Oscillating current clamp...how to stop oscillation?

[cupoftea]

Hi,
Fig 24, page 11, is a current clamp, using TLV431, but such simplicity may oscillate.

https://www.onsemi.com/download/data-sheet/pdf/tlv431a-d.pdf

Which of the attached circuits are least likely to oscillate? (LTspice and jpeg)
Note that there is no output capacitance, and here lies the harbinger of the danger of the oscillation

Here is a previous thread on it.

Current clamp oscillating at 50MHz

Hi We have a DALI power bus comprising a 15mA current clamp in a 15V rail. (as in attached schematic) As a test, I turned on the transistor, Q4. When I clipped a TA041 differential probe across the 82R resistor, I saw a 50MHz sinusoidal oscillation across this resistor, of about 250mV pk2pk...

www.edaboard.com

Also attached the TLV431 spice model etc

 

[FvM]

None of the circuit variants is providing useful loop gain compensation.

By connecting shunt directly to ref pin you are effectively shorting compensation, there must be a series resistor to increase ref node impedance. Secondly connect the compensation RC series circuit between cathode and ref pin. Set up a loop gain simulation.

 

[FvM]

Circuit B and C have friendly loop gain, even without compensation capacitor, presumed TLV431 model is correct.

 

[D.A.(Tony)Stewart]

Look up Type III or (3) compensation

 

[cupoftea]

Thanks, the attached appears to conform to your kind suggestions.
I would like to drop away from the type 3 bits to save component count.
(LTspice and jpeg as attached).

Actually i think C2 and R1 should lie across R14 instead...Would you agree?

 

[FvM]

The compensation scheme isn't bad now and could be used if the circuit has stability problems in a different implementation. With given shunt and load dimensioning it hasn't.

Only the first circuit in post #1 is unstable due to misplaced C3.

--- Updated Jan 13, 2024 ---

Actually i think C2 and R1 should lie across R14 instead...Would you agree?

Not at all.

 

[cupoftea]

Hi,
Regarding the above currrent clamp which uses a TLV431, why instead, would you ever use the attached type of current clamp which uses an opamp?
(i mean, the TLV431 version looks so much cheaper and better that i wonder why anyone would use the opamp version? There is the bias current in
the TLV431 which ends up in the clamped current too, but its just a few mA)

 

[D.A.(Tony)Stewart]

Bandgaps have been around since Jim Williams and Bob Pease created these designs.

The adjustable versions are more popular and price depends on volume or global orders and yours.

I would expect $0.15 by the reel or less.