Building the tap tempo tremolo

[Barcode80]

Hmm, strange. That's the same type I'm using and haven't experienced that. As you can see in the schematic, the tap switch connects directly to the PIC, with just a resistor pull-up. So I can't think of any error in your build that could be causing it.

From TTG's page on debouncing, the simplest thing to try would be a cap across the switch terminals, but this probably won't help much. The logic debouncing is pretty simple and would do a much better job. That would be what I'd try.

Did you just build this up, or have you had this issue for a while? I only ask because I have built quite a few of these and only had that problem at first because it was my first tap tempo effect, and I didn't have the hang of tapping solidly. I don't in any way mean to impugn your tapping skillz, just wanted to mention that as a possible place to look if you hadn't yet. It might be a total non-issue, but just trying to help.

No offense taken! I've actually built two of these, and the first was the same way. I chalked it up to the LED not charging up and down fast enough to accurately indicate the tempo. But this one is way more defined an issue than the first one.

[turunturun]

So, after executing what I thought was a super-clean build, I fire up the freshly completed Tap Tempo Tremolo and.....nothing. No LED, no effect, zero. So begins troubleshooting. Everything looks OK!!! Optocoupler in right (ie side with "dot" is facing the two trimpots...that is correct, right?)....power, ground appears OK....bypass switch ok.....all knobs set north/middle....what gives.

So after looking REAL close, I realize that my TL072CP is actually a TL022CP. My order was for a 072, and the bag it came out of was labeled 072, but the top of the chip says 022 (you can actually read it in one of the high res photos). Looking at the data sheets they appear to have the same pin assignments, but (slightly) different operating properties....

So could this incorrect op-amp cause the thing to not work at all?

Symptoms:
I hear the signal when pedal is bypassed, I hear the signal when pedal is on. No tremolo or LED though.

FYI, the two outer jacks are the audio in and out, and the two inner jacks are a CV in for multiplier, and clock out from Pin 7 of the TAPFLO. Those two inner jacks are hooked up with a single wire each that goes from the tip lug of the jack to the respective point in the circuit in a "piggyback" fashion. So for instance, the multiplier knob is a PCB mount pot, and I connected the CV wire to the center leg.

Also, all the jacks are pretty close to each other and to the pots below but I have made absolutely certain they are not touching each other or anything else and they are lock-washered snug and not going anywhere.

Here are some build pics:


high res here:
http://www.flickr.com/photos/jasonandlindy/5597246531/



high res here:
http://www.flickr.com/photos/jasonandlindy/5597246371/

Thanks for any help!

[.Mike]

Alright, so you get bypass and you get sound through the effect, but no tremolo.

Since you're getting sound through, it is not the opamp. The audio path and the LFO are, for all practical purposes, independent.

The dot on the NSL-32 indicates the negative side, so the way you have it oriented should be correct as well (Taylor can confirm this).

The one thing that stands out to me as being something to check is the resistor for the Wave Distort. Looking at the resistor that you soldered there, it looks like Brown-Black-Black-Orange-Brown-- 100k. It is supposed to be 10k-- Brown-Black-Black-Red-Brown.

I think that would set the wave distort all the way to one side-- in this case the dark side-- and make it so that the LED just can't turn on. And since the LED can't turn on, the signal goes straight through with no change in volume.

That's the first thing I would check.

Mike

[turunturun]

WOW! Good eye. I did indeed have a 100K in the wave distort spot instead of a 10K. Swapped that out, but it is still doing the same thing. Only now when the effect is engaged there is a "volume swell" from silent to less quiet than the bypassed signal. The swell occurs over about 1 second, and then the volume stays at the "less quiet level" without any modulation. Bummer. Still no LED.

If I had the LED wired backwards would that possibly result in the symptoms I am experiencing? I am pretty sure I wired it correctly but I don't have any way to check without dissassembling and redoing it.

Thanks,

[.Mike]

WOW! Good eye. I did indeed have a 100K in the wave distort spot instead of a 10K. Swapped that out, but it is still doing the same thing. Only now when the effect is engaged there is a "volume swell" from silent to less quiet than the bypassed signal. The swell occurs over about 1 second, and then the volume stays at the "less quiet level" without any modulation. Bummer. Still no LED.

So the sound drops out, and then fades back in (but quieter) when you engage the effect? Odd. Well, at least the right resistor is in there now-- a step in the right direction!

If I had the LED wired backwards would that possibly result in the symptoms I am experiencing? I am pretty sure I wired it correctly but I don't have any way to check without dissassembling and redoing it.

If we assume the NSL-32 is oriented correctly (and it should be, according to the datasheet and the PCB silk screen),  you can at least check that the external LED is wired properly without removing it. The cathode (negative) terminal should have a flat spot on it, like this.

Mike

[turunturun]

Yeah - thanks for the help Mike. (PS Hi from Atlanta!).
So unfortunately I positioned the LED under the circuit board and with PCB mounted pots....I will have to take everything out of the box in order to remove the PCB, in order to take a look at the LED.
A thought just occurred to me. Couldn't I test the LED as a diode with a multimeter's probes on the right parts of the circuit to confirm orientation? Lets give that a try....

As a side note for next time, I might get things working OUTSIDE the box on my next build, then stuff it all in there...! !!

[turunturun]

Tested the LED as a diode with my multimeter and its in there correctly (it even lit up with the multimeters current! who knew?!).

So it sounds like a botched job on the waveform distort knob pins could lead to the issue I am having? (ie accidentally reversed 1 & 2 or something???).
Or not?

[.Mike]

Well, the only reason I suspected the LED was oriented wrong is because you seemed unsure. When you're unsure, double-checking is in order. 

I think it's time to grab your multimeter, and take some voltages. The procedure is listed here.

In particular, make sure you get voltages on the PIC. 

Mike

[Taylor]

Are you positive you have the opto in right? It's important not only that the dot side face toward the trimpots, but that you get the resistor side (long legs) in on the "east" side (facing the trimpot above the TL072) and the LED side of the opto in on the PIC side of the board.

I don't know if the labeling of all NSL-32s follows a convention or is random, but if yours is like the ones I have here, then you have it in wrong. Looking at the ones I've built, if I oriented the "32" the way it is in your photo, it would put the resistor side of the opto on the left, and the LED on the right, which would be incorrect.

[turunturun]

Hmmm, I snipped the leads, so can't tell which is the long lead side. Seems like it should be possible to test it while its in the circuit tho, right? Like use the diode feature on a DMM and find out which side is the diode side?

[Taylor]

Right, that should work. Also, the LED side's leads are flattened, and the resistor side's leads are round in cross section. I'm looking at your hi res photo and it looks to me like you have the LED side in the resistor pads and vice versa.

[turunturun]

So now would be a good time for me to find a good tutorial on the best way to desolder a PCB. Every time I try it I lift a trace or ruin a pad or both.....( game over, man!!!!).

Thanks a for taking a look Taylor....who knew? I thought you could just stuff that 32 bug in there and cover it with solder.

[Hides-His-Eyes]

To desolder; yank the wire out with pliers while melting the solder.

[Taylor]

Indeed, but make certain that both the top and bottom sides of the pad are molten before pulling. Most of the problems I see people having with desoldering double-sided boards happen because people flow one side of the joint, but don't wait for the top side to melt, then they rip out the plated through hole.

Get a healthy amount of solder on your tip before flowing the joint. A desoldering braid or pump is cheap and helps quite a bit.

[turunturun]

Thanks for the tips! That is what I have been doing in fact, but to sufficiently wick the solder out enough to reopen the hole I end up burning the pad and making it too cooked to accept solder. Or at least that is how I have ruined a couple PCB's. I have built a couple amps with good success....I figured pedals would/should be easier!?!? I think the TTT has been a bit more difficult than the Spitfire clone I am building from scratch right now also....!!!! (not really....doing heater wires is a PITA).

Seriously tho, it is fun building the effects and the work that the DIY community has put into generating these incredible full-featured circuits and chips is amazing.

Thanks!

[Taylor]

Hmm, well that's part of the reason I prefer a solder pump to wick - never have any issue pulling pads. But I know others swear by wick/braid, so different strokes and all that.

[turunturun]

That makes sense regarding the pump. I never found it useful, but then I was trying it on amp sized components and cables where a fair amount of solder was involved. But the pump seems to make a lot more sense on a circuit board. Thanks!

Looked close, and yup....my LED and resistors are reversed. Bummer. It just seems like the 32 looked so good the way it was.... Time to play "operation". hehehe.

[Ben C]

Long time reader, first time poster. I just finished my build (pic below) - everything works great except for, like Ed, the ticking on some waveforms. On my build its the first and third ones that make the ticking - ramp up and square. It's very not noticeable when playing, but when muting the strings it's definitely there. I'm running an external volume pot that runs near the wave selector but not near the PIC. Cables are only as long as they need to be.

I can get rid of the ticking by cranking the trimpot near the PIC, but then the gain pot can't go about 9 o'clock without the pedal pumping out a waaaay too loud signal. Besides diming the trimpot, any other setting yields exactly the same ticking level when adjusted for gain. There also seems to be a loss in dynamics with the trimpot cranked, so currently I've got it set so I have unity gain when the Gain knob is at 12 oclock.

[Ben C]

OK - after having a mess with my build I've found a fix for my ticking issue. First I disconnected the LED to see that had any effect. Immediately the ticking disappeared. Next I tried messing with some resistors and found that I could remove the ticking by connecting a 10k resistor between ground and the positive LED pin. This dims the LED a bit and makes the LED trimpot more responsive - I can dial out the ticking while still having the LED blinking bright enough. All good so far. 

[turunturun]

Here are some debug readings for my TTT

Battery Terminal with new battery:
Pos: +8.9
Neg: 0

Circuit Boards 9V in:
Pos: +8.9
Neg: 0

TL072 (which is actually a TL022 b/c I haven't received the 72 yet...they appear similar in function per data sheet):
1: +4.2
2: +4.2
3: +4.2
4: 0
5: +4.2
6: +4.2
7: +4.2
8: +8.5

TAPFLO:
1: +5.0
2: +2.2
3: +2.3
4: +5.0 (drops to 0 when tap button is pressed)
5: +2.5 (stays steady...I would think this should be fluctuating???)
6: 0
7: 0 to 5 (voltage fluctuates pretty squarely and tracks the tempo that has been tapped in via button)
8: 0
9: 0 to 5 (appears to track movement of wave distort knob correctly)
10: 0
11: 0
12: 0
13: 0
14: 0

Wave Distort Pot:
1: 0
2: 0-5
3: +5

All other Pots:
1: 0
2: 0
3: 0

So it appears that only the wave distort knob is getting +5V on pin 1. From there it looks like it is supposed to go to the 10K resistor, then to Pin 3 of the Tempo knob. Since I don't get anything on P3 of the tempo pot I measured resistance from the "north" leg of the 10K resistor to P3 tempo and got 12K ohms or something....I looked over the solder joints and they all appear good, and the trace doesn't look damaged. Not sure why it would be reading out resistance.