So I'm using the ir2110 mosfet driver with to drive the ir3205 mosfet in a high side circuit. The problem is that the circuit is not working. I've rebuilt it many times so I don't think I've made any basic mistakes. I'm using an 22uf tantalum capacitor for the bootstrap operation, 10ohm gate resistors and a 14V power supply for the bootstrap. I'm driving a 12v motor. Like I already said, I'm only using 1 mosfet for the high side, I've grounded everything else.
I've had several problems with this circuit, one of them being that the motor isnt turning, but also that source is actually giving the mosfet power, not from my 12v supply for the motor, but from my 14V supply that I'm using for the bootstrap. When I'm writing this the Vb voltage is 14V in relation to ground but 0V in relation to Vss.
BTW, using this chip in a low side circuit works fine, but in the high side there aren't even a small signals indicating that the mosfet is getting turned on, as if the bootstrap capacitor weren't getting charged up to the proper voltage. I checked the difference in voltage of the capacitor and my voltimeter said 0.
My circuit is the one on the front page of the datasheet I've included (ir2110)
MarkT:
You cannot use a high/low driver to just drive the high side, then need the low side
switch to drive the charge pump.
What do you mean by that? He also shows a circuit using only the high side. Also, what do you mean by "need the low side switch to drive the charge pump? Do you mean that I can only charge the capacitor if 1 of the outputs (either ho or lo) is on at a time?
Well if your MOSFET's gates are damaged you might need to store more charge on the bootstrap cap,
but MOSFET gates should be very high resistance indeed, so the voltage on the bootstrap cap shouldn't
be drooping very fast.
The larger the bootstrap cap the heavier the current spike when its re-charged (which goes through
the diode).
Normally you'd arrange the voltage drop on the bootstrap diode on switching to be a volt or less, since
you want the gate to be at least 10V when on, and its charged via one diode drop from 12V. So you make the
value in nF the same as the gate charge in nC, so one volt drop on switching.
You want the decoupling cap on 12V to be >> the bootstrap cap value (so if the bootstrap is 22uF, the
decoupling cap should be 220uF or more to avoid the bootstrap cap being able to crowbar the 12V supply
and triggering undervolt cutout).
220uF is excessive, 10uF ceramic on the 12V ought to be enough...
johnfg:
Also, what do you mean by "need the low side switch to drive the charge pump? Do you mean that I can only charge the capacitor if 1 of the outputs (either ho or lo) is on at a time?
The switching on of the lower MOSFET is what charges the charge pump in these high/low drivers. So you
always need to park the half-H-bridge with the lower FET on, if you park it, or else continually use PWM with
a duty cycle that stays below 100%, so that the lower device switches on regularly to charge the cap.
Normally for motor control you'd be using PWM anyway, so you just have to arrange its never
goes above 96% or whatever allows enough time to charge the cap.
The very short times available for recharge that this implies is why you want the bootstrap cap no larger
than necessary, as it might be called to recharge from 10V back towards 12V in 1us or less. This is
also why the diode is normally ultra-fast-recovery or schottky. The current spikes into the bootstrap
cap can be several amps, so its important to have tight layout on the PCB around the charge
pump components too.
Thanks man, but I still can't get it to work. My first problem is that I don't know how to charge the capacitor. You said to use PWM but I don't know what you mean by that (I know what PWM is but I don't know in what way to use it). If you could show me a couple of lines of code so I can understand. And for some reason Vs is actually powering my load, even if its 2v. My high side mosfet is not turning on, only my low side one.
Btw, in Tahmid's blog there is a circuit that grounds the low side mosfet :
Talmud's high-side only circuit is a rather tricky circuit - it relies on the properties of the load to both
charge the bootstrap cap and avoid crow-barring the supply in the process.
It cannot handle indefinite on-time, and cannot handle an inductive load without frying itself and
the high side MOSFET.