3 or 4 octave up square wave ??

[gez]

Even if you don't want to share your circuit design, how about a clip so we can hear what your PLL approach sounds like during lock acquistion?

I found a bit of time this afternoon. 

The following samples were recorded with guitar but the effect is meant for guitar an octave lower or bass.  As such it sounds pretty uninspiring, ridiculous even - it's a stepped PWM effect intended for sustained bass notes not hackneyed blues licks!

It's been gathering dust on one of my breadboards for longer that I care to mention and sounds clunky as I didn't have the chips I wanted to use in my parts box.  I now have them but have been too lazy to wire them up (should be a lot smoother when they’re in).  There's no octave up effect but the VCO operates at frequencies many times higher (either 8, 9 or 10 times - I forget) than the input signal and various outputs could be tapped for octave up/harmoniser effects.

There's no gating circuitry as it's an unfinished module.  Because of this there are lots of blips and barps when there's no input signal (caused by finger/hand noise).  Easy enough to sort out though (see previous posts).  I've left these noises in where possible (I have a 1M upload limit) so that you can hear how notes end.   No EQ either so it sounds a little harsh. Any gliss you hear is a result of my fingers sliding up/down the fretboard, not the PLL (yeah...right!  ).

http://hometown.aol.co.uk/Gezpaton/Sample+1.MP3

http://hometown.aol.co.uk/Gezpaton/Sample+2.MP3

http://hometown.aol.co.uk/Gezpaton/Sample+3.MP3

[markusw]

Hey gez,

cool samples. What kind of mixture is this: fundamental square + fundamental*(8, 9 or 10) ??

Markus

[gez]

What kind of mixture is this: fundamental square + fundamental*(8, 9 or 10) ??

There's no mix.  The PLL slices up input signals into segments (8/10/16/whatever - depends what you use in the loop) which are then enabled/disabled to create 'stepped' pulse width modulation. 

One of the least inspiring circuits I've ever come up with, unless the guitar is an octave down then it's good for sustained synth-like bass lines.

[MR COFFEE]

Hi Markus,

just simulated the AN 41 doubler in LTSpice. Works really nice. After increasing the caps to adjust them to 5-string bass freq range (~30-300Hz) it works really nice. 
Some more Qs: the sims "say" that the integrator needs some time (depending on the freq range you want to cover) to "lock in" (i.e. to produce a symmetric triangle wave) . Is this just a Spice artefact?

Lock-in time is short if you don't overdo the fed back voltage reference filtering. You can also ac couple the triangle wave and use a 1/2 vcc for reference voltage in the second comparator. Your increasing the caps may need to be tweaked as an alternative.

I also tried to cascade these doublers and it seems although the circuit works fine for one, maybe two octaves up it gets problematic with more octaves. It seems the problem is caused by the integrator which produces a not-perfect-90° phase shift (or at least it does'nt produce exactly 90° over a wider freq range). Thus, pulse width is'nt exactly 50/50.

There is no reason for the tri wave to be other than a perfect shift of 90° at audio frequencies. I don't know why Spice is telling you different. Check the triwave output and see what it looks like - that should give you a clue what's throwing your Spice sim.

FWIW, I've found Spice oscillates more than any rat's nest wiring I've ever seen. It's sure given me more trouble than it's worth for anything other than filters, unless the sim has been tweaked by the IC manufacturer to work with a particular IC (like a switching regulator or such). And even on the manufacturer's specialized versions, if you put in crazy values for components it will try to run anyway, and sometimes Spice will tell you it will work when it won't and vice-versa. And the vice-versa is really perplexing if you've come to think Spice is usually right.

I love the Pease story about the engineer calling him and asking why a LM308 integrator oscillated when he simulated it in Spice. Bob asks, "Did you build it?" "Yes". "Did it oscillate?" "No." "So if the IC oscillated, I guess you'd be calling the EDA vendor?" Click. 

For one octave (and even two) it does not really matter but if you try to cascade the circuit 5 times (which would be necessary in order to get 4 octaves up plus to compensate for the 1 octave lost by the adaptive schmidt trigger I'm planning to put in front of it)  you only get a pulse wave with some kind of oscillating pw (sorry don't know how to explain it).

The pw oscillation sounds like more Spice artifacts.

I don't recommend the adaptive schmidt trigger. It doesn't tend to give any more stable waveform than a plain vanilla comparator driven hard if you've filtered it right first, and you lose an octave to start with. I've built those circuits and they don't really fitthis app well - they're more for shifting down like a bluebox. Lowpass Filter the guitar input (18-24 db\oct or better) and square the hell out of it with a comparator by driving it hard, like Gez says.  AGC is OK and helps with long sustaining notes, but on bass guitar, I doubt you need it.

So at the moment my theory is, that in order to get a reasonable squarish wave after the 5th octave up stage over the whole bas/guitar fundamental freq range, the phase shift had to be as close as possible to 90° for each stage.

The phase shift will be 90° if you drive it with a square wave, which the LPF and comparator will give you. I suggest a breadboard rather than Spice for figuring out the capacitor values, and build and tweak one stage at a time, looking at it on a scope while you play through the circuit.

I doubt you will like 5 octaves up, even starting with a bass; it will probably sound like a cheesy casio - remember, the output is a square wave. Maybe 2 to 4 octave up tops.

As I think you already figured out, DOME filters aren't for this kind of gizmo.

Hi Gez,
Thanks for sharing the clips! I might have to play around with a 4046 a little more when I get a few other things off the bench!

Have you looked at the waveform on a scope? I'm curious about what it looks like.

[gez]

Have you looked at the waveform on a scope?

Whenever I do logic square wave circuits I set up an oscillator to sweep the frequency range of a guitar, plumb it in, scope the circuit's output to make sure everything works, then troubleshoot (they never work  ), then plug in the axe.  Each to their own but I find troubleshooting easier this way, it saves time.

The waveform is just a square wave but its pulse width varies.  Instead of a nice smooth modulation its 'stepped', i.e. it 'jumps' from one width to the next sequentially.  It works fine across the whole range of frequencies but the effect sounds weak/is lost in the upper range...not really suitable for guitar, but fun with a bass.

[markusw]

Hi Mr Coffee,

thanks a lot for your explanations!

Always had the impression that it's better not to completely trust the results you get from Spice (even more since I'm definitely not an experienced user!). On the other hand, some predictions worked really well (e.g. the Ring Stinger simulation perfectly predicted the actual circuit).

Regarding the traingle wave from the integrator I did'nt explain properly. I meant that in the sims the traingle is not really a triangle in terms of symmetry. The rising edge is slightly bent upwards and the falling edge slightly downwards. At a first glance it is'nt even noticeable. Also the square wave after the first octave looks pretty "squarish". But after a second or third ocatve the triangle gets uneven and the pulse width of the "square" wave varies.

Anyway, I will order the missing parts for breadboarding 


Hi Gez,

thanks for the clarification!


Markus

[MR COFFEE]

Hi again Markus,

Regarding the traingle wave from the integrator I did'nt explain properly. I meant that in the sims the traingle is not really a triangle in terms of symmetry. The rising edge is slightly bent upwards and the falling edge slightly downwards. At a first glance it is'nt even noticeable. Also the square wave after the first octave looks pretty "squarish". But after a second or third ocatve the triangle gets uneven and the pulse width of the "square" wave varies.

Pulse width variation should be minimal at audio frequencies. But variation won't necessarily hurt the sound, just how many times you can multiply up cleanly. It's going to be pretty buzzy as is, whether a perfect square or not. And filtered, a little lack of symmetry gives you some even harmonics, which you don't get with a perfect square wave, so it may be a plus in the end.

The rising edge is slightly bent upwards and the falling edge slightly downwards.

Probably you are driving the integrator from a comparator chip with too high an output impedance for the size of the integrating resistor you are using. Assymetry can only come from 1) assymetrical , i.e., inadequately filtered input waves, or if you are using a symmetrical oscillator input, from 2) assymetrical output drive going into the integrator. You are using JFET or CMOS low input current op amps, right, not some dinosaur-era number bipolar input job off the AN?

If you are using a comparator with a pull-up resistor, decrease it to try and get a more even, low impedance drive, or switch to a comparator with a totem pole output (active pull up). Or are you running the output of the EX-OR CMOS IC directly into the integrator for the second stage? CMOS output impedance isn't very great or symmetrical.  And\or you can increase the input resistor going into the integrator (unless it's already huge). And you can see if you can change the value of integrating capacitor(s) for better results.

Or it could be comparator biasing isn't in the middle of the triwave where it belongs. Measure it and see.

Do you want to post your working schematic and\or waveforms from Spice so we can take a look at it and see what else could be the problem? If that's too much trouble, can you PM me a copy so I can get a better idea what's going on?

[markusw]

Hi all,

Pulse width variation should be minimal at audio frequencies. But variation won't necessarily hurt the sound, just how many times you can multiply up cleanly. It's going to be pretty buzzy as is, whether a perfect square or not. And filtered, a little lack of symmetry gives you some even harmonics, which you don't get with a perfect square wave, so it may be a plus in the end.

Maybe it really does'nt matter if the 3d or 4th ocatve is a perfect square wave. I just wanted to have the 3rd or 4th ocatve up square because it sounds cool with the Ring Stinger. But probably it's pretty OK with 3 octaves up and some more harmonics (it's pretty interesting looging at the FFT spectra of the various wave forms in LTSpice btw).

At the moment I'm trying to find apropriate opamps and comparators for breadboarding. What would be the specs you would search for??

Also I will post the schem and the wave forms I get in LTSpice. If you'd send me your email address Mr Coffe I could send you the LTSpice file.

If you are using a comparator with a pull-up resistor, decrease it to try and get a more even, low impedance drive, or switch to a comparator with a totem pole output (active pull up). Or are you running the output of the EX-OR CMOS IC directly into the integrator for the second stage? CMOS output impedance isn't very great or symmetrical.  And\or you can increase the input resistor going into the integrator (unless it's already huge). And you can see if you can change the value of integrating capacitor(s) for better results.

Thanks for the tips! Unfortunately, I only understand them partially.  :icon_redface:  Should be easier once I posted the schem.

I also added a 6dB/octave and a adaptive Schmidt trigger to my existing breadboarded 4046. It really could need further filtering. Also I'm pretty sure compression would help to make the response more even over the whole fret board. Unfortunately, the more I play with it the more I get convinced, that it will be too slow for bass. In the guitar range (80 Hz and higher) it works pretty fast. Will now try to replace the "damp" pot in the loop with one large and one small pot to allow more precise tweaking although I doubt that it will improve the situation significantly.

BTW, Gez I remember where I read about 2 cycles for lock-in. It was in a freq2voltage converter data sheet. Don't know of if PLLs are functionally at least in part compareable to F2V converters, but 2 cycles would fit to my observations with the 4046 too.

Regards,

Markus

[MR COFFEE]

Hi Markus,

At the moment I'm trying to find apropriate opamps and comparators for breadboarding. What would be the specs you would search for??

Lot's of JFET op amps will do fine for the integrators, but if you're buying some, it's probably a good idea to get JFET rail-to-rail output op amps so you can use then for the comparators as well. That will make the interface to the CMOS ex-or gates simple - as in just connect the op amp outputs to the CMOS inputs - and run everything off 9 volts. You don't need special comparator ICs at audio frequencies.

I'll PM you my email address.

I also added a 6dB/octave and a adaptive Schmidt trigger

In any case, I really think you need to get a better input filter for any multiplier, especially if you are trying the PLL approach. 6 db per octave is WAY too little. Try 12-18 db\oct above 500 hz, and then slope the response below 500 hz at 6 db per octave down to 80 hz or lower. It's the only way to stop slipping harmonics from screwing up the 50\50 duty cycle. Voice of experience here. And you still have to play clean

Without a good filter, your square wave drive will be assymetrical which makes it much harder for the loop to stabilize and actually lock as opposed to just semi-track. With the PLL, you'll hear the fundamental frequency and a semi-locked, sorta-harmonic multiplied buzz which is hunting around the multiple every cycle or two. Of course, since you like the ring mod thing... maybe simplicity and wild noise sounds is what you're after.

As I said before, I don't think the adaptive comparator is necessary if you filter the signal adequately. And if you don't filter adequately, the duty-cycle will change constantly as the harmonics drift relative to the fundamental, even with the adaptive comparator. Just fiter it well and square it up with a comparator (open-loop op amp). Run it open loop and the offset will give you enough "gate" to keep it quiet when you mute the strings.

[markusw]

Hi Mr Coffee,

thanks a lot for your support!

Lot's of JFET op amps will do fine for the integrators, but if you're buying some, it's probably a good idea to get JFET rail-to-rail output op amps so you can use then for the comparators as well. That will make the interface to the CMOS ex-or gates simple - as in just connect the op amp outputs to the CMOS inputs - and run everything off 9 volts. You don't need special comparator ICs at audio frequencies.

Thanks for the tip!

In any case, I really think you need to get a better input filter for any multiplier, especially if you are trying the PLL approach. 6 db per octave is WAY too little. Try 12-18 db\oct above 500 hz, and then slope the response below 500 hz at 6 db per octave down to 80 hz or lower. It's the only way to stop slipping harmonics from screwing up the 50\50 duty cycle. Voice of experience here. And you still have to play clean

Considering my 5-string bass guitar has a fundamental range from ~30 to 300 Hz I assume I could use a 12-18 (maybe 24) dB/oct filter for frequencies higher than 300 Hz. Would it also make sense to use 12 (or 18) dB/oct for the 30-300Hz range??

Another Q: is it possible by (heavy filtering and maybe some gain compensation) to get a more or less "perfect" sine wave from the guitar signal??

Regards,

Markus

[MR COFFEE]

Hi again, Markus,
Glad to help. Never got your sim.

Considering my 5-string bass guitar has a fundamental range from ~30 to 300 Hz I assume I could use a 12-18 (maybe 24) dB/oct filter for frequencies higher than 300 Hz. Would it also make sense to use 12 (or 18) dB/oct for the 30-300Hz range??

If you *only* want it to track bass, 300 hz is fine, maybe better. I never made one for a bass per se.

Considering my 5-string bass guitar has a fundamental range from ~30 to 300 Hz I assume I could use a 12-18 (maybe 24) dB/oct filter for frequencies higher than 300 Hz. Would it also make sense to use 12 (or 18) dB/oct for the 30-300Hz range??

I wouldn't *start* with 12 db per octave slope in the fundamental range. The point is to find a good compromise between  controlling low order harmonics on lower notes and not wimping out on the high notes.  Build it with 6 db\oct first, play through it, see what it needs to track the output you get from the way you and your bass - and there's a *big* difference between the neck and bridge pickups - the neck pu is better behaved for this purpose.

It may need a second rolloff for low frequencies, but you may need to shelve it for the upper range of fundamentals of your instrument. Different instruments have different characteristic waveforms in different registers and neck positions. That's why the old Tychobrahe Octavia really only sounds good when you play above the 12 fret. Guitars put out more sinewave-ish waves when played up high.

Another Q: is it possible by (heavy filtering and maybe some gain compensation) to get a more or less "perfect" sine wave from the guitar signal??

The only way you can get really superb tracking - i.e., get a sorta "perfect" sine wave - is to use more elaborate conditioning - that's where you get into more and more complex circuitry. Yeah you can tweak AGC, sliding or tracking filters to get ever more *perfect* oscilliscope pix and tracking, but how much circuitry do you want to build for a one-trick pony? And after you do a really good job of conditioning, you're back to the well-with-one-more-IC-it-could-also-do-this, and one more and it would do this too...

That's why it is a labor of love for people who are *really* into it.  One to four octaves up(double twice) is plenty for a useful sound - higher than that, put down your bass and pick up a six-string - maybe with an octave or two if you like. I play bass and six-string; you can only play so fast on those big bass guitar strings - if you want to play that high, you're better off changing instruments IMHO.

Hope this helps with design and finding your balance in your circuit design. Enjoy!

[markusw]

Hi Mr Coffee,

thanks once more for your help.

Will send the sims file again this evening when I'm at home.

Generally, I don't care too much if the circuit gets a little more complicated, even if it's just a one-trick-pony I just want it to work properly.

Do you think the following setup could work?
input buffer (gain stage) -> compressor (to reduce the peak transients  and get an even signal) -> LP filter (6-24 dB/octave or maybe a filter with different optimised slopes like you mentioned) -> AGC/compressor stage (to compensate for the volume drop at higher freqs) -> adatptive Schmitt trigger (or simply a plain comparator) -> PLL or any other doubler circuit

Also for the compressing stage what attack/decay and compression settings would be appropriate?? I suppose to catch the initial transient a rather fast response would be required. On the other hand, a too fast acting compressor will distort bass frequencies....

One to four octaves up(double twice) is plenty for a useful sound - higher than that, put down your bass and pick up a six-string - maybe with an octave or two if you like. I play bass and six-string; you can only play so fast on those big bass guitar strings - if you want to play that high, you're better off changing instruments IMHO.

True, but since I want to feed the passive ring modulator with my dry signal from my bass and a up to 3-4 ocatve up square wave (or a mix of the various octaves plus e.g. fifths) I can't change instrument

Regards,

Markus

[MR COFFEE]

Hi Markus,

True, but since I want to feed the passive ring modulator with my dry signal from my bass and a up to 3-4 ocatve up square wave

I'd suggest a plain mixer rather than ring mod. All the IM products every 80 hz. will sound horrible.

Do you think the following setup could work?
input buffer (gain stage) -> compressor (to reduce the peak transients  and get an even signal) -> LP filter (6-24 dB/octave or maybe a filter with different optimised slopes like you mentioned) -> AGC/compressor stage (to compensate for the volume drop at higher freqs) -> adatptive Schmitt trigger (or simply a plain comparator) -> PLL or any other doubler circuit

Yeah. With a bass, I don't think you'll need the compressor. Clip it instead if you need to get the signal up on note tails. I won't bore you by telling you to lose the adapttive Schmitt trigger again 

Also for the compressing stage what attack/decay and compression settings would be appropriate?? I suppose to catch the initial transient a rather fast response would be required. On the other hand, a too fast acting compressor will distort bass frequencies....

AGC should be fast attack 1-5ms, slow decay - 500ms maybe. Depends on your instrument. Infinite ratio. Distortion won't matter.

[markusw]

Hi Mr Coffee,

'd suggest a plain mixer rather than ring mod. All the IM products every 80 hz. will sound horrible.

Don't know about IM (intermodulation??) products (why 80 Hz??), just that the Ring Stinger sounds pretty cool if you set the internal oscillator to square wave exactly 3 or 4 ocatves up. Unfortunately, the Ring Stinger obviously keeps the same oscillator freq regardless of the note you play. Therefore, I came up with the idea of adding a circuit that produces the 3 or 4 ocatve up square wave.

Yeah. With a bass, I don't think you'll need the compressor. Clip it instead if you need to get the signal up on note tails. I won't bore you by telling you to lose the adapttive Schmitt trigger again

You don't bore me Thanks for the tip!

AGC should be fast attack 1-5ms, slow decay - 500ms maybe. Depends on your instrument. Infinite ratio. Distortion won't matter.

Stupid Q: AGC = compressor??? 

Regards,

Markus

[MR COFFEE]

Hi Markus,

Don't know about IM (intermodulation??) products (why 80 Hz??)

I'm not familiar with the ring stinger circuit per se. My interest in ring modulator circuits is kind of low these days. Too raucous for my taste from what I remember.

Ring modulators put out sum and difference frequencies, and cancel out the input frequencies. So you get a cluster of frequencies around each input frequency component of your 3-4 octave up filtered semi-square wave, consisting of those frequencies plus and minus the fundamental of your bass note. Also plus and minus all the harmonics of your bass note.

80 hz was just an example. For instance, with 3 octaves up, 80 X 2 X 2 X 2 = 640, and your output is 640 - 80 = 560 and 640 + 80 = 740 ... And 740/560 = 1.3something pitch ratio - I suspect pretty dissonant-sounding. Only you aren't filtering your octave signal down to a perfect sinewave (which would be pretty hard), you also get (third harmonic) 3 X 640 - 80 and 3 X 540 - 80, (fifth harmonic), and so on. And your bass isn't putting out a sinewave either, yada-yada-yada. A complex sound, shall we say.

You may like it more than I imagine I would. 

Stupid Q: AGC = compressor??? 

Not stupid Q at all. AGC = automatic gain control.

AGC is like a limiter, which is a basically a compressor with a hard threshold and infinite ratio slope designed to keep the output level constant irregardless of input level (above threshold).

I'll try to get LT Spice this week and check out your schematic and simulation output.

[markusw]

Hi Mr Coffee,

A complex sound, shall we say.

Definitely, even more complex probably considering I want to use a 3 or 4 ocatve up square wave which obviously has a lot of harmonics......

Re AGC: Thanks for your definition of AGC!  I found a cicuit that looks quite simple: http://www.elecdesign.com/Articles/ArticleID/6272/6272.html What do you think about it? Alternatively, could you probably target me to other AGC schems? 

Regards,

Markus

[MR COFFEE]

Hi Markus,

The AGC circuit you located is a good one; just run it on 9v instead of 5. It's a simplified version of the old standard "Orange Squeezer" circuit.

Of course, lose the unneeded R1 1K to ground across the input, and buffer your bass before you hit it - maybe a gain of 5-15x, depending on your bass. You can start to do some of your low pass or "slope down" filtering in that first gain stage amp.

I'd suggest putting a diode in series with the base of Q1 to get a higher output signal amplitude out of the AGC stage to drive your 18db\oct. LPF a little harder. I doubt it will increase the signal level across the JFET attenuator leg enough to make it distort excessively, ... you aren't going for pure tone out of your AGC anyway, just a stable waveform to filter and square up. 

A lower value for the series resistor R2 might do just as well or better - good thing to tweak by ear and\or with a 'scope. Note that the author of the circuit is talking about driving it with as high as 20 volts pk-to-pk. You won't be getting that hot a signal with a 9 volt supply on your buffer amp, eh? 

Is the LT Spice you are using the one that Linear calls Switchercad III?

[markusw]

Hi Mr Coffee,

The AGC circuit you located is a good one; just run it on 9v instead of 5

Great 

It's a simplified version of the old standard "Orange Squeezer" circuit.

Didn't realise that. Thanks for the hint!

Of course, lose the unneeded R1 1K to ground across the input, and buffer your bass before you hit it - maybe a gain of 5-15x, depending on your bass. You can start to do some of your low pass or "slope down" filtering in that first gain stage amp.

I'd suggest putting a diode in series with the base of Q1 to get a higher output signal amplitude out of the AGC stage to drive your 18db\oct. LPF a little harder. I doubt it will increase the signal level across the JFET attenuator leg enough to make it distort excessively, ... you aren't going for pure tone out of your AGC anyway, just a stable waveform to filter and square up. 

A lower value for the series resistor R2 might do just as well or better - good thing to tweak by ear and\or with a 'scope. Note that the author of the circuit is talking about driving it with as high as 20 volts pk-to-pk. You won't be getting that hot a signal with a 9 volt supply on your buffer amp, eh?

Thanks for all the mod tips!!! Will give them a try in LTspice first to get a better idea how the circuit works.

BTW, yes the LTSpice I'm using is SWCad III.

Regards,

Markus

[markusw]

Hi all,

here's a status report

In the meantime I breadboarded the AN-41 doubler. Basically, the schem can be described as follows:

<non-inverting opamp input gain stage> into

either

a: 6 or 12 dB/octave bp filter with Fc of about 30 Hz into LM311 comparator

b: 12 dB/octave bp filter with Fc of about 30 Hz into self adaptive Schmitt trigger (very similar to Boss OC-2) including the two flip flops in series (i.e. output is one octave down) into LM311 comparator

c: 12 or 24 dB/octave Buttherworth lp filter with Fc of 20 Hz  into LM311 comparator

a, b and c go into the first comparator of one of three consecutive AN-41 doublers. The three stages look exactly like in the AN-41 w/o the 100p cap (LM311 for both comparators, MC33171 for integrators). The values for C4 and C2 are

* 1st stage: 1µ and 100n for C2 and C4
* 2nd:        470n and 47n
* 3rd:         220n and 22n

Vref is provided by an opamp buffer and is heavily filtered as V+ (12V) is.

Basically the three octaves up work quite nicely even over the whole 5-string bass freq range (30 to ~450 Hz). However, three consecutive stages seem to be the maximum. After a fourth stage the sounds get too nasty (even for my taste ). When fed with a sine wave on the oscilloscope the wave shape at the output looks pretty squarish even after two octaves. Symmetry gets worse with each stage which seems to be the reason why 3 stages is the max. This is in accordance with the LTSpice sims although LTSpice exaggerates slightly. When looking at the scope trace of the triangle wave at the integrators' outs the edges have the same slight bump (break, don't really the know the proper expression) as in the sims.

All three variants (a, b, c) display some latency. Variant (c) is clearly the worst, the 24 dB version even worse than the 12 dB. As predicted by Mr Coffee the adaptive Schmitt trigger(b) doesn't improve the situation. The only advantage of (b) is the slightly more symmetric wave shape at the output of the second flip flop. After the first flip flop the wave shape is clearly less squarish than with a "plain vanilla" comparator .

Why variant (c) is that inferior I have no idea for. The signal is clearly heavily lp filtered, but the performance in combination with the three AN-41 doublers is depressing.

Obviously, since the whole mess is on breadboard and since there is no gate either there is a lot of noise when the strings are muted. I increased the offset voltage of the first comparator with a pot. Now the circuit is quiet when I carefully mute the strings.

When used as a carrier for my Ring Stinger clone background noise isn't really an issue since the RS gates anyway. The sound when used together with the RS is already pretty cool but the latency is annoying since immediately after plucking the string the sound is not immediately "present" like with the fixed stock RS oscillator.

The PLL in contrast has definitely higher latency (although I tweaked it quite a lot) but obviously yields a much " purer" square wave.

I also added a orange squeezer to the breadboard but it didn't improve latency of the AN-41 doublers even when heavily compressing.

Some Qs:

* what is that 100p frequency compensation cap for. do you think I need it??

* where does the latency come from? how can I minimise it?

* do you think that adding a gate in front of the first comparator of the first doubler would reduce latency??

* what can I do in general to improve the performance of the whole circuit

Thanks for your suggestions in advance!

Regards,

Markus

PS: Below is a pic of the whole mess. Have a good laugh

[toneman]

IT's ALIVE!!   IT's ALIVE!! 

now,


back 2 the...


regularly scheduled schedule...


:-)