volume 


Coming up next week: schemos for nothing, and vids for free! (no vero, though...) 

[Nasse]

I mail ordered two small dc motors with gearbox and I think the speed range could be useful. Of course those are toys. While I was thinkin if I had computer sequencer and audio editor and multiple output soundcard and just make ssuitable signals and rectify those and take the cable(s) to the pedal.

[StephenGiles]

 

[gez]

why trying to create a sinewave, and then chase it through an MF10 (or an LTC1060 - pin compatible...) ?

Create a square, divide it by 100, chase it through half of an MF10 (Low-Pass/clocked by the original square): out comes the sine...

(BTW.: filtering those staircase-ripples is frequency-dependant, again - and it does bother me - Rock`n`Roll or not...)

To get rid of the staircase-steps:
create a square, divide it by 1000, chase it through half an MF10 (Low-Pass/clocked by 1/10th of the frequency of the original square),
chase it through another half of an MF10 (Low-Pass/clocked by the original square): voila, off are the freaks!
[this way, the frequency influencing cap becomes a reasonable size for LFO purposes, too...  ]

I have an old EPE schematic that basically shapes a square into a sine by just using the MF10.  The MF10's centre frequency is the clock frequency divided by 50/100 (depends on which selection you choose).  So, the designer divided the clock by 50 (using some weird TL stuff I'd never heard of) and then fed that through the MF10's band-pass (running from clock).  Huge range, reasonably simple.  Would be even simpler just dividing down close to the 50 mark, then using the low-pass output (and probably taking a dip in amplitude...although it would be stable across the range of frequencies).

Although I had considered the above, I used the method outlined in this thread as I wanted the option of creating other wave forms (mainly 'inverted hyper-triangle') or one wave form against another.  Plus, it seemed a reasonably 'simple' way of creating the phase shift.  There are many ways of doing the shift, but when you start mapping out a working circuit it all starts to get complicated and there's not much in it.

Incidentally, creating quadrature with squares is very simple.  Feed your clock to one input of an XOR gate, divide clock by 2 and feed that to the other input of the gate.  The XOR's output is of the same frequency as, and 90 degrees out-of-phase from, the divided down clock.  Original square has to have perfect symmetry (usually divided down itself). 

[gez]

Forgot to mention, the MF10 in the EPE design was wired as a 4-pole filter (both halves used). 

[puretube]

Didn`t wanna rain on yer parade, Gez... 

Anyway, to explain what I wrote before, though the following is not continuously variable, yet... :

Create a square, divide it by 100, chase it through half of an MF10 (Low-Pass/clocked by the original square): out comes the sine...

(BTW.: filtering those staircase-ripples is frequency-dependant, again - and it does bother me - Rock`n`Roll or not...)

Coming up next week: schemos for nothing, and vids for free! (no vero, though...) 

second half of the dual switched-capacitor-filter-block IC4 unused...
(the first half is wired as LowPass for f/100, in mode 1a as noted in the National datasheet page9, fig.8 )

SinoStep_1

To get rid of the staircase-steps:
create a square, divide it by 1000, chase it through half an MF10 (Low-Pass/clocked by 1/10th of the frequency of the original square),
chase it through another half of an MF10 (Low-Pass/clocked by the original square): voila, off are the freaks!
[this way, the frequency influencing cap becomes a reasonable size for LFO purposes, too...  ]

Coming up next week: schemos for nothing, and vids for free! (no vero, though...) 

SinoStep_2

To get a quadrature (90°) pair:
create a suare, divide it by 100, chase it through the well-known 4013-to quadrature-flipflop-circuit,
chase 2  90° seperated outputs from the 4013 through half an MF10 each (Low-Pass/clocked by 1/4th of the original square)...

Coming up next week: schemos for nothing, and vids for free! (no vero, though...) 

well, err, the original square itself needs to be divided by 4 too , additionally (by IC5), to match the 100:1 frequency criterium, of course... 



the first divide-by-ten counter (IC2) above can be omitted, so the master clock goes into pin 14 of IC3 directly...

SinoQuad_3

[gez]

Didn`t wanna rain on yer parade, Gez... 

No problem Ton.  I haven't had a chance to look through what you've done, but it looks like a more elegant solution.

[puretube]

The "variable phaseshift"-experiment at the moment is in the stadium of having one fixed sine,
plus a +/- 90° shiftable sine of exactly half the frequency of the fixed one... 

[gez]

Ton, what on earth is your Avatar?? 

It's not some voodoo charm is it?  Mojo man! 

[puretube]

Ton, what on earth is your Avatar?? 

It's not some voodoo charm is it?  Mojo man! 

You know the original P-FUNK-sign? :

(nowadays often abused by the metalrockerz... )

That`s the way  BOOTSY & me

greet whenever we meet... :



making the P-FUNK-sign, and touching each other`s fingertips!

Pure P-Funk Energy Transfer...

(that pic in the avatar - see enlarged in above link - was shot when Bootsy saw me in the crowd right in front of the stage
at a concert in Nuremberg ~ a decade ago... yes, that naked arm is mine, there).

[gez]

Ah, I can make out your arm now.  And, if I get close to the monitor, I can make out a figure.

It looked like some bird's foot wrapped in/with something!

[puretube]

The "variable phaseshift"-experiment at the moment is in the stadium of having one fixed sine,
plus a +/- 90° shiftable sine of exactly half the frequency of the fixed one... 

Looks like I have stuck there where Gez had the same quirk in reply#49...

... I just realised that once the flip-flops divide down both squares, there's only a Max shift of 180 degrees.  I had one of those 'hang on!' moments when I was remembering the scope patterns.  A few sketches of the waveforms on the back of some scrap paper confirms this. 
In short, AGHHHHHHH!

... I thought of the following:
derive a PWMed (shiftable) rectangular wave [Fpwm] from the original square [Fo], which (similar to Gez`s 7556 oscillator) will be an almost 360° shiftable cousin of [Fo], however of varying duty-cycle
(that`s why Gez put flipflops after the oscillator`s outputs: to get symmetrical squares...);

now double the rates of each output, to get 2[Fpwm] and 2[Fo] (frequency-doubling),
and chase those two new pulsetrains through half a 4013 flipflop each,
to get a pair of symmetrical (50/50 duty-cycle) squares of identical original frequency [Fo]
with a shiftable phase-offset of ~0°...360° for [Fpwm].

These 2 signals to be filtered by an MF10 to obtain sinewaves...

The neccessary frequency-doubling could be done by RC-differentiating and triggering on the rising edge,
and do the same on the falling edge of [Fo], and combining (adding) the derived pulses,
so you get a pulsetrain of 2[Fo];
Same procedure for [Fpwm], to get the 2[Fpwm] pulsetrain.




BTW: to put out nice linear triangles out of the MF10 remains a little mysterious to me... 

OTOH, there`s yet another way to put out a nice quadrature pair of sines from squares out of one half of the filterchip...

[gez]

Ton.  Perhaps you could use a circuit similar to the one I posted earlier in this thread (the square-wave PWM one).  The PWM output could be sent to a 4046 PLL with all your divider stuff in the phase-loop.  At the bottom of the divider chain is your phased square, 50:50 duty cycle.  At the top is the clock for the MF10. 

The non-phased square could either be sent to the other half of the MF10 via another PLL for 'squarification' (50:50 duty cycle).  That would give you the best part of 360 degree phase shift.  If you divided down the non-phased square, you'd only have 180 degree phase shift but your two sines are of different frequency.

Or, you could use something along the lines of what I last posted (schematic wise), but change the resistor values to get a different waveform.  You could do an 8 step sine against a 16 step to get your different frequencies.  I've sketched out a circuit for a 16 step inverted hyper-triangle (the connoisseur's choice for filters!  ), but have been too lazy to breadboard it.  Could either use a 1-of16 multiplexer, or chain two 1-of-8 chips together.

Not sure if the above is of any use to you.  Just some thoughts off the top of my head (am probably missing something).

[gez]

PS  Nice quad schematic of yours (finally got some time last night to go through it).  The use of 4017s was inspired.  Also, I'd never before come across that wiring of flip-flops to get quadrature.  I still need to work through it with the data sheet to find out what's going on, but it looks much simpler than what I've been doing all these years.  Thanks!

[puretube]

Gez: a totally different approach to hypertriangling occurred to me 2 weeks ago while brainstorming/scoping stuff cc.  my "new project"...
will show soon, when I find that darned envelope-backside I sketched the schemo on...

and: yes, I`m aware of the R/2R ladder stuff and the up/down counting/multiplexing thing,
and the graphic oscillator approach,
but rather keep my thing "analogue" (though there`ll always be a comparator in there...  ).

[edit]:

hyper-sine (the connoisseur's choice for filters!  ),

and flangerz... 

[puretube]

PS  Nice quad schematic of yours (finally got some time last night to go through it).  The use of 4017s was inspired.  Also, I'd never before come across that wiring of flip-flops to get quadrature.  I still need to work through it with the data sheet to find out what's going on, but it looks much simpler than what I've been doing all these years.  Thanks!

Not neccessarily "mine"... 
(recently found the patent, where that circuitwiring that has been all over Elektor for decades, (as well as in the synth- and the radio-scene...)
had it`s origin  ).

[gez]

Just edited my post.  That should have been inverted hyper-triangle. 

[puretube]

Just edited my post.  That should have been inverted hyper-triangle. 

...didn`t even notice that slip... the intention was clear to those interested in the art... 

[gez]

I tip my cigar at you sir (and raise my brandy glass)! 

[puretube]

 ... from sketching the pulses on paper, it seems that my idea in reply#131 is another no-go...