Diy Flanger intermittent

[vinceg]

Flangers are made up of building blocks - broadly speaking, input amp, input filter, BBD, output filter and output amp/ passive mixing with dry signal, and out of harm's way to keep clicks to a minimum - the LFO and BBD clock.

What I would do is this:

1 Carefully remove BBD and keep it safe (I have never been particularly careful and never had a bad one!!)

2 Test for audio at input amp output and input filter output.

3 On the assumption you have used a socket for the BBD, insert a temporary wire link between its input and output pins.

4 Input filter should now be linked to output filter - again test for audio along output audio path.

5 If all is well, test for fluctuating voltage at LFO output.

6 Test for voltage of roughly half supply on each of the 4013 clock output pins (1 & 2), and again on pins 2 & 10 of the 4049 buffers.

I'm guessing that the problem is either the 4013 or the 4049.

Ok, I removed V3207 BBD and I inserted a wire link between pins 3 and 7. Then, I tested for audio output and it is ok.
Then, I tested for fluctuating voltage at LFO output (pin 7 of LM311) and I measured a variable voltage from about 0 to 8 volt. Also, I tested all pin of LM311 and voltages are:

pin 1: 4mV
pin 2: about 1.3-3.7 variable
pin 3: about 1-2.3 variable
pin 4: 4mV
pin 5: 8.9
pin 6: 8.9
pin 7: about 0-8 variable
pin 8: 8.9

Then, I tested pins 1 and 2 of 4013 and voltages are 4.5 on both pins, after i tested 4049 pins 2 and 10 and I found 4.5 voltage both pins

[Fender3D]

There's no need to remove BBD, unless your wearing electro-statics charged clothes... 

1st: it's not a EM-whatever clone. In order to be a proper clone it should have, at least, the same BBD.
since it uses another, longer, BBD they halved the clock capacitor so they'll get double the stock clock frequencies to compensate the longer delay line.
With such a low cap value you may easily exceed the max clock freq allowed for the chip (note this is not a MN32xx but a V32xx, and it won't behave the same as its counterpart).
I guess the "break" you notice when flanging is just this...

Try raising C18 value (EM has 47pF) and check if it has the same issue.

PS
suggest Omar to read DIYSB forum for... suggestions or advices

[DrAlx]

LM311 is the VCO, not the LFO.  The LM311 takes the LFO voltage (i.e. the "control voltage") at its input pin 3 and outputs a high frequency signal (in the 10s to 100s of MHz) at its pin 7.   That pin 7 signal triggers the 4013 to produce the complementary clocks that drive the BBD.
The lower the voltage at pin3 of the LM311, the faster the BBD clocks go.
If the voltage at pin 3 ever drops below about 0.8 to 0.9 V then the LM311 will freeze (there's a limit to how fast it can only oscillate) and that will be heard as a dropout in the flange effect (because the BBD is not being clocked).

Your LM311 voltages look OK.

I think you just have the wrong setup procedure. (BTW, i would also second Fender3D's recommendation to use 47pF instead of 22pF)

Do this...

First set BBD bias as follows:
1) Turn filter matrix off.
2) Turn color pot to minimum (i.e. you do not want any sort of bell sound to begin with)
3) Turn range pot to maximum.
4) Turn rate pot to about 75%.
5) Set clock trimmer to 50%.
6) Set BBD bias trimmer fully counter-clockwise.

The bias trimmer has a small useable range of settings, and you want to set the bias trimmer near the middle of that range.  Fully clockwise or fully counter-clockwise means nothing will come out of the BBD.  So...

7) While playing your guitar, gradually turn the BBC bias trimmer clockwise until you hear an acceptable flanger effect.  Take note of the position of the trimmer setting.  Then continue turning the trimmer until the flanger sound goes away.  Take note of that second trimmer setting.  You want to set the trimmer half-way between those two extremes.

Next stage is to set the feedback limit using the trimmer near the color pot.  The job of this color trimmer is to make sure that the color pot does not cause self-oscillation when set to maximum (i.e.  continuous screeching sound).

8 ) Turn filter matrix on
9) Set color pot to maximum.
10) Play something on the guitar while adjusting the color trimmer.  The trimmer determines how quick the bell-like sound decays.  You need to set it so that it does not decay so slowly that you actually get self-osciallation.  Once you've set the trimmer, do a check that it does not self-oscillate at other range pot settings too. 

Setting the clock trimmer.
12) Follow the process here...
https://www.diystompboxes.com/smfforum/index.php?topic=91981.msg1047491#msg1047491

[vinceg]

So, what is L So what is the LFO? At pin 3 of the LM311 there is the emitter of Q3 (2N5087). Is Q3 the LFO?

[DrAlx]

The VCO and LFO are circuit blocks, not single components.

LM311 and Q3 are part of the VCO.

Two of the LM324 op-amps are part fo the LFO.

The LFO is a triangle/square  osciallator.  Google "triangle square oscillator" and you will see images of some examples.

[vinceg]

The bias trimmer has a small useable range of settings, and you want to set the bias trimmer near the middle of that range.  Fully clockwise or fully counter-clockwise means nothing will come out of the BBD.  So...

7) While playing your guitar, gradually turn the BBC bias trimmer clockwise until you hear an acceptable flanger effect.  Take note of the position of the trimmer setting.  Then continue turning the trimmer until the flanger sound goes away.  Take note of that second trimmer setting.  You want to set the trimmer half-way between those two extremes.

About this point, the bias trimmer have a initial position where the sound there is, but after, if I continue to turn clockwise there is sound until it is fully. There is not a point where no sound.

[DrAlx]

Sorry, I was thinking of a similar circuit (not the one in your picture).
The circuit I am thinking of didn't have an R12.
So temporarily short out R12 (i.e. connect pin 3 of the bias trimmer directly to VC).
Once you've set the bias trimmer as I described, measure the voltage at pin 3 of the MN3207 (while you are not playing anything).  That is the desired bias voltage level that the BBD needs.
If you then unshort R12, re-adjust the trimmer so that you get the same bias voltage at pin 3 of the BBD.

Note that the overall sound will never fully disappear, but the flanger effect will.

Note: to be honest R12 is not necessary and could be left shorted out.

[vinceg]

Sorry, I was thinking of a similar circuit (not the one in your picture).
The circuit I am thinking of didn't have an R12.
So temporarily short out R12 (i.e. connect pin 3 of the bias trimmer directly to VC).
Once you've set the bias trimmer as I described, measure the voltage at pin 3 of the MN3207 (while you are not playing anything).  That is the desired bias voltage level that the BBD needs.
If you then unshort R12, re-adjust the trimmer so that you get the same bias voltage at pin 3 of the BBD.

Note that the overall sound will never fully disappear, but the flanger effect will.

Note: to be honest R12 is not necessary and could be left shorted out.

Wow, you are a source of knowledge. Tomorrow I'll do. Then we will discuss the clock setting method... I'm not sure to understand.

[vinceg]

Ok, I think it is right now, thank's you all. I've another question because in my opinion it is too much bell sound. If I use it togheter with Muff the tone change, more high.

[vinceg]

I noticed that there is a sort of intermittent breath/hiss. At first I thought it was my amp, but as soon as I turned off the pedal the intermittent breathe/hiss disappears. What could be the cause?

[Mark Hammer]

That design doesn't have much filtering of the delay signal at all.  Bucket brigade chips, and especially the clocks used to drive them, generate noise.  I suspect that is what you are hearing.

The way that different commercial designs control noise can be very different.  The A/DA Flanger used a noise gate that shut off the delay signal when you stopped playing.  The Boss BF-2 uses a combination of treble-pre-emphasis and de-emphasis before and after the delay chip to remove a lot of the hiss.  Some use a compander to dynamically compress and expand the signal before and after the delay chip, and nearly all will use several poles of lowpass filtering, whether they use one or more of the other techniques.

This design you showed doesn't seem to use any of that.  Sometimes, the lack of filtering is part of what makes a flanger sound more exciting, but as your case illustrates, it can also lead to noise you don't want.

Where to modify?  The simplest modification seems to me to be R15, which is 8K2.  Remove it and replace with a 4k7 and 3k3 in series (the 4k7 first).  Now run a 4n7 capacitor from where they meet to ground.  That will provide at least one pole of lowpass filtering at 7.2khz to the delay signal only.  That will not remove the hiss entirely, but will take some of it away, without reducing the ability of the flanger to show notches at the high end of the guitar spectrum, or change the delay/clean balance much.

I will also suggest increasing the value of C3 from 1nf to 3n3.  This will reduce the bandwidth of both delay and clean starting around 8.6khz.  Most guitar speakers roll off around 6khz so you won't be losing much treble.

The Boss BF-2 uses two poles of lowpass before the BBD and at least 5 poles after it, so what I'm suggesting here is small, compared to what typical commercial designs do.  BUt it should help a bit.

[vinceg]

That design doesn't have much filtering of the delay signal at all.  Bucket brigade chips, and especially the clocks used to drive them, generate noise.  I suspect that is what you are hearing.

The way that different commercial designs control noise can be very different.  The A/DA Flanger used a noise gate that shut off the delay signal when you stopped playing.  The Boss BF-2 uses a combination of treble-pre-emphasis and de-emphasis before and after the delay chip to remove a lot of the hiss.  Some use a compander to dynamically compress and expand the signal before and after the delay chip, and nearly all will use several poles of lowpass filtering, whether they use one or more of the other techniques.

This design you showed doesn't seem to use any of that.  Sometimes, the lack of filtering is part of what makes a flanger sound more exciting, but as your case illustrates, it can also lead to noise you don't want.

Where to modify?  The simplest modification seems to me to be R15, which is 8K2.  Remove it and replace with a 4k7 and 3k3 in series (the 4k7 first).  Now run a 4n7 capacitor from where they meet to ground.  That will provide at least one pole of lowpass filtering at 7.2khz to the delay signal only.  That will not remove the hiss entirely, but will take some of it away, without reducing the ability of the flanger to show notches at the high end of the guitar spectrum, or change the delay/clean balance much.

I will also suggest increasing the value of C3 from 1nf to 3n3.  This will reduce the bandwidth of both delay and clean starting around 8.6khz.  Most guitar speakers roll off around 6khz so you won't be losing much treble.

The Boss BF-2 uses two poles of lowpass before the BBD and at least 5 poles after it, so what I'm suggesting here is small, compared to what typical commercial designs do.  BUt it should help a bit.

This design should be electric mistress with some modification. Anyway, I turned left the clock trimmer and the   problem seems disappears.

[Mark Hammer]

I don't know what "left" did, but it probably increased the clock frequency range a bit so that the clock noise was above hearing range and/or what your speakers can reproduce.

Problem solved.

[vinceg]

I don't know what "left" did, but it probably increased the clock frequency range a bit so that the clock noise was above hearing range and/or what your speakers can reproduce.

Problem solved.

Yes, turn left increase the clock frequency.
Is possible decrease metallic/bell sound? I do not like much brigther tone.

[vinceg]

I noticed an intermittent volume drop that disappears when the Rate control is over halfway. I think there is something wrong with this circuit or I not be able to set clock trimmer.