Fra, try using science to explain my ex-wife. (or my current one for that matter!)
Hmmm, you don’t understand where I got those things huh? Impossible, says you. Given my readings are right, says you. Doubt, cynicism. And you with 7 posts. But yes, I’m aware that human nature prevails. Science has been misleading and dead wrong many times before. But that’s a whole nother story…
I’m a photographer, not an electrical engineer. I don’t know this stuff inside and out like some may. Basically, I’m a hack who follows the lead of others and breaks a little ground of his own sometimes while doing so. Richard says I’m a good guinea pig, because I have a way of stumbling into problems that haven’t cropped up before. Oh well. I’m moving and shaking, backwards and from cold perhaps, but still.
(and yes, maybe a little OCD in that I work pretty hard to eliminate resistance. The results, sometimes, speak for themselves.)
“Hmmm, you don’t understand where I got those things huh? Impossible, says you. Given my readings are right, says you. Doubt, cynicism. And you with 7 posts. But yes, I’m aware that human nature prevails. Science has been misleading and dead wrong many times before. But that’s a whole nother story.”
Well if I were to send you mine and you modifyed it the same as yours then got the same results, everyone would have to believe you! 0:)
If your willing.I’m not real happy with the green tint of the de-dolmed Xpg’s anyways.What tints are the bare Hi-Xpl’s available in? Oh wait, this would be done in the name of science. So the same temp. emmiters would have to be used.You know, to make this “test” valid.
Doubt is not the same as cynicism. And questioning someone's results is nothing personal where I come from (=science originally) but is actually a healthy thing. I admit this is not science but a hobby forum, and scientific proof of everything is luckily not required (that would take away a lot of fun out of the hobby), but nevertheless, science applies way better to flashlights than to wives, and I do not disapprove doubts every now and then. As I see it, Fra881 did not mean to disrespect your results when questioning them, and anyone's right to do that has nothing to do with a post count.
I build lights to play with, mostly for the build process itself as I’ve kinda become hooked on the mod. So day in day out, they’re just toys and it really doesn’t matter what anyone believes or disbelieves. For me,the UI on this one will keep it on the shelf more than it’s used, 5K, 10K or whatever it’s output is.
Still though, when someone gives a “fact” it’s not usually a good idea to rebuff that fact with statements like “that’s impossible”, because in a lot of circles that’s calling the person a liar. I, myself, could care less. Y’all aren’t here, y’all can’t see it, so it’s all just speculation in a forum.
It won’t be long until other’s put these emitters in this light and it’ll be interesting to see how those lights fare. I know not to take offense because I have a history of being very forgetful, I could well be doing some little thing wrong and have my results skewed. Pretty sure that’s not the case here as I’ve gone about testing it in various ways and everything pans out. I’m really curious myself as to how this one could be showing me these results when all indications are that it’s “impossible”.
So, what can I do to test it further in order for science to prevail? Y’all want a movie, walking into the woods and using other known lights and then this one? What can I do, as much for myself as for y’all? Want me to send it to manxbuggy1 and rdrfronty to have them test it? Send it to Richard? I’m willing to get some independent results to see how they stack up, because I too want to know why.
I’d send it to you djozz, but I’m afraid it’d melt your styro balls.
Edit: Post counts are indicative of experience, in most cases, so a noob to the scene questioning an “elder” is disrespectful in and of itself. Has all tradition gone to Hades in a handbasket?
to be honest, I’m a bit speechless. how did you get so offended, seriously.
first of all notice my first post in this topic is asking you about how fast it gets hot after 20sec, because I’m trying to decide whether to wait for the HI version or get the XPG2 now. doesn’t appear to be a post from somebody doubting you.
also, here is where you think I’m saying it’s impossible your light puts out 12k lumens.
if you’re still convinced I said that, please read it again.
“given Dale’s measurements are correct” means I take that for correct hypothesis, and I’m looking for an explanation that includes that as a result.
eventually, you could also go on CPF and read my last post in Vinh’s M43 topic and see for yourself what I write about your measurements out of this forum.
this said, I believe my only 8 posts will remain like that for the time being, at least in this topic. good day
Nearly all single-cell FET drivers or other direct-drive configs limit output based on voltage, not current. The cell voltage sags under load, the LED Vf gets higher under load and heat, and where they meet is where the output settles. This means a 4.35V cell will generally put out more light than a 4.2V cell, and causes the usual direct-drive output curve where it declines steadily until the cell is empty.
The Meteor uses a boost circuit though, so perhaps it has a fairly steady voltage and it lets through as many amps as possible until its output voltage meets the combined emitter Vf.
I only mention this because the charts someone posted earlier show XP-L with a lower Vf than XP-G2 at the same amperage, and that could explain part of why it appears to run at higher amps.
Edit: This is the chart I’m referring to:
(note how, at a Vf of 3.17, the XP-G2 puts out 519.7 lm at 1.2A while the XP-L HI puts out 613.6 lm at 1.5A)
Fra, not offended, it’s all good. I don’t understand it and certainly didn’t plan anything, so it’s doing what it’s doing on it’s own. And I have no idea about the hows and whys. (what I’m saying there is that if I’d worked hard figuring out sense resistors and resistance and planned to make 12K lumens then met with disbelief I might see it differently, but that’s not the case, I didn’t alter the driver at all nor make gains plans)
It doesn’t make any more sense to me than it does to y’all, I wasn’t expecting to see those kind of numbers. Somewhat higher than the XP-G2’s yes, but not like that. As it stands, Hi is making on the XP-L HI’s about the same as Turbo was on the S4’s. Gives a whole new meaning to Turbo.
I wish someone else would do this emitter swap so there would be some kind of verification…
Dale, thanks again for the modding info. It’s starting to look like when its time to do an LED swap on 12 LEDs sitting on one MCPCB, I better leave it to someone else who has the know-how. Unless I want to spend the extra $$$ and do some trial and error ‘experimenting’.
I appreciate you sharing your modding experience and that hot-rod M43 with the rest of us - all the measurements, photos, comments, etc. - its really a lot you have given us to make this thread very interesting. Thank you.
I gotta say, re-flowing 12 emitters on one mcpcb is a bit nerve-wracking!
For those old time soldering pro’s, when removing a dozen emitters from the mcpcb is it feasible to wipe the molten solder off with a damp paper towel or something? I’d much prefer to be using fresh solder paste as it is a visible seating of the emitter for a good confirmation. Existing solder doesn’t have that effect. Not that I’ll be doing this again anytime soon probably, but still…
Thanks cncyana. Wish I’d documented the XP-G2 S4 2B’s better with amperage and voltage readings at the mcpcb before making the swap so we’d have an accurate comparison.
Cutter in AU has bare XP-L HI in V2 1A bin, $99 order gets free shipping to most/some countries. 12 will get that. I just bought 10, thought about enough for the Meteor but really, why?
Those outputs are limited by only the entire resistance of the circuit, yeah, I thought you mean constant voltage circuits when referring to voltage limited.
The M43 driver has 4 parallel current sense resistors in series with the output for the LEDs (GND), the circuit always boosts voltage to meet the Vf for a particular current limit, I mean that is how a typical constant current boost circuit works.
If anything, the chart seems to show the differences in lumens output between the two LEDs for a given wattage is not that great. Of course, we are comparing a larger HI non-domed LED with lower output bin vs. a smaller domed LED with higher output bin. I hope Cree gives us an HI version of the XP-G2 real soon and Hank puts it into the Meteor.
DBCstm, You mentioned the M43 got searingly hot after 20 seconds on turbo. I assume this means:
1) The XPL-HI mod is probably consuming a lot more power from the batts then the XP-G2 LEDs and…
2) The extra output is not due to better efficiency of the LEDs but due to running at higher power levels?
Does anyone know if the factory XPL-DD also gets this hot this fast?
Also, I noticed your original cd measurement on the XPL-HI thread was 57.5Kcd. That’s decent throw although still less then the XP-G2-dedomed version of 70,000cd.
Note: mountain electronics has the XPL-DD version in stock.
I have used 2 MT-G2 in parallel because they have a very low Vf, 6V nominal (around 6.5V, based on my testings) compared to the 10Vf of how the MCPCB is setup on the M43. Using 2 in parallel means they share the current between them and I wont see a single MT-G2 possible dead 8A or ultra-overheating.
The XP-Ls have just a tiny bit lower Vf than XP-G2 and the output will still be 8A because with a massively lower Vf like the MT-G2 6V LEDs the current it is still the same, and any case it cannot be higher because that is the point of the current sense resistors, to regulate the current.
No matter the LEDs Vf:: 6V 10V or 12V the output is still regulated at 8A (after all that is supposed to be for a constant current circuit), as this does not work like a direct drive circuit. The output with the 6V MT-G2 in parallel wasn't exactly 8A, but 7.76A, that you can see in the photo, I dont think the driver can run as efficient closer to the Vin, especially because it is a step-up converter used (boost).
Both tests done with fully charged LG HE2 cells. The funny part is how cold the hosts is with the LEDs in front of it at 2-3cm. Both measurements zero-ed (important on current clamp meters due to magnetism).
So just to clarify your point, your conclusion from this test is that the lower Vf of the XPL will not cause a current increase, therefore, this is not the cause for the higher output of the XPL-HI mod being discussed earlier in this thread?
I don’t see what your’e getting at Hikelite, you’re testing different emitters but not the one’s in question so what is being proven? The MT-G2 shows different numbers than the XP-G2’s, so there is variance in the driver. Take all those XP-G2’s off and replace them with XP-L HI’s and do it again, which will still be relatively pointless because I already provided Vf and actual current draw.
We need to see a duplicate of the light I built in someone else’s light box. Preferably a different style box.
The light, in hand, still makes more or equal output than virtually anything else I have, including 9V MT-G2 Triple (not 6V emitters but 9V) and Triple XHP-70. I can’t blame anyone for not believing it because it’s hard for me to wrap my mind around as well, but it is what it is and in use it’s just a staggering amount of light.
(or my current one for that matter!)
