Is it possible to substitute ucc37321 with ucc37322?

[Egg]

Hello. I plan to build my first real sstc soon. The problem is that it is almost impossible to ucc37321 where i live. My question is, is it possible to use only ucc37322s? I heard all ucc37321 does differently is to invert the signal. Is there a way to make it work without one?
I see the uccs are fed with signal from two schmitt triggers one to get the signal and one to uninvert it. Maybe is it possible to feed one ucc with uninverted signal and one from the output of the first schmitt trigger?
Schematics from teslaundmehr.

[NyaaX_X]

I'm not a expert. So please check and listen to others answers.

I saw somebody use an ucc27424 that has dual input and dual non-inverting output dirvers in one chip. And the 74HC14 schmitt triggers was using in the same way as you're talking about. But you may have to check if the dead-time is fitting in your design.

[AstRii]

No you cannot only use UCC37322s, one has to be inverting and one non-inverting. You could theoretically (as you stated) use UCC37322 instead but tie its input to pin 12 of the 74HC14 schmitt inverter.
The only problem with this solution is that external inverters usually have higher propagation delay than those built in the chip itself and so the internally programmed dead time of the UCC3732x chips may not be enough to prevent cross-conduction problem.
I advise using different gate drivers like for example a pair of IXDD609/IXDI609

[thaumatichthys]

The propagation delay is a good point. I could be wrong but I think as long as you get the phase correct, it should be fine, since the 74HC14s usually have a pretty low propagation delay, under 20ns typ. at 5v. Also the 74AHC series are a bit faster if necessary. As NyaaX_X mentioned, the UCC2742x contains two 4A drivers as opposed to the single 9A driver in the UCC3732x, and the 27425 has one inverting and one non inverting driver, which you could use if 4A is sufficient.

I could be totally wrong here, but according to the NGTB40N60 datasheet, the turn off time of the IGBT itself is a lot longer than the 74HC14 and the RC time constant of the 4.7R and the gate capacitance, so in my opinion the propagation delay of the inverter would not be that significant.

[Egg]

Thanks for the replies. Since it is not possible for me to get that ic at the moment anyways I will go ahead and try it this way. Just to make sure, do i need to change any other connections besides connecting one input pin to output of the first schmitt trigger?

[NyaaX_X]

I guess you have to check the driving stage output waveform after you connect one inverting schmitt to a 37322 and two series schmitt to other 37322 first. Maybe making a testing oscillator to generate signal that frequency is close to your coil designing. If the two square wave are nearing 180 degree phase shift and the overlapping are as little as well (?). That will be fine. Other things well be better is to connect those non-using input pins of 74HC14 to GND or VCC (5V~7V). And according to https://www.richieburnett.co.uk/sstate3.html " The requirement for "free-wheel" diodes " The bridge may live longer if you add some extra components.

[davekni]

Maybe is it possible to feed one ucc with uninverted signal and one from the output of the first schmitt trigger?

Yes it is possible.  I think there are one or two other threads here of people doing that.
If delay is of concern, you could substitute 74AC14 instead of 74HC14.  However, the much faster 74AC14 requires good board layout and supply bypass capacitor connections due to its higher and shorter power supply current spikes.
With a good low-leakage-inductance GDT, delay through the inverter can actually be useful.  Adds a bit more dead-time between IGBT turn-off and opposite IGBT turn-on.  So I'd stay with 74HC14.