• You must be a Supporting Member to participate in the Candle Power Forums Marketplace.

    You can become a Supporting Member.

The New Luxeon III Generation Issues

Kiessling

Flashaholic
Joined
Nov 26, 2002
Messages
16,140
Location
Old World
This thread is about data points regarding the new generation of LuxeonIII LEDs with the different placing of the LED die resulting in a different photon management.
As I don't know anything about it, I'll re-post Don's comments of some other threads in this one place here so that we have a good starting point for further discussion.

bernie


Don said:
These PD's have the "new style" Luxeon LED's and I have recently had it come to light that the "new style" does indeed have a new image location relative to the package and consequently the beam is different than that of the previous LuxIII. :banghead: :mad:

Fortunately for me and my customers, the only lights I have built with "new style" have been the HD's and now this group of PD's. The good news is that the beam is OK in terms of white wall. It is actually better if you are like me and prefer more light in the corona and more of a divergent flood beam with the spot. It is not good news if you want max lux and throw from the light!

I saw a post a few weeks back about the revised LED with the new round heat sink and apparent "lowered" die but there was little follow up comment and it was also shown, as I recall, that the new style was working fine with existing optics and lights. I will post my preliminary observations in a new thread here in this forum later tonight if I can get to it. In the mean time, I have discovered that it would appear that the reflector's curve needs to be dropped down about .015" to .020" to return to focal allignment. This is a PITA as the reflector's geometry as well as the heads geometry is not open to such realignment without some modification! In the case of the PD, I need to shave off the rear end of the reflector as well as shave off a shoulder on the reflector that is designed to reference off a mating shoulder within the PD head.

As is, these lights here will give you a brighter corona and larger spot but with less definition in spot and less intensity in the spot. Ironically, this is more in keeping with recent reflectors I have been designing but none the less, this comes as a rude surprise that was not bargained for!! :rolleyes:


Don again said:
The flux or luminous output of these PD's will be exactly what it should be with these U bin LED's. However, as assembled, the beam from these will favor a softer blend in hot spot to brighter corona as compared to if they were assembled using what one would assume to be the "standard" Luxeon. The lux of these lights takes a hit as a result. I have tested a couple set up and one was in a very controlled test where everything tested was the same with the only exception being a modification to the same exact reflector and then a re assembly. In the case of the 20 mm reflector, you can expect about a 50% increase in lux if the reflector were put in sharper focal allignment with the LED. On the other side, consider this about a 35% reduction in lux from what it could be. I.E. 1000 lux in current flood/blended state and 1500 lux in sharper spot with reduced corona intensity. Since this batch of PD's is the only set of PD lights I have used the new style Luxeons in, I am willing to offer these lights either as is (good flood beam and better than normal for corona intensity) or I will take them apart and modify the reflector to match this new geometry.

This is ultimately going to get pretty confusing and I know that I have no desire to spend time cutting perflectly good reflectors and especially when I prefer the beam that results with this new "de focus". I don't consider the PD as a light designed for throw and I think this new distribution of light is actually an improvement over the properly alligned reflector with previous "standard" Luxeon. I have been playing with a pair of PD's with the only difference being the reflector adjustment and the difference is visible and the foreground and mid distances are improved with the missalignment but the distance is reached much better with the adjustment made.

This alert is only in regards to these lights here and all of the previously build PD's, inclusing those with U bin LED's were of the "standard" Luxeon and not this new joker! Those of you who have already been shipped one of these have the option of a reflector modification if you choose and any of these to be purchased come with the option. Beyond these, I have not decided what I am going to do but since the TY0J and TW0J LED's I am building with are of the "standard", the point is moot.


Don can't stop said:
However, after looking at the lux measures of this powerful UX1J behind the 27 mm reflector, I knew something was goofy! I had noticed that the spot was rather smooth and not sharp but didn't give it any thought. Well, the LuxIII in this CPF standard is of the new packaging which has been briefly mentioned but not really explored to my knowledge; or at least really delved into here on CPF. To date, I have only used the new style LED's, to my knowledge, in my HD45 and very recently, some U bins in some PD's of color. I am now aware of the fact that these LED's do in fact have an image location different than their predecessors and focal allignment is not the same!!! :banghead: I have attributed the less than stellar lux measures of the HD45 to the orange peel of the reflector when it now seems that a possibly stronger influence is the fact that the reflector is in effect forward of the focal point by probably .015" to .020" at a guess. Ironically, I have been trying to design and effect optics which cause the beem to diverge more and get more light in the corona and it seems that Lumiled's, in their infinite wisdom has implimented a push in the same direction without any alert to us lowly customers! The good news to flood lovers and white wall hunters is that at least in the case of the lights I have been making, the beam will not disappoint nor does it look bad. However, if you are in fact looking for high lux measurements and long throw, these new LED's will require modifications to optics designed for the previous packaging and/ or placement changes within their hosts! WTF!!!

I have been sitting on this post while I did a quick build and light test. I put one of the new version (we need a friggin name for these damn things!) Luxeon's (UX1K) in a Light Engine with DB917 driver. I put the LE in an Aleph 2 head with stock reflector and no window and put it in the integrating sphere as well as took a lux measurement with it. 80 lumens and a lux of 1030. I then took this same reflector and faced .015 off the rear of the reflector and put it back with the same LE and again took flux and lux measures. 80 lumens once again but the lux jumped up to 1525. There was essentially an increase of 50% in the lux measure! There is no loss of photons but the deployment is certainly effected! :rolleyes:

A quick and late query to the Future Reps I have been in contact with resulted in a comment of essentially, news to them! I suspect that there is going to be mixed inventory coming down the pike in production units and some manufacturers may catch wind of something not quite right or the same from customers who may notice something "off" about the lights. :shrug: In addition to the fun lottery, Lumileds it would seem has added blind siding of the customers as an additional bonus and part of the game. :crazy:
 
Last edited:

McGizmo

Flashaholic
Joined
May 1, 2002
Messages
17,290
Location
Maui
OK, thanks Bernie! This issue goes well beyond the McGizmo forum I suspect and we need to find the previous thread where one of our members from off shore posted his observations about this as well as some photos and perhaps a link to Lumileds documentation?!?! I am supposed to get a technical response back next week from a fellow at Future. Frankly, in my limited experience with this new "joker", it seems to me that this LED is different enough in its composition and image location, that it is very disrupting and misleading for it to have been slipped in as the same part as the previous Luxeon. I could be wrong and off base here but I sure see a dang difference in the small population I have messed with!!!! :rolleyes:

The fact that Lumileds may now designate the image location or geometry is somewhat meaningless if they don't also provide the previous LED's image location. :shrug:
 

Kiessling

Flashaholic
Joined
Nov 26, 2002
Messages
16,140
Location
Old World
I found this picture posted by chimo in this thread:

http://candlepowerforums.com/vb/showthread.php?t=93139&highlight=luxeon

newslugs6gt.jpg


bernie
 

whammy707

Newly Enlightened
Joined
Sep 6, 2005
Messages
13
I KNEW something was wrong with my newest light (not one of yours Don) and now that I know what it is I can fix it. Thanks Don for finding and letting us know this.
 

Sway

Flashlight Enthusiast
Joined
Aug 25, 2003
Messages
1,721
Location
North Carolina
This is a most excellent thread and confirms some of my suspicions, the last Quad-Lux I made using UX1L emitters with the new round sink had more flood and less throw. This effect was noticeable using the IMS20 and the McR20 reflector's.

Now I can get a good nights sleep and know it's all not me, thanks! :thumbsup:

Later
Kelly
 

McGizmo

Flashaholic
Joined
May 1, 2002
Messages
17,290
Location
Maui
Cool guys! It would be interesting and helpful to know which lights are turning up with the jokers and how they are behaving. Unfortunately the best clue as to which are the jokers is seeing their sinks and once mounted or on a star, one doesn't have a visual clue unless you can view the LED from the side. :shrug;
 

McGizmo

Flashaholic
Joined
May 1, 2002
Messages
17,290
Location
Maui
Thanks Larry!

I expect to see smooth reflectors and optics which are designed specifically for throw and "true" image projection to be most effected by the joker and its new image location. Ironiclly, some might come to appreciate the benefits of more of a flood beam and greater distribution of light beyond the spot due to these LED's but in principle, the defocusing or movement out of focal allignment should be done intentionally and not as a result of an unannounced change in the "rules".

I just tried a joker with unaltered McR-27L and I am realy happy with the results!
 

HarryN

Flashlight Enthusiast
Joined
Jan 22, 2004
Messages
3,976
Location
Pleasanton (Bay Area), CA, USA
Hi Don - I feel your pain - no really. I had a batch of custom reflectors (12.5mm for my cr2 light) built specifically for the "old normal" Lux III s I had. With small optics comes tight tolerance requirements.

I am sweating this one out.
 

Kiessling

Flashaholic
Joined
Nov 26, 2002
Messages
16,140
Location
Old World
Interesting comment in the above-linked thread:

wasBlinded said:
I recently received a LuxV which has a different slug shape than an older one. I didn't try to measure die height, but I put it into a U2. This U2 had a donut hole in the beam with the original LuxV, but not with the new LuxV. That would be consistent with a change in die height and consequently the light focus.

It does seem that some of the K2 technology is being slipstreamed into the existing Lux3 and LuxV lines.

Warren stated that LumiLeds had made an announcement in March 2005 about it, but only for the LuxI.

bernie
 

wasBlinded

Flashlight Enthusiast
Joined
Dec 14, 2004
Messages
1,222
Location
Oklahoma
In my unscientific comparison, using the MarkIII eyeball, the new LED in the McR20 reflector results in a a hotspot that is larger and not as intense compared to the older, higher die height Luxeon.

In the Aleph3 the new LED doesn't seem to make much of a difference in the hotspot size.
 

hifipenny

Newly Enlightened
Joined
Oct 17, 2004
Messages
3
Location
Dalian, P.R.C
Hi Don

Has you tried this "new style" emitter with the MCR-19 reflector? I am interest with the performance!

thanks,

Penny
 

NewBie

*Retired*
Joined
Feb 18, 2004
Messages
4,944
Location
Oregon- United States of America
So, for most folks, is this going to be just a reflector tweak?

I did notice on the Lux I drawing, they were showing the slug sticking out the bottom at 0.25mm vs. 0.2mm of the Lux III drawing.
 

HarryN

Flashlight Enthusiast
Joined
Jan 22, 2004
Messages
3,976
Location
Pleasanton (Bay Area), CA, USA
Hi Newbie - it depends what you call a "tweek". If you have to modify several hundred custom made, vacuum aluminized reflectors, with delicate interiors that are made to exactly fit, then that begins to seem like more than a tweek.\

I usually go by the data sheets, and perhaps I am blind today, but I certainly did not see any notification of a change in focal point. (in the data sheet) I have somehow missed this entirely in their docs.

Interestingly, if I am reading the data sheet mechancal diagram correctly, the die height is not specified.
 
Last edited:

gregw

Flashlight Enthusiast
Joined
Jun 7, 2004
Messages
1,511
Location
Hong Kong
From the first post, I guess that this also means that to keep the original focus with the new Luxeon III on a McR20, I would have to file off 0.015" on the bottom of the reflector.. right?
 

NewBie

*Retired*
Joined
Feb 18, 2004
Messages
4,944
Location
Oregon- United States of America
I just got looking at the pictures, and combined the two emitter halves into one:

chimo.jpg


Not only is the die a different height from the slug bottom, it is also optically oriented differently within the dome "lens". It appears to no longer be oriented in the center of the half dome lens. This too could be corrected, but would need a prescription to be applied to the reflector.

Anyone got better, and clearer, and larger more detailed high resolution photos of the two, so we could make a better comparison?
 

McGizmo

Flashaholic
Joined
May 1, 2002
Messages
17,290
Location
Maui
Hi guys,

Like Harry said, this goes a bit beyond "tweak" as a .015" reduction in the component stack is typically not within the tolerance range of the build design without some effect on assembly fit and function. The fact that the optic or reflector needs to get closer to the LED is more problematic than if it needed to be shimmed away from the optic/ reflector. In many designs, the optic/ reflector is already in contact with the LED and further, it may be seated on a fixed shoulder or mount in the head. Material may need to be removed from the optic/ reflector both on its rear face where it contacts the LED as well as a reference or mating surface corresponding to its fixturing in the head.

It is very difficult to turn down a reflector without having the chips contact and scratch the delicate reflective surface. Even holding the reflectors for the modification process can require custom collets be made.

Moving forward, the obvious solution is to design new optics and reflectors based on the new geometry as well as intended beam patterns. The fact that the die is now lower in the package actually allows for a slightly greater range of focal length solutions. The existing deeper reflectors are more forgiving in that they will provide reasonable beams with a greater range on the Z axis. The orange peel also allows for slop on the image focus adjustment.

Due to tolerances which need to be allowed in the components, I have been fudging in the direction of having the image at or behind the focal point of the reflector all along. This .015" displacement goes beyond what I had in mind but is not catastrophic. It is enough to significangly effect the collimation and lux of the design though!! The smooth reflectors and optics designed with the intent of projecting a sharp and "tight" image will be most effected and the difference will be obvious to anyone with a sample of previous and the new Joker in comperable lights.

hifipenny,
The McR-19 flush against the Luxeon has always been a case of the image being well behind the focal point of the reflector. The LuxIII in this set up provided a good flood with a sort of starburst artifact in the center. The joker now provides a small spot of obvious brighter intensity but more in keeping with a "typical" beam in appearance. In short, the joker with the McR-19 is an improvement and actually quite effective, IMHO. For close up inspection work as well as near to mid range concentrated flood, this is a great package! From physical contact eith the target on out to meters away, the spot is pure. Many lights have a null or black + when they are too close to the target; no so with the McR-19. I think this particular combination has now become a more acceptable and viable option.

gregw,

Filing the rear of the McR-20 down by .015" will allow it to mate properly with the LED provided the reflector is not constrained by a reference stop or shoulder seat in the head.

******

Just got a call back from a couple Future guys. They were completely unaware of this and preliminary speculation is that since Lumileds has never stated image location, the part is quite possibly still within specification. Yeah, I can understand this and I also know that the allowance for image movement has now been given a new and lower starting point and the change not only can be measured but seen as well. :shrug:
 
Top