LT1764 adjustable current source

[XgN]

Hi guys,

Wanted to implement adjustable current source which is shown in LT1764 LDO datasheet https://www.analog.com/media/en/technical-documentation/data-sheets/1764fb.pdf at page 16. The only change I did that placed OPA196 (higher min supply voltage and lower offset currents) instead of LT1366. But the output current is not adjusting only a ~50 miliamps at some point on R1 potentiometer. The output current is always at maximum according to the load/supply voltage. Is this schematic true? Seems interesting feedback capacitor C3 instead of a resistor. Or something else here I'm missing.

[Capernicus]

Just use a good ol' potentiometre.   

[XgN]

The potentiometer in position R1 is brand new, 1k multiturn. As for CC source would like to push to the limits for transients and try to achieve a a few of ten microseconds stable constant current pulse at the output. The problem that this schematic is not working at all so far.

[porter]

I think the external op amp completes the feedback using the regulator via the sense/adj pin.

"SENSE (Pin 5/Pin 6): Sense. For fixed voltage versions of
the LT1764 (LT1764-1.8/LT1764-2.5/LT1764-3.3), the
SENSE pin is the input to the error amplifier"

[KT88]

Just a stupid question: did you check the inverting and non-inverting inputs of the amplifier - I tend to exchange them accidentally from time to time just to annoy myself...

[XgN]

Just a stupid question: did you check the inverting and non-inverting inputs of the amplifier - I tend to exchange them accidentally from time to time just to annoy myself...

Double-checked a few times..

I think the external op amp completes the feedback using the regulator via the sense/adj pin.

"SENSE (Pin 5/Pin 6): Sense. For fixed voltage versions of
the LT1764 (LT1764-1.8/LT1764-2.5/LT1764-3.3), the
SENSE pin is the input to the error amplifier"

The error amplifier should be inside LDO and yes, this is the input pin to it. So far as I understand they decided to make another 1.2V reference and change "sensed" voltage according to the drop on the shunt. Nevertheless, in this schematic the non-inverting input of the opamp always gets power supply voltage (e.g. +5V) and not that 1.2V which was intended. Also the opamp should be in unity gain mode or something.

[SiliconWizard]

Just the LTSpice simulation. It's absolutely horrible. It starts at the max current (> 3 A) and takes a long time to settle, time which depends on R1 setting... and was not able to get lower than ~957 mA.
Haven't investigated it much more than this.