Have you considered using an n channel mosfet for Qinv? Are you sure you're driving Qinv hard enough to get it to switch at the speed you need? Hook a scope up to the input and output of Qinv and verify that is the problem. -Adam On Wed, Jul 27, 2011 at 11:25 PM, Forrest Christian wr= ote: > I'm basically using a circuit based on the schematic from Figure 20 on > page 20 of http://focus.ti.com/lit/ml/slup169/slup169.pdf to drive a > high-side mosfet. Vin is around 30V. > > My application requires a fast (~1uS) turnoff of the FET. However, I'm > seeing actual turnoff times much longer than that. The faster time seems > definitely doable with reduction of Rgate to an appropriate level and a > quick shift of the drive level into the totem pole drivers. With that in > mind, I seem to be having problems where the level shifting transistor > (Qinv) isn't changing state very fast. I suspect it's related to the > issues described in the text - and specifically the saturation of the > level shifting transistor. The text recommends moving a resistor around > - but as the purpose of Qinv is to level shift the level from the 3.3V > logic I'm working with up a level between 30V and 18V depending on which > state it's in - putting half of the voltage divider below the transistor > prevents the 3.3V from turning the transistor on. > > I'd like to find a circuit which will work for this which is fairly > non-complicated. I'm not completely stuck on the P-channel device, but > driving a N channel device seems even worse (even with a driver intended > for this). Of course, there is every possibility I just haven't found > the right driver yet. > > Ideas? Hints? > > -forrest > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .