2N7002 Questions

[dodadew]

I have a dilemma. I am building a circuit to switch 2 separate "loads"...one with a switched ground and one with a switched positive. I am using an ATTINY to switch a pair of 2n7002 Mosfets. I have attached a schematic. I am using a resistor to step down the voltage from +12 to ~ 4v to run the ATTINY. The question I have is if Q2 would run consistently in this setup or if you have a better idea of what to run? I am using the 2n7002 to sink < 1 amp in both directions. Ultimately I have another circuit monitoring the output of these pins checking for -12 on one and +12 on the other. The circuit works but I had to swap the Source and Drain of Q2 to get the functionality that I want. I am hoping this is proper use, and not an undocumented feature that will soon fail. Does anyone have any insight?

EDIT : I think I have a bug in my schematic I will follow up in a moment.
EDIT2: Lol. Lets make this easy. I need to switch +12v from a +4ish v signal (the attiny going HIGH). Any ideas on what I could do this with?

[eccerr0r]

Re: 2N7002 Questions

Q2 won't work the way you expect it to work. As you found, reversing D and S works because now you're using the body diode as the conduction device, and you might well just short out D and S and save a little power this way...

I'm not sure how you're using a resistor to 'step down' 12V to 4V but surely that's also questionable use of resistors to produce a constant voltage needed for digital electronics. You're best off using a cheap linear dissipative regulator like a 78L05 or something to power the attiny.

[dodadew]

Re: 2N7002 Questions

This circuit basically sends a few pulses quick (~ 5 seconds of off and on) then shuts down (~10 microamps). I was going to use some sort of 5v regulator, but to keep cost down and mainly size (final product is less than .5in x .5in) but with the tiny current pull, the resistor kept the ATTINY within operating voltage. The q1 side works perfect. Right after I posted I noticed q2 was not doing what I wanted. I don't know where my brain was. I think what I need to use is a P channel instead to get the functionality that I want. Does that sound like English, because I am sorta following what I am saying.

[stj]

Re: 2N7002 Questions

you need an N type fet and a P type fet.
i dont know what your driving but know this,
fets take time to turn off.
if your doing motor drive, you must put a small software delay between turning one fet off, and turning the other on.
it's called "dead-time" if you dont, you can end up with both fets conducting at the same time and exploding.

[dodadew]

Re: 2N7002 Questions

I'm actually using them to signal another device. Both are on separate circuits but they need to signal +12 and -12. In a nutshell this circuit powers up, waits 6 seconds, goes high on pb1, waits 500ms, goes high on pb0, waits 1s, goes low on pb0, waits 1s, goes high on pb0, waits 500ms and goes low on pb1.

Since I need 12v signal I am switching the fet to to get this. No motor drive. . Pb0 is signaling -12 and pb1 needs to signal +12. The -12 side works perfectly. I just need a way to signal +12 from a sot-23 package. That's where I'm stumped. As I am not 100% sure of the current draw from the other device reading the signal (small but unknown at this time) I chose the 2n7002. Got half working at least.

[stj]

Re: 2N7002 Questions

find a p-type equiv of the 2n7002

[budm]

Re: 2N7002 Questions

The SW-12V is the same as GND when SW1 is turned on, but since SW-12V is the same as -12V (which is +12V return) so why do you need to switch the GND that goes to another circuit?
The SW+12V will feed another circuit (The LOAD) and has to return to power source negative which is GND, so what is the purpose of SW-12V(GND)?
Are you using dual power supplies, +12V & GND, -12V & GND?

[eccerr0r]

Re: 2N7002 Questions

I still think even if you want to reduce cost/size, using a single resistor to "create" your +5V for the attiny will result in reduced lifetime or poor operation for the MCU because of regulation problems. Not sure how critical for your device to work but I'd not let it work like this for any extended period of time with any reliability.

Even if you do use a pfet you still have to be able to shut it off somehow which you won't do with a +5V supply for the attiny, you may have to go use a two stage pfet/pnp driver and a nfet/npn controller for this.

btw, this is confusing, -12 as in -12 wrt to GND (meaning 24V rail to rail) or just 12V rail to rail and -12 is actually GND as the schematic seems to imply? The -12 should also not work if it's truly -12V with a 2n7002 nfet. Just trying to make sure your conventions are correct...

[budm]

Re: 2N7002 Questions

"btw, this is confusing, -12 as in -12 wrt to GND (meaning 24V rail to rail) or just 12V rail to rail and -12 is actually GND as the schematic seems to imply? The -12 should also not work if it's truly -12V with a 2n7002 nfet. Just trying to make sure your conventions are correct..."
That is what I am trying to understand too as to what he is trying do to the load (the other circuit).
Also there is no cap for the VCC, at least use Zener diode and cap there on VCC.

[stj]

Re: 2N7002 Questions

if you want to reduce cost, use a pic chip instead of an atmel.
something like 12f629 or even a 6pin 10f series.

[dodadew]

Re: 2N7002 Questions

Sorry I have not been around for the last few days. Ill follow up with explaining everything this evening. Thanks for the input.