Happy new year as well.
The Fluke 17B+ and latest version of the free Harbor Freight meter appear to have the longest battery life based on my Wiki battery data.
You won't find a lot of cheap meters, because they were recycled after they could not be repaired, or in the case of the UT61E and ZT-102, they were modified to the point where I expect it would corrupt the data. A bit tired of dealing with the 121GW prototype and left it off as a result.
I used to get a lot of negative comments about that 17B+. Seeing it do so well in my transient tests, switch life cycle testing and now what appears to be a very long battery life, if you can live with the features it has, it appears to be a pretty decent meter.
There is a previous link to a thread showing some of the test jig. The BM869s was one of the meters I used during my testing. I mentioned a few details about the cutoff voltage. The software is currently looking for the current draw to be less than half of the nominal. In the case of the BM869s, a low battery alarm was set long before this.
You can see how they behave as the voltage is ramped down. The cutoff is not always a nice sharp edge.
If anyone has some ideas on how they would like to see the data measured or presented, feel free to make your suggestions.
**********
In the last plot, I have added a cursor readout. The Brymen BM869s is selected and the cursor is placed where the alarm is just ready to sound. We can see the voltage is 5.7, much higher than the 2.8 volts shown in the table for the half nominal. The meter is still running down at 2.8 but as I mentioned, the results are not predictable. I think as long is it is clearly stated how the measurement is taken, it should be fine.. Then again...