i am looking to make my own batt voltage monitor. the LM358 is on the table.
LM358 is rated 20mA(typ sink)

in a 8pin dip, any concerns about sinking 10-12mA ?

it will sink a piezo buzzer (~10mA) and a low power LED (~2mA).

 

The datasheet says it will work.

 

LM358 is rated 20mA(typ sink)

in a 8pin dip, any concerns about sinking 10-12mA ?

What is the full part number and manufacturer?

For On Semiconductor LM358A and Fairchild LM358, there are conditions where the minimum sink current drops to 5mA.

 

TI LM358-N

here's my ckt for doing source current, of which the datasheet says 40mA(typ). i just not sure i need hysteresis, and if so where would it go. when the alarm sounds the batt gets placed onto a batt charger. the X1 is a piezo rated 12v@10mA, and D1 is 2mA(typ)@ fwd voltage.

the purpose & goals are:
1) batt voltage monitor
2) minimal amount of discrete components
3) lowest power consumption, within reason, this sits on a 12v 100Ah batt, etc.

 

TI LM358 are guaranteed to source more current than they sink; 20mA vs 10mA. If you swap the inputs on the opamp, you can high side drive the buzzer and LED.

EDIT: OP *is* high side driving. Wasn't paying sufficient attention and missed it due to circuitous routing nonconventional layout. And the fact that the first post mentioned sinking current, not sourcing...

You're using an opamp as a comparator, which is generally discouraged. In your case, it's a don't care because you don't need the speed that a comparator offers.

Most comparators can only sink current, and can't sink much; so would require the addition of a transistor to be able to sink or source a minimum of 12mA.

How to implement hysteresis depends in how the inputs are connected. In your case, you just need to feedback a portion of the output to the appropriate input.

 

hi dl324

the 358 can src and sink current, it can src about 2x what it can sink, thus is why i did connections as i did. i am ref'ing at the + input and then allowing the - input to drift down as the batt depletes. when - reaches below the 6v zener the output should go high driving the buzzer and led.

do i have this wrong?

 

the 358 can src and sink current, it can src about 2x what it can sink, thus is why i did connections as i did.

Then why did you say this?

in a 8pin dip, any concerns about sinking 10-12mA ?

 

hi dl324

sorry, i 1st created the ckt sinking, then swapped inputs because the 358 can src about 2x more mA.

where would you tie hysteresis?

 

maybe some clamping feedback with a reset switch? the output goes hi and then clamps itself high until the batt gets onto a charger and person presses the reset alarm button? but then again, i dont really think i need the feedback to clamp it, it doesnt matter if it chirps on/off because that alone is the start of low batt indication, etc.

 

i am now up to this as far as schematic goes. you see anything faulty? the ckt will be wired to copper perf board about 1.3" square, it will sit in a pill box and i will pot it. finding tall enough jumper headers for potting was a no-go, but i found some pigtail style jumpers, so i will just pigtail out the jumpers and then pot the whole thing (less trim pot screw and piezo hole).

the ckt and J1/J3/J5/J7 (pill box) sit on top of a 12v 100Ah group 30H AGM batt. the led, switch, and Murata meter are mounted into the cover of a std plastic marine batt box. the jumpers allow me to remove the cover completely w/o having to undo any batt terminal bolts.

power is fed to the ckt, a PPTC resettable fuse protects ckt and meter. the switch is just a silence rocker. at alarm level voltage the 4kHz piezo buzzer will run at ~5-6mA, and the LED will be under 20mA when it flashes.

 

my final ckt, built, works great, low amps use. if anyone wants component #'s just ask, i'll list them. soldered to copper lands perf board and then potted.