I've only seen one AR coated lens out of 5 or so of batch 2's. I'm thinking they are fairly uncommon.
For the blind spot thing, I dunno - don't think I've ever seen anything like that - can't tell much from the pics. The LED surface itself looks fine, maybe the dome has something on it on only one side.
At first glance, I was wondering if it’s sitting flat on the mcpcb maybe too much solder paste under one side?
How does it look against a white wall on low/medium power, does that LED look different from the other 3??
For what it’s worth both of the batch 2 lights I received have AR lenses.
I don’t have the light with me right now (no EDC light 
) , but the blindspot didn’t seem to be on the LED itself.
It seems to be on the reflector.
It looks, like when the reflector was manufactured, this blindspot somehow made it on the surface… (or the surface never was polished at that spot)
I tried to find a pattern when flashing against a wall, but could not really recognize a spot there, that reflects the spot on the reflector…
But what I think is, that the second Q8 (the one without AR coating and without a blind spot) is a little bit brighter, than my first Q8 (with AR and with the blind spot)
(but I will check this again, to be sure to rule out any mistakes, because of different battery brands, I have used here…)
Finally the BLF Q8 arrived.
I revised the light and dissasembly and all looks ok. From the treads to soldering work all are ok(perfect). My unit has 18awg short wires instead of 20awg thats say in the first post. So good upgrade.
The bad thing is for unknow reason the desing its not the same that its supose to be. The ledboard its not fixed to integrated self with the screws, instead of this, the screws screw in the reflector so all the force goes trow the gasket positioning leds (made of plastic) and I think that it is a BIG FAIL and the original desing its better and i see no reason to change a thing that its perfect already. So, whats the gain in change this???
Despite of this, the ligh its awesome and thanks to all involved in the creation.
Btw- i want to buy a old reflector in orther to be able to screw the ledboard to the integrated selft and comes this light to the original form. Any ideas?
Thanks.
I’m comparing Q8 with Imalent DT35.
XHP35 HI tends to get roughly 50% higher cd/lm than XP-L HD.
Q8 has ~10 cd/lm from its SMO reflector, so XHP35 HI swap should bring it to ~15 cd/lm.
The difference in reflector area is probably slightly less than 50%, so scaling it up to DT35 size would give 22 cd/lm. DT35 is measured at 30 cd/lm despite having OP reflector. Why is it so?
The change in how the ledboard is clamped down has been noticed by many. Your opinion is that it is a big fail, I assume because the possible clamping force via the plastic gaskets is less than when the ledboard is screwed directly into the shelf. Others, including me, see advantages of the new situation as well:
*the ledboard is clamped more evenly, at the 4 most important positions where the leds are. Deformation of the ledboard from screwing it down is less likely
*the reflector is in a fixed position when the lens and bezel is screwed in, this makes it more likely that the leds are centered well.
So for me it is not a ‘big fail’ but a design change that has positives and negatives, and I tend to see the change as positive.
FWIW, at least no one has seen a performance drop from the new design 
You are correct, the clamping force is less this way, and this its very important in the heatsinking of the flashlight. And not only that, the new way put all the force trow a plastic gasket… to win what?
The ledboard its clamped equally evently. This is the work of the Big center screw.
Neither deformation will occur with any of two desings.
This sentence doesn´t make any sense for me… probably because of my english. The reflector is in a fixed position in the two desings, because this is the work of the big center screw. The leds will well centered in the two desings because the ledboard and the reflector are designed for this (without taking into account the screws).
This change its not make any sense…all is bad and nothing is better to doing this way.
Probably you are correct yes
“The ledboard its clamped equally evently. This is the work of the Big center screw.
Neither deformation will occur with any of two desings.”
That would be correct if the center screw fixes against the shelf, but instead it is fixed against the ledboard so it does not clamp down the board.
“The reflector is in a fixed position in the two desings, because this is the work of the big center screw.”
The center screw allows rotation of the reflector over the ledboard, in many of run1 of Q8’s you can see the leds slightly off-centered all in the direction of the rotation, this probably happened when closing the bezel. The run2 design change makes that rotation impossible.
I think the batch 2 screw re-design in the head has more to gain, less to lose from batch 1. Totally disagree with the "BIG FAIL". I've worked on about 10 batch 1's, about 6 or so batch 2's, and from direct experience, I can center the LED's better in a batch 2, get real nice even compression of the MCPCB on the shelf, and overall batch 2's seem to perform slightly better in terms of output and holding output, with all the same driver and tail mods.
I've had many struggles centering a batch 1 LED's on the center reflector screw - the 2 long screws now eliminate those struggles. Of course I replace all 3 screws with brass ones, so that helps as well, but also replaced batch 1's with brass screws.
Is there any warrenty for this light?
Thanks Dale. I appreciate that. I will sit down at some point and go through the documentation and see what I can do. I don’t want to get rid of ramping. For me that is one of the slickest parts of the light.
The Miller - One item you might suggest to the builder is to make the black feed wire as long as the other wires attaching to the bottom of the processor board. That short wire really limits access to the processor board.
Ordered 15 brass screws but got closer to 30 (bag says 15). Poor fellow (or lady) must have gotten tired of counting and just started tossing them in a bag. Most screws well made, very few with imperfections.
If I understand correctly, the first batch had longer wires. The second batch seems to be a bit random about how long the wires are, so some are long enough to flash the driver without soldering, and others aren’t. (edit: Tom checked several and found the wire length to be pretty consistent, just barely long enough for reflashing)
I would definitely not mind if “wires long enough for easy reflashing” was a design requirement… but I recognize that longer wires reduce maximum brightness and can potentially increase electrical noise on the driver. So there are good reasons for shorter wires. And I’m, uh, a bit more reflash-happy than most people, so what I want isn’t typical. But still, I love when I can mess with the code without having to otherwise mod a light.
It seems I got lucky my results are pretty normal for batch 2… wires just the right length, reliable and balanced switch LEDs, screws made of metal instead of butter, generally no significant issues. I had some AR coating to remove, but that’s resolved now too, thanks to an entire day worth of abrasive scrubbing and harsh chemical treatments.
My code changes only took a short time though, so I haven’t opened up my Q8 in a while. There isn’t much left for me to optimize.
Of course, my needs aren’t really emergency-related. I’m more interested in it as an entertainment device than as an emergency tool. It’s fun to close the door, turn off the lights, and take a shower with a modded Q8 running in lightning storm mode. Maybe I’ll finally add a candle flicker mode too. That might be nice for a bath.
I realize the group buy is long past, but if anyone hasn’t paid or if there’s just any extras, I would like to purchase one.
Thanks and no worries if none are available.
TK, have you published lightning mode yet? What version? And Yes Please on a candle flicker mode.
I found the following:
Batch 1 wires: 62 mm long (over 10+ Q8's checked)
Batch 2 wires: 45-46 mm long (5-6 Q8's checked)
I've re-programmed several batch 2's with the stock wires in place and didn't have any problems. My preferred length for replacement wires is 48 mm though - I seem to need the extra length, might have to do with how I strip/solder the ends. But Yes - the shorter the better. There's no good reason to go over 48 mm. I haven't seen any randomness with the wire lengths, least in the few pieces I've worked on.
Weird though you see the black wire shorter than the red - sure it's not how you have the driver positioned?
Group buy coupons have expired, but the M4D M4X code works as of today, and it's a pretty good deal, for $46.50 here: https://m4dm4x.com/blf-q8-arrived/
That sounded nice, so I did it. And now for a bath…
(not sure how people normally do a candle mode, but I implemented it as the sum of 3 waves at different randomly-changing frequencies)
would it be indelicate to ask for pics?