passing signals over galvanic isolation without an opto coupler

[Marco]

I don't know what you muean by simple split planes. I put the capacitive CAN transciever on a 4 layer PCB with 330nF each side and it passed.

Just one ground plane on either side of the isolator, no overlapping planes or other capacitance between the sides, power traces instead of power planes (ie. 2 layer). Even in such a realistic worst case I doubt it will be a problem.

[Simon]

Yea that is what i had but it was 4 layer and it past military grade EMC. I think the gap was 0.5mm

[schmitt trigger]

Are they all called pulse transformers? Basically what sort of search terms lead to what I am looking for, what do people call these things?

If you decide to go the pulse transformer route the key requirement is the maximum volt-time product the transformer may sustain.

Which essentially is the applied voltage (5 volts in your case) times the maximum “on” period.
This occurs at the lowest frequency and highest duty cycle.

Pulse transformers require a pulse by pulse core reset. Fortunately pulse transformers have been employed since Adam and Eve to pass signals thru isolation boundaries.  There are many proven circuits.

Unitrode (nowadays TI), had an excellent app note.
Google or DuckDuckgo is your best friend here.