I'm looking for some guidance, to tell me if I'm headed in the right general direction, or if there's an obvious better way that I'm just not considering.

My goal is to build a single axis inclinometer with 0.1-degree precision, that can tell me if I'm perfectly level, between 0-0.9 degree, 1-1.9 degree, or 2+ degrees off of level, in either direction. The output method will be 7 small LED's, one for each range. It has to fit within about 1 cu-inch and run for 20+ hours on a pair of coin cells (240mAh capacity) when level.

I ruled out a micro controller for efficiency. This Inclinometer (muRata SCA61T, click for data sheet) provides a ratiometric analog output. My thought was to use 6 voltage comparators (like this one, 2 per package) to tell me which range the output voltage falls into, using voltage dividers and 25 turn trimmer pots to calibrate the inputs for the comparators. Then use the comparator outputs to drive the small LED's. The LED's draw 2mA/ea, the inclinometer 2.5mA and the comparators 0.45mA/ea.

Does this sound reasonable, or am I headed in the wrong direction?

 

Does this sound reasonable, or am I headed in the wrong direction?

Sounds very reasonable to me. If you want to maximize your battery life, there are comparators that draw less supply current, such as the LMC6772 (among many others).

 

Thank you!

 

You're welcome.

 

You might consider a bar-graph circuit such as the LM3914 which has 10 comparators.

 

Thanks I will check that out!

Another question, maybe worthy of another topic.. but what type of circuit would I add on if I wanted to be able to put this analog circuit to sleep, and wake it with a button push, for the purpose of extending battery life? Maybe charge a cap that activates a n-channel MOSFET to power the entire circuit, with a big resistor connecting the gate to ground for the delay?