half bridge mosfet switching

you will only see that problem when using insanely low threshold FETs, at elevated temperatures, with a lousy gatedrive :)

thats a fairly high gate resistor, the IR2110 has a fair bit as well. Try a diode across Rg, anode to FET, so turn OFF is a lot faster than turn ON.

say Vth = 3V

If you have a +12V gate voltage, then (carefully ignoring Cmiller and nonlinear Cgs)a fully discharged gate will take -ln(1-3/12) = 0.29 time constants to rise from 0V to Vth = 3V.

By contrast, discharging a fully charged gate from +12V down to Vth =

At Vg = 12V, Qg = 140nC so Cg = 11nF or so. Ignoring the IR2110 Rout (which is small c.f. 25 Ohms), the gate time constant is about 250ns.

And that completely ignores the flat spot caused by high Rgate, which will further increase the switching time.

So it doesnt sound like your 500ns dead time is really enough.

its actually an IRFp460 too.

I had the same exact problem with building a low power switcher that needed to operate off 200-400 VDC (half bridge). By low power I mean 20mA off the vbuss at full load. Noload current draw was 4mA. The IRFP460 was too much capacitance and suffered from the same problems you are describing. my no load was 60mA.

I had better luck by pushing up the Rds on rating and getting parts that had less D-S capacitance. (irfbg30, bg20, some Ixys fets also had high voltage and low capacitance). Also drive the gates harder with < 10 ohm resistors.

If you're trying to do this on anything other than a PCB...forget it.