Q8, PMS SEND TO THOSE WITH ISSUES BLF soda can light

BLF Special Edition Products
5521

yep :blush:
But honestly this very driver was not built by me.
Due to lack of experience and tools have to start with smth more simple and so while I was working on Convoy S8+triple Nichia 21C+preparing driver and flashing custom firmware one local guy was helping me with that pre-meteor driver for SRK.
Iā€™ve already removed some extra solder while soldering contact rings.
There is still some work left - waiting for half transparent button. Must be ready in 2-3 days.

Now Iā€™m looking for some good single 26650 host with side button to put XHP50 led and similar driver
and still not decided whether I need to order one more SRK clone (besides Q8 :slight_smile: ) for modding. Wanna insert the same pre-meteor driver and 4 XHP50 (or 4 XHP35).

5522

OK, thanks strango. :slight_smile: ā€¦ What was his reasoning for the solder, better ground? ?
Just curious as I just have never seen one done like that.

5523

The Lumintop SD26 is a pretty nice single 26650 light. I got mine running NarsilTriple, of course, on a TA triple, but with a XPL2 V6 3B, nice switch LED in there too.

(Click on either pic to go the SD26 mod picture album):

Anyways back to the Q8, I took the XPL V6 3D LED board from proto #1 and transferred it to proto #3, swapped the 18 AWG wires to my own from IOS. Their stock wires are thicker strands, very stiff, and swapped the switch resistor from 4.7K to 10K. Everything else remains the same, in stock form.

On SANYO GA protected @4.21V, lumens: 5,600 @start, 5,230 @30 secs, throw: 50 kcd @5m

On Samsung 30Q BT's @4.20V, lumens: 6,110 @start, 5,580 @30 secs, throw: 53 kcd @5m

This is lower at 5,580 from 5,880 compared to the original CW LED's, equals about a 5% drop. Our cheap meters will tend to read NW's lower than CW's from tests djozz has done. Other than that, I'm not sure. Taking amp readings would help here but that requires some rework on the tail of the light to get a wire in there to clamp around for a clamp meter.

Still these are impressive numbers for the stock configuration.

This is a little lower on the 30Q's then with the original CW LED's

5524

Do not know for sure :slight_smile:

And wonā€™t even try to express my assumptions because of bad English lol :laughing:

5525

thx, will check tomorrow (actually today as it is already 3:30 AM on my clock).

5526

I understand my friend. :wink: ā€¦ But may I say, your English seems pretty good to me. :+1: ā€¦ :slight_smile:

5527

Hmmmm, well apparently not The Miller. I found the answer to what I mentioned earlier.

I see where FastTech first listed them (UF 4-L2) for sale 3-30-2016ā€¦ before this thread began.

https://m.fasttech.com/products/0/10015804/4521601-ultrafire-4-l2-led-flashlight

Seems to be 6 different models toā€¦ with either 3, 4, 6, 7, 8, or 9 LEDā€™s.

Looks like UltraFire got the jump on it. :wink:

They might even say they inspired this threadā€¦ā€¦ :person_facepalming:
Hopefully not thoughā€¦ā€¦ :slight_smile:

5528

Nice Tom, thanks again!

Ha that UF is older yes, I hadnā€™t seen it (if I would have that red bezel is too much of turn off to give it more then a quick glance anyways :wink: )
We were not at the stage for designing the housing then even.

5529

About what current the Q8 can handle (as in: work fine without damage, not as in: can still touch it) in continuous use: considering an S2+ can continuously handle 2.8A without problem, and considering the transfer of heat to the air is a pretty direct function of outside surface area which is for the Q8 about 3.5 times more (estimation), I expect that the Q8 will survive 10A current continuously (half power) but while being untouchable at close to 100 degrees Celsius (as far as I can think of, all components are fine at 100 degrees). Too hot to touch (60 degrees) starts at about half of that so 5A. This is all at room temperature with passive airflow so without active cooling.

Of course this estimation needs testing at some point :innocent:

5530

In the 4 led model Ultrafire it is also available with a blue or silver coloured bezel :wink:

5531

You beat me to itā€¦ā€¦ :+1: ā€¦ :slight_smile:

5532

Djozz I think the surface area could be even more since we have a lot of fins besides a basic tube that fits a S2+ for times
I recall 1200 lumens without the Q8 getting hot but donā€™t know where I get that from :wink:

Ah well they should have shown silver as first pic and if I had seen it I would have given it a second glance :smiley:

5533

Strange enough I accounted for that already, if you do a ballpark calculation simplifying the two lights to being plain tubes, the difference is just a factor 2.6, which is completely counter-intuitive :slight_smile: .

5534

I recall, and Iā€™ve always kinda hoped I could do something similar to mine. But then I remember how much hair-pulling you had to go through to mod yours. I donā€™t care that much, especially with the Q8 coming out. :slight_smile:

The Meteor is reflashable, but its code (a prototype version is available) is rather opaque and, apparently, according to its author, carries a high risk of physically exploding if there are any bugs. Iā€™m told that it can only really do a small specific set of brightness levels due to the way the hardware is designed, and if used wrong, the driver itself could self-destruct. Iā€™m told true ramping isnā€™t feasible on it, aside from very brief ramps between fixed levels. So I havenā€™t really tried to change its firmware.

In general, boost drivers have more heat issues the higher the power is, because converting voltage creates heat. And in general, linear-regulated drivers like led4power makes have heat issues at around half power (but theyā€™re fine at both the low end and high end).

If I understand correctly (which is questionable), the Meteor has both of these issues to deal with.

Meanwhile, the Q8 has neither limitation. The Q8 driver doesnā€™t generate a significant amount of heat no matter what mode itā€™s in; virtually all the heat comes from the LEDs. So the worst you could do with it is get it stuck in turboā€¦ which will take quite a while before anything fails, and even then itā€™ll probably only damage the LEDs.

And if people want to change anything, the code is available and mod-friendly.

5535

We are probably going a little off track, but I really would like to know more bout what you guys are discussing on the Meteor - reflashable? Is there a thread on this?

5536

Fair point but with the PCB being so thin, itā€™s also possible the screw is stopped by the screw head reaching the screw hole in the lightā€™s head. If they sized the PCB hole correctly, there can be minimal pressure on it. There does not need to be much as the only purpose of the screws is to keep the driver from falling out when the light is opened and to keep it from rotating. When everything is closed, they are useless.

Anyway, as has been pointed out previously, PCBs are pretty hard too. Try cutting one just for fun. I expect theyā€™ll strip the head before they break the PCB.

5537

For the screw issue, we are gonna address it with them, but don't think it's show stopper, myself anyways. The board is thick, the hole is only slightly narrower than the head of the flathead screw. Screws are typical non-high qual types. I didn't yet, but I could crank down on them. Hhmmm, I'm @work now so can show it to our super tech here, maybe an engineer or two, see what they think. These guys are very much experienced with PCB's and mechanical designs - they are all 30+ years doin this stuff.

5538

Yeah, itā€™s probably better to go to the Meteorā€™s thread. Some of the background for this is about two years old now, on page 15 or 16-ish there:

5539

Thanks TK!

Spoke with and showed one of the prototypes to the tech who has 45+ years experience, been through about the entire digital, not quite industrial, revolution (), NASA/mil certified for soldering, etc. He said the new fire retardant G10 material PCB boards are made of is very tough, difficult to break or crack, unlike the older PCB's used in the past that could crack easier. He sees no concerns there, and has seen flathead screws used on PC boards before, some countersunk, some drilled straight and trimmed at the top edge, or just straight drilled. More important is the wear on the ground ring from tightening up the battery tube - we will bring this up and ask them about it. Might only be a concern after hundreds of tightenings though, but from frequent mech. lockouts, that could happen fairly quick on an actively used light.

5540

If you were to ask a QC guy, he might suggest some sort of durability/reliability testing if he thought you were concerned about the wear on the ground ringā€¦ :wink: