Thanks d_t_a, fixed it in the original misguided post… they actually stacked capacitors not resistors, I think that was an early fix but DEL actually found a better more stable workaround by tombstoning an 100 Ohm resistor between the diode and 10 Ohm Cap. Trickier to do, for sure.
They used the original Bistro, not the final revision. So it has 7 or 4 modes, not the programmable table. Timed step down, not thermal. This is the same orange colored BLF X6 board that was in the X6/X5 group buy lights. Or, that’s what’s in the sample I got, times two.
d_t_a
(d_t_a)
42
7 or 4 programmable modes = this the BLF A6 driver, not Bistro isn’t it? You mean your X6S came with BLF A6 driver instead of Bistro?
(easy check if BLF A6 driver or Bistro driver, is to test the battery-check: BLF A6 will blink 0 to 5 blinks to indicate approximate battery level, whereas Bistro will blink out the digits, like 4.1v will be 4 blinks, pause, then 1 blink, then long pause and repeat.)
You’re kidding me, right? I guess you joined after all this was done. If you could only know how many of these I’ve built from scratch…
A6 had a straightforward Strobe mode, Bistro has a tactical strobe that changes frequency. And then of course there’s the 4 blinks to show 100% and 3 to show 75% and so on of the A6 as compared to the actual voltage blink of Bistro.
Yes, it’s Bistro. An early version straight off the shelves of backstock alley…
lfowner
(lfowner)
44
Hey @Newlumen,
could you please take a look at these pictures? I’ve got my X6S NW a few days ago, but I’m very disappointed! For me this is really not NW, it’s CW in my view. Perhaps you can say something to it’s colour…
http://members.ping.de/~dirk/foren/blf/
Left is Astrolux S1 (XPL 3D / 5000K)
Middle is Sofirn Q8 (XP-L HI CW / 6500K, they sell no other opinions)
Right is Mateminco X6S.
Tally-ho
(Tally-ho)
45
Your XP-L HI is 1A tint = cool white.
lfowner
(lfowner)
46
Yes, I know, but Neal complained that the sticker is wrong.
d_t_a
(d_t_a)
47
The actual LED used in the X6S neutral-white is XPL-Hi 3B or 3A? (label says 3B, but I read somewhere it is actually 3A, anyone can confirm?)
This is a nifty little light… How does it handle the heat…?
It has huge output, but for how long? What does it step down to or settle at?
Thanks
DB_Custom
(Dale)
49
d_t_a, both samples I got say 3A on the box. Pretty believable when judging the tint.
DB_Custom
(Dale)
50
justanotherguy, with a rested Sony VTC5A cell at 4.20V I put the stock X6S in my lightbox. It started at 2980.8 lumens and at 30 seconds was down to 2701.35. At 40 seconds it stepped down to 2287.35 lumens and maintained this through 1 minute of run time. At 1 minute and 30 seconds it dropped to 1787.1 lumens. By 2 minutes it was down to 1521.45 lumens. The cell was at 4.04V when I stopped the test.
Sitting in the lightbox with no air circulation and running for 2 minutes it got hot, almost too hot to hold, but that’s to be expected sitting down in the foam.
I didn’t realize until it was uploaded to YouTube that the phone hadn’t set video in landscape orientation. Once in YouTube I couldn’t figure out how to correct this. I don’t do vids much. lol Sorry.
d_t_a
(d_t_a)
51
I’d like to ask if the thermal control was turned Off or was it just set to a high temperature level for this test?
DB_Custom
(Dale)
52
There is no thermal regulation. Older version of Bistro. You can choose 7 or 4 levels, last mode memory on or off. That’s it.
Edit: This may seem to be taxing my memory, but I built so many Bistro drivers from the beginning that this is deeply ingrained… on the original group buy, when ToyKeeper first wrote Bistro, she upgraded from the UI of the A6. The primary difference between the A6 firmware and Bistro at that time was that Bistro had a Tactical strobe and the Batt Check mode in Bistro blinks out actual voltage in tenths where the A6 firmware only has 4 blinks… 4 for 100, 3 for 75, 2 for 50% and 1 for below 25%. Then she wrote the new Bistro with the table and before it could be acknowledged they started production early. So the early Bistro drivers had this A6 adapted form of Bistro. And this is what they seem to be using in the X6S. They’ve resurrected the old original driver from the X6. And added a pair of stacked capacitors between the ground ring and the power (red wire) lead to the emitters to stabilize the driver’s functions. And in some way, this has enabled the hidden modes to be locked in last mode memory… if you turn the light on then reverse to … say… Batt Check, watch it blink out the voltage then turn it off…. when you turn it back on it will still be in Batt Check. This is not supposed to happen. The hidden mode group is not supposed to be visible to last mode memory. But it is. And in this way this driver is different from the original X6 driver.
I can ask ToyKeeper to step in and verify this, and I believe I will, but I’ve built the Bistro driver so many times I feel I could do it in my sleep while dreaming of umbrella drinks in the tropical sunshine. 
SKV89
(SKV89)
53
Sometimes two units of the same flashlight can perform very different. Your R90C measures 4,000 lumens higher than mine and both of our TA spheres are calibrated with Maukka lights. I even cleaned all contacts multiple times with alcohol on my R90C to make sure it wasn’t grease getting in the way. I’ve measured duplicate lights at different output too. It’s luck of the draw sometimes.
Therefore I don’t trust measurements from any lumen spheres calibrated using flashlights with manufacturer’s specs.
DB_Custom
(Dale)
54
Ah, but that’s where the law of averages comes in! My light box was calibrated with no less than 25 lights, not just one low powered unit.
I would love the chance to take my light box, a few dozen lights, and a big box of cells and chargers and sit down next to a big high dollar sphere for some comparison testing. Maybe a few hours of side by side would have some illuminating consequences…
SKV89
(SKV89)
55
Well you can always just get a set of Maukka calibration lights, which is measured using a high dollar sphere
DB_Custom
(Dale)
56
And what does Fenix use? And Nightcore? And Lumintop? How many times have we seen that big green ball pictured in the corner of the lab? They all use expensive equipment, much more so than our PVC P Traps. How is Maukka’s better than theirs?
I’ve been doing this here for a long time, don’t need any calibration lights, already been done. And if I only had a dollar for every time this has been discussed…
If it has the following traits…
- 4/7 mode only, no other mode groups
- Only two config options
- 2-speed strobe
- Volts+tenths style battery check
… they must have made their own version. I never made one like that.
The early version which got used on the BLF X5 / X6v2 had 8 mode groups, even though I also sent a 9-group version.
Anyone know how to contact Mateminco? They’re legally required to publish the source code for what they’re shipping in the light, so I probably need to have a talk with them about how the license works.
DB, did you manage to get a picture of the driver? It’d be interesting to see exactly which chips they used.
DB_Custom
(Dale)
59
I cut this one out of the stock X6S just now, images straight out of the Canon 1Ds MkII with 100mm Macro Hybrid IS lens.
DB_Custom
(Dale)
60
Oops! Sorry! My Bad! The first blink in configuration mode doesn’t do like I expected so I ASSUMED the second blink was for last mode memory. It in fact is not… it goes through 8 blinks for the table like the original Bistro. Sorry about that.
The first blink, if selected, changes 7 modes into 3 with no moon. I was confused there and made the assumption. Since I never engage last mode memory, I didn’t go further. 
I have used the 9 level table from the beginning, since you (TK) sent it to me before the initial release.
