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

And yes the integer resolution is a bit low in the inverted read mode. It's about 0.06V per integer, so precision can only be within +/-0.03V if tabulated/calculated carefully. If the math is done on the fly in 8 bit rounding errors can make it worse. I think maybe Tom E is doing the math on the fly, unlike bistro where it's pre-tabulate (although in HD pretabulated by, I guess 32 bit, compiler math from calibration parameters). Still I doubt that's going to make things worse than half a tenth.

FWIW, if I check the cell voltages as my 30Q’s come off either of my chargers they usually read about 4.17 to 4.19. The two Q8’s I have will both give me a blink of 4-1 when the cells are placed in them and a blink read called up. C;ose enough for me and on the conservative side by a tad.

Is this light IPXX rated? Thought I saw it at one time but can’t find it now.

OK, refreshed my memory on this. DEL wrote this code - I was quite impressed and went with it to drop into NarsilM. It's an equation and seemed to be dead-on. I found this though:

#define D1_DIODE 2 // Drop over rev. polarity protection diode: 0.2V normally, 0.3V for Q8

DEL may recall this better than me, but I think we determined the diode they were using had a higher voltage drop the the usual ones we use. As you can see though, I left the value at 2 for 0.2V instead of 0.3V. I think the margin of error turned out to be under 0.1V, so again as was mentioned, better to be on the safer side.

This is the equation:

wAdcVal = (11264 + (wAdcVal >> 1))/wAdcVal + D1_DIODE; // in volts * 10: 10 * 1.1 * 1024 / ADC + D1_DIODE, rounded

Received 2 of my Q8’s on 10-18-17, order dates were 9-7 and 9-12. Both lights are in near perfect condition.
The NARSIL software is AWESOME! This is fast becoming my favorite light, big thanks to the Q8 team and especially The Miller for making this light a reality.

Guess what I got today
It’s beautiful…flawless and if you press the white button correctly Disco mode!
I cannot thank the team that created this enough, great work people

Ah, so you're using the full 10 adc bits, so potentially 4x more resolution. Wow, non-constant runtime division, it's like living in the digital age. What a luxury.

Product name: BLF Q8 4X XP-L 5000LM Professional Super Bright LED Flashlight
Brand: BLF
Model: Q8
Emitter Type: XP-L HD V6 3D
Light temperature: 4750K-5000K
Light color: Neutral White
Mode: Smooth ramping or mode set operation
Material: Aerospace grade aluminum alloy construction
Total Emitters: 4x
Battery Configurations: 4 x 18650 or unprotected button top (not included)
Working Voltage: 3-4.2v
Switch Type: Electronic switch
Switch Location: Body side of the flashlight
Peak Beam Intensity: 50,000+ cd
Impact Resistant 1.2meters
Max Output: 5000+ lumens
Range: 450+ meters (max)
Lens: High transmittance glass
Reflector: Aluminum alloy mirror reflective cup
Waterproof: IPX-8 waterproof
Color: Black
Weight: 400g without battery
Size: 132mm x 59mm x 50mm(length x head x body diameter)

I’ve been wondering about that but didn’t ask… why “+ (wAdcVal >> 1)” ? It doesn’t match the expected equation and it doesn’t make a lot of difference to the result.

In mine, I took it out and replaced it with a slightly higher value for the diode constant (0.25V added instead of 0.2V). Can’t really tell from the graph, but the brown line is hiding behind the green line, indistinguishable. It compiles a bit smaller and gives the same values.

I wonder if they really checked the waterproofness. At least it must be bucket-proof, so I checked a few months ago. After lubing all o-rings I put a switched-on Q8 prototype (the body has no differences to the production model) in a bucket of water. It kept working for a minute and then I checked all seals and no water had entered for that minute.

So it should at least be good for all types of weather.

Yeah before dipping it water I recommend removing all lube and relubing it

But I wont be dipping mine in. It should at least work in rainy weather and that is good enough for me

In a SRK the metal switch is not waterproof, the rubber Q8 cover ought to be much better and Djozz confirmed this

Although tracking said “Departed Shipping Partner Facility, USPS Awaiting Item” a few hours ago, I went out to check the mail and found a Q8 waiting.

So of course I brought it inside, made sure it worked, and opened up the driver. I tried my other Q8 firmware on it, added indicator LED support, and pushed code up into the repository. And by the time I checked again, tracking was updated to say it was delivered. :slight_smile:

I found I can get the indicator LED into three states: high, low, and off. The power draw for these is 95 uA, 30 uA, and 0.34 uA. Each of the three is lower than stock because I turned off brownout detection (which uses about 23 uA). Each measurement varies with cell voltage though, and should get lower as the cell does.

The way I have mine set up now (using alternate firmware), it uses ‘high’ for regular operation (always on), ‘low’ for the soft lockout state, and it never turns the indicator off. However, the main and lockout levels are configurable to any of the three indicator states. I’ll probably add a fourth level at some point too, for blinking.

The Q8 itself seems flawless, though I haven’t really inspected the hardware in detail yet. Its shipping box was certainly a little beat up, but the contents seem intact.

NICE!

TK is this other Q8 firmware an upgraded variant of Andúril?

+1, also interested to know. Sounds awesome!

I thinked about two level of indicator led because stock Q8 button light is too bright at night but that is solved now :smiley:
Can I have a link? And today or tomorrow I will try it in my Q8.

GREAT! I’ve had my Q8 a week-and-a-half and TK has already made it obsolete! :person_facepalming:

…but I still like it. :wink:

No David, the MCU has ample space around it for the flash clip and maybe you can even get the driver out far enough without the need of soldering.

The Q8 was obsolete the day it was conceived . Always a problem - updating/upgrading software and firmware because no matter who, it's outdated the day it ships. We've looked into and discussed better, easier means to upgrade flashlight firmware, just nothing yet developed to be practical and economical. It's currently do-able of course, but requiring special tools, etc.

If I'm not busy modding/supporting/researching, I'm bizy making your flashlight obsolete