Blues Breaker + DOD Chorus = Birds?!?

[Rockhorst]

Ok, so I built this high octane Blues Driver for a friend, working on 18V using an ICL 7660 S CPAZ like in a Road Rage. So I get a phone call two days ago that his chorus pedal is making weird ticking noises in combination with it. I can best describe it as a birdlike chirping, like waking up on a nice spring morning. Cool yet undesired. I made a recording of it whilst fiddling the depth and speed knobs:

Recording

This happened with his DOD Super Stereo Chorus FX68 (sadly, no schematic is available, though there is one for the FX67) but not with an Arion Stereo Chorus SCH-1.

This is the schematic for the Blues Breaker:


Anyone ever experienced something similar or have any clues/tests/suggestions?

[Cortexturizer]

I have experienced such a thing with my Sunking build and a MN3207 electric mistress build!
It happens only in one room of one studio in the town that I live in. Nowhere else! It's just what you are describing, and I have a ICL7660SCPA in the Sunking, mistress is a time modulation pedal like that chorus, so...there has to be something that can be explained. Sadly, I possess no knowledge nor time to try this myself
Tell your friend to rehearse in a different room! Hehe

[RobA]

In relation to another thread, I have a boost/overdrive on breadboard with a switchable 9V/18V power supply and a whole mess of various charge pumps. I also have a Current Lover I can mix with it. So, I put them together and then ran them into the spectrum/FFT thing on my audio interface.

I assume that this is running the BB into the chorus(?). I can get very similar sounds. Using different charge pumps changes the sound characteristics and volume some with the 7660SCPAZ being the worst and the TC7660H being the least noisy. But they all do it to some degree.

With the FFT, you can see the noise bands and the oscillator noise being swept by the BBD well down into the audible frequency range.

With the TC7660H and the LT1054, the oscillator bands are high enough that they aren't getting swept down low enough to be heard. But, there is other noise that the charge pump is at least somewhat responsible for that also gets swept well down in frequency and sounds much the same.

I say it's from the charge pump because if I flip the switch and move to 9V the noise is much reduced. The charge pump is still in the circuit and running, but the op amp's rails aren't being feed by the charge pump. If I completely remove the charge pump and go to using an 18V supply directly, there is no noise at all.

I don't know how much of this noise is being feed directly on the audio chain and how much is coming from the coupling of the power supply. My power supply isn't isolated and that might help, but I don't know.

[Scruffie]

Sounds possibly like heterodyning to me, try shielding the charge pump from the chorus (or anything with a clock in it) and maybe adding extra power decoupling.

Dunno if he's using a switching power supply but it may be better on a transformer based one or battery.

[Rockhorst]

try shielding the charge pump from the chorus

How?

By the way, he's also using a J201 based OD with a charge pump regulated at 12V which gives no problems with the chorus.

[RobA]

In the working as typical universe, when I plugged everything back together this morning, the chirping I get is much reduced. It's still there, so I was able to experiment a bit, but it wasn't nearly as strong. Kinda to bad too because the Current Lover in matrix mode made some really cool noises with it.

Any way, I tried moving the Current Lover to a separate power supply and that didn't help.

So, heterodyning? It's basically beat frequencies between the two clocks? The chirping happens because the clock for the BBD is moving and |f_bbd - f_pump| moves around in the audio range? Is it noise on the rails or noise in the audio path?  Even though my experiments this morning were a bit less than convincing because it wasn't nearly as present, it seems like the clock noise from the charge pump is on the ground and this is mixing with the clock noise from the BBD clock and this is messing with the op amps in the audio path of the chorus/flanger. Does that make sense? I'm trying to figure out where filtering/isolation would work.

If this were happening with a non-sweeping clock, like a second charge pump, then the issue wouldn't be nearly as prominent (fixed at a single beat frequency). And, you could try to rig the frequencies of the two clocks so that |f_1 - f_2| was outside the audible range, right?

[Rockhorst]

Hm, so the reason why it doesn't give any problems with a different chorus might be that it's got a different clock frequency. I'll fiddle around with different power supplies. The next version of this is going to have a split rail powersupply from -9v to +9v, maybe that helps too.

[RobA]

Hm, so the reason why it doesn't give any problems with a different chorus might be that it's got a different clock frequency. I'll fiddle around with different power supplies. The next version of this is going to have a split rail powersupply from -9v to +9v, maybe that helps too.

That would be my interpretation. You may want to try the Microchip TC7660H in the current pedal. The spec sheet for it claims an oscillator frequency of 120kHz. That should take the beat frequencies of the two clocks into the inaudible range (if I've read the spec sheets for the BBD's right). I really don't know if this will work, but it would be an interesting test and the TC7660H is only in the $1 range from Mouser and Avnet if you plan on making any orders from them.