I have a circuit that's running off of a CR2032 (Digikey, P-189-ND). It runs an IR beam break (emitter and sensor) and a comparator that relays the beam break state via another IR LED (transmitter).
I empirically measured the current draw as ~15mA nominally and up to ~20mA when the transmitter LED is active (reasonable?). The battery is rated at 225mAh. So, I expected 10+ hours of constant transmitter illumination. In practice, the transmitter should only be active intermittently, for seconds or tens of seconds at a time. Therefore I really expected 15+ hours of on-time per battery. However, if I could even get 30 minutes of use out of it at this point, that would be acceptable, if not ideal. As it stands, even that doesn't seem to be feasible currently.
When I put the circuit into the entire system, my receiver was having trouble detecting the relay signal after a few minutes. I traced this back to "low" battery voltage (still >2.7V). I actually think that should be a sufficient voltage, but apparently not.
I measured the LED emission with a power meter (Thor Labs S121C) and the transmitter signal power drops at around 0.01mW/s (a fresh battery puts out about 1.7mW). This does not recover after interruptions in the transmission, it is linear and monotonic. I also measured a concurrent drop of about 1mV/s in the battery's output. When I bypass the battery with a bench power supply (set for 3V and 200mA maximum), the transmitter LED emission power remains steady.
Any ideas why the battery seems to be discharging faster than expected? Is this actually in line with the expected behavior? Were my expectations out of line? Is there any way to resolve this without a complete redesign?
I'm looking at other battery options, but the physical limitations for the height profile and weight don't leave much room for anything else.