I think it would be better to just update the OP with the new information. Keeps it more clean so we don't have to scroll down to get the information. Thanks for your hard work Don.
Please note the various qualifications and caveats in the first post. There are likely errors in the method and measurement, not to mention calibration but the lux numbers for comparison are reliable for comparison with other lights measured on the same lightbox as the method is consistent.
Light
Battery
Lightbox H
Lightbox M
Lightbox L
Estimated lumens on high
Fenix E01
AAA
37
8
MDXL "3W"
NiMH
77
17
TiKey
3xLR44
81
18
BLF DeLight
NiMH
85
19
Powerlight
NiMH
94
21
Aurora SH0031
Alk. AAA
142
32
Ultra-budget C3 clone
2xNiMH
163
36
Akoray K-106
NiMH
222
49
Golston "7W"
2xRCR123
390
87
Trustfire XP-E F23
AAA NiMH
396
33
88
Smallsun ZY-C85 full zoom
14500
456
101
Uniquefire S10
NiMH
575
128
Trustfire F22
NiMH
586
249
55
130
Romisen RC-C6
RCR123
618
166
137
Aurora SH-035
NiMH
644
201
26
143
Hugsby P32
NiMH
650
189
144
Trustfire R5-A3
NiMH
650
360
144
Smallsun ZY-C85
14500
654
145
Uniquefire AA-S1
NiMH
665
148
Ultrafire C3 SS Q5
NiMH
718
249
49
160
Aurora SH-033
RCR2
744
221
33
165
Solarforce L2R Solarforce low voltage R2
2xNiMH
779
173
Aurora SH-035
14500
785
250
36
174
Trustfire XP-E F23
10440
850
82
189
MX Power ML-310
14500
857
220
190
Trustfire Z1
RCR123
889
82
198
Ultrafire WF-502B R2 single mode
18650
940
209
a first batch Aurora SH-034
RCR123
1016
304
39
226
Akoray K-106
14500
1077
239
Ultrafire WF-504B with the 5 mode R2
18650
1102
449
168
245
Uniquefire S10
14500
1112
247
Trustfire F22
14500
1210
535
115
269
Trustfire F22
14500
1254
563
122
279
Ultrafire RL-118
RCR123
1260
145
280
Ultrafire C3 SS Q5
14500
1391
620
128
309
ROP Low
6D NiCd
1401
311
Uniquefire AA-S1
14500
1445
321
Uniquefire R5 5 mode XP-G R5
18650
1565
164
84
348
Trustfire R5-A3
14500
1600
576
356
Ultrafire WF-503B DX MC-E
18650
1636
675
364
Piritlight SG-L8 MC-E
18650
1735
665
237
386
Solarforce L2M DX P7
18650
1956
199
435
Ultrafire Wf-504B KD MC-E
18650
2310
1575
61
513
Ultrafire WF-501B M bin MC-E
18650
3220
1237
165
716
ROP High
2x32600
3250
722
Varapower prototype
4C NiMH
8290
1842
Edit: By calculation, the R5-A3 should be no more than 280 lumens which is what I think it should be. From the DX spec it is driving the LED at 1300mA. The R5 bin is 463-493 lumens at 1500mA. So, by calculation 13/15 x 493 = 427 lumens though it'll be less than that as the output is not linear with current. The conventional reckoning is that 70% of the LED's output makes it out the front so at best the output is 280-299 lumens rather than the 356 lumens my lightbox/meter/calculations come up with.
Makes sense. I'll see if anyone wants any more measurements - if I've got it I'm happy to measure it - the measuring just takes a few seconds to do.
Once that's done I'll tidy up the first post, add comments and qualifications and put in the best data I can get. I've had to work every day for the last month or so, but a colleague has kindly offered to cover this weekend for me so if it works out, I get my first weekend off since June. Then I'll do the tidying. If not it'll get done shortly.
Very nice Don! This is an excellent resource. It's on the frontpage and sticky'd.
Quick suggestion: If you select all the cells in the table and click on cell properties you can set the border color to #000000. It would look like this:
I'll certainly clean it up - at the moment it is just copied and pasted from a spreadsheet - I hope I can make a much better looking table once I've got all the data together.
The biggie for me is the errors - I got bitten hard by that once in a lab job. If my results had been correct, someone was going to jail - it turned out my results depended only on how long the spectrophotometer had been switched on... :(
I plan to go through the light collection and get numbers for most of them. Since I'm doing instantaneous peak readings they only take a few seconds each - it takes longer to make sure I'm using fully charged cells in everything than it does to take the measurements.
Don't you think it would be more useful to take the reading after at least 30 seconds or so - not going for the highest output, but real-life performance?
Great work, Don! You are a real sport to take everyone's input and re-do the tests. It is kind of neat to see how some of them sag after a couple of minutes. I agree some of the lumens readings seem high, but it's still a great comparison of one light vs. another and, man, you have a lot of lights!
If I were to do all of them, it'd take quite a long time. What would be fun would be a way of collating all the information in one place - probably need a lot of hyperlinking to make it useful or the result would be a monstrous mess. I think I have around 100 lights now - will try to get them all in one place and count them some time soon.