LM311 circuit not working

Further to my two, other questions about Zener diodes to detect when an automotive battery is being charged by the alternator, I've developed the following:

Note: The Fritzing tool I used doesn't have a breadboard symbol for a 10-14v supply, so I've used a battery symbol to make clear which my positive and ground rails are.

The schematic looks like this:

My main question is, is this circuit correct?

If I understand things correctly, (unfortunately not the results I am seeing in the real world):

1. The Zener diode should break down at 12v, it doesn't - it's always letting through < 0.1v regardless of the input voltage in the range of 1-12v. Past 12v there's a marked increase in the throughput of the zener diode, consistent with what one would expect.
2. It's not immediately obvious from the data sheet whether I should wire my reference, or the variable source to the - or + of the LM311. I've done it the way below.
3. I'm told, as this is an open collector IC, I need a pull-up resistor to

I also have some questions:

1. How can I calculate how much power this circuit will use "at rest", i.e in order to do almost nothing until the alternator is running, without placing a demand on the battery?
2. Is this the only/best way to detect a running alternator?
3. Given that my reference is 1.26v from the rechargeable button cell, I will eventually need to connect something to recharge that when the system is running.

Lastly...

Where does one find a freelance electrical engineer who could design such a circuit to a list of constraints and produce a simple schematic/PDB that meet the requirements?

• Hello I am trying do find a solution for the same problem for my motorcycle. For now I created this i971.photobucket.com/albums/ae198/aniobash/bike_schema.png It works at home using stable 12V power supply but doesn't work on the bike. Im working on filtering the input. Mar 19 '14 at 6:27

A zener diode needs some current (a few mA) trough it to 'drop' its designed volatge.

If you want the + input to be the zener voltage below the positive rail, connect the resistor from the + input to ground.

If you want the + input to be the zener voltage above ground, connect the zener between + input and ground, and the resistor between + input and the positive rail.

To sense if current is flowing one direction or another you could use a resistor in the positive lead. This resistor could simply be a part of the cable carrying the current (very low impedance is good as less power is wasted).

Measure the voltage across this "resistor" with a part like the TI INA213A-Q1 to find if the current is going one way or the other.

You need to filter noise from the generator etc... and you need to be vary of things like the "load dump test" if you want the thing approved...

1) Don't worry. This circuit uses "nothing" compared to what's available on a car battery. No calculations needed. Verify with measurement.

2) Definitely not the only way. Not even sure it would work well - so not the best. What the "best way" is... not sure.

3) Not a question?

Lastly: You can find freelancers at elance to do your project. Maybe however you should find your local hackerspace and find someone there to help you.