Why do my NiMH rechargeable AAs die so fast in my smoke detectors?

The issue you're seeing with the detectors is likely the load of the device, not just the battery chemistry. Smoke detectors are designed to run for a decade on an alkaline battery, which is optimized for a high initial current draw and a steady 1.5-volt delivery over a long period without leakage. NiMH batteries, even high-quality Eneloop style, have a lower internal resistance and are optimized for devices that draw current over weeks or months, like a remote control or a digital clock.

Detectors are essentially drawing power in bursts or maintaining a constant low voltage, but the way they are rated for life assumes a traditional cell performance. NiMH cells genrally have a lower available voltage when fully discharged and they have a diferent internal resistance profile than alkaline cells. This difference means that while they last forever in low-draw electronics like a BT amp, the continuous or pulsed load of a sensor will drain them much faster until they hit the specific low-voltage threshold that triggers the chirp.

Switching to NiMH for these types of sensors is always a risk because you're forcing a battery meant for low-drain applications into a high-demand application. I had similar issues with a few older carbon zinc packs in my smoke detectors, and even those seemed to struggle when the unit started chirping early. The alkaline cells were the gold standard for that specific function for a reason.

If you really want to try to keep NiMH, you might need to look for specific devices rated for rechargeables, but for standard detectors, the initial cost of the alkalines is generally lower than the hassle of swapping them out monthly.

I've tried this switch myself with a digital clock and it was a little surprising. Mine worked fine, but it really makes you wonder about the longevity of the cheap rechargeable packs versus the higher end versions like those Eneloops. I guess even if one works for a while, you're still trading a known lifecycle for an unknown one. Do you think the issue with the detectors is mostly just the high burst current from the sensors, or is it something about the voltage sag once they start chirping? It's a bit of a headache figuring that out.

I totally agree with the assessment that it comes down to the device’s specific power draw, not just the battery. It’s definately about how those sensors are calibrated to recognize a voltage drop. I had the exact same expirience when I tried this with a digital thermostat sensor I had installed last winter. It ran fine at first, but about two months later, the display started flickering because the NiMH output didn't match the specific voltage curve the unit was expecting til it acutally died.

I used these batteries in a motion sensor near the garage first. They lasted longer than the small clocks, but then the sensor just died completley instead of giving a warning sound. It feels like the way the power is pulled out of the battery is the issue, not just the amount of power. I wonder if using a slightly larger battery size for these things, like AA instead of AAA, woud help keep the power stable? I don't want to mess up the device, but I want to keep trying to use recharges.

I had a similar thought when I started testing my weather station outside. I was running it on those high-end rechargeable sets since I figured it would be a hassle to keep replacing them, but it started reporting completely frozen temperatures even when the temps were mild. It makes me wonder if the sensors just pull too much power at certain points. I've been looking into wheather some of these sensors require a specific voltage curve, since even if the battery holds a good charge, the way the device samples the data might be stressing the cell. Do you know if there are any recommendations for specific devices that explicitly state they are optimized for NiMH usage?

Update: finally got it sorted. The point about the detectors being calibrated around the alkaline voltage curve was the answer. I put one alkaline back in the chirpiest detector and it's been running normal for two months now, no false low battery warnings. Clocks still have the rechargeables and are doing fine, so I'll just leave those be and keep alkalines for the detectors.