Ideas to toggle supply (battery powered product)

[ixps]

Hello everyone,

I designing a circuit are battery powered and I would put a push button to turn on / turn off battery, to avoid discharge without use.

My supply is between 3.4V .. 4.2V (when batterys are fully charged).

I thinked to use the IC LMC555 (because its minimum supply is 1.5V) to toggle output (when press a simple push button), and the are works perfectly in simulation, see:



However, I need to use either MOSFET or RELAY to turn on / off the supply of circuit and i'm doubt about which select. Anyone has an idea? With mosfets, the circuit a little bit, and with relay need the coil supports 3.4 ... 4.2V.

Thnks

[PartialDischarge]

Use the circuit at 15:50 here
https://youtu.be/Foc9R0dC2iI

[Ian.M]

The LMC555 will drain of the order of 100uA due to its quiescent current and your R1:R2 divider will draw another 150 to 200uA, so unless you have large parallel array of LiPOs such that 250uA is small compared to their equivalent self-discharge current^* the use of a 555 based latch circuit is highly undesirable as there is a serious risk of deep and possibly unrecoverable discharge if the device is stored when the battery is already low.

For solutions that vastly reduce the quiescent current and also allow you to save a MOSFET, see Eblc1388's suggestions for two inverter toggle latch circuits in https://www.eevblog.com/forum/beginners/help-on-how-to-choose-an-ic-for-a-project/

The 'EEVblog #262​ - World's Simplest Soft Latching Power Switch Circuit'  MasterTech suggested uses eight components (not including switch, battery and load) and is much fussier to get working reliably, vs five for the two inverter toggle latch if you use a dual inverter IC , e.g 74LVC2G04 or 74LVC2G17, which will give you an on *and* off state quiescent current of only a couple of uA (under 5uA worst case unless your pass P-MOSFET is notably leaky).


*  Equivalent self-discharge current calculated from capacity and their percentage self-discharge, which is typically in the 1% to 5% per month range depending on a number of factors including among others temperature, terminal voltage, age and condition.

[coppercone2]

You want to shoot for 5uA or less

[Ian.M]

You want to shoot for 5uA or less

Ideally, but not necessarily.   1% self-discharge for a month for an 18650 with a capacity of 2200mAH is approximately equivalent to a 30uA continuous discharge current.  Preferably your power switch would have a significantly smaller quiescent current, but as long as there is sufficient reserve capacity between your chosen low battery cutoff voltage and the manufacturer's rated minimum voltage below which the cell will be damaged and probably become unsafe to charge, to handle the quiescent current for a long period of storage, and your charging circuit or external charger implements undervoltage safety lockout, you can tolerate higher quiescent current.

Another option that may be attractive if your design needs a MCU anyway, is to use a MCU with a wide supply voltage range and sufficiently low sleep mode current consumption to implement the power management to the rest of your circuit.  If the chosen MCU has an ADC with internal reference, its also usually possible to compare the reference with the battery voltage on its Vcc pin,  which after calibration can let you implement low battery cutoff.  Software based charge control systems are more risky - get them wrong and 'halt and catch fire' is likely!

[ixps]

Use the circuit at 15:50 here
https://youtu.be/Foc9R0dC2iI

Thank you, MasterTech. I simulated this circuit and it works perfectlly.

The LMC555 will drain of the order of 100uA due to its quiescent current and your R1:R2 divider will draw another 150 to 200uA, so unless you have large parallel array of LiPOs such that 250uA is small compared to their equivalent self-discharge current^* the use of a 555 based latch circuit is highly undesirable as there is a serious risk of deep and possibly unrecoverable discharge if the device is stored when the battery is already low.

For solutions that vastly reduce the quiescent current and also allow you to save a MOSFET, see Eblc1388's suggestions for two inverter toggle latch circuits in https://www.eevblog.com/forum/beginners/help-on-how-to-choose-an-ic-for-a-project/

The 'EEVblog #262​ - World's Simplest Soft Latching Power Switch Circuit'  MasterTech suggested uses eight components (not including switch, battery and load) and is much fussier to get working reliably, vs five for the two inverter toggle latch if you use a dual inverter IC , e.g 74LVC2G04 or 74LVC2G17, which will give you an on *and* off state quiescent current of only a couple of uA (under 5uA worst case unless your pass P-MOSFET is notably leaky).


*  Equivalent self-discharge current calculated from capacity and their percentage self-discharge, which is typically in the 1% to 5% per month range depending on a number of factors including among others temperature, terminal voltage, age and condition.

Hello Ian, thank you very much. I got it and learn with you.

I will try to implement this circuit too and ill decide.

[bdunham7]

Not to criticize the creative and learning processes, but why go to all that trouble when there are mechanical latching switches that fully and reliably perform the same function?

https://www.mouser.com/ProductDetail/CW-Industries/GPTS203212B?qs=sajaCoHCXPRInbSJg1kuog%3D%3D

[ledtester]

One source for a latching power switch are the battery-powered "puck led lights":



which have a latching switch in them like this:

"8 x 8mm Torch Latching Tact Tactile Push Button Switch DPDT 6 Pin DIP PCB"

https://www.aliexpress.com/item/32698356157.html