Comparator - Battery Cutoff

[Xenobius]

Hi all,

I have a simple requirement and I had a go at it using a simulator and it worked. I thought I understood what was going on so I built it and .... nothing worked. I would like to share it with you so that you may help me understand better what I did wrong.

I am using an LM258, power directly from battery which may be anywhere from 3.7v down to 3.1v.
What I need is to output low when the voltage input reaches anywhere between 3.1v to 3.3v.

The diode in the circuit is a 3.3v zener diode.

Thanks
X

 

[kellys_eye]

R6 is too high a value. The zener requires a minimum current to reach the zener voltage. 100k mean only 33μA flowing..... get it to the mA range.

 

[Alec_t]

If you read the datasheet for the LM258 you will see that the common mode input range is from 0V to Vcc-1.5V. If your supply voltage (Vcc) is 3.7V or less, then the voltages at the inverting and non-inverting inputs must be 3.7V-1.5V = 2.2V or less for the opamp to work correctly. You need a voltage reference less than 2.2V when Vcc is 3.7V, but if Vcc drops to 3V then the reference would have to be less than 1.5V.

 

[dorke]

I agree with the above notes.

More than that,
I would say there is no need to reinvent the wheel...

Here is 3 pin under-voltage detector that will do what you need in every way better:
smaller,cheaper,more accurate,less power and way simpler.

There are many others,
Enjoy

 

[Xenobius]

R6 is too high a value. The zener requires a minimum current to reach the zener voltage. 100k mean only 33μA flowing..... get it to the mA range.

I took it down to 10k but still same thing

 

[Xenobius]

I agree with the above notes.

More than that,
I would say there is no need to reinvent the wheel...

Here is 3 pin under-voltage detector that will do what you need in every way better:
smaller,cheaper,more accurate,less power and way simpler.

There are many others,
Enjoy

Oh dear ... if only I knew before... this is what I need. Only the boards are ready now
I was trying to order some of these, specifically the 3.1v but I cannot find it much. Is this brand new from ST?

 

[dorke]

It isn't new,the datasheet is from 2006.
The design you have is faulty.

You can't expect to have 3.1V on the zener when you feed it 3.1V through a 10k resistor.
lets' say you feed it with 3.15 V,the current would be (3.15-3.1)/10k=5μA,that is way too low.

You should use a lower voltage zener, say 2.5V, and bias it properly(a few mA),
and scale down the "sampled voltage" on the other leg of the op-amp.

Another thing ,
you should use a rail-to-rail(at least the upper rail) op.amp that can work down to 3V (or a bit lower if possible) .

 

[Xenobius]

It isn't new,the datasheet is from 2006.
The design you have is faulty.

You can't expect to have 3.1V on the zener when you feed it 3.1V through a 10k resistor.
lets' say you feed it with 3.15 V,the current would be (3.15-3.1)/10k=5μA,that is way too low.

You should use a lower voltage zener, say 2.5V, and bias it properly(a few mA),
and scale down the "sampled voltage" on the other leg of the op-amp.

Another thing ,
you should use a rail-to-rail(at least the upper rail) op.amp that can work down to 3V (or a bit lower if possible) .

Yes I agree that it is faulty I just want to learn and when you recommend rail-to-rail I am still a bit lost. I will do some more research.

 

[Audioguru]

The 1N4728A zener diode is 3.3V only when it has a whopping 76mA(!) of current in it that will quickly kill the battery. There are many 3.3V zener diodes available that draw only 5mA. This is a horrible faulty circuit.

On another thread or another website there is a 3 pins IC that draws a very low current and gives an output when a battery voltage is low.

 

[Xenobius]

The 1N4728A zener diode is 3.3V only when it has a whopping 76mA(!) of current in it that will quickly kill the battery. There are many 3.3V zener diodes available that draw only 5mA. This is a horrible faulty circuit.

On another thread or another website there is a 3 pins IC that draws a very low current and gives an output when a battery voltage is low.

Yeah horrible indeed I used 100k pull ups just to lower the current consumption as much as possible only to put a zener like that. I am at the moment removing this part of the circuit and finding an alternative ready made solution.

This was going to drive a low power voltage regulator which has a power on and off feature. When LOW it shuts down power and this is why I need a LOW from this circuit.

Thanks everyone for your valuable input.