Will answer my own question now :eekster:
Originally Posted by Whitedog1
Did a short review on the battery pack (better tell it a battery hoder cause the deliverd cells inside are crap)
You can find it here (post #133)
Finally got my light. Please track your package more closely than I did. From China to Pennsylvania:
Ordered = July 4th
Delivery attempt with notice supposedly left and never seen = July 19th
Checked the tracking online = August 1st
Re-delivered = August 5th
If I had not noticed it for 5 more business days it would have been sent back to China. Anyway, I checked the light down in my basement. Super bright compared to my Lumina 650. Best comparison I can make is that with the KD unit set on Low, it was brighter than the Lumina on Medium (400 lumens according to manufacturer). If I had to eyeball estimate the differences on High, I would say that the KD is close to 2x the brightness of the Lumina (for whatever that is worth).
Unit = 135 grams (as advertised)
Battery = 250 grams
2 battery straps + large O-ring = 8 grams
Was leaning towards not riding today, but I guess that's out of the question. Gotta see how this looks on the trail. Eventually I will do run-time tests and beam shots (on the trail).
So did you order from KD or DX? You seemed to get it pretty quick.
Originally Posted by steelhmr
Could you tell us your impressions of the beam shape of this light. Someone else described this a being more floody than the Solarstorm X2, which has been described as somewhat spotty. Of course there was disagreement on that point, but without beam shots it's difficult to compare the two.
Originally Posted by varider
This was from KD. The pattern is definitely more of a flood, at least compared to my Lumina. Of course, that has to be expected to some degree since this is a 2x LED setup. I took it out on the trails last night. My time was limited so I only rode one trail. Also, I left right at sunset and got on that trail about 10 minutes later. Even though the sky was cloudy, there was still a lot of light hanging in the sky. I ride in the forest though under foliage so it still was too dark to ride without a light in there.
Setup was the KD on the bars and a Lumina on my helmet. Very soon in I could tell that I was going to have to order another light for my helmet. The KD unit is so bright and full that it completely dwarfs any light being output from the Lumina. As stated earlier, it was far from 100% dark so I didn't get the maximum contrast needed to see the full effect of the light. What I did see was very impressive though. I am very very pleased with the light output. I will try to get beam shots sometime this month. Hopefully sooner than later, but gf is leaving for out of town today for a week and I'll need her to help me with this project.
Just to be clear, this is the exact one that I ordered: http://www.kaidomain.com/product/details.S021521
Runtime Test #1
1. KD light facing the air AC/heater in my office.
2. Fan (only) running parallel to the front profile of the unit.
3. 75-80 degrees F in the room.
4. 'High' setting used throughout entirety of test.
Before I go into the results, some notes about the test. As seen from the setup, the light was going into the fan so I could not see if the light pattern was changing. That being said, everything was running smooth. At the 2:07 mark, I went up to see if the unit was getting hot. I swear right when I was putting my fingers toward it, something happened. The light on the UI changed color or shape (bear in mind that I am colorblind so these changes are not as obvious to me). It continued to run and I didn't notice if the light output changed or not.
It kept running and I cycled the UI to confirm that I was indeed still on the High setting. As it was approaching the 3-hour mark I again cycled through the UI, but it was again still on high. After the 3-hour mark I feel that I definitely noticed that it was not as bright (this was the first time in the test that I specifically noted a change in brightness). The unit continued to run until at the 4:15 mark when I had to leave for the day (obviously I wasn't expecting it to run this long). The light output was very weak by the end, but again, it was still running. I even cycled the UI one more time and it was still on High, as the other two setting were significantly weaker.
In summary, I will have to test again and set it up so I can distinctly notice if the brightness changes. So far though, I am very very very pleased with the runtime. I need to quantify though when exactly it deviated from the max brightness (I was unaware that it was built in to run on lower power when the battery was low).
Great info. Thank you.
Since this runs almost four hours, which is about the same as a Magicshine 808, I don't see how this could be any brighter.
It ran for 4+ hours, but the light output at the end and sometime before the end was significantly lower than the maximum brightness. Again, I will have to quantify that with future tests. Also, this is a clone of the 880 with U2, not the 808.
Originally Posted by varider
What I was trying to say is that they use the same amount of energy in the same amount of time. The two xml led light can't be putting out that much more light than a single xml led light if it drains the battery in the same amount of time. That's if the leds are the same between these lights and driver efficiencies are the same. Of course this light uses the newer U2 light, so that already makes it different.
To make the light brighter it needs to drain the battery faster. The 880 comes with a 6 cell battery and it only runs 2.5 hours on high.
Anyway, I think this light probably has a better beam than a single xml (808 or clone) light. That makes it useful in my opinion. It probably not as bright as bright as the 880 though. I do like the two button user interface.
The description of these lights have always been works of fantasy that merely hint at the truth . I quite like the lights, but you would do well not accept any of the words and numbers in the description as being true. Just to give a few back of the envelope examples
- The specs published by the manufacturer, Cree, on an XM-L U2 is that the maximum lumens produced at 2.8A is 936 x 2 = 1872
- The only Cree LED that produces 1100 lumens at 2.8A is the XM-L2 and that's in the U3 bin.
- The power consumed by an XM-L at 2.8A is 3.33V x 2 = ~18.6 W per hour. The available power in a li-ion cell rated at 2500 mAh is ~9.2W x 4 = ~37W. Even if you could achieve 100% efficiency, 4 x 18650 2500mAh cells could only run at 2.8A for less than 2 hours.
- to run 2 XM-Ls at 2.8A for 4 hours would require in excess of ~75W or more than 8 x 2500 mAh cells.
- working the other way, if 4 x 2500 mAh cells last 4 hours 37W / 4 hours = 9.25W per hour / 2 XM-Ls = 4.625W = ~1.45A each XM-L = less than 570 lumens each XM-L.
Given these "exaggerations" in the description, I would be utterly astounded if they were using XM-Ls from the highest (& most expensive) U2 bin.
One of the things I like about using 2 LEDs is that they produce more light per watt than a single LED driven at twice the current. Ignoring questions of driver efficiency,
- 1 XM-L U2 at 2.8A = a maximum of 1067 lumens
- 2 x XM-L U2 at 1.4A = a maximum of 623 x 2 = 1246 lumens
Like I said I quite like the lights, just don't believe the words & numbers quoted for them.
I don't feel that I am explaining myself well enough. This unit DID NOT produce the maximum brightness for the entire 4:15. For all I know, and I am hypothesizing here, it ran with max brightness for 2-3 hours and then switched to some emergency power-saving mode for the rest of the test. I am not equipped to perform the test well at work since the setup I used directs the light into a fan instead of a wall. That is to say I can't see the beam well enough. I will re-test this with pictures at home to show what I mean when I say the the light output appears to change.
Additionally, Kir measured the current of this unit on high at 2.36A, and not the 2.8A listed on the site.
The only thing one should believe about these Chinese lights is, the housing will be some kind of aluminum and it will have some kind of LED(s) in it.
Originally Posted by find_bruce
Yeah, that's the question I really had. I know the current to lumen curve isn't linear, so I thought there would be an advantage to using two leds instead of one. It's good to have the actual numbers, though. Thanks.
Originally Posted by find_bruce
So how does this extrapolate out to three or four leds? Those are becoming more prevalent in the chinese market.
Ok, I get is now. It doesn't last 4 hours on high. I wonder if Kir measure the current going through the leds or coming out of the battery.
Originally Posted by steelhmr
Well I'm struggling with find_bruce's numbers, but from what I gather this light is probably emitting 1100 lumens or more and lasts less than 2 hours on high. If Kir measured the led current, than it's emitting way more than 1100 lumens (ignoring driver efficiency).
I think this light is probably a good buy, if the beam is as floody as I imagine.
I'm the only one here with basic electronics equipment and some chinese bike lights? :)
All these numbers in the description of all chinese lights are very exaggerated. For example Fenix BT20 supplies 3A to LED on turbo mode and specifies 750 lumen output - that sounds correct. But these chinese "1200-1800" lumens from single xml are obvious lies. Same with "6400" mAh batteries, you will never get more that ~4000, maybe 4400 from a chinese battery.
I will measure the current on leds/battery in a few days/weeks...don't really have any spare time with work and bike riding and some other stuff like flying R/C helicopters :)
For now I will explain the change in brightness. Yes, this light (and probably most other 2 XML lights) will have lower brightness in end of the battery life. The reason is simple - it uses step-down/buck driver with 2 leds connected in series. When battery (input) voltage drops lower than the voltage of 2 leds in series (Vf, depends on current/temperature/bin, usually about 3.1-3.2V per led, 6.2-6.4v for 2 leds) driver will slowly stop working and the current on leds will slowly drop to zero.
Frequently Asked Questions its expained in more details here (2nd question)
Buck converter - Wikipedia, the free encyclopedia and here.
So basically when battery voltage drops to ~6.8v and lower - leds brightness will begin to drop too. Then there will be a linear drop in brightness until battery is fully discharged.
And thats actually a good thing - you will get slow brightness decrease, longer battery life in the end of the discharge cycle, noticable indication of discharged battery, etc.
Imo this slow brightness drop is much better than sudden cutoff in 1 XML lights which have full regulation in entire discharge cycle.
Here is a quick video to show how it works (sorry for the quality, used my old cellphone). Numbers on the left are current draw from PSU, numbers on the right are output voltage, D-50 light in middle mode. This is a very simple DC psu, so numbers aren't very accurate - but good enough for demonstration.
As you can see light output is fully regulated until about 7v, at which point light draws 1.4A from psu. Then as I decrease the voltage - current/brightness drops too, driver slowly stops working and then completely turns off at 5v (another good thing, light will turn off even if protection pcb in battery will fail to activate).
If I increase/decrease voltage between 5 and 7v - brightness will change too.
So thats what steelhmr noticed in his discharge test, sorry for the kinda long explanation ;)
Run Test No. 2
1. KD Light facing away from fan.
2. 82 degrees F.
3. 'High' setting used throughout entirety of test.
1. I started this test too late.
2. 2:09 of runtime until UI lights changed color.
3. 2:45 abandoned test as I needed to sleep.
4. Some change in light output noticed between 2:30 and 2:45 (might not have come across as well in the pics).
Note: Unit was facing a wall with a mirror during the test so I turned unit towards a different wall for pics. Sorry for mild inconsistency in pics. Just used my cell phone (Samsung S3 with 'low-light' setting). Wall was 13' from the light. I believe you can see the light output drop slightly between 2:10 an 2:30 (look toward the outside of the pattern). Same for 2:30 to 2:45. I really just wanted to see when the first change was on this test. Images are probably best viewed by opening first link and then hitting the left arrow on the pic(s).
(Sorry for living in the stone age with photo bucket. I'll try to host these somewhere else as I just noticed the ads, yeesh.)
My thoughts about the test done by Steelhmr: Something very odd going on. Even with auto-power down no two-LED lamp should run for 4hrs unless the lamp was on mid-mode at the start and then powered down at some point. Even then the output of the lamp has to be called in question. Typical run times on high should be between 2.5-3 hr. and that with good batteries. Even with the lamp on mid-mode 4hrs would be excellent run time.
Nice that this lamp has auto-power down. Most Chinese made lamps don't have this feature.
Well, I'm not keen on running too many more tests, but for the next one I will setup a better location where I don't have to reposition the light for pics. Additionally, I will take a reference pic with the Lumina 650 to compare it against. One comment about the pics, the hot spot comes across exaggerated in those pics. In real life the unit doesn't cast a virtual full moon against the wall.
Something about the 880 clones that I noticed tonight while looking at both the Kaidomain and D/X web sites.; The K/D models claim to have a 2800ma driver, the D/X ( Ultrafire D-50 ) claim to have a 1800ma driver and output 580 lumen. This could be a misstatement on D/X's part but there is really no way to know for sure.
All things considered, if i was going to buy one I would most likely buy from Kaidomain because theirs has the more powerful driver. Then again if Kaidomain runs out of stock who's to say they don't restock with a lamp that might have another lower-powered driver. Stuff like this happens. If Steelhmr has one with a less powerful driver than all his tests are with a lamp that is outputting something around 600 lumen. I'm just mentioning this because it is a possibility.
Originally Posted by Cat-man-do
Isn't it equally possible that they just supplied this light with a very strong battery pack? It essentially runs on high for slightly over 2 hours and then gradually fades out. It's hard imagine that this is a 600 lumen light since I have a Lumina 650 (rated and tested at 650 lumens) to directly compare it to. The KD unit blows away the Lumina in terms of brightness (and every other property one would want in a light). Sorry that I don't have the equipment to test the output, but Kir already tested it and gave his findings on the first page of this thread. I'm just reporting my experience from a less electronics-savy and more from a user-experience perspective. My next task will be to take shootout-style pics with the proper camera and terrain settings. That should tell a better picture (no pun intended).
Possible, well anything's possible. Probable, no way. These are cheap bike lights that have cut lots of corners to get a very low price point. Great value all in all but not top level output nor top quality components.
Originally Posted by steelhmr
I have not worked on this particular light model but have repaired a couple MS lights so I have seen how cheap lights are made.
Sorry, I didn't want it to sound like I didn't believe what you were saying. If you say it blows away the Lumina I believe you. At least it's nice to hear positive feedback even if the results are somewhat unusual. All things considered if I were in the market for a good Chinese made lamp I might choose one of these ( or one of the Solorstorm X2's ) Good positive feedback from both lamps.
Originally Posted by steelhmr
Originally Posted by Cat-man-do
No offense taken at all. I can only report what I have observed. I share the same tempered enthusiasm for the light. During the first test at 2:00 I was all fistpumps. At 2:30 I think let out an audible "woo-hoo". By 3:00 I switched to "hmmm" and at 3:30 I knew something wasn't adding up. On the 2nd test my goal was to see if the light output visually changed through the first two hours on the high setting and my observation is that it did not. After 2:09, it does begin to change but I can't begin to quantify it without any additional electronics equipment. I hope that when I can produce shootout pics it will be obvious to everyone that this light is putting out a very respectable beam, at least compared to the Lumina 650. And before any questions begin to surface about the integrity of THAT light, I have two of them to use for comparison ;)
looks like a tactical flashlight beam pattern. aren't there bike mounts for Fenix tactical flashlights ? you would be able to use your own 18650 cells and charger, and you could take it hiking, or look for stuff under the bed, or survive a blackout.
its crazy though how they were able to include so much in the package at the price - it's like they are selling the product for the cost of what the components cost - how do they make a profit ?
the Chinese never cease to amaze me - from the amount of clones they put out you would have guessed it must have taken 10 planets the size of Earth to produce it all.
and considering the real products that they are cloning are also made in China it makes you wonder how are they able to undercut the real deal by 4X in price ? the Chinese seem to defy any sort of rational analysis - or maybe i'm just not smart enough to understand them.
i do think the established companies suffer a bit from over-engineering. since they don't compete on price their engineering efforts end up directed at problems that don't really exist. this is especially true of German companies who spend 90% of their energy solving imaginary non-existent problems. for example the Porsche 918 has 55 computers just to control the drivetrain ... apparently somebody in Germany was bored.
I just want to say that I wouldn't take too much away from those pics in terms of the actual beam pattern.
Originally Posted by androgen
1. I mainly used to pics to compare among themselves as to whether the brightness of the light was changing over the course of the 2nd test.
2. It was indoors with white walls from 13 feet away.
3. The pictures seemed to exaggerate the contrast between the hot spot and the rest of the beam. I don't feel that it is as drastic of a difference in person as it appears in the pics.
Perhaps it does mimic that of a tactical beam pattern to some degree. I just wanted to state that my purpose of those indoor pics was simply to compare the brightness during different times of the test. I'll try to get outdoor pics soon, but I just had a crash today and (slightly) bent my rear rim so it just dropped down a notch on my to do list :mad:
Wish they sell the head only :)