How to measure 3V battery lifespan?

[circuitmaker]

Hi everyone,

I am using an IRF4905 P-channel Mosfet to control the main circuit power supply. When the voltage is applied to the gate of the Mosfet the main circuit is disconnected from the power supply. Most of the time the "Power switch" is in a closed position, applying a voltage to the gate. I am curious how long the CR2032 3v battery would last in this case. How do I measure the current consumed by the IRF4905 Mosfet in a "sleep" mode? When I use a tester to measure the current, it shows zero amp usage. I can tell that the battery voltage is dropping within a week from 3.1V to 3V. Looks like it is discharging pretty fast. Ideally, it should last at least a year in a "sleep" mode. Will resistors help to extend the battery life?

 

[Bluejets]

R1 your gate resistor..??? Are you sure..??? Switch position...???
Would be more obvious if you marked "load".

 

[circuitmaker]

R1 your gate resistor..??? Are you sure..??? Switch position...???
Would be more obvious if you marked "load".

The load is labeled by "VCC". R1 is a pull-down resistor for the gate. The gate resistor is not shown on the schematic and should be much lower than 1M.

 

[circuitmaker]

I retested the current and it shows 2.9uA. The battery drained probably because of the testing I have done several times.

 

[Harald Kapp]

How do I measure the current consumed by the IRF4905 Mosfet in a "sleep" mode?

A MOSFET "consumes" no current. The static gate current is negligible. Most of the current in the gate circuit is through R1.
3 V / 1 MOhm = 3 µA, that's the 2.9 µA you measured.

Most of the time the "Power switch" is in a closed position, applying a voltage to the gate.

No.
When the switch is closed, V_GS = 0 V.
When the switch is open, V_GS = -3 V via R1.
The MOSFET will be off with the switch closed and vice versa. Typical case.
However, the thrreshold voltage of the IRF4905 varies between -2 V and -4 V:



Depending on the specific part and factors like ambient temperature, the -3 V from the battery via R1 may not suffice to turn on the MOSFET reliably.

So why use a MOSFET here at all? The power switch could directly switch +3V0 to Vcc without the additional losses incurred by the MOSFET's R_DS.

 

[circuitmaker]

So why use a MOSFET here at all? The power switch could directly switch +3V0 to Vcc without the additional losses incurred by the MOSFET's R_DS.

This schematic is part of a wire break alarm. When the wire connected to the "Power switch" is cut, MOSFET should turn on the alarm. I don't know any better solution to implement this.

I am considering using two or even three AAA batteries instead of a tiny CR2032.

 

[Harald Kapp]

When the wire connected to the "Power switch" is cut, MOSFET should turn on the alarm.

O.K., that's reasonable but not obvious from your first post.
There is still the issue of reliability. The MOSFET will not turn on if V_GSth < V_bat.

With the 1 Meg pull-down resistor you can't get rid of the 3 µA static current. This sums up to ~ 25 mAh of losses every year.
A CR20232 has a capacity of approx. 200 mAh, give or take a few mAh depending on the specific make. The self discharge is on the order of 1 % per year (datasheet) which is negligible compared to the static current through R1. Thus the lifetime of the battery is limited by the static current to [imath]\frac{200 mAh}{(25 mAh per year} = 8 years[/imath].
This is a best case estimation. other factors like temperature, humidity and pollution also play a role. To be on the safe side, you should test and possibly swap the battery every year.

I am considering using two or even three AAA batteries instead of a tiny CR2032.

An alkaline battery has a much higher capacity. This is - in principle - good for a longer lifetime. But self discharge is higher. Shelf life is only 5 years, see e.g. this table. On the other hand 3 alkalines will deliver 4.5 V down to 3.6 V and will thus make the system much more reliable with respect to variations in V_GSth. Plus they are easily available. And the discharge by the static current will not drain the batteries noticeably. Changing them every 4 years (with 1 year safety margin) should suffice.