Phase 90 bias ?

[jayp5150]

I've searched, but didn't see anything that addressed my question directly.

I completed the Phase 90 the other day, using a board from Tonepad. Mods as follows:

-Added a 500k pot in series with the 1M to vary depth (this way, cranked one direction gives me the stock value)
-Vibrato switch (removes the 150k resistor to cut dry signal)
-added an on-off-on switch to give me feedback resistance of 22k, 47k, or no contact.

I used 2N5952 fets, matched (all around 1.80-1.84 on the Geo matcher).

So, when adjusting the trimpot for a good phase, I can find a happy medium for all settings, but if I adjust it for a nice, full phase when the feedback resistor is bypassed, when I click back over to the stock 22k, I get R2D2 beeping in time to the LFO,

Now, I can tune this out with a small adjustment, but then the phase really isn't where it 'should' be, if that makes sense.

I just hadn't heard anyone else really comment on this. Should I just change that 22k to a higher value and forget about it? I know a lot of people are using a pot there to vary resistance down to as low as 18k, but still seem to get a deep phase out of it.

Has anyone else encountered this? I mean, where you have to compromise to get all feedback resistances to sound equally good (I understand that the resonance changes with this mod, but shouldn't I be able to tune this to 'perfect' then flip the switch through the settings, and not get R2D2 through my amp?).

Any input is appreciated. Thanks ahead of time.

[jayp5150]

Just a bump... I deliver this to my cousin tomorrow, and I don't mind telling him he has to turn the depth down on certain settings, lol, but he might.

Otherwise, he's getting a phaser that's a bit more mild that what I believe he expected.

On a side note, it has become tradition to drop off/pick up pedals, guitars, etc., at this Mexican restaurant between us (we live an hour and a half apart), so I'm stoked about some Mexican lunch! My wife won't try it... 

Anyhow, If anyone has anything, thanks.