Other Parts Discussed in Thread: TPS2660,
Hi,
I am considering using the 25921x or similar device as a load controller in a 13V8 (nominal) system.
This question concerns the device's behaviour under EMI events and not steady-state voltages.
Circuit overview...
25921x IN terminal is powered by the 13V8 (nominal, 15V5 abs. max) rail which has significant bulk and ceramic capacitance.
Its OUT pin connects to the load: No bulk capacitance; adequate ceramic caps and protection devices for the ESD & surge threats.
Test conditions & questions...
The load supplied is subject to:
ESD (HBM, +/-6kV contact, +/-8kV air), and
Surge (61000-4-5, Class 2: Rg=42R, 1kV --> 24A peak current, 5 pulses per polarity, 5s interval).
Our load protection fails if connected to DC >18V (a longstanding acceptable product limitation), so I am not concerned[*]) about the 20V DC limit on the 25921x.
Our ESD protection aims to limit the resulting voltage rise to <20V (charge balance calcs), suggesting that the device's 2kV un-aided rating should suffice.
The surge event is of much longer duration (~30us).
Front line protection will clamp the pulse to around 38V, and this voltage will persist for the pulse duration (1st order estimate).
If the 25921x's is ON during the event:
The FET channel will shunt the remainder of the pulse energy into the 13V8 bulk capacitance which will maintain the voltage well below 20V.
If the 25921x's is OFF during the event:
Will the FET body diode conduct during the 30us surge event (thus similarly clamping the pulse at the OUT terminal to the IN side bulk capacitance)?
Does TI have any specifications or test data relating to the behaviour of the body diode?
Does TI have any test data relating to the behaviour of the device under 61000-4-5 testing at the 42R level?
[*] Parts with this price point and capability at higher voltage ratings are of course welcomed :)
Kind regards,
George Tzanatos.
