active low mosfet switch circuit

Thread Starter

emoney123

Joined Jun 4, 2012
32
Hello,

I am looking for ideas for a mosfet circuit to switch on my project battery. Ideally I need a zero to low current draw/leakage circuit for extended shelf life of the battery and I would like the mosfet circuit to connect and disconnect the circuit ground. Is there a rugged way to do this with virtually no current draw while in the off state?
Any suggestions are appreciated.
 

GopherT

Joined Nov 23, 2012
8,012
Hello,

I am looking for ideas for a mosfet circuit to switch on my project battery. Ideally I need a zero to low current draw/leakage circuit for extended shelf life of the battery and I would like the mosfet circuit to connect and disconnect the circuit ground. Is there a rugged way to do this with virtually no current draw while in the off state?
Any suggestions are appreciated.
You need almost any standard p-channel MOSFET. It would help to know your current requirements and if you are using anything other than a simple switch to turn on-off.

I added the resistor divider since you are over 15V. Some MOSFETS may let you get away with connecting them directly to ground (-18 volts to gate)

An IRF 9540 or similar for a simple/cheap through-hole part.

image.png
 

GopherT

Joined Nov 23, 2012
8,012
He said, "low side switch".
Pretty much the same circuit in N-channel.
I was assuming he wanted his switch to connect to ground. Either way, just take a vertical mirror image of the circuit and use a N-Channel mosfet.

@emoney123
Simulator is iCircuit for iPad.

Also, the pair of resistors will consume 9 mA when the switch is on (and a complete waste of energy). Use 10k or 100k if you want that down to 0.9mA or 0.09 mA respectively. Since you are doing very low frequency switching with a hand-operated switch, the 100k ohm resistors will be fine (or anything from 10k to 100k as long as they are roughly the same value as each other).
 

Thread Starter

emoney123

Joined Jun 4, 2012
32
I was assuming he wanted his switch to connect to ground. Either way, just take a vertical mirror image of the circuit and use a N-Channel mosfet.

@emoney123
Simulator is iCircuit for iPad.

Also, the pair of resistors will consume 9 mA when the switch is on (and a complete waste of energy). Use 10k or 100k if you want that down to 0.9mA or 0.09 mA respectively. Since you are doing very low frequency switching with a hand-operated switch, the 100k ohm resistors will be fine (or anything from 10k to 100k as long as they are roughly the same value as each other).

I'm sorry this may be trivial, but I tried this circuit in icircuits and when the switch is in the open condition I still get current flow through the pmos drain and source. Am I missing a parameter? This will drain the battery while in the off state.
 

GopherT

Joined Nov 23, 2012
8,012
I'm sorry this may be trivial, but I tried this circuit in icircuits and when the switch is in the open condition I still get current flow through the pmos drain and source. Am I missing a parameter? This will drain the battery while in the off state.
I should have told you it's a bit wonky sometimes.
First, check that you have a p-channel MOSFET.

Tap (double tap?) and set the beta to 1 or more.

Then tap on the lamp and set the nominal voltage to 18V from 120 V.

Check the current flow, it should be 180 nA or nano amps.

0.00018 milliamps will let your battery run forever (500 years for a 1000 mAh battery).

Also, check the DATASHEET of the actual MOSFET you plan to use - look for quiescent current.
 

GopherT

Joined Nov 23, 2012
8,012
I should have told you it's a bit wonky sometimes.
First, check that you have a p-channel MOSFET.

Tap (double tap?) and set the beta to 1 or more.

Then tap on the lamp and set the nominal voltage to 18V from 120 V.

Check the current flow, it should be 180 nA or nano amps.

0.00018 milliamps will let your battery run forever (500 years for a 1000 mAh battery).

Also, check the DATASHEET of the actual MOSFET you plan to use - look for quiescent current.

Also, the specs on a real MOSFET at 18V drain to source will be 100 uAmps for the IRF9540.

If you go with something modern and a voltage closer to your needs, you can find something with a quiescent current of 1 uA (0V gate-source) .
 

AnalogKid

Joined Aug 1, 2013
8,948
At 18 V and 0.5 A, there is no advantage to having a MOSFET switch over just using the mechanical switch to switch the load directly. Secrets, secrets, secrets...

ak
 

#12

Joined Nov 30, 2010
18,217
there is no advantage to having a MOSFET switch over just using the mechanical switch
I thought about that. "If you're going to walk over and turn on the switch, why not just use a switch?"
But then I would be playing the 20 question game, and I get so tired of that.:(
If people are going to ask nebulous questions about indescribable purposes, they can try to work with whatever answers they get.
 

GopherT

Joined Nov 23, 2012
8,012
At 18 V and 0.5 A, there is no advantage to having a MOSFET switch over just using the mechanical switch to switch the load directly. Secrets, secrets, secrets...

ak
If he wants a small tactile momentary switch, they are normally limited to 100 mA and some less.
 

AnalogKid

Joined Aug 1, 2013
8,948
True, but zillions of small toggles and rockers from C&K, NKK, Alps, Alco, etc. can do this. Plus, there's nothing about momentary in post #1. Hard to beat the standby current of a 7101.

ak
 

GopherT

Joined Nov 23, 2012
8,012
True, but zillions of small toggles and rockers from C&K, NKK, Alps, Alco, etc. can do this. Plus, there's nothing about momentary in post #1. Hard to beat the standby current of a 7101.

ak
I'm just trying (grasping for straws trying) to explain why he is going this route.
 
Top