Convoy L6... XHP70 Beast!

J-Dub74 · May 23, 2016, 4:05am

Dale thank you for all the info and thorough investigation. I truly hope that the issue you had with the side switch was just a “beta” fluke. Mine works flawlessly. In fact tonight it had its longest run. Mars being viewable tonight without a telescope led to my son and I taking a walk after dark.

Warning: I’m going to spin off-topic a bit but it does involve the L6…

As evening rolled in the first light in the sky was Mars. How cool! It looked like a star but brighter… and orange… and visible before all the rest. As it got truly dark the full moon rose out of the tree-lined background, abnormally large and an unusual yet beautiful rosy-orange color. This unusual situation inspired my son to ask of we could go on one of the nighttime flashlight walks we sometimes take in the summer. He brought his S2+ and I brought the L6. We have a pond across the street that we like to walk around. It’s full of bullfrogs and small fish and as we found out tonight also a handful of crayfish. The L6 could light up the pond enough that we saw a lot of what was going on under the water that we don’t normally see. Also well worth noting is that the L6 has a much lower low than the S2+ and was beautifully usable at all levels. In our walk I ran it for roughly 25 minutes continuously, using turbo a lot but also the lower modes. After a few minutes in turbo it had a nice warmth but not hot in the least. I must have cycled through all the modes 30 times on our walk and I still smile at how well the modes are set. Another tip-of-the-hat to Dale there. I setup the initial mode spacing but ran it by him first and he tweaked my suggested mid range numbers to an even better mode spacing that we can now all benefit from. During my night walk I found that after being in turbo for a bit, dropping to low I actually wondered if the light had turned off but after a few seconds my eyes adjusted and a realized it was just that nice low. My son’s S2+ is a nice 5000K neutral and seeing them side-by-side he was saying how much he enjoyed the tint of his flashlight.

I love the variety of tints we have in flashlights now and no company I’ve seen brings as many stock tint options to the table as Convoy. It also makes me smile that my 10 year old son is talking about neutral tints as compared to cool white. I don’t know for certain without having the DTP version of the L6 everyone else will but it’s possible the S70 might bring a few more lumens to the table. That said they’re both so crazy bright I don’t care which one wins that number game. They’re both excellent performers but I know which one I will use more. The L6 just puts a smile on my face every time I turn it on.

DB_Custom · May 23, 2016, 4:17am

FWIW, my S70 made 3636 lumens straight out of the box, as compared to the 3474 of my L6. Virtual tie. After I played with both, the S70 is making 4276 while my L6 is making 4609. No longer a tie. Beam profiles look similar, except for that vague shadow in the middle of the S70’s.

Yes, it’s great when a big honkin light has a low enough output to be that useful. Well done there for sure. With the bump mine is now doing…

0.06A for 23.5 Lumens
0.20A for 183.54
1.81A for 1649.1
6.36A for 4609.2

Useful levels, each one.

SCEman · May 23, 2016, 4:23am

Really looking for more beamshot comparisons to help me decide.
Mukcep’s images show the S70 to be brighter IMHO; even with the “invisible” dark spot.

Mukcep · May 23, 2016, 5:00am

Beamshots on 1.4 meter distance (f/7.1, exposition – 1/100sec, ISO100):

S70:

L6:

I still believe that the brightness of these lights have parity, they just have different beam patterns.
And I like L6’s beam more.

I can see “petals” on S70’s shot. And i can see blurred hole (S70) on distance 0-5 meters (on shots you can not see it because exposition is too big in this case).
Both of lights are stock-versions, without upgrades and modifications.

DB_Custom · May 23, 2016, 4:54am

I took beamshots and video last night, but my internet won’t allow me to do anything with em.

djburkes · May 23, 2016, 1:04pm

DB Custom:

Decided to play around some with the new L6, recut the emitter shelf to accept a 31mm MaxToch copper DTP mcpcb for a new top level XHP-70 from Cutter. P2 1C binning, rated at 1830 lumens as compared to 1710? of the N4. Then I did a wee bump on the driver by stacking a R140 on top of the two R082 sense resistors (reducing resistance from .041 to .031) for an increase tail amperage reading from 4.83A to 6.33A, it’s now doing 4505 lumens out the front. Sweet! The L6 is a very nice light, love the design and overall perspectives of it, Simon nailed it with the reflector, no donut hole, good hot spot area from the big 4die emitter, gotta love that, his first (I think) in this arena and very well done.

Initially I was seeing a low of 17 lumens, Turbo at 3500. I’m not at my notes at the moment but I think 2nd level was around 249 and third was 1500. Low, being only 17 lumens, really shows the lack of dark shadow in the center. Hot spot is about all you get in this mode. Very cool.

Side by side with the S70, also modified with spring bypasses (Simon already did these on the stock L6) the hot spot area is a bit tighter, no donut hole in the L6 beam but my S70 shows a definite shadow in the center which is somewhat deceptive as the mind expects more intensity in the middle of the beam, not less. For all practical purposes though the donut hole is or can be difficult to see. Shining it on anything really solid, especially lighter in color, at a distance of out to 50 yds or so, moving the light around, the dark center is very obvious. Shining the S70 directly at, say, a rabbit at 30 yds, you look at the rabbit and not the light quality, so you don’t notice really as the rabbit is darker anyway. So it depends on what you’re shining it on, how you’re using it, as to whether it’s a nuisance in the S70. Also, side by side the L6 is just a visually nicer light. The matte black ano looks more high end compared to the shinier surface of the S70. Reflector surface appears similar in both lights, but the base is executed differently, with the Convoy having a slight taper down toward the emitter while the S70 base is flat. This might be what solved the donut hole issue, not sure exactly. I don’t know that the dark center is enough to truly be disappointed about in the Thorfire light, but the beam profile of the L6 is certainly pleasing and well received.

At any rate, the shadowed center is not present in the L6 so even shining it on pure white fresh snow while scanning, searching, you wouldn’t see that shadow. (like I know all that much about snow! )

I’ll have to get another of these emitters now, as I got this one with the intention of putting it in JDub’s light. lol Sorry Jared, all in the interest of experimentation don’t ya know!

I’m gonna be straight up with you guys, my L6, being one of the first 3 out the door, had an issue. This seemed to stem from the non DTP mcpcb that Simon used in these first 3 lights. Replacing the mcpcb/emitter made the issue go away. But it was weird as it involved the side switch not working the levels consistently, like the output would get stuck in Turbo. Removing the emitter and putting it on a SinkPad didn’t solve the issue. Replacing the emitter with a new one on a SinkPad made the light work as expected and I haven’t seen the glitch since. I don’t understand it, but it definitely seems related to that initial mcpcb/emitter combination. JDub’s own light was one of these 3, and it didn’t exhibit this problem. Kawi’s L6 was the other of these 3 and I don’t believe he had the issue either. So I’m assuming the emitter was shorting on the mcpcb on this particular sample. I tested the momentary switch and it was perfect. I tested the driver outside the light, running an MT-G2, and ran the modes by touching the two solder bumps on the driver where the switch wires go with tweezers, worked flawlessly (I, uh, might have played with it a bit long doing this, the MT-G2 started to smoke a bit, free floating and with dust on it from sitting on my workbench for a while. Oops!). All the testing I could do pointed at the emitter/mcpcb itself, and changing that pair out solved the issue.

A note on the big emitter. The XHP-70 has divided positive and negative contacts on the bottom. This is what allows it to be used in 12V configuration. I find it easier to use too much solder paste and cause a bridge under the emitter with this set-up. AND, everybody needs to be aware that in 12V configuration the negative contacts are NOT separated or isolated from the ground… they share the thermal pad to make it work and the thermal pad is grounded on the emitter shelf on any light that has ano cleared from this shelf. Something to be aware of, as a 12V set-up this way could be shorted at ground and only run in direct drive, bypassing the driver. (I say this for those that like to play with a light and do mods, this emitter requires paying a bit more attention to detail) So I highly recommend getting copper DTP mcpcb’s in the 6V format only for these, the dual format 6V/12V SinkPad’s have smaller traces, the split contacts, and are much more finicky.

Convoy L6 XHP-70, a big WIN in my book! :+1:

What would happen if you just removed the resistors and bridged the connection with solder? Also, when your internet starts working right, can you post a pic of how you stacked the resistors. My light will be in today so I’m going to try the resistor mod too.

DB_Custom · May 23, 2016, 1:17pm

I simply removed a R140 from one of the LuckySun D80 drivers and laid it on top of the 2 side by side R082 sense resistors. I was only planning on using one so I stacked it right in the middle, pyramid style. Not much to see, every kid has stacked dominoes or lego’s right?

ZozzV6 · May 23, 2016, 1:21pm

Or like bricks in the wall :+1:

kwarwick · May 23, 2016, 1:29pm

Here’s a better picture of the 6/12V XHP70 sinkpad board that shows the traces. Looking at the layout of this board, I don’t see the thermal pad being in the circuit at all… am I missing something?

DB_Custom · May 23, 2016, 1:39pm

Thank you for that. I’ve probably seen that before myself and probably took a macro image to get there, might have even sanded off masking to reveal it, but danged if I hadn’t completely forgotten how it worked!

I had one short on me recently and so I assumed, based on the datasheet that shows the thermal pad connecting those two dies in series, that the thermal pad on the star was also doing the same. My mistake, Sorry about that!

Kawiboy is trying to explain to me that it does indeed work, even flat down on a brass pill, and it was boggling my mind trying to figure it out based on the datasheet. Looking closer clears it right up. A maze of tiny traces. DUH!

KawiBoy1428 · May 23, 2016, 2:33pm

I had the same problem when I first got to building lights. I built the driver, reflowed the LED to mcpcb, wired it all up, wouldn’t change modes?? Spent a couple HOURS messing around tore the driver out, tore it down, changed the FET, changed the MCU, tried another New Driver, same thing? Finally looked at the emitter (10X glass) sitting up a little on one end? Rang it out with my DMM, sure enough grounded to the base, F&*K!! :person_facepalming: A simple tiny microscopic sliver of solder jumped the isolation trace, from (-) LED contact to DTP pad? Reflowed , checked with a battery jump, fired up, checked it with the DMM to the mcbpcb base (no continuity) wired it up, good to go!! A year and a half later, same thing, (I forgot about it), and started the same debug process again going nuts for a couple hours, and then it dawned on me, F&*K! :person_facepalming: . :person_facepalming: , :person_facepalming: ………

djburkes · May 23, 2016, 3:09pm

If you would have stacked a R082 instead of a R140, would it have given you more output?

DB_Custom · May 23, 2016, 4:23pm

Well, yes of course. But I had a R140 and I didn’t have a R082….

Edit: parallel resistor calculator , open it, scroll down a bit to see a 10 resistor pad. You have to clear it each time and re-enter values for accurate results.

DB_Custom · May 23, 2016, 4:25pm

There are formula’s for converting that resistance value to amperage, using the cells output power and the driver efficiency with the emitter Vf and all that, but I could never in a million years remember all the variables.

djburkes · May 23, 2016, 4:38pm

Thanks a lot, can’t wait to get home and play now!!

DB_Custom · May 23, 2016, 4:46pm

I’ll say this, reason needs be applied.
If you remove the sense resistors and bridge the pads, sure, it’ll make bonkers light output, like nearly 6000 lumens. BUT, it’ll fall super fast, like to the tune of 400-500 lumens the first 30 seconds. So the gains are fleeting, lost to heat and in the long run there will be damage done shortening the life of the emitter and at a very minimum yielding very short run times from your devoutly sought after cells.
Some compromise will bring a balance that can be lived with. The stock resistance is .041. I chose .031 and the resulting 6.36A from Basen cells works for me, as does the 4609 lumens. It doesn’t get burning hot super fast, the cells are still, kinda, in their range, and all is well. You could probably go to .025 and get over 7A and it wouldn’t be so horrible, but 9A or more is really not the best yield here, which is of course completely up to you.

J-Dub74 · May 23, 2016, 4:57pm

I don’t really understand all the resistor stuff but making a few changes there seems like the key. The driver originally had a R100 and a R082. After I talked over with KawiBoy he suggested going to either two R050 or for a little tamer version two R082 resistors. Simons driver guys are still being a little cautious but steadily improving. I can tell you for certain that this driver would have been putting out significantly less without his input. Thanks to Dale and KawiBoy1428 assisting me in the technical side of my advisory role, the L6 is better than it would have been and came out sooner that it would have. Cheers boys!

JakeDjanitor · May 23, 2016, 5:12pm

maybe DB custom can make us a video walk through just in case some feel comfortable doing what he did to up the lumens a bit? what do you think DB?

1dash1 · May 23, 2016, 6:31pm

I think you guys have done a marvelous job of optimizing:

output
run times
heat build-up
form factor (size, battery configuration)
style
UI (switches and modes)
beam (profile and tint)
quality construction
practicality (all of the above relative to cost)

Kudos to all!

djburkes · May 23, 2016, 6:34pm

I went home during lunch and swapped the emitter to a P2 1C and stacked a R050 resistor on top of the R082s. It seemed to be a bit brighter afterwards. I have a crappy meter so current measurements won’t do me any good.