LT1962 misbehaving - why?

[stolbovoy]

I'm building balanced to unbalanced converter on OP727 -> optoresistor volume control. I'm using LT1962 and LT1964 in the power supply. I don't see any problem with LT1964 so far, but LT1962 oscillates.
LT1962 oscillates at about 80Hz (sawtooth) loaded on 20uF OS-CON, it also oscillates with higher ESR capacitors (BG) as well. Oscillation amplitude is about 0.2V.

It is also oscillates loaded just at TL074 + 2 LEDs. Oscillation in that case depends on LEDs current.

LT1962 used in the standard circuit from datasheet with 18v input ant 14v output.

I checked my PCB number of time it should be correct.
Is there are some specifics in using LT1962? Are they generally behaving this way, or I just got ICs from bad batch?

Thanks,
Viktor

 

[h_a]

No way 0.2V oscillation is normal behaviour for any regulator that deserves the name.

What is the dropout voltage in your supply?

Have fun, Hannes

 

[stolbovoy]

Oscillation is worse under load (30mA)
Dropout voltage according to design is about 4V.
It is actually second stage of the regulator. The first is simplest transistor + zenner + resistor regulator to make 18-20V (so input will be within operational limits of LT1962) from 60v power amp input/driver stage rail.

I probably will replace LT1962 with LT1761 or LT1086. LT1962 package is inconvenient to use with DIY PCB anyway.

 

[dmills]

What value (and what type) is the cap across the input of the regulator?

An excessively high source impedance can cause squegging which may well be what you are seeing.

Regards, Dan.

 

[stolbovoy]

Sorry, couldn't reply sooner...
Capacitor at the input didn't help. Got LT1761 in the yesterday mail.

 

[stolbovoy]

Project had been put aside because of absence of time till 2 weeks ago.
Solution was pretty simple: WIMA capacitor on BYP picked up interference from Mastech "lab" PSU. Changing the capacitor to 1206 one helped a lot