Time to play everyone's favorite game...

[Govmnt_Lacky]

NAME THAT NOISE!! 

This is a short video with the SH!T noise that I have been battling with for 6 months!

The pedal is an MXR Micro Flanger and a Micro Chorus in one box. The BBDs are RD5106A for the flanger and a RD5107 for the Chorus. I am out of ideas on where the noise is being generated. They both have some form of this noise however, I only recorded it from the Flanger. You might not be able to see but, I have the Regen at max. I am changing the Rate pot during points of the recording to show that the noise changes with it. Oh yeah.... this is on BATTERY POWER!

So..... NAME THAT NOISE! And hopefully, give me a bloody idea of how to get rid of it PLEASE!!! 

[haveyouseenhim]

Are there any drop in replacements for the BBD chips? I get a little bit of that squeak on my Gilmour flanger. Not quite as percussive though.

[Govmnt_Lacky]

Are there any drop in replacements for the BBD chips? I get a little bit of that squeak on my Gilmour flanger. Not quite as percussive though.

Some say that the SAD512D is a quieter BBD than the R-series but good luck finding those!

I honestly don't think it is in the BBD. I don't get any noise from my Gilmour unless the amp is cranked. Even then it is negligible.

[SISKO]

heterodyning? Do you have a scope?

[WaveshapeIllusions]

I bet it's the clocks for the two chips intermodulating. Depending on their frequency the difference can be in audio range. Try cutting power to one of the circuits and see if it goes away.

[Govmnt_Lacky]

heterodyning? Do you have a scope?

I do have a scope but, I wont have access to it until Friday.

Suggestions??

I bet it's the clocks for the two chips intermodulating. Depending on their frequency the difference can be in audio range. Try cutting power to one of the circuits and see if it goes away.

I will try this tonight and report back!

[Mark Hammer]

The fact that swept squeal changes as you vary modulation rate suggests very strongly that it is a heterodyning thing.

1) Much of what we find on the output of BBD chips - not just the lowpass filtering but the trimpots adjusting the balance between dual outputs - is intended to eliminate as much potential clock noise as possible.  The schematics for both the Micro-Flanger and MIcro-Chorus show no mxing resistors on the output, just a direct tieing together of the outputs.  Maybe you ought to consider inserting a low-value trimpot, like 1k or 2k.  I realize that it is hard to get better matching than two zero-ohm straight wire paths, but there is an underlying assumption - perhaps naive - that the two outputs from the BBD are already equivalent in level, hence optimally clock cancelling.  One pedal at a time, that may be  entirely plausible, but with two stuffed in a box, it may not be enough.

2) Layout, layout, layout.  Are these two circuits merely rehoused?  Are they on PCBs, vero, perf?  Do the clock circuits sit anywhere near each other?

[earthtonesaudio]

Heterodyning +100000000000

[Govmnt_Lacky]

Layout, layout, layout.  Are these two circuits merely rehoused?  Are they on PCBs, vero, perf?  Do the clock circuits sit anywhere near each other?

Mark,

Both circuits are on self-etched PCBs. They are not rehouses. These layouts were done for me by another forum member with the intent being to utilize the R510X series BBDs that I had. The PCBs are right next to each other within the 1590BB enclosure. So, you could DEFINITELY say that the clocks of each circuit are close. Think of (2) 1590B PCBs stuffed into a 1590BB  

[Fender3D]

It looks like heterodyning...
If this is the case, pull out the 2 chorus clock caps (thanks God, they are socketed (and I know why  ) ),
and something will change (on flanger side, BTW).

Otherwise, I heard that BBD noise already...
When the null trimmer is set badly (as Mark said)...
SADly R51xx chips got one out pin only...

[Govmnt_Lacky]

It looks like heterodyning...
If this is the case, pull out the 2 chorus clock caps (thanks God, they are socketed (and I know why  ) ),
and something will change (on flanger side, BTW).

Otherwise, I heard that BBD noise already...
When the null trimmer is set badly (as Mark said)...
SADly R51xx chips got one out pin only...

Thanks for the input Fender...

I really dont think it is "entirely" the BBD chips. I have already used the RD5106A chip in my Gilmour build and the noise is nowhere near this 

I will try pulling the Chorus clock caps (say that 10 times fast!) and see what happens. If that is a no go, I will move on to disconnecting input power from the Chorus. I will report back results as soon as I get to this! 

[Govmnt_Lacky]

OK...

I pulled out the (2) 22pF timing caps in the Micro Chorus circuit AND.... the high pitched whining went away in the Flanger BUT....

I am still getting a continuous static-y Whooshing sound that is quite loud and varies with the flanger Rate pot (LFO?)

I am doing this testing with NO guitar plugged into the input.

Also, I was getting a thumping "heartbeat" from the circuit EVEN WHEN THE FLANGER WAS TRUE BYPASSED! This also varied with the Flanger's Rate knob.

So...

1) Any ideas on how to prevent the heterodyning AND keep them in the same box?

2) Any ideas on how to get rid of the sonic Woosh and thumping?

[mistahead]

Food for thought from a comparative newb...

[PirateParrotVoice]
You've got digital and analogue grounds in the those circuits, my understanding is that its worth doing a check on the gnd planes etc.
[/PirateParrotVoice]

Next week I'll know what that means lol.

[Govmnt_Lacky]

Food for thought from a comparative newb...

[PirateParrotVoice]
You've got digital and analogue grounds in the those circuits, my understanding is that its worth doing a check on the gnd planes etc.
[/PirateParrotVoice]

Next week I'll know what that means lol.

Digital grounds 

[WaveshapeIllusions]

Is there by any chance a charge pump in the circuit as well? That could be another source of heterodyning. Extra bypassing as close to all supply pins on all ICs may help. Individually shielding each board and using shielded wire for all interconnects may help as well.

[mistahead]

Think of the issues with the PT2399 chips in some of the delay designs, there was a digital gnd reference and an analogue one on different pins.

Those that had a low R between the D-gnd and the A-gnd showed far less issues with noise, and just to mess with us there was a run of chips that had the low R internal between them, but then later chips didn't.

As I said - food for thought- not a Guru answer.

[Fender3D]

1) Any ideas on how to prevent the heterodyning AND keep them in the same box?

2) Any ideas on how to get rid of the sonic Woosh and thumping?

1) Star ground... and RC cell on positive going to each PCB, maybe a copper shield foil (as in oldest MXR delays).

2) thumping is there in MXR's micros also... Sometimes, LFO thumping dims varying power supply...
Try TL072, 5532 or OPAs for signal op-amp, otherwise low-pass filters may be lowered...

[earthtonesaudio]

Shield input+output wires at the very least.  Maybe also shield one of the boards entirely (wrap with insulator, then wrap with conductor, ground conductor).

[Govmnt_Lacky]

1) Star ground... and RC cell on positive going to each PCB, maybe a copper shield foil (as in oldest MXR delays).

2) thumping is there in MXR's micros also... Sometimes, LFO thumping dims varying power supply...
Try TL072, 5532 or OPAs for signal op-amp, otherwise low-pass filters may be lowered...

Shield input+output wires at the very least.  Maybe also shield one of the boards entirely (wrap with insulator, then wrap with conductor, ground conductor).

Could you point me in the right direction on good ways to "insulate and shield" one of the PCBs? Possibly some pics as an example...

[Mark Hammer]

If you're like me, then what you'll often do to test out modulation circuits is crank the regeneration/feedback up so that the sweep is more obvious.  Want to know if the speed is varying properly or the sweep width adjusting in a desired manner?  Crank up the regen to hear the sweep better.

There's just one problem with this: higher regen usually means more noise.  It's clearly not going to be the only source of noise, but whatever noise may be inherent in the circuit is going to be exaggerated by higher regen.

There are several solutions to this.  The least desirable is to never apply any regen/feedback.    The 2nd path is to track down any controllable sources of noise, and tackle them so that there is nothing for the feedback path to exaggerate.  The third path is to apply the noise reducation to the feedback path itself.  Indeed, in some flanger designs - the A/DA comes to mind - there is more lowpass filtering applied to the feedback loop than to the delay path between BBD and output mixer.

Some circuits, like phasers, allow one to apply a bit of lowpass filtering to one or more phase shift stages so that noise does not accumulate.  It is not uncommon to see in 6-stage units (or more) a cap in the feedback loop of one of the phase-shift stages.  Alternatively, the feedback path may itself consist of an op-amp with some "light" lowpass filtering applied to keep a lid on accumulating noise.