noisy LT1054 inverting configuration

[Dimitree]

I'm trying to use a LT1054 to provide -9V to a Whammy 4.
The original design used 9V AC input, rectified and regulated to -5V (that is used to supply the negative rail of 4x opamps).


I'm trying to remove the 9V AC requirement and use a standard 9V DC supply, simply replacing D6 with this LT1054 circuit:





it works but is really noisy. If I disconnect the LT1054 pcb, the noise disappear (but of course the audio distort because of the lack of negative supply on the opamp).
I think I'm doing something wrong, since I am aware of many who uses LT1054 with no issues.

[Keppy]

What's the purpose of CF? I haven't seen that cap in any inverter configuration before.

What's the noise like? Whine, buzz, white noise, crackling...?

[MikeA]

CF is to raise the switching frequency of the LT1054, it's on the data sheet.  Wouldn't hurt to try without it and see if anything changes.

I see two things, one is to remove C7, 1000 uF is way too much capacitance for the LT1054 to push.

The other is to add the startup transistor shown on page 6 of the AD data sheet (or page 23 of the TI version.)  It keeps the load from pulling pin 5 higher than ground on startup.  I use a 2N2243 and a 3.6k resistor to ground.

[GGBB]

Try using the regulation feature - TI figure 15.

[GGBB]

Also - Whammy IV wants ~800mA 9VAC - LT1054 can only give you 100mA max.

[Dimitree]

I see two things, one is to remove C7, 1000 uF is way too much capacitance for the LT1054 to push.

thanks for pointing that to me! I totally forgot about the presence of C7.
but unfortunately I just removed it, and it didn't solve the issue.
this is an audio sample of the noise (first whammy is off - true bypass, then is switched on)

https://soundcloud.com/user-32380672/whammy-noise-wh4

Also - Whammy IV wants ~800mA 9VAC - LT1054 can only give you 100mA max.

please check again the schematic. The LT1054 is only replacing D6, so only used to supply the -5V rail, and that rail is requiring "only" 20mA at -13V. Indeed, except for the noise, the pedal is running great (voltages are spot-on, and the regulators/LT1054 are far from hot).

[Rob Strand]

thanks for pointing that to me! I totally forgot about the presence of C7.
but unfortunately I just removed it, and it didn't solve the issue.
this is an audio sample of the noise (first whammy is off - true bypass, then is switched on)

Using the numbering on your schematic:
- Make sure both leads of C2 are close to the charge-pump pins 3 and 8.
- Make sure the ground leads of C1 and C4 are close to charge pump pin 3

You don't want any audio passing through the grounds:
  DC in ground to Charge pump ground, or, Charge pump ground to effect ground