A/DA Flanger refuses to flange. any advice?

[andrew_k]

I finally got around to building the A/DA flanger and have been attempting to get it working for more than 6 hours with minor progress but no success 
I've read every page of every thread I can find related to this circuit and have tried everything I can think of before posting this plea for help..

mods: dual opamp is TL072, R13 changed from 10k to 4k7 to allow more room to move on TR1. Everything else is stock, including all jumpers (pads C, D, E, K and S+Sw on JK1)

symptoms/status clean signal comes through fine. Playing while adjusting TR1 results in a different tone at approx 3/4 rotation; much like a subtle chorus without any LFO. I left TR1 there.  Sounds like there is definitely some mixing of altered signals in there as adjusting TR4 while holding a chord gives a hint of flange/chorus, similar to the sound of adjusting the time control on a delay while playing -- that pitchshift/catch-up sound.

The voltage readings were taken with the following settings: Threshold 100%, Manual 0%, Range 100%, Speed 50%, Enhance 50%, all trimmers at 50%.

supply: 18.3v
V+: 14.9
Vb: 7.4

4007  (CD4007UBE)
1: 1.29
2: 1.03
3: 0
4: 0.73
5: 0.84 - 0.95
6: 0
7: 0
8: 0.68
9: 0.03
10: 7.7
11: 8.53
12: 0.1
13: 14.9
14: 14.9

4047  (CD4047BCN)
1: 14.8
2: 0
3: 8.53
4: 14.9
5: 14.9
6: 14.9
7: 0
8: 0
9: 0
10: 7.46
11: 7.46
12: 0
13: 14.82
14: 14.9

4049  (CD4049UBE)
1: 14.9
2: 7.45
3: 7.45
4: 7.45
5: 7.45
6: 7.45
7: 7.45
8: 0
9: 7.45
10: 7.45
11: 7.45
12: 7.45
13: 0
14: 7.45
15: 7.45
16: 14.9

SAD1024
1: 0
2: 9.43
3: 7.45
4: 0
5: 5.73
6: 14.88
7: 14.88
8: 7.46
9: 0
10: 7.46
11: 0
12: 7.68
13: 0
14: 7:46
15: 9.44
16: 0

IC1  (LM324N)
1: 7.4
2: 7.4
3: 6.7
4: 14.76
5: 5.77
6: 5.8
7: 5.84
8: 7.49
9: 7.38
10: 7.39
11: 0
12: 7.42
13: 7.49
14: 7.49

IC2  (LM324N)
1: 4.99
2: 4.99
3: 4.96
4: 14.76
5: 7.38
6: 7.39
7: 7.94
8: 7.32
9: 7.28
10: 7.38
11: 0
12: 7.36
13: 7.38
14: 7.38

IC3  (LM324N)
1: 5v - 9v sweep
2: 7.24
3: 7.35
4: 14.82
5: 7.39
6: 7.36 (unstable)
7: 7.8
8: 7.43
9: 7.21
10: 7.21
11: 0
12: 0
13: 0
14:3.64

IC4 (TL072)
1: 13.23
2: 7.83
3: 3v - 12v sweep
4: approx. 3
5: 7.38
6: 7.39
7: 4v - 12v sweep
8: 14.52


I know i'm asking a lot by requesting assistance with this monster, but I've hit a brick wall were the problem and circuit far outreach my knowledge (despite many, many hours of research to try and get some understanding of what's going on).

Thanks for reading this far, and any advice you have to offer

[StephenGiles]

Did you build it all without testing as you go as I have been suggesting for months - you cannot hope to have success on first fire up if you don't, because there are so many sub units to consider. That sounds hard, but I do it and I have been at the soldering iron since 1970!

[andrew_k]

Did you build it all without testing as you go as I have been suggesting for months

I'll give you two guesses, and the first one doesn't count...    Stupid me stuck all the resistors in first and then worked my way from smallest components to largest. Even now that I've studied the schematic for hours on end I don't think I'm knowledgeable enough to have built it as you suggest anyway. I did probe audio at all the pad points before putting in the jumpers and they *seemed* fine, but it wasn't until I tried to follow mossapotamus' guide to calibrating by ear and couldn't get past setting TR1 that I knew I had trouble.

I've tried the obvious things like ensuring pins 10 and 11 on the 4047 match and dropping R13 in hope of finding the right spot on TR1, but alas, the LFO doesn't do the O bit...  Any advice at all... please?

[StephenGiles]

I'm short of time right now, but can someone please help out here, we can't have a non working ADA!!

[slacker]

From your voltages it looks like the LFO is working, pin 7 of IC4 is changing voltage so that's a good sign. If you want to double check the LFO get an LED and a 1k - 10k resistor and hook them up in series between IC4 pin7 and ground, the LED should flash at the speed of the LFO.

I think the problem is that the LFO signal doesn't seem to be making it as far as IC3 pin 9 because that pin should be oscillating at the same rate as pin7 of IC4. This could just be because you've wired the range pot up backwards so that you've really got it set to 0 not 100% try turning it and see if anything happens. I appreciate that you've probably already done that so sorry if that sounds like stupid advice, but it's the sort of thing I'd do.
If that doesn't do anything measure the voltages on all the components between the 2 pins and see where the LFO signal disapears.
Once you've got the LFO signal on pin 9 of IC3 you should then get the signal on pin 8 of IC3 and pin 10 of IC5, once you get that the LFO should be driving the SAD1024. Then it should just be a case of adjusting T1 to bias the SAD1024 until it flanges and then setting the other trimmers to taste.

[andrew_k]

WOOO  Thanks slacker, the LFO disappeared at the Range pot; replacing the range pot made the whole thing come alive.

...except...

It's vibrato only, it would seem. The even cutting the jumper at pad C and probing the left of the two pads gives the same vibrato sound. Pad B also gives the same sound. Lots of pitch-bending warble, but no flanger whoooosh

[andrew_k]

Here's a 15 second clip of the flanger vibrato to demonstrate that I'm not completely insane

http://notinteractive.com/stuff/guitar/fx/flanger-vibrato.mp3

[StephenGiles]

That means that the dry signal is lost somewhere. Check that the 2 68k mixing resistors for flanged and dry signals are both 68k - I think that is the value, from memory only though.

Just checked the circuit, one 68k comes from the opamp with the funny limiting diodes. I would unsolder the end of that resistor which connects with the other mixing 68k from the BBD and check that there is dry audio there. If you have vibrato, all the clocking /BBD and output circuitry must be working.

[andrew_k]

Both of those resistors are 68k. Referencing this schematic, you're suggesting I probe the audio at pads C and D with the jumpers removed from the board to break the circuit? I hope so, as I just did that -- the 68k at pad D has super wide vibrato, with clock noise (?) at the peak of the pitchbend. The one at pad C is cleaner, but I would still classify it as vibrato. Pad B has essentially the same sound as pad C.

At least the SAD1024 works!

[flo]

The signal at pad B and C (with jumpers removed) should be the clean signal without vibrato as long as the "enhance" pot is set to zero (wiper to ground).
How is the signal at the input and at pad A (should be the clean signal regardless of "enhance" or other pot settings)?

[andrew_k]

Pad A is crystal clean. Enhance was originally set above zero but while probing the signal I remembered what enhance does and set it to zero. The mild vibrato is present with enhance at zero.

[StephenGiles]

Remove the BBD and see what appears at the 68K.

[andrew_k]

Remove the BBD and see what appears at the 68K.

Pad D is silent, pad C is clean signal.

[flo]

> "Pad A is crystal clean."
Good.

> "The mild vibrato is present with enhance at zero."
This is on pad B, right?
That is a bit strange. I think it should be clean now.

>> "Remove the BBD and see what appears at the 68K."
> "Pad D is silent, pad C is clean signal."
Good.
Pad B is now also clean while the BBD is removed?

So if you now put the BBD back again, pad D should have the vibrato signal, pad B and pad C should be clean (with "enhance" on zero).
If so, put back the jumpers for pad C and D, so that these two signals are mixed, and listen for the flange...

[andrew_k]

So if you now put the BBD back again, pad D should have the vibrato signal, pad B and pad C should be clean (with "enhance" on zero).
If so, put back the jumpers for pad C and D, so that these two signals are mixed, and listen for the flange...

Progress! When I put the BBD back in, with the jumpers removed, pads B and C are now clean signal, as expected. However, once I put the jumpers back in across adjacent pads at C and D, the output is still the heavy vibrato demonstrated in the sample above.

I have to go to work now, but I'm thinking perhaps that even though the two 1% tolerance mixing resistors have the 68k colour bands, they might not be perfectly matched? I'm going to replace them tonight with a pair that measure exactly the same... I've long ago demonstrated that the workings of this circuit are over my head, but I'm especially confused as to how the signal mixing could be so out of whack and how re-seating the BBD could have fixed the vibrato mixed into the clean signal?


My sincerest gratitude to the three of you for taking the time to work through this with me!

[andrew_k]

A picture is worth a thousand words, and so is an audio sample:

http://notinteractive.com/stuff/guitar/fx/flanger-vibrato-2.mp3

So where it's at -- still vibrato, but pad B is clean and pad C without a jumper is clean. Excellent Both 68k mixing resistors have been replaced with resistors that measure 67.9k.

Definite progress!! This thread has resulted in the LFO coming to life, the clean signal making it through to the mixing stage... now I've just got to get the mixing right and get rid of some of that noise. There's all sorts of noise in there, as you can hear in the above sample. THe EQ of the noise is dependant on the position of the harmonics switch and the threshold pot, but never goes away.


?

[flo]

For now I'm not focusing on the noises. I just want to get the mixing to function so that you can hear it flange...

Although the flanger will sound "better" with a "perfect" mix, it is not extremely critical for the flanger to function. So, resistors that match within 1% are fine.
Now that you have measured that the correct signals that are given to the mixer-stage, delayed (=vibrato) and clean, it seems that that mixer-stage is not functioning well because it does not provide the expected flanging output. The clean signal (not delayed) seems to be left out of the mixed signal somehow.
Are you positive that the flanger output signal on the mixer-stage output, IC2b-pin7, is the same as the delayed (vibrato) signal on IC1c-pin8?
If so, remove the jumper for pad D. Is the flanger output signal on the mixer-stage output, IC2b-pin7, now the same as the clean signal on pad B?

[StephenGiles]

Unsolder both 68k resistors where they meet at the mixing IC you should have flanging there, if you don't, the likelyhood is that IC2 is faulty. You could try 2 different value resistors - say 47k.

[andrew_k]

Are you positive that the flanger output signal on the mixer-stage output, IC2b-pin7, is the same as the delayed (vibrato) signal on IC1c-pin8?
If so, remove the jumper for pad D. Is the flanger output signal on the mixer-stage output, IC2b-pin7, now the same as the clean signal on pad B?

The output of those pins isn't the same, but they are very close. IC1 pin 8 is slightly louder, w/ more treble. The following mp3 snippet has three notes - the first is the Output pad, the second is IC2 pin 7, the third is IC1 pin 8:

http://notinteractive.com/stuff/guitar/fx/flanger-3.mp3

To me it sounds like the clean signal is present, but it's completely overpowered by the delayed signal. That's a guess though.

Stephen: Replacing IC2 made no difference

[StephenGiles]

Try connecting a 100K pot in series betwen the flanged signal 68k and the mixing IC to reduce it's volume. You should get flanging then. Otherwise there is something very odd going on.