Unity gain amp

jakekooser · Post by **jakekooser** » Thu Jan 22, 2015 1:38 am

I did some more research on how to connect RCP to my existing fuel level and oil pressure gauges without affecting the readings on the original gauges, and built myself a dual unity gain amp to isolate them. Here's the schematic attached. I started with the less-critical fuel gauge, which is a little below 1/2 tank, and I read 3.6 volts at the output of my amp going into the RCP. If I put 12V onto the gauge input, it pegs at "E" and the voltage at the RCP goes up to 5V. If I ground the input, it pegs at "F" and the RCP reads 0 volts. I'm a bit confused about how that happens, unless I screwed up my inverting / non-inverting inputs to the op-amp (quite possible) but shouldn't I be able to calibrate the RCP to show max level at zero volts?

toga94m · Post by **toga94m** » Thu Jan 22, 2015 1:46 am

If your circuit is really as shown, it should be up=up.. what's the power supply to pin 8? It should be vehicle +12V, rather than the 5V ref from RCP. Even though you're only feeding a 0-5V signal to RCP, you'll need about 7V or higher supply to get 5V from the output - opamps need a little headroom to work properly.

The two resistors on the front end make a divider, so the output will be (4.7/(4.7+6.

) = 40.8% of the input signal. For a real 12.00V, this will give you +4.90V into the opamp buffer and then to the RCP, which is perfect.

jakekooser · Post by **jakekooser** » Thu Jan 22, 2015 1:54 am

That's the schematic right from Eagle, as I drew it. I made the board - after I tried it once with perfboard and thought something was shorting out because I was only seeing about 1/2 of the voltage at the RCP that I was expecting to see (sorry, don't have numbers in front of me right now). I was convinced the perfboard model was somehow shorting out because of some sketchy jumpers but the nice board is doing the same thing.

It occurred to me as I was typing the first post that my scaling is wrong, the top row is as follows:

5 | 3.75 | 2.5 | 1.25 | 0 but the bottom row is
0 | 1.25 | 2.5 | 3.75 | 5

I think I realized that the bottom of the chart should be more like 0 | 25 | 50 | 75 |100 to give me the display I want.

Either way, got some feedback on the circuit and I'm offering it up to anyone that wants it. I used the dual op-amp chips because that is what I had laying around the house. I only had to buy the zener diodes.

I am feeding the opamp with +12v from the back of the gauge, since it's right there.

toga94m · Post by **toga94m** » Thu Jan 22, 2015 2:48 am

Assuming you've used a meter and checked voltages right across the 4.7k resistors, which should be scaled to the 41% of input signal... and from there, the opamps should be buffering those voltages 1:1 without complaining. Also assuming you've checked power supply voltage right at pins 4/8 on the chip - that's one of the first things we check when a new board doesn't work... supplies.
https://www.youtube.com/watch?v=RB2GboGOuTI