tps27081 load switching

Any problem besides that the P-Channel is reversed?

Fred,

That is a creative usage. I find it very interesting.

If both inputs are on then both devices will damaged.  

I also see another problem, the load will be much less resistance than the pull up resistors. Therefore, when one NMOS starts to turn on both PMOS will have about the same gate voltage, so both PMOS will turn on and overpower the much smaller NMOS and cook it.

The circuit works for both inputs off and it works once one TPS27081A has both MOSFETs fully on and the other TPS27081A has both MOSFET cutoff. However, I suspect that the transistion between off and on will not work well.

Tell me about the VCC and the load; then I can suggest a device to power the load.

Load is a latching relay controlled by an FPGA.  Static, both sides off.  Then one side pulsed on.   Under normal operation conditions (that I contol) only one side is turned on at a time.  There is usually a significant delay between pulses on opposite control lines.

Fred,

I see.

What is the VCC supply voltage and coil current? I need also to know the FPGA logic high voltage.
Is there just one relay?

VCC = 5V

Drive Voltage = 3.3V

Coil Current rated at 33 mA

Relay is latching, so current is pulsed.

Fred,

I'm having a difficult time tyring to find a single IC that accepts 3.3V logic and can drive a 5V relay coil bidirectionaly.
Can you switch to a 3V latching relay coil?

No.  The only problem I can see with the circuit I posted is that some vendors do not spec the current rating for the N-FET.

Fred,

I found the TS5A22366 that can run the relay using 5V and accept 3.3V logic input.
I attached the the pin connection. VCC (not shown) is 5V.
To clamp the inductive turn off, I suggest adding a diode to VCC and a diode to ground on each relay coil pin.