String pluck detector??

[gez]

Just had a brief look at the 4093 data sheet. It includes the calcs for exactly what you suggested. A 1µ cap plus a 100k to 1 Meg pot should be fine to tweak the sample time.
Also it would be retriggerable which I prefer anyway. So incase you would play faster than the "sample time" it would extend the sample time (if I understood it correctly  ).

Markus

Sure you're not confusing this chip with the 4098 (retriggerable monostable)?  Or does the data sheet for the 4093 (which is a 2 input NAND with schmidtt inputs) show some applications?

Incidentally, you could probably use the 4070 to do what I suggested (replace the inverts & disable pulses to S&H), just that I think the 4093 is a better chip for this purpose.

[markusw]

Sure you're not confusing this chip with the 4098 (retriggerable monostable)?  Or does the data sheet for the 4093 (which is a 2 input NAND with schmidtt inputs) show some applications?

Incidentally, you could probably use the 4070 to do what I suggested (replace the inverts & disable pulses to S&H), just that I think the 4093 is a better chip for this purpose.

Now I'm sure I'm confusing something.  Obviously it doesn't work the way I want. Don't know if the correct expression is retriggerable.
The idea is: let's say the sample time is set to 100 ms. Now let's assume you play two notes with just 90 ms between them. With the NE555 wired like it is at the moment the second note wouldn't trigger. If I wire the NE555 like in the Electrax synth (as a attack/decay generator) it would be retriggered, thus the sampling time would be extended to 190 ms.
Is there a way to make it "re-triggerable" with a 4093?

Thanks again for your help!!

Markus

[gez]

OK Markus, understood.

You might be able to wire up a 4093 as a retriggerable monostable (with a few additional components) but it won't be pretty.

The 555 could easily be made retriggerable, but it would require DC coupling.  No big deal as R23 & C6 seem superfluous.  However, I'd use a 4098.  It would replace the 555 and the inverters as it has outputs that oppose one another (when one output is logic 1 the other is logic 0).  You can also use the reset pin to disable pulses (which you can also do with the 555 incidentally - don't know why that didn't occur to me earlier).

Only draw back with the 4098 is that its outputs can't source/sink much current so you'd need to use a discrete FET/trannie for the LED.

It might be possible to wire the other device in the 4098 as an inverter/buffer to do the job of the 4070, but I doubt it would perform that well...who knows?

[StephenGiles]

http://hammer.ampage.org/files/EH-Guitar-Synth.ZIP
See board 1 in the zip file for an example of 4098 use.

[markusw]

The 555 could easily be made retriggerable

How would you do this? 
The only way I found so far was the attack/decay generator of the Electrax synth but it requires a comparator to provide low and high signals, at least a comparator was the only thing that came to my mind 

as it has outputs that oppose one another (when one output is logic 1 the other is logic 0).

Definitely, an advantage over the 555.

You can also use the reset pin to disable pulses (which you can also do with the 555 incidentally - don't know why that didn't occur to me earlier).

Thanks for the tip! 

Only draw back with the 4098 is that its outputs can't source/sink much current so you'd need to use a discrete FET/trannie for the LED.

This would still be fine since a trannie plus a resistor or two doesn't take much space.

http://hammer.ampage.org/files/EH-Guitar-Synth.ZIP
See board 1 in the zip file for an example of 4098 use.

Thanks!! Will have a look at it. 

Markus

[gez]

The 555 is negative edge triggered, so you need to find a way to discharge the cap everytime the input goes low.  If DC coupled, the easiest way to do this is with a diode.  Connect the cathode to the input (pin 2) and the anode to the junction of the cap and resistor (pin 7).  This is the cheapskate way of doing things and it has three drawbacks:

1. The cap only discharges to 0.6V so timing is affected (doubt its critical here) and you have to compensate by making the resistor slightly larger in value, and

2.  You have to watch that the device preceding the 555 doesn't sink current to the point where it's damaged.  With a 30k resistor, you shouldn't have a problem (but always best to check data sheets).

3.  You'd need to fix the preceding comparator to be 'high' when in its quiescent state

The better way to do this (the method you find in text books) is using a PNP trannie.  Connect the base to the input, its emitter to pin7 and its collector to ground.  A p-channel MOSFET could also be used.  This is the best option for you as you won't have to bias the comparator to be high - keep R23 and C6.

[markusw]

Thanks a lot, Gez!!! 

Markus

[gez]

PS  Stick a Schottky diode in parallel with R23 - cathode to V+, anode to junction of C6, R23 and pin 2 - to give protection to the trannie & the 555 (shouldn't be any need for the additional protective resistor with the diode wired in there).

[markusw]

PS  Stick a Schottky diode in parallel with R23 - cathode to V+, anode to junction of C6, R23 and pin 2 - to give protection to the trannie & the 555 (shouldn't be any need for the additional protective resistor with the diode wired in there).

Thanks again! 
Could I also use a 1N4184 instead of the Schottky?

3.  You'd need to fix the preceding comparator to be 'high' when in its quiescent state

I'm confused now. The precceding comparator is high when in quiescent state.

Markus

[gez]

3.  You'd need to fix the preceding comparator to be 'high' when in its quiescent state

I'm confused now. The precceding comparator is high when in quiescent state.

My apologies, I didn't look at your schematic in detail.  If that's the case, you can simply use a diode - pull C6 and R23 and DC couple the 555 (no need for protection diodes/resistors) - or the trannie if you want more precision (again, pull all the other stuff - it's superfluous).

[markusw]

My apologies, I didn't look at your schematic in detail.  If that's the case, you can simply use a diode - pull C6 and R23 and DC couple the 555 (no need for protection diodes/resistors) - or the trannie if you want more precision (again, pull all the other stuff - it's superfluous).

Thanks for the clarification! 

BTW, I found a schem on how to convert a 555 to retriggerable. Unfortunately, it's in german but there are schems: http://www.elektronik-kompendium.de/public/schaerer/retr555.htm

It seems to do exactly what I want  but it is far more complex than you suggested.
Will check out your suggestions for sure because they are much simpler. LTSpice will help me understanding what's going on 

Markus

[markusw]

What do you think about this possibility? It's the attack/decay generator from the Electrax plus some 4093's....



It would be retriggerable and I could use one of the 4093's for cleaning up the squared guitar signal.

Regards,

Markus

[gez]

It would be retriggerable

Maybe I'm missing something, but I don't see how it is. 

As I've already mentioned, if you've biased the preceding comparator so that it's high when no signal is present, you don't need C6 and R23.

[markusw]

Maybe I'm missing something, but I don't see how it is. 

As I've already mentioned, if you've biased the preceding comparator so that it's high when no signal is present, you don't need C6 and R23.

It's taken from the Electrax schem page 17 http://experimentalistsanonymous.com/diy/Uploads/electrax.pdf. The description how the NE555 stage works is also in the file. It's for sure better than I could explain 

At least according to the LTSpice sims it should be retriggerable. C6 and R23 are also derived from the Electrax. If I understand it correctly C6 and R23 just make very short pulses from the comparator's output. So in comparison to feeding the precceding comparator directly into the 555 the pulse generated by the 555 should be more precise (time-wise). E.g. depending on picking strength the pulse length from the 555 would vary when fed directly with the comparator's out (e.g. let's assume the pulse generated from the comparator are 100 ms and the 555 is set to give a 100 ms pulse it would be 200 ms in total). Maybe it's just rubish 

Markus

[gez]

The description how the NE555 stage works is also in the file.

What page?  I'm afraid I don't have the time to wade thru all that!

If I understand it correctly C6 and R23 just make very short pulses from the comparator's output. So in comparison to feeding the precceding comparator directly into the 555 the pulse generated by the 555 should be more precise (time-wise). E.g. depending on picking strength the pulse length from the 555 would vary when fed directly with the comparator's out (e.g. let's assume the pulse generated from the comparator are 100 ms and the 555 is set to give a 100 ms pulse it would be 200 ms in total).

OK, from what I gather, you're trying to make it as close to positive edge-triggered as possible by reducing the pulses to a slither, so that triggering (which happens on the negative going edge) occurs quickly.  Still seems unnecessary to me.  I don't have time to go thru you circuit in detail, but if it's vital that triggering happens in phase with another side chain then you could probably just invert the phase of a preceding stage to the 555.  Maybe I'm missing something.

[markusw]

What page?  I'm afraid I don't have the time to wade thru all that!

It's on page 6 under "Envelope generator". BTW, the pages numbers are the total page numbers given by Adobe Reader not the ones printed on the pages.

OK, from what I gather, you're trying to make it as close to positive edge-triggered as possible by reducing the pulses to a slither, so that triggering (which happens on the negative going edge) occurs quickly.  Still seems unnecessary to me.  I don't have time to go thru you circuit in detail, but if it's vital that triggering happens in phase with another side chain then you could probably just invert the phase of a preceding stage to the 555.  Maybe I'm missing something.

Basically, I just want the pulses to have the same length irrespective of the pulse lenght generated by the preceeding comparator.
Sorry, why would inverting the phase of the comparator help?  It goes negative with each pluck. So I could leave C6 and R23 off but then sampling time would depend on picking strenght (at least to some extend).
BTW, I already added a protection diode from the 555's input to V+.

Markus
 

[toneman]

hey M,

haven't seen that Electrax article in a LONG time!!   
thanx for reminding me   
good article!!!   definitely a "modular guitar"   
and....there's your VCA!!!   U could sub a dual OTA like the 13700/13600 easily.

but, reguarding:
missing pulse detectors, ADSRs, ADs, retriggerable one-shots, LED pulse stretchers, threshold comparators and VCOs--
the 555 will do it all   
There are CMOS versions as well.
afn
T

[swt]

I've also built the electrax in perfboard. It's triggered is slow as a turtle. But the attack realease function works well. It would be nice to get it working with the ams trigger. By the way markus...can you tell me what can be wrong in the lfo of the electrax?. it's square wave it's somewhat strange. it also ticks like hell. Thanks.
By the way, the harmonic generator also works fine, and the octaves down tracks better than some pedals, but have to do some mods to get it right. Filter is amazing, and also vca with ring mod type sounds, although carrier bleeds a little, specially with squarewaves.

[gez]

Basically, I just want the pulses to have the same length irrespective of the pulse lenght generated by the preceeding comparator.
Sorry, why would inverting the phase of the comparator help?  It goes negative with each pluck. So I could leave C6 and R23 off but then sampling time would depend on picking strenght (at least to some extend).

How would it depend on picking strength?  A comparator only has two states: high and low.

I don't think I fully understand what it is you're trying to do.  Why do you need all the pulses to have the same time Length from the initial attack of a note to the finish of the one-shot's timed period?  If you're going to make the 555 retriggerable its output will always be high, and the circuit always sampling, until the signal drops below a threshold set by the comparator (I'm presuming you've done this); only then will the 'hold' part happen.  I somehow get the feeling this isn't what you want to achieve?

[markusw]

I've also built the electrax in perfboard.

Wow! 

By the way, the harmonic generator also works fine, and the octaves down tracks better than some pedals, but have to do some mods to get it right. Filter is amazing, and also vca with ring mod type sounds, although carrier bleeds a little, specially with squarewaves.

Good to know! Thanks for the information! 

By the way markus...can you tell me what can be wrong in the lfo of the electrax?. it's square wave it's somewhat strange. it also ticks like hell. Thanks.

I'm not sure that I'm the right one for asking  but I will have a look at it!

How would it depend on picking strength?  A comparator only has two states: high and low.

If picking strenght is higher the comparator would stay high (or low) longer. 

Why do you need all the pulses to have the same time Length from the initial attack of a note to the finish of the one-shot's timed period?

I want it to sample just for as much time as necessary to achieve lock-in (should be around 120 ms). So sampling should be done immediately after the initial attack.

If you're going to make the 555 retriggerable its output will always be high, and the circuit always sampling, until the signal drops below a threshold set by the comparator (I'm presuming you've done this)

Assuming a sampling time of 120 ms it should only retrigger if one would play faster than one note each 120 ms. If you play slower, it samples for 120 ms and then stays at this frequency for at least some time. The idea is, that due to the four 4066 switches the PC2 out stays at the level of the last note. This way the lock-in  for the next note (provided it's not 2 octaves above or below) will be faster (or at least less noticeable). This works pretty well. If I have the PLL running at idle frequency and play a note the lock-in is more noticeable than if I play the same note a second time (or another close note).

If it would sample as long as the signal is above a certain threshold PC2 out voltage would drop at the end of notes, because the envelope follower, retriggerable monostable or whatever is used for triggering simply isn't fast enough for turning to hold. With a defined sampling time this issue doen't exist.

Hopefully, my explanations are not too confusing. 

Markus