EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: girishji on January 16, 2023, 09:42:23 am
-
I am using a logic level N Mosfet (https://www.mouser.pl/datasheet/2/115/DIOD_S_A0009865515_1-2543348.pdf (https://www.mouser.pl/datasheet/2/115/DIOD_S_A0009865515_1-2543348.pdf)) for switching a load. It switches on very quickly (few nanosec) but takes ~1.2us to switch off. Gate capacitor is not too unreasonable though. The driver circuit I am using is simply connecting GPIO pin to Gate through 160R resistor. It looks like fig 12 without the diode https://www.ti.com/lit/ml/slua618a/slua618a.pdf (https://www.ti.com/lit/ml/slua618a/slua618a.pdf)
I cannot include turn-off diode as in Fig 12 since the GPIO pin cannot sink >20ma. It is not recommended by MCU I am interested in (both RP2040 and nrf52x).
There were bunch of other circuits they recommend (fig 13, 14, 15) but they all have drawbacks or needs -ve supply. I need to improve only a little bit (0.5-0.7us turn off will be fine). If you have any ideas please chime in.
-
One can use a circuit like in Fig 13 (or inbetween Fig 12 and 13) of the Ti text. So a PNP transistor to replace the diode. The extra current for a fast turn off would come from the transistor instead of the MCU via the diode.
The Diode for the turn on part could be replaced by a resistor.
P.S. for just a little improvement, just a smaller gate resistor coudl work. For a short pulse the µCs are usually OK with a somewhat higher current.
-
Because it is a single ended drive, the turn off time must be slower than the turn on time, and choosing a smaller pull-up resistor will improve it.
But if you get turn off times >1us, maybe that's not the gate capacitance at play, it's more like the capacitance of your oscilloscope probe, switch it to the X10 position if it's on the X1 position.
I did a simple test with the low threshold MOSFET FDN337N that I mentioned in another topic. The gate is connected in series with a 160R resistor, and the drain is pulled up with a 1K resistor. It can be seen that the turn off time is about 250ns.
-
A smaller gate resistor is acceptable because the current flow is not continuous; those are DC limits.
If that's still not enough, add a driver. This can be a bunch of gates in parallel (e.g. 74LVC3G17), or a gate driver proper.
Tim
-
Problem solved. It now switches off in nanosecs. I just increased the current at Drain to a value I will be eventually using at. I(ds) has some bearing on switch off time it seems. When I read "1k pull up resistor" it occurred to me.
First pic is when Drain was carrying ~2mA (switch off time ~1ms), and second one for ~20mA. Yellow signal is Vgs inverted.
-
Oh, that "switching" time... yeah, transistor switching has to do with current, the voltage you see needn't reflect that. :-+
Tim