TPS22946 trip point hysteresis

I think its just a "hard limit" ie no hysteresis. If the OC condition persists, the switch shuts off. Note the wide variation (min to max) for the CL setting. This is not a precision current limit.

Thanks - I think I'm the one who instigated this question through Arrow, San Diego  . . .

For whatever the device's threshold is, is there a mechanism to prevent chatter in a slowly increasing load that might be noisy?  If I slowly turn a potentiometer to increase the load to the trip point, how much do I need to back off the load to reset the 5 second timer?

Referring this to BU Apps. and Device Engineers for comment.

Hi Jeff, 

Has there been any word on this?

Device Apps. has not responded yet (I will keep asking), but as far as I can tell this would be a "moderate overcurrent condition" which is sampled for .5 ms every 63.5 ms as the Data Sheet states:

"If the constant current condition still persists after 10ms, the device shuts off the switch and pulls the fault signal pin (OC) low. The TPS22946 has an auto-restart feature which turns the switch on again after 70 ms if the ON pin is still active".

See Fig. 28 for start sequence ad OCL timing.

Jeff, Joshua,

We are currently testing this out in the lab to see if there hysteresis in the device. We will keep you updated as we go through the experiment. 

Could you provide more details on "noisy load" condition mentioned above?

Just ran a test in the lab at 4.4V, CL pin grounded.  I used a fixed 121 ohm load and put a 100 ohm potentiometer in parallel.  I slowly decreased the potentiometer resistance until the TP22946 entered moderate overcurrent mode.  That potentiometer resistance was 68.1 ohms, or ~101mA total load.  I immediately reversed the potentiometer direction and pulses did not stop until about 80 ohms, or ~91mA total load.  (This test was done with only 1uF load capacitance.)  Looks like I had about 10mA of hysteresis.  When I re-ran the test with same resistances, but using our actual capacitive load (~94uF), I was not able to reset the timer  - the load switch opened 6 seconds after I reached 68.1ohms as if the full load was still present, event though I increased the parallel resistance to 100 ohms (80mA total load).

The good news is that there seems to be a significant hysteresis (10 mA in my test with 1uF capacitive load), and even more with a heavy capacitive load.  But it does bring up the question: What is the maximum recommended capacitive load for the TPS22946?

The noisy load in question might be one that has 100us current spikes of 4-5 mA with a DC average approaching the current threshold of the TSS22946 (50 to 120mA for our application).