# Unity Gain Inverter or Op-Amp Comparator?

Discussion in 'General Electronics Chat' started by tempest411, Dec 29, 2012.

1. ### tempest411 Thread Starter New Member

Dec 29, 2012
6
1
Hello,

For my first post I would like to ask a question about something I'd like to make that will have one input and two points at the output which will vary in potential based on the input. The circuit should function such that at 13.7~14.2 volts at the input there is no output, at ~14.2+V at the input a bias is seen at the output, and then at input voltages below ~13.7V an equal, BUT OPPOSITE bias is seen (current traveling in the reverse direction).

What I am trying to do is develop a circuit which will control an ammeter via voltage level. The as-is configuration is a standard shunted design, which is horribly inaccurate as the ammeter only sees <1% of the current. If I could reference it to voltage it would work more reliably I am sure. I think some kind of op-amp comparator or a unity gain design might work somehow...

Rick

2. ### Ron H AAC Fanatic!

Apr 14, 2005
7,018
682
You don't say what the value of 'bias' (voltage or current?) is.
By "no output", do you mean zero volts, or open circuit?
Also, what supply voltage(s) are you wanting to use?

3. ### #12 Expert

Nov 30, 2010
18,076
9,678
It's called a window comparator. You set up 2 opamps with a supply voltage that has both a more positive voltage than ground and a more negative voltage than ground, make a reference voltage with a zener diode (or something better) and use 3 resistors to make a percentage of the real voltage numbers you want, like 6.95 volts and 7.10 volts. Attach those voltages, one to each opamp, and let the other input of each opamp feel the voltage you are detecting (divided by 2).

Wiring it is the next problem. You want the upper voltage opamp to say nothing until the detected voltage is higher than its reference input...but what is, "nothing" in this case? A low output voltage? I think you decided that. When it detects a high input voltage you can have it switch on hard or put out a proportional voltage. I think you wanted it gradual.

Then we have the lower limit opamp feeling a voltage that is usually higher than its reference voltage and putting out, what? A high voltage? When the detected voltage gets below the lower limit, the lower opamp switches to a low output. Again, output a gradually lower voltage? So then you have a dead zone, a positive increasing output, and an output that gradually goes lower.

All the while, I think you are wrong about your accuracy belief. If your ampmeter sucks, it isn't because it senses 1% of the total current, it's because it senses the 1% badly. My speciality is precision analog and I have no problem with comparisons of a thousand to one. My first whack would be to try to fix the original design so it works right. Only if I find the crud I have to work with doesn't stand a chance, then I scrap the inaccurate parts and figure a different way.

Spill the beans. This might be easier than you think.