1.5v - 100V input.

[Brandon1293]

Hi, I have a source voltage that I am wanting to sample and a steady supply for power available, 12v or 5v, doesn't matter. What I want to do is energize a relay when the sample is over 1.5v. Simple enough except that the sample can go as high as 100v for sustained periods of time. I already have 12v 5w zeners if that helps. I'm just not sure how to deal with the really high AND the really low voltages in one circuit.
Any help/insight would be greatly appreciated, Thanks!

Edit: Source sample is currently 60hz AC but that is easily rectifiable.

 

[Bluejets]

Rectify....resistor divide down to 10:1 .....feed into voltage comparitor with a reference voltage on the other leg set below 150mV.
Output to a transistor driven relay. Comparitor should be capable of handling over say 15v and driven off your aux supply.

 

[Brandon1293]

Thanks for the quick reply Bluejets. This solution sounds good to me, Thanks!
I'm thinking LM339N for the comparitor. I think that it can handle the high volts (10 or so) but is it sensitive enough for 100-150mV? Thanks.

 

[Alec_t]

The LM339N common mode input voltage includes ground, so 100mV would be no problem.

 

[GPG]

There is another way...
http://www.tradeofic.com/Circuit/11577-Norton_precision_comparator.html
No divider necessary

 

[cjdelphi]

I think everyone failed to notice "AC"

You can use capacitor and it's reactance to lower the voltage to a safer level, i just designed this circuit..

 

[cjdelphi]

For added safety an optoisolator could be used instead ....

 

[Brandon1293]

GPG, Good Idea, however I really don't demand that much precision for my application. Thanks though.
cjdelphi, I really like your idea, however I'm having a little trouble following how your circuit works?
Also, user safety isn't really an issue (all enclosed) unless you mean component safety? Is and optoisolator going have that large of an input range or do I have to fix that first?

Mainly though I'd like to keep it simple yet reliable.

 

[cjdelphi]

A 100v 400-600nf film capacitor will drop the voltage then rectify it via a diode, the other diode is there so the capacitor can discharge on each cycle...

I'm not sure it's a great technique because if the cap fails, the base of the transistor will see 100v ... hence the optoisolator which fry's the led, but the low 5v voltage side stays low voltage

 

[GPG]

GPG, Good Idea, however I really don't demand that much precision for my application.

Its not so much about the precision as the ability to accept input well above the amp supply.
An LM359 could be used http://www.ti.com/lit/ds/symlink/lm359.pdf
A little more setup reqd

 

[AnalogKid]

OR, since you don't care what the input voltage is above the 1.5 V threshold, bring it into the input of a comparator through a series resistor and have a shunt zener to GND. For example, have a 100 K series resistor and a 5.1 V zener. It acts as a zener regulator when the input is above 5.1 V, and a non-attenuator when the input is below 5.1 V. The series resistor is not critical. It is there to limit the peak zener current. However, it should be much less han the input impedance of the comparator. You will have much better noise performance compared to attenuating everything down by 90%.

ak