Thermal Performance Tests on Popular Light Models

LED Flashlights – General Info
1

In response to a post by Old-Lumens where he did a tear-down of the Nitecore EA8, I did some less destructive testing of my own on some lights. The responses were such that I felt the need to expand the testing to more lights. So here it is, my Thermal Performance test thread. In the original thread I compared the EA4W to a heavily modded HD2010. First off, those lights are not in the same class so I moved the HD2010 to a separate graph (since I have not tested another big light, I left the EA4W in there).

Power is supplied with a bench power supply. Voltage is such that fully regulated output is obtained for the entire test, but not such that excessive driver heat would occur. Supply voltage is indicated.

Small single XM-L lights

Nitecore EA4W (U2, 5.5V, Turbo), Convoy M1 (2.8A T6 3C, 3.8V, High), and Convoy C8 (2.8A U2 1A, 3.8V, High)

Time EA4W
Temperature (Deg C)		 EA4W
Output (Lumens)		 EA4W
Current (A)		 Convoy M1
Temperature (Deg C)		 Convoy M1
Output (Lumens)		 Convoy M1
Current (A)		 Convoy C8
Temperature (Deg C)		 Convoy C8
Output (Lumens)		 Convoy C8
Current (A)
0:00 24.0 772 2.21 23.8 673 2.79 23.8 718 2.8
0:30 27.0 685 2.19 26.8 616 2.80 26.6 682 2.8
1:00 28.8 677 2.19 29.8 604 2.80 28.8 678 2.81
1:30 30.6 671 2.19 31.8 603 2.81 30.6 674 2.81
2:00 32.2 667 2.19 34.4 597 2.81 32.4 669 2.81
2:30 33.6 664 2.18 36.2 590 2.81 34.0 666 2.81
3:00 36.0 659 2.18 38.0 586 2.81 35.2 663 2.82
3:30 37.2 657 2.18 40.2 584 2.81 36.8 658 2.82
4:00 38.0 652 2.18 41.8 579 2.81 38.0 656 2.82
4:30 39.2 649 2.18 43.4 577 2.81 39.6 652 2.82
5:00 40.6 644 2.18 44.6 575 2.81 40.4 651 2.82
5:30 41.8 642 2.18 46.0 572 2.81 41.6 648 2.82
6:00 42.6 639 2.18 47.2 570 2.81 42.4 645 2.82
6:30 43.4 635 2.18 48.6 566 2.81 43.6 643 2.82
7:00 45.0 633 2.18 49.4 564 2.81 44.6 640 2.82
7:30 45.6 630 2.18 50.0 563 2.81 45.4 637 2.82
8:00 46.4 626 2.18 50.8 562 2.81 46.2 636 2.82
8:30 47.6 624 2.18 51.6 560 2.81 46.8 635 2.82
9:00 48.6 622 2.18 52.6 558 2.81 47.4 634 2.82
9:30 49.2 621 2.18 53.4 557 2.81 48.0 634 2.82
10:00 49.8 618 2.18 54.0 556 2.81 48.8 632 2.82

And here is the graph, current is not plotted as it is virtually flat.

The EA4W has a significant initial lumen sag during the first 30 seconds (87lm / 11.3%). In fact, the majority of this occurs in the first 7-10 seconds. I believe this is due to poor thermal contact between the emitter - star - heat sink - body. The emitter - star junction would contribute to the immediate drop (within the first second), and the other junctions for the rest. Once heat starts reaching the body, things begin to ramp at a reasonable level for a light of this thermal mass. I suspect that the emitter is running quite a bit hotter than the body. Overall drop after three minutes and 10 minutes was 113lm / 14.7% and 154lm / 20%, respectively.

The Convoy M1 has a significant initial lumen sag during the first 30 seconds (57lm / 8.5%). In this case, the majority occurs within the first 15 seconds. Intend to open this light up to look for potential issues. If anything is found, I'll note it and retest. Based on the results, I suspect a poor star-pill contact. Once heat reaches the body thing ramp as expects for a light with this small thermal mass. Overall drop after three and 10 minutes was 87lm / 13.0% and 117lm / 17.4%, respectively.

The Convoy C8 results are very impressive. It exhibited a comparatively low initial thermal sage during the first 30 seconds (36lm, 5.0%). They are very close to what I measured from a light using a copper SinkPad. I should note that with respect to power input, the C8 and M1 are identical. The thermal mass of the C8 is greater, which shows in the latter portion of the graph (final temperature is lower). Overall drop after three and 10 minutes was 55lm / 7.7% and 86lm / 12.0%, respectively.

Medium Single XM-L lights

TangsFire HD2010 (XM-L2 T6 1C, Cu SinkPad, 4.10V), Nitecore EA4W (for reference, same as above)

Time EA4W
Temperature (Deg C)		 EA4W
Output (Lumens)		 EA4W
Current (A)		 HD2010
Temperature (Deg C)		 HD2010
Output (Lumens)		 HD2010
Current (A)
0:00 24.0 772 2.21 23.8 1146 4.03
0:30 27.0 685 2.19 25.8 1104 4.03
1:00 28.8 677 2.19 27.4 1099 4.03
1:30 30.6 671 2.19 29.2 1092 4.03
2:00 32.2 667 2.19 30.8 1087 4.03
2:30 33.6 664 2.18 31.6 1082 4.03
3:00 36.0 659 2.18 32.8 1080 4.03
3:30 37.2 657 2.18 34.0 1076 4.03
4:00 38.0 652 2.18 35.4 1074 4.03
4:30 39.2 649 2.18 36.4 1071 4.03
5:00 40.6 644 2.18 37.2 1069 4.03
5:30 41.8 642 2.18 38.4 1066 4.03
6:00 42.6 639 2.18 39.2 1063 4.03
6:30 43.4 635 2.18 40.0 1060 4.03
7:00 45.0 633 2.18 40.6 1057 4.03
7:30 45.6 630 2.18 41.4 1055 4.03
8:00 46.4 626 2.18 42.6 1052 4.03
8:30 47.6 624 2.18 43.0 1050 4.03
9:00 48.6 622 2.18 43.6 1048 4.03
9:30 49.2 621 2.18 44.2 1046 4.03
10:00 49.8 618 2.18 44.8 1043 4.03

Graph

The HD2010 performs quite well. Even with the relatively high power input, initial sag is kept low (41lm / 3.6%). The SinkPad is doing it's job to quickly get heat away from the emitter. The star-pill junction is also optimized by lapping and the use of silver thermal compound. Beyond that, the pill-body contact area is significantly large (and has compound as well) so it is not a thermal bottleneck. Overall drop after three and 10 minutes was 55lm / 7.7% and 86lm / 12%, respectively.

More to come, including some bigger lights.

2

Just in case…

3

I am surprised about the Convoy M1, my biggest thought is that mine has a blob of thermal compound between the star and pill, if that were removed, the surfaces lapped and the bare minimum of compound used (as in a few mg), it may take care of the problem?
I would like to try the Convoy S2 in your lumen box

4

I think I found the problem with the my M1… retesting…
And you are not putting any thing in my box! :stuck_out_tongue:

5

Nice work, thanks for the info.

6

Hi,

I’m still relatively new, but isn’t the kind of general consensus that the C8s are pretty good thermally?

Is the Convoy C8 different from other C8s?

I’d really be interested in seeing the same data for a “generic”, el-cheapo C8!

Thanks for doing all of this!

Jim

7

sorry about poor quality

8

Nope, same results. I think I improved one thing a little, but there’s still another bottleneck somewhere.
What I found was one of the screws holding the star down was stripped. I have not fully addressed this yet, but I did improve it some. I reassembled and tested it quickly with virtually identical results after three minutes. I need to sleep on this and get back to the M1 some other time.

9

Non related question, at what voltage does the C8 start going under 2.8A

10

C8: drops out of regulation around 3.55V, hits 2.0A around 3.40V
M1: drops out of regulation around 3.50V, hits 2.0A around 3.30V

11

So it's tested on cells right?

12

Thanks so much, i’ve been wanting that information ever since i bought my first XM-L light :beer:

13

Tested on a power supply which indicates voltage and current. I recorded the voltage at which the supply current begins to drop below the regulated value. For these lights, that’s right close to 2.8A.

14

its no wonder you can’t get full power for long with the XM-L2
i assume your numbers are about average for a T6 and U2 with 105C driver

15

On my QL355 I tested the KD V2. I measured the output with a clamp meter. At 3.40V input voltage there is still an 3.01A Input current, At 3.30V there's an input current of 2.97A.

16

relic38. You are doing some fantastic work lately. This is pretty orsm info you are giving us all. Thanks.

17

Nice work. Good to see how they do over the 10 min.

18

Great work! These results are always interesting to watch so thanks for taking the time! Highly appreciated!

If you have a P60@2,8A it would be interesting to see that in comparison with the other “small single XM-L lights”. Maybe “foil” wrapped and not foil wrapped… I think many would like to see that. I would! :slight_smile:

19

Same here. Thanks relic!

20

I’d be interested in this also, and as I said earlier, maybe a generic C8 (vs. the Convoy C8)?