HF noise with Duo-Phase (SOLVED)

[onevibe93]

Hi everyone !

I have a very high-pitched noise, a shrill whistling sound, faint but present to the ear. It sounds like a TL072 oscillation, HF noise. However:

- the noise persists (less loud but persists) when I remove all the TL072s
- it disappears completely when I also remove the TC1044s

The phaser works well, but this noise is present and can be heard. It remains present even without a transistor, LFO, LED, etc. It is not generated by the send-return (I'm currently testing it in a Denon mixer). When I bypass the phaser, it reduces considerably but remains audible.

It is only present with TC1044-A or with TC1044-A and TC1044-B, but it is not present with TC1044B only. I have tested different TCs, the problem is the same. I have already assembled this PCB and I have never had this problem...

I don't have an alternative to test the power supply yet...

 

[Cybercow]

The called for charge pump is a TC1044SCPA. The "S" in the suffix is important for this application.

 

[JohnnyCreepy]

Remember to post pictures of your build to help troubleshooting, but if what @Cybercow pointed out doesn’t remove the noise it’s good to look into lead dressing. Are in and out wired running near noise emitting components and is shielded i/o wiring needed?

 

[onevibe93]

sorry for this inconvenience... But yes they are TC1044SCPA and I have 4 of them to test this layout in this small box... I didn't use shielded cabling the first time, I didn't have this HF noise. This is my second setup, I planned 3 others but due to this difficulty they are on hold until I find the origin of the problem. I have the noise with or without/outside the box...

 

[JohnnyCreepy]

Could be nothing, are the silent ones are without pot lug extension wires?

 

[onevibe93]

I use same wire for every pots... Now I have removed all the potentiometers, and If I unplug power supply during the HF noise is produced by TC1044, I can hear the noise disappear like an engine turned down, from the pump turned down maybe. But I Dont understand why is so present....

 

[onevibe93]

The noise is more important in the dark... It seems to be related to the dark resistance of the LDR network?

 

[JohnnyCreepy]

A picture from pcb underside to see component solder joints would be golden. It’s easy to spot solder bridges, cold joints etc from a good picture!

Have you double checked component values around charge pumps?

Have you set led trimmers according to this: Do one side at a time. Set the rate to almost all the way CCW. Now adjust the 10K trimmer so that the light doesn’t quite turn all the way off. This will avoid any annoying ticking sounds. ?

 

[onevibe93]

Ok so, I solved my problem but I would like to understand why lol

I didn't have a second 9v DC power supply to test, so I did all my tests with the same power supply... As I said previously...

So first I tested a 12V power supply, and immediately the pitch of the TC1044 charge pump's RF noise changed; it became less high-pitched (and therefore less noticeable), and a little quieter (less noise gain). So I tested a second 12V power supply, the chippest lol, by the way, and bingo: zero noise.

Why can power supplies indirectly produce noise that they don't generate themselves in this specific case?

@JohnnyCreepy about the leds setting : Shouldn't the LEDs turn off completely? I think I prefer when the light/dark cycle is as extended as possible.

Edit : So the TC1044SCPA in this pedal is is sensitive to power disturbances... and this circuit seems to work well under 12v, no noise

 

[lowpitch]

Ok so, I solved my problem but I would like to understand why lol

I didn't have a second 9v DC power supply to test, so I did all my tests with the same power supply... As I said previously...

So first I tested a 12V power supply, and immediately the pitch of the TC1044 charge pump's RF noise changed; it became less high-pitched (and therefore less noticeable), and a little quieter (less noise gain). So I tested a second 12V power supply, the chippest lol, by the way, and bingo: zero noise.

Why can power supplies indirectly produce noise that they don't generate themselves in this specific case?

@JohnnyCreepy about the leds setting : Shouldn't the LEDs turn off completely? I think I prefer when the light/dark cycle is as extended as possible.

Edit : So the TC1044SCPA in this pedal is is sensitive to power disturbances... and this circuit seems to work well under 12v, no noise

TC1044SCPA is rated for 12V maximum input voltage. I would advise against running it that close to max specs continuously.

Also, not all power supplies are created equal. If the external power supply isn't filtered well, that noise can get into the pedal circuit. Also, running the charge pump on higher voltage can increase its switching frequency to where it isn't as audible anymore, although you say the whine became less high pitched, so it may very well just be poorly filtered external power supplies.

 

[jimilee]

I was wondering about that. I had a phaser build that drive me nuts until I finally tried a different one.

Some power supplies are just noisy. I’ve found optical phasers to be sensitive to thst.

 

[onevibe93]

Today I just tested a second circuit (I'm assembling 4 for friends). I put everything together with the same series of resistors and capacitors so if this is all a mistake, it was probably repeated 4 times lol

The second circuit makes the same noise with the same power supply. Slightly less loud I think. And same, the noise disappear with the chip 12v power supply.

The V max of the TC1044SCPA is 13v so I think the tolerance is respected. It's strange, I have the impression that it's not the power supplies that I hear but the charge pump, the pitch variations make me say that it's a noise in reception of the power supply, but not emitted by the power supply itself.

Here is an example of my soldering !

About power supply on photo : the HQ Power are affected by this noise problem. The cheap one is the one on the far left of the photo, it must come from an old battery charger or something...

Some power supplies 12v/9v/5v to propose to me ? Thanks everyone !

EDIT : I tried ICL7660 on this second layout, same noise !

 

[z2amiller]

I might try adding some additional capacitors to see if additional filtering helps.
You can tack these on to the back side of the PCB using the same pins for testing things out.
This probably the order I would try things:

- Add a 100nF capacitor in parallel to the input cap, ceramic if you have it. The large electrolytic is better at energy storage but the smaller ceramic cap does better at filtering higher frequency noise.
- Try the same on C18(A|B)
- Double the input bulk capacitor (e.g. tack another cap of the same value in parallel on the back, watch the polarity!)
- Double the charge pump caps (same)

 

[Erik S]

Those are all center positive type power supplies.

Standard pedal power is normally set up on 2.1mm with center negative.

While it's possible to wire your pedal correctly for an odd supply, if you plan on giving these away to friends, there would be a high chance these are going to get plugged into a "normal" supply at some point and get fried.