What did you mod today?

I prefer the S2 triple over the S2+ and the S2+ over the E2L. Simply because of the thermal mass. S2 Triple uses a good hunk of Copper.
Hard part for me to understand on my single Luxeon V S2+ is that the outside of the light does not feel hot enough to justify the thermal shutdown on highest possible settings. I have the Fet on driver and the thermal spacer installed and I am starting to wonder if its the driver getting too much heat from the emitter. I should try to remove the thermal driver spacer.

^ I hope you do not confuse mass with thermal mass. Per volume, copper and brass can store more heat, but per weight aluminium is better. So the mass of a flashlight can give a good feeling, but lightweight aluminium can store a surprising amount of heat too (I do not have the numbers at hand, I believe it was: copper has 2.5 times the weight of aluminium, but just 1.4 times the heat capacity).

That said, regardless of which metal is used, an S2+ triple with a spacer to compensate the short optic just has a lot more metal under the leds, and thus takes more time to heat up than a single led S2+ without extra spacer.

For those who having problem with setting custom OTP threshold, I just updated LD-x4 datasheet, you can download it from my page: LD-x4 datasheet

Since someone already asked me about OTP setting on LD-x4 driver over mail, I will just copy my answer here, which explains difference between LD-3 and LD-x4 OTP setting:

"But I will describe it here for now:
If you want to change OTP threshold, you should turn on the
flashlight on highest mode, wait until is around temperature that you
want to be new threshold,
and then enter into config. settings - next select EXT. OTP menu and
choose "CUSTOM" sub-menu. After that driver will set mode to max. and
flashlight would get hotter. When temperature is what you want to be
new threshold - fully turn the flashlight off. On the next turn on new
temp. threshold value will be stored.

The difference between LD-3 OTP setting and LD-x4 setting is that you
must "pre-heat" flashlight outside config. setting on LD-x4 drivers
and then set final value of OTP in config. setting; on LD-3 drivers it
was possible to set OTP directly in config. setting. Reason for that
is that LD-x4 has 30sec "no-press" turn-off timer while in config.
settings as safety feature,if you don't press switch for 30sec, driver
would shut down without storing anything, so if you don't "preheat"
flashlight in normal mode, there is a good chance , your flashlight
would not get hot enough during those 30sec.

If you have problem with setting desired OTP level, please try to
follow upper procedure:
"Pre-heat" light in normal mode -> enter into config. settings -> go
to EXT. OTP menu ->CUSTOM sub.menu ->driver would go to max. mode ->
wait until temp. is hot enough (should be in <30sec!)-> turn the
flashlight off ->turn the flashlight ON ->DONE "

Today I built up a Jaxman E2L host, using a BLF-A6 (FET+7135) driver and 4000K Luxeon V leds on the new led4power 4040 triple board. In fact, I did this build to check some real world performance of this new board type.

Special about the new boards that he sells now is that they are not DTP, so none of the solder pads are electrically connected to the core, but the dielectric layer is so thin (it is just an anodisation layer on the aluminium core) that the board should perform just as well as a DTP board.

I won a number of these boards in a small quiz that l4p held to promote them, and I happened to be online () just as he posted it so I was the first with a correct answer. And I even received some bonus boards, thanks!

The 4040 triple board is exactly 20mm diameter, as it should be, but the E2L host annoyingssely just allows 19.5mm, so it needed some slight shaving and that can only by just be done while still clearing the circuitry on top. But I do not know how I would have done that without my disc sander. The shaved board:

Reflowing the leds was no problem, and since I'm not using the FET-option on the ledboard, the ledminus wire was soldered on the FET-drain pad.

The leds fit tight but nicely under the Carclo optic.

Both springs were bypassed with 20AWG wire, the rather long tail spring replaced with a shorter one to make a 18350 cell fit less tight in shorty form. I replaced the switch because I melted the stock one during the spring soldering. The BLF-A6 driver received the 'voltage spike-fix' (not really needed for this driver, an OCD-thingy), a 680 Ohm bleeder resistor and 20AWG ledwires. A lighted tail is planned at a later moment.

The only real problem was soldering the ledwires on the ledboard. It was immediately clear that this dielectric layer is something special because even though the pads and wires were pre-tinned, even my super-hot 80W solder iron (admittedly with old rotten tip) struggled to get the solder hot enough to join, the heat was just sucked away to the core: I had multiple cold solder joints (the wires just sprung loose when the light was on max for a few seconds) before I finally got good connections. I ordered a new tip for the solder iron to get better work done next time.

So the low thermal resistance dielectric layer which is the key quality of these boards, has a nasty side-effect: not just under the led but also under the solder pads the heat is sucked away to the core with record speed.

So I got the light done and did some measurements. I also measured my modded Emisar D4 with Luxeon V leds again for comparison, even though it is a bit different light, and it has 4 Luxeon V leds compared to 3 for the E2L. Ignoring other differences, having 3 leds the E2L could theoretically put out 75% of the D4. But probably a bit less because the E2L uses a clicky switch and a bit lower quality FET, both adding some extra resistance. I measured output on a fully charged 30Q battery, at 3 seconds (first stable reading of my luxmeter after an output change), 12 seconds (first stepdown of the D4) and 30 seconds for the E2L.

D4:

3 seconds: 5200 lumen

12 seconds: 4850 lumen

E2L:

3 seconds: 3650 lumen

12 seconds: 3450 lumen

30 seconds: 3170 lumen

The D4 even performs better than when I measured it first a few weeks ago. It looses 7% output in 12 seconds. The E2L looses just 5.5% in 12 seconds, already a good sign for the ledboard. 3170 lumen at 30 seconds is also very good for this very small triple. Only the output (at 12 seconds) is not 75% of the D4 but 71%, a small difference that can easily be attributed to FET and switch, I do not blame the ledboard. The longest runtime that I checked on turbo was 40 seconds, which is the fixed stepdown time for the BLF-A6 driver. No angry blue light and no damage could be observed.

I give this board a very good rating for performance, when doing this mod I did not not see any worse performance compared to a DTP board (better??). The only challenge is getting the ledwires soldered, a problem inherent to the good thermal properties of these boards.

Thanks for your test and thoughts on these mcpcbs Jos .

Also , you should get some Sony VTC5A batteries if you haven’t any , they give a bit more power for these quads and triples :wink:

Forgot which ones I got on the way, but anyone know any substantive differences between the 5/5A/6?

SinkPAD lists the thermal capacity of their copper MCPCB’s as well as aluminum ones, both DTP, the copper has virtually twice the capacity. Al 210.0 W/m.K Cu 385.0 W/m.K, so while it’s not exactly half for the Al it’s not that far off. These numbers are given on a 20.5mm triple MCPCB with DTP for each metal core.

If you were to make a flashlight head in Aluminum to weigh the same as one in Copper, well obviously the Aluminum one would be very bulky. lol

5A’s ROCK! :smiley:

VTC5 is 2600mAh , up to 20A continuous discharge current “(test by mooch)”:Sony VTC5 20A 2600mAh Bench Test Results...a good 20A cell | E-Cigarette Forum
VTC5A is 2500-2600mAh , up to 25A continuous discharge current “(test by mooch)”:Sony VTC5A 2500mAh 18650 Bench Test Results...a fantastic 25A battery! | E-Cigarette Forum
VTC6 is 3000mAh , up to 15-20A continuous discharge current “(test by mooch)”:Sony VTC6 15A 3000mAh 18650 Bench Test Results...a great battery, equaling HG2/30Q | E-Cigarette Forum

VTC5A seems to be the 18650 battery with the lowest internal resistance up to date (aka it will give the most current when used in a direct drive setup , in comparison to any other 18650 cell)

Looked up my orders, yeah, VTC6.

https://www.reddit.com/r/electronic_cigarette/comments/5avri8/sony_vtc_battery_shootout_which_one_is_the_best/

I figure I’m not going to be vaping or DDing a quad or anything, so I went for capacity. 3AH sounded good. :smiley:

But yeah, lotsa people like the 5 and 5A, so am still wondering why.

That was the gist of what I was getting earlier, albeit in a roundabout way, but this about sums it up nicely. :smiley:

I modded another nitefighter C8 style light today and decided to see what the SST-40 would do if I put it on the aluminum mcpcb that comes with them. I did put down good thermal grease to give it it’s best chance of surviving the low resistance FET driver, heavier wires and spring mod. Surprisingly it did not kill the emitter even with a LGDBHE4 which is a pretty good high amp battery.

I did this because I was curious how many amps they can handle without being put on direct copper mcpcbs. Ton’s of lights like the SRK’s and other multi-emitters are a lot easier and cheaper to mod if you re-flow onto the original mcpcbs. From this I would say that there is little to worry about even when using high amp batteries.

Does anyone know how many amps these SST-40’s can handle without putting them on copper mcpcbs? I wouldn’t think they could handle this much and now I’m curious what the max is.

^ DTP boards perform pretty close to each other (a few years ago relic38 and I tested a number of types that all came pretty close to each other), but I have the impression that in non-DTP boards there is a huge variation in performance. The best I have seen is a cheap (copper core) 20mm XM-L board that easily allowed to run an XM-L2 led at 5A while performing only 5% less than a Sinkpad. But on the safe side I think 4A is no problem for almost any ledboard, also with a SST-40.

Thanks for the info.

Heres djozz’s test on the SST-40.


Today I modded a LuckySun D80 with a BLF-A6 driver and a Black Flat on led4power’s new, promising, Oslon board.

Unfortunately the modding attempt took all day and turned into an investigation of this new type of board, and this is what I found:

  • to compare this new mod to, I have a driver-less (so led directly connected to battery, with only the switch in the circuitry) ‘Mitko-thrower’ with a Black Flat on a XP-DTP board, current is 4.5A and output is 700 lumen after 30 seconds, which is pretty consistent with my Black Flat emitter test.
  • But this D80 mod, very neatly performed, did just under 500 lumen (at 4.2A, close enough), so there is 200 lumen missing somehow. The light was clearly more purple on the highest setting compared to the lower levels, a bit more purple than the (higher output) Mitko thrower.
  • I thought maybe the driver is off, so I put a new BLF-A6 driver into the D80. Result: still a bit under 500 lumen.
  • I reflowed a new (my last :frowning: ) Black Flat on the board to see if it was the led that was wrong, result: still under 500 lumen.
  • I suspected that my D80 has a really bad reflector, so I took a brand new Convoy C8 host with beautiful smooth reflector and transplanted driver and board+led into it. Result: 4.2A and just under 500 lumen.
  • then I tried driver-less, connecting led-minus thoroughly to ground. Result: under 500 lumen.
  • then as a last way out, I reflowed the led on a 20mm Noctigon and soldered it into the C8 (with the BLF-A6-driver, but with the DTP-board the driver is bypassed and led is directly connected to battery). Result: clear and present 620 lumen (and a small black speck on the led from all the transplants).

In all steps above I made sure that the led was perfectly centered and in focus, they were finished mods.

Conclusion: I can still think of other explanations, but for me the most plausible one at this moment is that this new type of board (at least the one that I did the test on), when heat-shedding really counts with a severely overdriven led that has the small Oslon footprint, performs not as good as a DTP board. I hope that I will be proven wrong by others as I really want that easy to use well-performing Black Flat board.

Black flats also do not come in flux bins they put them randomly in the tapes
likely this ones were on the low end of efficiency

Yeah, but reflowing another LED yielded the same result while switching MCPCBs brought an improvement. This is a strong indication that MCPCB was the reason for low performance.
That’s indeed a disheartening result.

Yeah you definitely want DTP if you want to get the most out of this LED.
Thermal paste is a must.
If you want to do 5-6A then you need a proper heatsink, not a flashlight host, to dissipate the heat.

Thanks Steve.