H Bridge driver using discrete components

[mrinalmani]

I have built an H-Bridge driver using discrete BJTs only (no driver IC). The driver gives a pretty clean simulation output. However, BJTs as we know have lower surge handling capability as compared to MOSFETS.
The BJTs that I am thinking to use are rated 1A continuous and 2A peak current with a turn-off delay of 17nS. (Priced 0.1$ @ 100 units)
MOSFETs with similar pulse capability with a delay of 11nS are available for 0.05$.
Now, my question is, can the BJTs be replaced by MOSFETs without causing reliability issues?
The high side circuit is attached below.

 

[kabeer02]

What do you really want to drive?

 

[neddie]

What is the voltage of your driving signal? If it's only 3V3 you will need logic level mosfets etc.
Unless you are running your bridge fast , I would not bother about the difference between 17nS and 11nS.
Remember the current into the fet gate is quite short compared to the on time in most cases.
BJT's will be easier to build and test with I'd imagine. Is this for mass production where the cost difference is an issue?

 

[mrinalmani]

@kabeer02: The right most MOSFET is being switched

@neddie:
yes, you are right. It's the mass production that is driving my focus to MOSFETS. Each driver will have to use 8 BJTs (one pair for each power switch)
Upper thresh hold of switching frequency is approx 150kHz. And the driving signal is indeed 3.3v
Surely the current through the gate is for a small time. Does this translate to crossing the absolute maximum rating of the BJT for a few nano seconds.
"BJTs wiil be easier to test" I think the same, but cant justify it. Is this thought justifiable?