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

Doepfer A142 delay length
MUFF WIGGLER Forum Index -> Eurorack Modules  
Author Doepfer A142 delay length
MrBiggs
I got me a Doepfer A142 a while back so I can CV the length of the gates coming outta my MFB Seq-02. The idea was to send trigger from the Seq-02 to the A142, which would then send a gate or it's A/D envelope to my VCA or whatever. But I'm having dick of a time working this.

The Doepfer page says that the shortest decay time is about 2 ms, but that's just not the case. It's more like half a second when the delay knob is all CCW. However, when I CV the delay, I can get it much shorter -- just a snap. So maybe that's what the 2 ms is referring to. Anyone have similar behavior? I use CV for this most of the time, but I can't believe the difference is so much. It would be nice to have manual options shorter than the knob gives me.

Secondly, the CV from my Seq-02 doesn't seem to have enough oomph to affect the CV control of the A142 without pumping up the signal somehow. The A133 polarizer does a good job with this. But since I use the A133 for other things, would a small straight-up amplifier like the A183-3 work? Alternatively, is there some adjustment that might be made on the Seq-02 to strengthen its CV, or to the A142 to make it a bit more sensitive? Doepfer mentions that the "value" of the capacitor can be changed to modify the shortest time, but I don't know if that's what I'm trying to do here.

The process works seamlessly on my Model 10, so I'm tempted to merely find another one. But that's really overkill to just control these gate lengths.
REwire
Are you using the Decay Out or the Gate Out? If you use the first it won't get that short but the second is also governed by the Threshold control which will make the gates extremly short when a notch over disappearing

Dan
MrBiggs
REwire wrote:
Are you using the Decay Out or the Gate Out? If you use the first it won't get that short but the second is also governed by the Threshold control which will make the gates extremly short when a notch over disappearing


In this case I'm talking about Decay Out, which is what Doepfer describes as being 2 ms at its shortest. With a Gate Out I'd have to add on another ADSR which is ok, but since there is this envelope already on the 142 that's supposed to be really quick, it would be nice to not need another module to get it to work.

(The gate also has an odd response where there is a very small space on the dial where one gets a usable short gate between "really long" and "nothing.")
MrDys
To me, it sounds like your 142 is malfunctioning. Without CV and with the threshold set appropriately, I can get really quick envelopes and gates.
MrBiggs
MrDys wrote:
To me, it sounds like your 142 is malfunctioning. Without CV and with the threshold set appropriately, I can get really quick envelopes and gates.


I was afraid of that.

Since the threshold only affects the gate, you're saying that merely with the Decay knob turned counter clockwise, you're getting a fast envelope?

Fug. waah
MrBiggs
Well you learn something new every day.

I pulled the A142 out to check the capacitor that Doepfer writes about, and noticed a blue offset trim thingie on the PCB. With my trusty tiny screwdriver I nudged this thing clockwise miniscule amounts and it made drastic changes to the delay time. It's still overly sensitive in that, if the 5 position is a nice quick popcorn decay, nice for quick bloopy beepy stuff, past 4 to the left gets unusable clicks and past 6 makes for very long decays that are basically slow whole notes. Don't know how to fix that.

The CV response is more predictable now as well, and this makes the gate threshold more usable as well.

The problem still exists that I have to amplify my trigger from the Seq-02 to use this, so maybe I'll email Manfred and ask him if this can be rectified.
Suburban Bather
MrBiggs wrote:
I have to amplify my trigger from the Seq-02


Are you using trigger or CV mode for triggering the A-142? I'm not sure what the voltage level is for the trigger mode, but if you use CV mode you can increase the voltage up to 9volts. 5volts should do it though. As for triggering with the Seq-02, triggers are 32nd not gates and CV's are 16th note with variable voltages. I don't see anywhere in the manual saying that you can adjust the voltage level for triggers like you can with the Seq-01.
MrBiggs
Suburban Bather wrote:
MrBiggs wrote:
I have to amplify my trigger from the Seq-02


Are you using trigger or CV mode for triggering the A-142? I'm not sure what the voltage level is for the trigger mode, but if you use CV mode you can increase the voltage up to 9volts. 5volts should do it though. As for triggering with the Seq-02, triggers are 32nd not gates and CV's are 16th note with variable voltages. I don't see anywhere in the manual saying that you can adjust the voltage level for triggers like you can with the Seq-01.


Triggers. Triggers trigger the A142. The CVs on the Seq02 are used to control the CV of the A142. I haven't figured out yet whether the Seq02's triggers are weak or whether the A142 is insensitive. I would guess the latter, since the Seq02 triggers things like my ADSR and other EGs just fine.

Do you know what is the calibration trimmer on the back?
Suburban Bather
I know nothing about the trimmers.


I'm suggesting that you try using the CV's to trigger your A-142 instead of the triggers. My Seq-01 would not trigger my Maths until I upped the voltage level to 5volts. Maths was the only module I was having trouble with prior to increasing the voltage level of the Seq-01 triggers. Give CV mode a try and don't forget to increase the voltage of that CV via the data knob on the Seq-02.
MrBiggs
Yeah that don't work. The CVs of the SEQ-02 act like gates. I suppose maybe if this were an A/R instead of A/D envelope that might do it. But it takes me back to the original problem of the decay too long.

Since I have this A132-3, the second VCA on it switched to linear does a nice job of amping it up for now and I suppose if I keep doing this and don't find a way to solve it in the 142 itself the 183-3 will be it's friend.
Tim Stinchcombe
OK, very easy to get confused with all the controls and terminology here, so I shan't attempt to cut-and-paste quote sections from the above, but try to start from scratch...

First, the trigger. The A142 is triggered by a sharp rising edge - this needs to be a step-change of at least 0.8V for it to trigger properly, so let's say a nice round 1V or more. If yours isn't triggering well with a voltage above this, then there may be something wrong - mine easily triggers down to 0.8V, and even below that it still generates an envelope (though not 'full height') until the trigger step-change is below about 0.5V, when it will no longer trigger.

Next, the gate out length. This depends on two things - the decay setting and the threshold setting. Look at fig 1 in the manual:
www.doepfer.de/a100_man/a142_man.pdf
For a given decay setting, as the threshold moves up, the gate will shorten, as it moves down, the gate will lengthen. Now consider the threshold fixed: as the decay is lengthened, then the gate will become longer, and v-v.

Thus to get a 'smooth' voltage-control of the gate length, you will probably need to experiment with where the threshold setting is: if too low, then the slightest change to the decay will make the gate length shoot up; if too high, you might need to provide lots of CV to make any noticeable impact.

The trimmer. The circuit contains an exponential converter, which presumably translates the CV/pot position into a nice tractable decay length, in some sort of pleasing way. The trimmer determines where the 'start' of this affect occurs. You'd have to ask Doepfer exactly how they set it up in the factory, if you want to return it to the 'factory settings'; on mine, with the 'decay' and 'CV' pots fully CCW, the envelope length is about 5ms, if you have some way to measure /estimate this!

(Much of this becomes much easier (literally) to see if you have an oscilloscope...)

Hope this helps, and doesn't confuse too much - if you need any other measurements/timings, let me know, or I could provide some scope shots.

Tim
MrBiggs
Thanks for the all the info Tim.

The Seq-02's triggers in the mode I am using are 9v, so by all estimations plenty enough to zap the A142 just fine. Are you really meaning .8v? As in 8/10ths of a volt? If so then there is almost certainly something amiss with the A142 I have, if more than ten times that voltage isn't working.

I'm good with the gate out and threshold and how they work. It takes some twiddling as you describe, but since it hinges on the decay working correctly it's not always a breeze.

The trimmer on the PCB seems to adjust where on the DECAY pot the usable range sits. Like I wrote, I adjusted it so that the length I like best sits at noon, or 5. This is sans CV. The problem with it is that if I move it to 4, it's so short as to be less than a tiny barely audible click, and at 6 it's more than a second long -- way too long for my use. So the trimmer is really an offset, as it is labeled. It would be nice to have a way to spread this range out rather than merely move it around.

Even with these issues, when I amplify the Seq-02 signal and use CV also from the Seq-02, I get some nice results. It's finicky but it works. As the sequencer runs through its steps some of the steps are all legato-like and some are short and sharp.

I plan to test some other triggers out with the A142 to find some consistency. I have Volta, so I'll use that. And I'm taking delivery on a Machinedrum next week and will use the triggers on that as well. I wonder as well whether the clock out of the uLFO will trigger the A142.

I do have an oscilloscope. It's a MOTU software oscilloscope and while I've had great joy in looking at the waves of my VCOs, I've not worked it with an EG yet. I'll try it out. I also have a multimeter but I have no idea how to test the voltage coming out of a module.

thanks again
DGTom
def. sounds wierd man, neither of my 142s work anyway like the way you are describing that. Maybe someone already swapped in a bigger timing cap or something?

The Decay pot is never that either / or, goes from super short CCW to long @ CW seriously, i just don't get it

Sucks, take some pics & get in contact with Doepfer see if they can spot anything out of place.
MrBiggs
DGTom wrote:
def. sounds wierd man, neither of my 142s work anyway like the way you are describing that. Maybe someone already swapped in a bigger timing cap or something?

The Decay pot is never that either / or, goes from super short CCW to long @ CW seriously, i just don't get it

Sucks, take some pics & get in contact with Doepfer see if they can spot anything out of place.


Yeah will do. Rage!
thx
Tim Stinchcombe
MrBiggs wrote:
The Seq-02's triggers in the mode I am using are 9v, so by all estimations plenty enough to zap the A142 just fine. Are you really meaning .8v? As in 8/10ths of a volt?
Yes, absolutely - that small.

Quote:
I'm good with the gate out and threshold and how they work.
Cool.

Quote:
The trimmer on the PCB seems to adjust where on the DECAY pot the usable range sits.
Yep, that is what it is doing.

Quote:
It would be nice to have a way to spread this range out rather than merely move it around.
I'll double-check how that part of the circuit operates, but it should be possible with a single resistor value change to achieve a 'less agressive' exponential curve.

Quote:
I do have an oscilloscope. It's a MOTU software oscilloscope and while I've had great joy in looking at the waves of my VCOs, I've not worked it with an EG yet. I'll try it out.
Never heard of it, but if you can choose a level for it to trigger at, then it will probably prove useful.

Quote:
I also have a multimeter but I have no idea how to test the voltage coming out of a module.
Less useful in this scenario as the voltages of interest are changing so quickly.

I've also attached a snap of the PCB, to identify the two main caps of interest - if your module is second-hand and potentially 'modded', it would be worth checking the values of these two. The 'input cap' C4 gives a brief pulse from the rising edge of the trigger signal input - quite a wide range of values may work here, but if it has been changed for one way off that originally specified, it probably will cause grief. Mine is 10nF, as the silkscreen says.

The timing cap C3 obviously controls the decay length, but again, if it has been changed, it may not rise quickly enough, which will look like the same symptoms as not triggering properly. Mine is 100nF (=0.1uF, again as silkscreen). Also check the value of R20, immediately to the right of the CA3080 - is it still a 1k? (again, the timing cap charges through this, so if bodged about with, will have an impact). And as I write 'CA3080', I realise it's possible the module may have been re-designed, so yours might not look like this. Ho hum.

Tim

Edit: PS - just discovered I have a complete SPICE simulation of this circuit, so I can easily see about flattening the response curve out, but this will have to wait until after I sort my guttering out = well scary prospect working 20 foot up a ladder...
MrBiggs
Tim you are a god among Wigglers, I don't care what they say about you in the subforums. razz

This is GREAT information. The A142 is sitting at my studio and I am at my breakfast table so I cannot look at it right now. But I will stop by this afternoon and peek behind the scenes. My PCB is the same design -- I do know that -- but looking at it yesterday I could not tell whether the timing cap was the specified one from the Doepfer site. I do not remember seeing the red input cap at all but that doesn't mean it wasn't there.

The oscilloscope is part of the Ultralite's "CueMix" software package and upon reading about it last night it should be capable of telling me what I need to know re: length of the envelope's decay. I don't know if I can get to that this weekend but I'll attempt to carve an hour or two out.

Thanks again for all your time and help.
Tim Stinchcombe
MrBiggs wrote:
Tim you are a god among Wigglers, I don't care what they say about you in the subforums.
Stop it, you're making me blush! oops

Quote:
My PCB is the same design -- I do know that -- but looking at it yesterday I could not tell whether the timing cap was the specified one from the Doepfer site.
I sought out that Doepfer DIY cap page - had I seen that before, it would have saved me the photo! Presumably that there it shows a CA3080 too suggests they either have lots of built modules in stock, or more likely he bought truckloads of CA3080s before they stopped production, so he may have several years supply before he needs to do that re-design.

Quote:
I do not remember seeing the red input cap at all but that doesn't mean it wasn't there.
It will probably just be a different make & colour.

Quote:
The oscilloscope is part of the Ultralite's "CueMix" software package and upon reading about it last night it should be capable of telling me what I need to know re: length of the envelope's decay.
I watched their promotional video, and it definitely has the kind of triggering capability that is needed!

Tim

(Needless to say I survived the DIY up a ladder - now all I need is some rain to see if I have actually fixed the problem, and I'm sure it will do that soon enough without wishing for it!)
MrBiggs
Okay, haven't done the oscilloscope yet. But I have my A142 here.

The Input cap is labeled like so: 10nK100
Is that similar to the 10nF you describe?

The Timing cap: .1K63
Yours says u10J63. These seem like different sets of code than the 100nF and .1uF you write about. I should maybe refer to my electronics book.

The resistor at R20 looks is indeed 1k and look like this:


Unless you catch something that isn't right in all that, I'm guessing that these values are correct. And if this is true then it is to Dieter I shall write.
Tim Stinchcombe
MrBiggs wrote:
The Input cap is labeled like so: 10nK100
Is that similar to the 10nF you describe?
Yep: 10n = 10nF, K is probably the tolerance, 100V.

Quote:
The Timing cap: .1K63
Yours says u10J63.
.1 = 0.1uF, again K for tolerance, 63V ('u10' is just shorthand for 0.10uF).

Quote:
The resistor at R20 looks is indeed 1k and look like this
Seems good too.

Quote:
I'm guessing that these values are correct.
Yes, they all seem good.

It occurred to me you should check R18 too, at 47k, right next to C4 - this also affects the length of the charge pulse, derived from the rising edge of the trigger. But it is starting to look like component values aren't the problem...

Tim
MrBiggs
Yeah 47k.
Thanks Tim.

B waah
Tim Stinchcombe
Having run a few simulations I discovered a copy-error in my circuit tracing, and this in turn allowed me to understand how the module is probably set-up in the factory: with no CV in, decay fully CCW, adjust the trimpot until turning it further doesn't give any decrease in the envelope duration (so the fastest is around the 2ms quoted). (My error was to miss out a current-limiting resistor in my 'tidy' redraw of the circuit, as opposed to the messy, raw tracing). From your descriptions above, I'm not sure if you actually hit this point (?).

Anyway, to flatten out the exponential curve effect controlling the envelope durations, the thing to do would be to reduce R9, perhaps like halving it, by soldering another 4k7 across it (and then re-adjusting the trimpot for the max envelope once more).

And another thing to check might be that IC3 is actually a 78L10, and not one of the transistor types used elsewhere in the circuit - simulation suggests it might run, but not whether anything would get blown up with too-high currents/what-have-you. And if it is a 78L10, measuring that it is giving 10V might be worth doing, best accomplished across the outer two pins of the threshold pot.

Tim
MrBiggs
Tim Stinchcombe wrote:
with no CV in, decay fully CCW, adjust the trimpot until turning it further doesn't give any decrease in the envelope duration (so the fastest is around the 2ms quoted). (My error was to miss out a current-limiting resistor in my 'tidy' redraw of the circuit, as opposed to the messy, raw tracing). From your descriptions above, I'm not sure if you actually hit this point (?).


Your suggestion led to a slight adjustment which ended up solving one of my problems, but not the big one. By adjusting the trim, the trigger on the Seq02 now easily triggers the A142. I had the fastest time ending at around the 12 o'clock position for simplicity sake, which made the triggers seem weak. I shoulda caught that. So now it triggers wonderfully but still goes through its entire useful range between the 1 and about 2.5 positions of the dial. By 3 it's at more than 12 seconds, while 1 is a 1ms click (measured on the oscilloscope).

Tim Stinchcombe wrote:
Anyway, to flatten out the exponential curve effect controlling the envelope durations, the thing to do would be to reduce R9, perhaps like halving it, by soldering another 4k7 across it (and then re-adjusting the trimpot for the max envelope once more).


I am purchasing a 10w-40w soldering iron and clearing space on the desk. Time to learn me some new skillz.

I made a little (maybe longish) video today showing the use of this A142 with the Seq02. You can see when I adjust the Decay knob that the adjustments are tiny with big changes in the envelope. But then you can also see that in the end it works. I just don't have the room for minor adjustments that I'd like.

Thanks again for the help.

MUFF WIGGLER Forum Index -> Eurorack Modules  
Page 1 of 1
Powered by phpBB © phpBB Group