if the IO on the uC has open collectors like most do until you initiate them to include the internal pullup/down, the gate of the fet can be stuck floating aronnd and maybe in a damgerous zone or not in a working zone.
Jamie
I partly agree, I hope those input are now safely sealed into a box made of steel.
True, two inputs with no output is not a good idea.
Depending on the drain current, 3.3 volts could be marginal gate drive for a 2N7002, especially over temperature. You might consider a "logic level" fet.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
So you need 5 volts on your MCU inputs? Normally 3.3 volts is enough for any logic family up to and including 5 volts unless it is true CMOS.
The current needed to control a BJT can be made very, very tiny. Is this a battery powered app?
-- Rick
The 2n7002 is *the* most static sensitive part I've ever handled. I blew one after another building "hobby" breadboards, until I got really serious about static control for that one part.
I was thinking about that this morning, coincidentally, as I designed a non-critical power-up delay for a toy:
Q1 FMMT717
+4.5V >--+-----+----. .----+------------> | | \ ^ | | | ----- | .-. .-. | .-. R1 | | | |R2 | | | R4 1M | | | |10K .-. | |100K '-' '-' | |R3 '-' | | | |1K | | | '-' | | | | | | '------+ | | | | | | | | Q2 ||-' | | 2N7002 ||
Or if the I/Os come up as inputs/tristate (as most, IME, do).
We use boatloads of Vishay 6A27s on power inputs. That and a 100uH inductor and a couple of thousand uF. ;-)
Power-up delay, yes, nice! With a diode across R1 to rearm it quickly at power-off? I fixed Q1. Could be a p-MOS. R3 could be smaller, depending on I-load.
--
Thanks,
- Win
I was looking at power switches recently, the ones designed for USB are
*really* cheap. E.g. AP22802A, $0.05 each on reels. This part is so cheap I feel I have to consider it for every power switch.Controls slew-rate, current limit, thermal cutout, UVLO, logic level enable input. 50mOhm, 2 amps. All for the price of the similarly rated P-FET. 5V only though.
-- John Devereux
Not bad. What makes it a 5 volt only part? The enable threshold is 1.5 V max and the internal UVLO is 2.4 max and recommended operation conditions is down to 2.7 V. They even give specs for switch resistance at Vin = 3.3 V. Funny, the current limit spec has a foot note indicated, but no foot note.
-- Rick
Thanks for fixing my PNP. Yes, it could be a P-FET, but I've got a reel of the FMMT717's, and two reels of the 2n7002s.:-)
This is a two-off demo, so simplicity and stock were the key motivators. A uC will do the job in real life.
John Devereux's load switch looks pretty sweet...
I hadn't intended to permit quick reset, but now that you mention it, it might be a nice feature. Load current is perhaps 200mA peak, pulsing.
Cheers, James Arthur
Why is the pulldown resistor R4 tied to R5 rather than ground? Is this for hysteresis to prevent oscillation or sensitivity to noise?
-- Rick
Yes, R4-R5 gives 10% +FB to help everything snap on once the output sees some dV/dt. It would be defeatable by a very large capacitive load, or a short; but it would seem, it's not intended to be (and certainly is not capable of) short circuit conditions for very long, so the lack of hysteresis under that condition wouldn't seem to matter.
Personally, I'd put R2 across Q1 E-B, and maybe a 5V zener for Q2, just in case (hey, then it's good for almost any input voltage!).
You could continue by adding more common-emitter stages to handle logic functions (like the missing on/off / set/reset function), active current limit, self reset and so on. (Or go out and buy the $0.02 integrated switch version. :) )
Tim
-- Seven Transistor Labs Electrical Engineering Consultation Website: http://seventransistorlabs.com
Yes sorry I meant 5V max, it is fine for 3.3V operation too AFAIK and very useful ideal for switching 5V supplies from a 3.3V micro. I use it e.g. to gracefully power-up off-board daughter modules in an instrument.
-- John Devereux
Yep. Without it the slew rates are painful.
The main current is drawn by LEDs, so the load is pretty light until the output hits Vf(LED).
Yes,we could add all those things. But, sometimes a switch is just a switch.
Cheers, James Arthur
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.