It had just been runnning on medium for seven hours so the pill was slightly warm before the start of the test. IT wasn't any warmer at the end of the test. But then it wasn't any brighter than it was with an 18650 anyway.
Here's a graph of the output for the first minute.
Thanks for all the hard work Don. I really do enjoy these light, as do all the others I have given them to. Do you have a data logging light meter to record your findings or do you do it the old fashioned way? My hat is off to you if you have to hang out all day with a stopwatch, pad and pencil. ;)
Is there a thread where you explain your reporting process? Fun stuff!
I don't think there is a thread on it. I'll do one soon. It certainly isn't manual - the lux measurements are, the output over time ones certainly aren't. I actually have two cheap datalogging meters though I need to use two PCs to get both of them running - the software that came with them doesn't seem to let me use more than one meter at a time. I run two copies of it, both of them see the same meter. Annoying...
Hey Don, when doing these kinds of tests, do you turn them off after a while because of the heat produced? Do you leave them on during the whole runtime test test period? Just curious. Thanks.
I don't expect it to still be giving light in the morning - 8 hours from now when I have to go to work again. But maybe it will.
And it is an unprotected 18650 so i hope I am not killing a cell.
However a 16 lumen (25.5 hours) thrower is not the most useful light in the world - a flood light would be much more useful at that light level. I am wondering how long it will stay lit for though - I will keep it running till it goes out. I thought it would give up about 10 hours ago.
I've not run an instrumented temperature measurement, but I can. I keep a fan or two pointed at any light running on high or medium but once the cell finally gives out - 26.5 hours and it is still at 3.69V on low I'll do a temperature test. Problem is that Windows chokes on trying to run more than one meter at once. I'll dig out the Linux laptop which can handle more than one meter at a time and log temperature against output once I've got the software working properly. This will not be before next weekend as I have two machines to build and set up immediately. I hate building and setting up Windows boxes!
And given that I have had to work at the job that pays my mortgage every day for the last 49, I don't get a lot of free time to do the stuff I want to do.