Would you like a higher sustained lumen for your TS21? I have discovered a hack that allows me to sustain more than 1000 lumen for 15 minutes! The exact lumen number is unknown to me. At 15 min it was still noticeably brighter than an SP10 Pro with a fresh battery on Turbo. I don’t have a light meter or a way to measure current. However, I do have a timer and a charger. Using the supplied 21700, fully charged, I set the timer for 15 min and put the light on turbo. A full recharge took 1379 mAh. This would work out to an average of about 5.5A. Certainly enough to corroborate my eyeball observation that it was still brighter than an SP10 Pro turbo with a fresh battery. Certainly, based on predicted run time, way more than the review charts showing the light only sustaining a couple of hundred lumen.
What was the hack that caused the light to sustain such an output? It was a very simple hack. All that was required was an iron railing on my back step for the magnet in the tail cap to secure the light. Worked like a charm due to the fact of the railing being in Minneapolis in December;-) Happy Holidays everyone.
I made a magnet mount that screws onto my Astrolux MF01S. I can open the sunroof on the truck and stick it on the metal right in front of the sunroof. I use it out here in the sticks as a high beam when there is no traffic. With the wind flowing over it while I drive, it will stay as bright as it gets for quite some time. Usually the batteries start going low before there is any real noticeable difference in light output.
I won’t mod the Astrolux, this one is too nice but, I am thinking of putting together a proper voltage source, make a cord and mod a extra Q8 I have to accept my voltage source instead of the batteries. I would be able to run it my entire drive instead of having to swap out cells when they get too low. Granted I have been saying this for 2 years now and I haven’t even started. I am a master procrastinator.
I’ve always thought the disagreements about usability of “30-sec Stepdown pocket hand warmer lights” (eg the Emisar D4 etc) might just stem from the typical weather conditions of where people are located.
In the tropics, even at night, I found my D4 with 219a’s got too hot almost immediately, but the starting temp would have been about 20C…
In the UK, ambient temperatures of 15C or lower mean the light stays brighter for longer.
I wonder how low the temperature has to be to keep one of the high output pocket lights from overheating…
Christmas got cancelled here due to ice in Seattle and the weather is staying in the roughly –20C range so I decided to see if it could sustain 30 min. (Wind chills actually make it much colder than –20C but unfortunately are too variable to quantify.)
For this run I let the light sit out for a few minutes before starting. At 20 min I took my SP10 Pro out and the TS21 was still maintaining brightness on par with the fresh SP10 Pro on Turbo. Decided to set the SP10 on the ice and see what it would do for the final 10 minutes on turbo.
Both lights use Anduril 2 but got quite different results. Wonder if someone will understand the differences and be able to explain it to me? At the 30 Min mark the TS21 was still very bright. For some reason the SP10 was much dimmer. I had to kick the SP10 free from the ice it had melted into then frozen to. Once I kicked it loose and checked the temp it was flashing 26 — go figure. Why the output was noticeably low I have no clue. The TS21 flashed 34, had great output, and had been running for a full 30 min. The SP10 started at room temp while the TS21 started quite cold. Would that explain the difference? However, the light was well below throttling temp, shouldn’t it have ramped back up even if initially, due to the warm start, it did get warm enough to throttle?
Both batteries are now on the charger. Clearly the SP10 Pro did not maintain much for current output. The TS21 clearly did. Anyone want to make a guess as to how close to 5A the TS21 will average over a half hour?
I’d suspect that you’re looking at simple thermal performance with the cells themselves, compounded by current demands and the (relative) insulation of battery tube thickness, etc. 14500 cells aren’t quite the same animal as just about any 21700 on the market (China or major OEM), plus there’s so little in there to start with. Asking for turbo on a 14500 with Anduril puts a lot of stress on the cell, even with high drain cells like the H10 vs most other 3A cells, whereas 21700 handles that a whole lot better. The larger cell and thicker host metal probably adds appreciable thermal warming on the TS21 to aid performance a little. Temp sensor is just for the driver and it doesn’t know what’s happening upstream with the cell/battery tube.
Be sure to let the cells warm to room temperature and check that they’re dry (condensation) before charging…best practice, anyway.
It’d be interesting to see a side-by-side with two different hosts (thick/thin) using the same driver topology and same cells, same emiters, to see if that thermal insulation/dissipation effect matters a lot. There’s a point where the cell heats enough to help itself out, so to speak, in cold temps, but I think that matters more when the cold has really saturated the device over hours (like an e-bike in the garage or something). I think I did notice a little of this when using a Convoy L2 and the Astrolux FT03 together one night around 18°F, where the FT03 seemed to sag much earlier, but that wasn’t a great comparison really…did make me wonder about the thick tubing on the L2, though.
TS21 started about 5 min after being placed outside 21700 30 min on Turbo. 34 degree flash from light after turn off. Charge 2640 mAh. Average 4.9 Amp draw.
SP10 Pro warm from pocket start 10 min on turbo. Light placed on ice. Measured 26 degree after I had to kick it out of ice! Charge 272 mAh. Average 1.6 Amp draw.
Pretty large discrepancy in performance but both A2 which is what I thought was responsible for thermal control. Both batteries were low IR 20 Mohms. Obviously the mass of the batteries would be a factor. Hardware in the lights? Don’t really know what is going on here.(FWIW the SP10 had been very lightly used in the prior day so true draw would be even a touch lower.)
(Waited overnight to charge cells so the charge was done with batteries starting at roughly 15 degrees give or take a few. Certainly no condensation.)
I guess it’s not heat that caused the stepdown, but voltage sag from Turbo run instead. The small battery was underpowered to sustain it more than a few minutes even if the thermal limit had not been reached.
Contrast that to a 21700 battery, very big difference in capacity and discharge capability.
Yea I expect that was what correllux was trying to point out as well. The 14500 must just not be able to maintain the draw. That would explain the freezing in as well. Warm it was able to put out enough to melt but as the battery sagged it cooled down. FWIW, it is the cell that came with the light. Seems good quality to me with low IR and stated capacity.
