MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

A-162 Trigger Delay mod (was: 'monoflop to ping filter?')
MUFF WIGGLER Forum Index -> Eurorack Modules  
Author A-162 Trigger Delay mod (was: 'monoflop to ping filter?')
Navs
I normally use Maths channel 1's EOA to do this, but it seems an 'expensive' use of the module i.e. just to convert a gate into a trigger to ping a filter.

Does anyone use their A-162 for this and, if so, can you tell me whether it can be set for zero delay i.e. instantaneous and how hot the output is?

Maths EOA is ca. 10V, which works well. The VCS EOC is 5V, which is a bit weak for the task.

Thanks smile
Reptil
Yes this works.
As for the output voltage, can't measure it right now (gnomes stole my probe cables), but the module takes a tiny bit of the input signal. No problem pinging my a-121, generates a nice sinus.
Kodama
Yes the a162's length is great for getting just the right ping from my a101-2
Klopfgeist
Yes I use the A-162 for this and it works great.
Navs
Thanks for the answers, guys smile

I picked one up today and it outputs over 10V!

Great module - it doesn't re-trigger until the gate has run its course, so it can be used for cool clocking and audio applications.

However, the downside for pinging filters is that if I move the length setting a fraction off zero, the filter double triggers i.e. the gate is too long. I get more throw with Maths and this can have a marked effect on the filter's ring.

The Doepfer DIY page lists the timing caps used, so I would like to mod mine:

http://www.doepfer.de/DIY/a100_capacitors.htm

The standard value is 10 uF. If I replace this with a 1uF, I would reduce the time by a factor of 10, which would be too much.

I couldn't seem to find any 5 uF caps in the right format - are these even made? Is there a way to halve the time, similar to running resistors in parallel? The Doepfer page seems to allude to this in regard to the A-119, but I don't understand it fully.

Thanks again!
DGTom
add a diode to one of your passive slews so it only slopes down, the voltage drop might mean you are abit more restricted in what will drive it, but as long as you have one of Doepfers ~10V gate outs handy you should be ok to ping filters etc.

Simple Decay generators are a fantastic utility to have around (even better paired with a comparator or 2)
flashheart
I'd expect the caps form part of a CR timing circuit, so halving the resistor values should do the trick, but this may have other unseen effects on the circuit.
What I'd like to know is how you found that page, and are there any others like it - I know the main DIY page but where are the other hidden gems?
felixer
Navs wrote:
Is there a way to half the time, similar to running resistors in parallel?

sure. but it works out just the other way around: putting two identical caps in par will double the value. in series is half.
maybe make a little switch? i'm sure there is some room between/under the jacks .... pm me if you need help. i've got the tools & done this before Mr. Green
flashheart wrote:
What I'd like to know is how you found that page, and are there any others like it - I know the main DIY page but where are the other hidden gems?

the timing caps page is a link just at the start of the main diy page hihi and dieter has always been extremly helpfull if i email about some mod .... or just ask around here 8_)
tim stinchcombe's site also has a link there. many really well based mods/ideas. he's around here too ....
Navs
Cheers!

DGTom, the diode might prevent re-triggering, but wouldn't get me the needle-pulse I want. However, I think the quad slew + A-162 are going to be good mates.

Flashheart, the page was linked on the A-162 page, so no Holmes-ian research on my part! I've also had very helpful, detailed responses from Dieter to my technical inquiries.

Felixer, thanks - that's what I needed to know! I'll give it a shot bzw. mail you.
Navs
Tis done - bought and modded in under 24 hours lol

http://navsmodularlab.blogspot.com/2011/02/patch-of-day-ping.html

1 uF didn't give me much more range, so I used a 0.1 uF blocking cap!

This is better, but I wonder whether the pots might also play a role i.e. lin vs log?
felixer
congrats! sounds beautifull thumbs up
fm-ing with a short puls is something i do a lot on the vco to emphasize 'plek/stick impact'. mostly with the gate out on the a142. which is a bit fiddly to set right, but it works. and the lenght changes with the release time cv 8_)
concerning pots: hard to see but looks like 'A1M'? which would mean log 1 megaohm. that seems good. anyway: replacing the pots would require some surgery .... adding a range switch is easier.
A Dingleberry Monstrosity
Navs wrote:
However, the downside for pinging filters is that if I move the length setting a fraction off zero, the filter double triggers


I imagine you could easilly add a fine tune knob. hmmm.....

...well not easily in terms of realestate
MrBiggs
I still get nooby now and then: What exactly are you "pinging?" That is, is this resonance/self-oscillation we're hearing? What's the patch? Is the trigger hitting the frequency input?
Navs
Ah, I thought that might need explaining ...

Take a filter that self-oscillates, turn the resonance up to the point where it starts, then back-off just enough to silence it.

Now apply a short trigger (Maths EOA, A-162 etc. at min settings) to the *audio* input.

This should give you a Ping! (or plock, umpfh or bong, depending on tuning).

The tuning is relevant as you might have to adjust the resonance. Higher frequencies will ring earlier. Resonance sets the length of 'decay' or dampening.

The Roland TR-808 kick, for example, is a pinged band-pass filter. Experiment with filter types (LP, BP or, if you have an MMF-1, dual peak BP love ) and also filter slope (12dB, 24dB).

Once you've got the basic sound, you can also apply a short envelope to the filter's CV input to 'twang' your drum. Experiment with +ve and -ve envelopes to imitate the natural pitch bend that occurs when you bang a drum.

Do you still have your M12? It won't self-oscillate, but thanks to its wet and rubbery character, also sounds great when pinged. It won't 'ring', but it will pock & plop!
MrBiggs
Thanks Navs. Yeah I still have the M12 but I've been thinking of trying out the Sem20 as a possible replacement. I also have a Z2040 which will work well here. I'll try them this weekend. I've got a uStep and uScale on the way so there shall be timed and quantized pinging...

nanners
Tim Stinchcombe
Navs wrote:
This is better, but I wonder whether the pots might also play a role i.e. lin vs log?
After reading this it occurred to me some simple simulations could hopefully answer the question, and I believe the answer is 'yes'.

The A-162 circuit is pretty simple: each half of the 556 timer chip (a dual 555) is used identically, and they are cascaded, so the first controls the trigger delay, and the second the trigger length. The incoming trigger triggers the first 555 circuit: the (555) output goes high, and the timing cap then starts to charge via the 1 meg pot; when the cap gets to the threshold voltage (about two thirds supply, or around 8V), it'll turn the 555 off, and the cap discharges; the second 555 sees this falling edge, and it triggers in turn, setting its output, and this is the one output from the module. The second cap charging time determines the length of the output pulse.

Thus simulating a cap charging from 10V, through a 1 meg pot gives an indication of how long either the delay or pulse will be. I have models of the 'standard' JIS pot laws (including the little 'dead zone' at either end of the travel, of which you can see the effect in the following plots by the last two traces being very close). The following curves show a 100n cap charging for 11 settings of the pot (0, 1, 2,...10); read the times off where the curves cross the 8V line. The top set are for the 'A' log law pot, then underneath the 'B' linear law. Since the 'A' law is merely two different linear slopes, you can see two distinct sets of spacings between the curves where they cross the 8V line - the longer times have a larger gap between them. For the linear pot, the spacings are more even across the whole lot, suggesting that you may have better control of the faster times if you were to change it, because the curves are not so closely spaced:


[I think (and hope!) that this makes sense, and is the correct way round for the law 'A' vs 'C' pots - in any case it will be easy to check by simply measuring either the delay or pulse width at pot settings 0, 1, 2, etc. and comparing them against these plots...]

Tim
Reptil
nice, thanks for sharing, know what to do now.
thumbs up

and yes, that filter sounds good, pinged
vav
Hm...i love the sound and want to try this, but i dont really want to buy a second A162 as i do use it as is...any enterprising person wanna make an expander module for it? Maybe like a 4HP with a switch for each to change from Factory to Modded and a multiple for the rest of the panel?
felixer
Tim Stinchcombe wrote:
For the linear pot, the spacings are more even across the whole lot, suggesting that you may have better control of the faster times if you were to change it

thanks tim thumbs up so in navs case a 'C' antilog pot might even be better ... if you can find one ... been thinking about putting a ten-turn pot into my a142. not sure if it would physically fit hmmm.....
vav wrote:
any enterprising person wanna make an expander module for it?

it would take an equally enterprising customer since it would involve some soldering. get your local tech to mod it for you if you can't diy. better yet: start to learn diy Mr. Green
vav
felixer wrote:
Tim Stinchcombe wrote:
For the linear pot, the spacings are more even across the whole lot, suggesting that you may have better control of the faster times if you were to change it

thanks tim thumbs up so in navs case a 'C' antilog pot might even be better ... if you can find one ... been thinking about putting a ten-turn pot into my a142. not sure if it would physically fit hmmm.....
vav wrote:
any enterprising person wanna make an expander module for it?

it would take an equally enterprising customer since it would involve some soldering. get your local tech to mod it for you if you can't diy. better yet: start to learn diy Mr. Green


Oh i can do the mod, i just dont know how to make circuits myself. Replacing the caps is the easy part razz
felixer
replacing the caps pretty much IS the mod. then put the original part par to the new/smaller value. with one leg going thru a switch ... or is it the metalwork that puts you off? often the most expensive part of diy is a slick looking frontpanel .... if plain alu and sharpie markings don't put you off it gets a lot easier Mr. Green
Reptil
I was thinking two small switches on the Doepfer front panel.
There's plenty of space for another (small) capacitor?
Navs
Thanks for taking the time to investigate, Tim!

I was toying with the delay offset last night and, of course, now realize this needs to be modded too. The range is fine for long delays, but unsuitable for subtle timing offsets or 'swing'.

What I don't understand is how, despite reducing the length cap by a factor of 100, a setting of '10' is still close to a second long. Looking at your plots and description of the circuit, am I right in thinking that the two parameters, delay & length, are linked i.e. the one influences the other?

Another mod that might be interesting would be to implement a re-trigger option, so that it triggers on each incoming impulse, but I have no idea if this is possible.

For anyone wanting in a comparator-based solution, this might be of interest:

http://navsmodularlab.blogspot.com/2009/03/synth-diy-trigger-twin-dual -comparator.html

Peter's trigger delay, and my adaptation, will follow changes in timing, where the A-162 only allows fixed delays.

Also, MrBiggs, the filter-pinging/ drums description is straight out of Florian Anwander's essential 'Synthesizer' book:

http://www.schneidersladen.de/florian-anwander-synthesizer

Packed with useful information and practical examples, but sadly only in German - I've been badgering him for years to get it translated!
Reptil
yes, very good! your comparator is an excellent Idea.
I got two Ken Stone Slope detector PCBs and doepfer VCLFOs planned for this purpose. but having CV-able trigger delays in one module is much neater.

the Florian Anwander articles in KEYS magazine is what raised my interest in 1996 for the doepfer systems. very clear and inspiring to go and experiment.
thumbs up
Chuck E. Jesus
Navs wrote:

Also, MrBiggs, the filter-pinging/ drums description is straight out of Florian Anwander's essential 'Synthesizer' book:

http://www.schneidersladen.de/florian-anwander-synthesizer

Packed with useful information and practical examples, but sadly only in German - I've been badgering him for years to get it translated!


i'd buy this in a second...
frozenkore
Just tried out pinging my M11 and M12 and all I have to say is MY ASS IS BLEEDING . I'm definitely going to be messing with this more. Thanks for all the info!
Tim Stinchcombe
Navs wrote:
What I don't understand is how, despite reducing the length cap by a factor of 100, a setting of '10' is still close to a second long.
That does seem a bit long, and I have little suggestion as to why that might be. I can only suggest trying a different cap type - it sounds like you used a ceramic one, which according to one of my books doesn't suffer from particularly high DC leakage, but a different type might behave differently.
Quote:
Looking at your plots and description of the circuit, am I right in thinking that the two parameters, delay & length, are linked i.e. the one influences the other?
They are linked only in as much that the 'length' only starts once the 'delay' has finished, i.e. the one follows the other. Other than that there is no interaction between the two.
Quote:
Another mod that might be interesting would be to implement a re-trigger option, so that it triggers on each incoming impulse, but I have no idea if this is possible.
This would depend on what you'd expect the 'retrigger' effect to be. A simple mod, adding a transistor+resistor, might allow a second incoming pulse during the delay period to reset the delay, so that it starts timing from the second pulse (and so the output pulse takes longer to appear). If a second pulse arrives whilst the output is high, as the circuit currently is, if the delay is longer than the remaining pulse length, a second pulse should start again soon after the current one finishes. If the delay is shorter than the remaining pulse, then the new delay which has been initiated won't lengthen the pulse, though again I think the addition of a transistor could do this...

Quote:
Also, MrBiggs, the filter-pinging/ drums description is straight out of Florian Anwander's essential 'Synthesizer' book
In the past I have had occasion to look at the odd Analogue Solutions drum module, and from memory, I think this is a fairly 'standard' technique for producing drum sounds, as used by many of the Roland circuits that the ASol modules are based on.

Tim
Navs
Thanks for the confirmation, Tim.

I could have another look at the cap. It's curious, but the range is fine for what I want.

I was thinking more in terms of having the gate re-initiate, rather than re-setting the delay time, similar to the way, IIRC, the A-142 works. A mod might be interesting, but its current behaviour also has its uses (clock division, syncopation, delayed accents).
felixer
for more precision there's always this:

http://www.reichelt.de/?ACTION=3;GROUP=B27;GROUPID=3139;ARTICLE=42140; SID=40TUi4hn8AAAIAAFtpuQwf72545b5f0fa1306470def3db84e6785
pricey and you still need to drill a small hole for the locking pin, but neat .... and will fit if 10turn pot is too big. or isn't available in the desired range (largest i found was only 100k).
Navs
Looking at the board, replacing the pot would be a PITA.

The modded A-162 (0.1 uF) gives me a ca. 1.6 ms pulse at a length setting of '1', which is great (according to Florian Anwander the TR-808 uses ca. 1.5 ms pulse).

Maths offers finer control when sweeping the response with R&F @ zero: ca. 0.4 - 1.2 ms. I can hear the difference these minimal adjustments make on the tone of the 'drum'.

Is there any danger in using an even smaller cap value to get me the additional range or am I really going to have to tackle the pot?
Tim Stinchcombe
Navs wrote:
Is there any danger in using an even smaller cap value to get me the additional range or am I really going to have to tackle the pot?
Thinking about an answer to that question has led me to the reason for the previous anomaly:
Quote:
Navs wrote:

What I don't understand is how, despite reducing the length cap by a factor of 100, a setting of '10' is still close to a second long.

That does seem a bit long, and I have little suggestion as to why that might be.
Longer time settings correspond to the larger resistance values from the pot, and this in turn means we get little current. At the end of the charging period this current is down to about 2 micro amps (with the pot at max); the 'input bias current' drawn by the threshold input of the '555 chip is around several hundred nano amps, and thus it 'robs' current from the cap, making it take longer to get to the threshold level. Presumably with a larger cap, as the times are so much longer anyway, this 'slowing' effect isn't that noticeable (I'm kind of guessing here, as you can probably tell - I can't be bothered to simulate it, which would tell the correct story...), but with a smaller cap, this effect will be a greater part of the whole, and so seemingly is noticeable.

Thus reducing the cap even more, in itself, won't pose any danger, but it might just start behaving a little weirdly. The solution of course would be to lower the pot too, and thus bring the current levels back up, so the proportion bled off by the threshold input doesn't overly impact on the cap charging.

Tim
flashheart
If you wanted to lower the value of the pot could you just put a resistor in parallel with the current pot rather than replace it?
Navs
Thanks Tim & flashheart - that's exactly the information I needed.

This means I could possibly increase the cap to a 1uF, which should give me a one second span.

Do you have any suggested values for the parallel resistor?

Also, which lugs should I connect it to e.g. the outer lugs or a > b/ c > b?

http://doepfer.de/A100_pictures/doepfer_a162.jpg

help
Tim Stinchcombe
flashheart wrote:
If you wanted to lower the value of the pot could you just put a resistor in parallel with the current pot rather than replace it?
Yes, good idea - it will completely mess up the log response, but it is so simple to do compared to swapping the pot out, it is probably worth a try. So, simulation to the rescue once again: I re-discovered the function that reads across a series of traces, so here we have the times read off the plot above, for the standard 1meg pot with the 100nF cap (red); the green is a 1uF cap, and with a 100k resistor across the wiper of the pot (x-axis is pot from 0 all way round to 1 = max clockwise):

The red displays the typical two-linear segment approx to the log pot; the green is very much more linear.

Which legs to solder the resistor to is fairly obvious: on each pot on the module two of the legs are clearly joined with a trace - the resistor goes from one of these to the third leg!

Tim
Navs
Thanks for the plot, Tim.

As you predicted, the parallel resistor played havoc with the pot's response. I tried 10 different values, from 1M to 10k. The latter reduced the span to the range I wanted, but settings above 3 yielded little change.

I left the pot alone and replaced the 0.1uF/ 100nF with a 0.01uF/ 10nF cap, thus reducing the max length to about 13ms. The first 50% of the pot's throw now covers settings up to ca. 2ms - perfect Guinness ftw!

I'll post some pics and sounds later.
Navs
http://navsmodularlab.blogspot.com/2011/02/162-trigger-delay-modificat ion.html

smile
komyta
Navs wrote:
Take a filter that self-oscillates, turn the resonance up to the point where it starts, then back-off just enough to silence it.

Now apply a short trigger (Maths EOA, A-162 etc. at min settings) to the *audio* input.

This should give you a Ping! (or plock, umpfh or bong, depending on tuning).


I just listened AGAIN (hundredth time, maybe lol ) to these pinged filters tracks !

I need to give it a try with one of my A-140 today.
Navs
Delay cap swapped:

http://navsmodularlab.blogspot.com/2012/03/simple-mod-tuesday.html

Much better Guinness ftw!
Navs
Modded again!

http://navsmodularlab.blogspot.de/2012/10/a-162-cv-mod.html

Added voltage 'control' grin
kozepz
Navs wrote:
Modded again!

http://navsmodularlab.blogspot.de/2012/10/a-162-cv-mod.html

Added voltage 'control' grin


Informative thread! thumbs up
Fantastic demo, never enough percussion! Really looking forward to cv demo!
hyper
felixer
Navs wrote:
Modded again!

http://navsmodularlab.blogspot.de/2012/10/a-162-cv-mod.html

Added voltage 'control' grin

applause
Estes
hi all I was just wondering if you managed to make a mod of the gate lenth for both up and down of the A-162 ?

I did the exact same thing on both but somehow only one is working. I did first the upper one and it worked and then I made the same change und the lower one but no only the lower gate delay does the job.
MUFF WIGGLER Forum Index -> Eurorack Modules  
Page 1 of 2
Powered by phpBB © phpBB Group