Building the tap tempo tremolo

[thehoj]

I've got it built, but I've got ticking. It's not too bad but I can't completely get rid of it. The biggest issue I have right now I guess is that I get the ticking noise even when it's bypassed..
How could that be happening?

[thehoj]

So it would seem that the LED being connected through the 3pdt switch even when disengaged causes ticking in the bypassed signal.

I think What I'm going to do is add a second LED which is only there to be on/off if the pedal is engaged/bypassed, the other LED connected to the PCB will be directly connected.. That might be a bit weird since it will always be on when there is power, but I think it will be handy for setting my tempo and seeing that tempo in the pcb led even when the pedal is disengaged.

As for the ticking when engaged.. I guess I'll have to start checking components. Would it be worth upping the values of the trimpots I'm using so that I've got more play there? I suppose I could wire a resistor in series with the pots somehow to bring their values up a bit..
Or is it more likely that my optocoupler is noisy. I did buy a second one of those as well.
I only really hear the ticking when I've got the waveform set to the hard on/off settings.. When it's set to the sinewave setting I don't hear the ticking.

[Taylor]

I find replacing the opto i the easiest way to fix ticking. But the audio section of this is identical to the tremulus lune, so searching for ticking fixes with that circuit will turn up lots of things you can try if the new opto doesn't fix it.

[thehoj]

Well, this is frustrating. I've tried my new opto and the ticking didn't change as far as I can tell. It is pretty mild and happens when it's doing the square wave, but it's still there.
I've looked thru some of the tremulus lune forums. A couple things I tried are putting a .001uf cap over the gain pot, and a 100uf cap from v+ to ground. Those things didn't seem to make any difference. Am I just being too sensitive, should there be absolutely no ticking at all regardless of waveform and volume?
Anyone else run across this?

[Taylor]

I do have some ticking with some optos. Others seem to have no ticking in square mode. It is a pain to have to sort them, so I've talked to Tom about doing a version of the code where the square and ramp waves are just barely angled. This is something we may have together at some point in the future.

[thehoj]

I think I've got it pretty good now. The volume at which I need to have the amp when on the square wave and still hear ticking is so loud that you would never notice the ticking anyways. That and I'll be primarily using the sinewave anyways, which I have no ticking with.
One thing that I noticed is that (in my case anyways) to reduce any extra noise when the pedal is engaged, having that .001uf cap over the gain pot helps.

[Taylor]

That's actually what the 330p cap is already doing, so what you've done by adding that cap is to up the capacitance there to 1.33n. I may make a suggestion in the PDF to use a bigger than 330p cap there.

[.Mike]

I believe the 330pF cap is what was originally in the schematic for the Tremulus Lune.

When I built my (non tap) version of this, after I figured out that too much current to the LED was causing a lot of ticking, I ended up using a 1nF cap in that position to get rid of the ticking once and for all.

Mike

[joinpobob]

First off, thank you Taylor  and .Mike for all of your hard work (and to the others that contributed).

I have a question about the pdf under Modifications. It says, "Any pot can be modded for CV input easily." Is that referring to adding an expression pedal? I think that makes sense but would like to verify.

Thanks!

[Taylor]

Sort of. What I'm referring to there is that, for guys with modular synth systems, or more experimental folks, you can send a control voltage (CV) to the pot inputs directly. So synth guys could use this with their modular systems which send all control signals as CV (and since 0v-5v is, I believe, a common CV format, it works perfectly with the TAPLFO chip). This way you could have a sequencer change the waveform on every beat, or an LFO scrolling through multiplier settings for extreme animated weirdness.

If you just want to add an expression control, you don't need to think about voltages or anything. Simply wire a jack to the 3 pads of the pot you want to replace, making sure that you wire it to be in compliance with the expression pedal you're using (they're all wired differently). Use a switching stereo jack so that when you unplug the expression pedal, the pot becomes active.

[KazooMan]

Taylor:

I am finally getting around to building the tap tempo tremolo using the board and chip I got from you.

I have a question about the multiplier that I can't seem to find an answer for in the forum thread or your build pdf.

If I install the SPST momentary contact switch at the EXP pads on the board to use this to step through the multiplier stages what do I do with the multiplier pot?  Do I leave it in or replace it with resistors?

[Taylor]

You can have both a switch and the pot if you want. If you use the switch, it steps to the next multiplier, then if you move the pot, it overrides the switch setting. If you don't want a multiplier pot at all, I guess you'd want to stick 5k resistors from the 5v and gnd pads and have them both meet in the center pad.

[KazooMan]

Thanks,  That's interesting.  I understood how using the rotary switches in place of the pots would work, but I wasn't certain about the multiplier.  I'm don't know how they manage to do it, but from your explanation the SPST switch takes precedence and then if you adjust the pot the chip goes back to using that as the control.  I'll build it with the SPST and the pot and see how that works.  It's sort of like wearing a belt and suspenders at the same time. 

[joinpobob]

Taylor,

Thanks for your reply; that makes sense. I don't do any of that, so I guess I'm fine without it.


I am proud to report that I finished the build, and it works perfectly. I was easily able to tune the circuit, using the trim pots, to get the hum out of the circuit. I will get a full report with pics and comments up soon.

One thing to note (and I think you said this earlier, Taylor), I would keep it simple and not add the rotary switch. It is very easy to figure out which waveform you have selected. Also, the multiplier with both pot and switch function just as said, each takes precedence over the other when changed. I am going to play with it more, but I am leaning towards just going with the pot. The switch doesn't let you switch back and forth between settings; it just lets you move through the series. I don't think that it would help me "on the fly" (but I guess it would save me from having to bend over and change settings), and I think I would rather have more foot room, especially if I'm tapping.

[Taylor]

It would be pretty easy to rig up a stomp switch to swap between 2 multiplier pots if that's what you wanted. You only need to swap the center pin so you could use a DPDT stomp (or SPDT even). Then you could preset them as you please and switch back and forth mid-song.

[likefireflies]

Hi all,

I have my tap tempo completed, but I am running into a snag that can't be lived with.  When I send a signal generator it works fine, but when I send a passive signal(guitar) all I get is a really weak signal and massive ticking... I have put a second optocoupler in and it still is not mitigating the ticking. I have also adjusted the two trimpots all to no avail.  Any ideas? And thank you Taylor for all your help thus far.
Cheers,
Eli

[Taylor]

I'd start by swapping the 330p next to the opamp for something above 1n, as noted a few posts up.

[.Mike]

If a stronger signal seems to work but a weak signal doesn't, maybe the input signal is being reduced. Maybe the signal generator can overcome the loss, but the guitar can't.

Check the values of R11 and R12 and make sure they are correct (220K). If R11 is larger (I think) than R12, it would attenuate any incoming signal.

Mike

[likefireflies]

I'd start by swapping the 330p next to the opamp for something above 1n, as noted a few posts up.

Of course if I would have read first I would have seen that huh?  I can be impatient at times, I am sure it is what lead to the problem I am struggling to fix now. I will try the cap change, also I will check those resistors. Thanks guys.

-Eli

[thehoj]

Since I'm putting an order in at Smallbear for a few things, I figured I'd try a different optocoupler, just to see if I can completely get rid of the ticking I have.
I sent a few PM's to taylor regarding it, so I've posted them below my further questions regarding optocouplers, in case anyone else was thinking of doing something similar.

The datasheet for the NSL-32 shows 500ohm on @ 20ma and 500K off resistances, and 3.5msec rise /  500msec fall times.

In looking at those Vactec datasheets,
I see from the graph that the VTL5C10 shows about 50ohm on @ 20ma and 400K off resistances, but a shorter rise time and longer fall time.. 1msec rise and 1.5sec fall.
The VTL5C7 also looks interesting, in that it's 6msec rise and 1sec fall (maybe this is more what I'd be after, but again the on/off resistances are different.. what bearing does that have on things?) It looks to be about 50ohms on @ 20ma also, and 1Mohm off.

I'm leaning towards trying the VTL5C7, but I am hoping to get some suggestions on what to try. Basically I'd like to try something that will soften any ticking that might occur.

I'm needing to order a few things from smallbear for something else in the next few days, and so I was thinking (even though the ticking I was seeing on the trem is so minimal and only on square wave) maybe I would try another optocoupler.. Is there a different one all together that might be worth trying, that smallbear would stock?

Hey, I haven't tried any others yet, but the best thing to do would be to check my PDF to see what resistances I recommend (I forget at the moment) and compare the NSL-32's rise and fall times to those of some of the vactecs. If you find a vactec with longer time, it should tick less.

datasheets:

http://www.alliedelec.com/Images/Products/Datasheets/BM/SILONEX_INC/Silonex-Inc_Actives-and-Passives_6995014.pdf

http://www.elenota.pl/upload/VTL5C2.pdf

You mention 500ohm on and 500Kohm off.
The rise and fall times of the NSL-32 are 3.5 and 500msec respectively.
What do the rise and fall times translate to in this circuit? Will having a longer rise or fall time affect the speed of the tremolo?

Each cycle of the LFO, the opto rises (goes from zero volume to max volume) and then falls (goes from max to zero volume). This creates the tremolo effect. If you find an opto with slower times, it will round off the hard edge of the square wave slightly, which will alleviate ticking. It doesn't change the speed of the trem, only the time it takes to go from silence to full volume, and at the times we're talking about, thousandths of a second, you don't hear the difference really. It just makes the transition a little softer so the momentary current pull is spread out a little bit which should fix the ticking.