someone had to do it: dedomed Nichia219highCRI :-D

48 posts / 0 new
Last post

Pages

djozz
djozz's picture
Offline
Last seen: 3 hours 23 min ago
Joined: 09/07/2012 - 17:04
Posts: 15781
Location: Amsterdam
someone had to do it: dedomed Nichia219highCRI :-D

There is absolutely no point dedoming this wonderful led with the best CRI around and beautiful neutral tint, causing the tint to go warm and the output to go down, other than: I was curious about it Smile

I had a 219 leftover from the DIY-built contest this summer that had lost the minus solder pad. However there was a small shred going to the inside of the led still hanging about so I reckoned that if I reflowed it, the solder would adhere to that bit and the led should work. So it was a fine led to try dedoming it. I did it with a scalpel while the board was still hot from the reflow (on youtube it is in HD):

So that worked well, no phosfor left on the dome at all. The feel was different from Cree leds though, the Cree silicone is more flexible, gives way and comes off slowly, this dome just snapped off, feeling a lot stiffer. this was the result:

The led-tester showed that it worked (joehoe!), I mounted the board on my usual block of aluminium with a reflector on top, soldered wires and connected the power supply so I could do some beam comparisons. I do not show white wall shots because they give the wrong idea about the tint. The tint temperature is in between the stock Nichia 219 and the xpg highCRI 2700K, a bit more on the warm side of that so I'd say 3200K. As you would have expected there is a bit more yellow in it, but not at all as yellow as some Cree leds go after dedoming, this is still a very nice tint. I did some beamshots of a pile of laundry lying about (mainly my 2year-old's Wink ), the differences are subtle and not exact the true colours you see in real, and the beams are also not exactly the same type, but you may get an idea. Top is Nichia 219 (a UF C3 ss mod), middle is Nichia219 dedomed, bottom is XP-GhighCRI2700K (4sevens miniAA highCRI).

That was fun, the led will even find its way in a flashlight some day Smile

Edited by: djozz on 09/17/2013 - 18:22
GottaZoom
GottaZoom's picture
Offline
Last seen: 1 year 1 month ago
Joined: 04/12/2012 - 03:14
Posts: 2179

Center of the dedome looks a bit more intense, wonder if you will find it throws a bit more than domed, and if this makes it a better aspheric candidate.

Pulsar13
Pulsar13's picture
Offline
Last seen: 5 months 3 weeks ago
Joined: 02/05/2012 - 10:23
Posts: 1909
Location: Malaysia

Thanks for the post. Now I don’t have to waste $5 for my own curiousity, nor do anyone else Wink

Seems the CRI visibly drop after the dedome. Too bad. And I bet the hotspot would be more centric and less floody, which is also NOT what I would want in a high CRI light.

Seems Nichia 219 is already quite perfect as it is – with the exception of low current. If only they’d take the hint and put bigger die in there…

MRsDNF
MRsDNF's picture
Offline
Last seen: 5 hours 23 min ago
Joined: 12/22/2011 - 21:18
Posts: 12965
Location: A light beam away from the missus in the land of Aus.

Nice mod djozz. How’s it compare outside?
Its good to see your two year old is as good as my 16 to 20 year olds. At least a two year old has an excuse being mum does everything. Thinking about it mine, their mum still does everything for them. Funny nothing changes. Thanks for the comparison.

 

djozz quotes, "it came with chinese lettering that is chinese to me".

                      "My man mousehole needs one too"

old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".

Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.

 

flashination
Offline
Last seen: 9 months 3 weeks ago
Joined: 09/19/2011 - 04:08
Posts: 828
Location: Texas

Who will be the first to redome after a dedome? Wink Silly

 

 

 

 

 
 
'I pledge not to get drawn into cpf/blf arguments, & just understand that both places have their own merits'.

RaceR86
RaceR86's picture
Offline
Last seen: 3 years 7 months ago
Joined: 02/22/2012 - 08:32
Posts: 3777
Location: Norway

Djozz, sounds like you had a perfectly good reason to de-domed one! :bigsmile:  I might de-dome one of these some day too just to see it myself. I have been thinking about it...  Smile

I think your picture comparison have limited value as a CRI comparison. The Nichia 219 picture is exposed darker, making all colors deeper and nicer. With same exposure I seriously doubt no one could tell them apart in terms of CRI.

What Pulsar13 thinks he is seeing in CRI difference is just the exposure difference as far as I can tell.

Thanks for sharing various experiments! Highly appreciated! Smile

BLF LED database – collaboration spreadsheet and latest news about where to buy LEDs
http://budgetlightforum.com/node/19342

Chloe
Chloe's picture
Offline
Last seen: 3 years 2 months ago
Joined: 01/30/2013 - 14:14
Posts: 4595
Location: 联合王国

Thank you for trying this and sharing your findings!

ImA4Wheelr
Offline
Last seen: 1 hour 53 min ago
Joined: 02/03/2013 - 14:51
Posts: 7909
Location: SC

You seem to always be pushing the limits and offering potential sacrifices to the almighty emitter gods.  Not sure they appreciate being teased so though. 

Looks good to me.  Looks like you may have started a new trend.  I could see some dedoming some of these for more throw.  Like that 12 Nichia emitter SRK.  It has flood in spades.  One could make one of those RaceR86 style and dedome half of the emitters.

djozz
djozz's picture
Offline
Last seen: 3 hours 23 min ago
Joined: 09/07/2012 - 17:04
Posts: 15781
Location: Amsterdam

I find it difficult to registrate flashlight output quality, with my own eyes I see quality differences (tint,CRI) better than it appears on photographs, and even then it is very subjective (when I switch on my Nichia219 light  always feel how nice and clear everything looks, but when I try to point at the several colour differences compared to a non-highCRI light,  every individual difference is very subtle, while the impression of the whole beam is quite clear to me)

Something I can still do is make pictures of the different emitters shining on a colour checker cart. Don't expect miracles from that, the differences will also be subtle.

If I was texaspyro I could put just the emitter in the sphere and it would just spit out tint and CRI.

Perhaps the best I can do is put the emitter in a flashlight and report a subjective impression, my impression for now is that the CRI is not at all that bad Smile

djozz
djozz's picture
Offline
Last seen: 3 hours 23 min ago
Joined: 09/07/2012 - 17:04
Posts: 15781
Location: Amsterdam

I have put the led into a flashlight, but first an issue, look at this picture of the led, I could have noticed it directly after the dedome, but I just noticed it today:

The die is highly reflective, so it looks like there is a silicone film leftover on the die (or those Nichia dies are different from Cree dies in every aspect). And that tells me that I just may have been a lucky b*stard with this dedome; will every dedome leave that film, ot will the next one just take off half of the phosfor like in most reports of (accidental) Nichia dedoming? And would the tint-shift be different (more) if all the silicone was removed? Time for a gasoline dedome! Anyone?

Anyway, the led is in a AA-host I had waiting idle for a while already (from cnqg), with a BLF-mini driver (boost on AA with meh output, sort of direct drive on Li-ion, 2%-25%-100%, no memory, ±1000Hz PWM). I have a Nichia 219 mod (with dome) with the same driver and AA-size, so although there are differences (e.g. the dedomed mod has a AR-lens, that is 5% more light compared to non-AR), some sort of comparison can be made:

 

I tested with a fully loaded (brand new from IS LOL Kinoko 14500 IMR battery. I would have expected the same current draw from both drivers, but the light with domed Nichia drawed 1.35A, the one with dedomed Nichia 1.46A, so there's a small difference. The output of the light with dome is 215 lumen, the output of the light without dome 223 lumen. Hey, that is not much different, even considering that the dedomed light draws a bit more current and has a AR-coated lens. It can have something to do with the silicone film over the die, but the 1000K tint-shift tells me that there must be a considerable light loss. The throw of the dedomed light is 5.4klux at 1 meter, the domed light has half of that (as expected from the about double effective surface area of the die).

I have walked around with this light for a day now, and I am not impressed by the output (my current EDC-light is considerably smaller and has double the output on a comparable current), but I am delighted by the tint Smile Smile , it gives everything a sunny day feeling, and I suspect that the CRI is still fairly (or very) good . It is also nice that it is a bit more throwy than most of my small lights.

djozz
djozz's picture
Offline
Last seen: 3 hours 23 min ago
Joined: 09/07/2012 - 17:04
Posts: 15781
Location: Amsterdam

Hmmm, lost contact in this thread, talking to myself now. Well, I do that a lot anyway, kind of used to that Laughing.

I tried to make pictures of the colour checker card with different leds, and after putting them on the computer I corrected the colours on the monitor, I failed in that for several reasons: the exposures are not corrected well enough, the colours do not match the real life colours enough (you just can not catch that with a camera, the worst are bright red and bright yellow, can not get them the vivid colour of reality) Further, if I change the angle that I view my monitor at, all colours change again Cry And I do not even suffer from the bleach colour representation of the crappy monitors you out there are looking at now Tongue Out . To put it all even more in perspective: on my computer screen I like the cooler colours more, in reality the 'warm neutrals' look much better to me.

I get you the pictures anyway because I spend too much time on them to waste it, and although the colours are not the same colours that you see with your own eyes, they do show some of the differences between the leds. Here they are:

EDIT: I redid these charts with a better camera in this new thread.

ImA4Wheelr
Offline
Last seen: 1 hour 53 min ago
Joined: 02/03/2013 - 14:51
Posts: 7909
Location: SC

Very nice pics.  Thank you!  I will study them later as I have to run.

Yeah, it's kind of sad.  Someone makes a contribution, the thread has a high read count, but only a few responses.  It is especially true of non-glowing reviews of lights.  You know they are helping a lot of folks not waste money, but only a couple members bother thanking the OP for the effort.  I'm a huge fan of you substantial contributions to the forum.  You are appreciated by many.Smile

dct73
Offline
Last seen: 7 months 2 weeks ago
Joined: 06/29/2012 - 21:40
Posts: 384
Location: wisconsin

Don’t be discouraged. I think your write up is great. Especially the color charts!

RaceR86
RaceR86's picture
Offline
Last seen: 3 years 7 months ago
Joined: 02/22/2012 - 08:32
Posts: 3777
Location: Norway

+1 to what ImA4Wheelr said.

Djozz, you said you corrected the colors. What exactly did you do?

Do you have these images in RAW? (either way, PM incoming)

BLF LED database – collaboration spreadsheet and latest news about where to buy LEDs
http://budgetlightforum.com/node/19342

JimmyTheChooch
Offline
Last seen: 3 years 8 months ago
Joined: 09/05/2012 - 12:33
Posts: 38
Location: The Great White North

Very interesting thread. Thanks for the work and effort you put into it.

“ It was the Law of the Sea, they said. Civilization ends at the waterline. Beyond that, we all enter the food chain, and not always right at the top.” 
Hunter S. Thompson

ImA4Wheelr
Offline
Last seen: 1 hour 53 min ago
Joined: 02/03/2013 - 14:51
Posts: 7909
Location: SC

Without having one of those charts in 100cri daylight included to compare to, it appears to me that the dedomed nichia has the best cri of the whole bunch.

djozz
djozz's picture
Offline
Last seen: 3 hours 23 min ago
Joined: 09/07/2012 - 17:04
Posts: 15781
Location: Amsterdam

ImA4Wheelr wrote:

Without having one of those charts in 100cri daylight included to compare to, it appears to me that the dedomed nichia has the best cri of the whole bunch.

Here you go,

just made a picture (camera on daylight setting, no corrections on screen done this time) with a slightly overcast sky, that gives the best quality daylight imo (that is very subjective of course). But still the camera does not capture the true vivid colours of reality.

ImA4Wheelr
Offline
Last seen: 1 hour 53 min ago
Joined: 02/03/2013 - 14:51
Posts: 7909
Location: SC

Thanks again djozz.  Not sure what to make of that.  Scrolling up and down, it now looks like the xml2 0D is closest to that.  Do you think that is because you used a different setting (daylight) for the picture in Post 16?  I apologize if I am asking dumb questions.  I'm not very knowledgeable in this area (cri and tint).  I do know that, despite it's relatively low output, I love the Nichia 219 B10 H1.

Boaz
Boaz's picture
Offline
Last seen: 17 hours 53 min ago
Joined: 11/07/2010 - 09:31
Posts: 7122
Location: Birthplace of Aviation

Sure...de-doming is easy ..

 

  Lets see someone put the dome back on .:P

       καὶ τὸ φῶς ἐν τῇ σκοτίᾳ φαίνει καὶ ἡ σκοτία αὐτὸ οὐ κατέλαβεν

                            

       Dc-fix diffuser film  >…  http://budgetlightforum.com/node/42208

lightme
lightme's picture
Offline
Last seen: 3 years 7 months ago
Joined: 11/12/2012 - 19:51
Posts: 615
Location: WV

Yeah, thanks for the charts. Now, I know what to expect from my 5B1 Smile . The colors aren’t as vibrant as the nichia but it isn’t as yellowish in the greyscale.

mattthemuppet
Offline
Last seen: 2 days 10 hours ago
Joined: 01/04/2012 - 16:20
Posts: 1268
Location: WA

thanks for all the charts djozz. To me it shows that the 4000-5000K emitters have most of the colour reproduction fidelity of the Nichia 219, but without the yellow tint that the <4000K Cree high CRI emitters have. Although the 219 appears to be the best of both worlds (CRI and tint), for higher outputs the 3-5 Cree tint bins seem to be good enough. I have a XP-G2 5B on a Sinkpad coming from LED-DNA (less than $4 after discount!), so it’ll be fun to compare that to the 3C XM-L lights I have.

GottaZoom
GottaZoom's picture
Offline
Last seen: 1 year 1 month ago
Joined: 04/12/2012 - 03:14
Posts: 2179
djozz wrote:
just made a picture (camera on daylight setting, no corrections on screen done this time) with a slightly overcast sky, that gives the best quality daylight imo (that is very subjective of course)

Using side by side monitors (verifying no significant color shift between monitors), my wife and I both pick the 5B1 as the best overall match – dedomed nichia was 2nd, though each has strengths depending on color. As you say, seeing it in person may be different.

lightme
lightme's picture
Offline
Last seen: 3 years 7 months ago
Joined: 11/12/2012 - 19:51
Posts: 615
Location: WV

djozz wrote:

ImA4Wheelr wrote:

Without having one of those charts in 100cri daylight included to compare to, it appears to me that the dedomed nichia has the best cri of the whole bunch.

Here you go,

just made a picture (camera on daylight setting, no corrections on screen done this time) with a slightly overcast sky, that gives the best quality daylight imo (that is very subjective of course). But still the camera does not capture the true vivid colours of reality.


That chart appears a little too blue. Greyscale should be equal amounts of each color for each tile. From left to right it should be 0,0,0 for black, 255, 255, 255 for white and 128, 128, 128 in the middle. It’s best to use the actual chart for true colors, not a captured image.

This one should be accurate. If the greyscale still looks off, it’s your monitor.

Edit: New picture.
Edit: There’s a 18ish% reflectivity that I didn’t think about so black should not be 0, 0, 0. So, for comparison, all tiles should be brighter.

From x-rite: Since they exemplify the color of their counterparts and reflect light the same way in all parts of the visible spectrum, the squares will match the colors of representative sample natural objects under any illumination, and with any color reproduction process.

UliBär
UliBär's picture
Offline
Last seen: 2 months 1 day ago
Joined: 05/22/2013 - 16:08
Posts: 389
Location: N52° 16' 9.444" E10° 31' 26.105"

Nichia 219 original is the best – sorry! Wink

You can clearly see it on the dark blue square (just below the black one in the top left corner) and by the purple square (the third square from the left in the third row from the top). The dedomed Nichia clearly lost a big amount of blue color by the shift to yellow/warmer tint. Comparison to your daylight shot proves this even more.

Aloha, Uli

Chloe
Chloe's picture
Offline
Last seen: 3 years 2 months ago
Joined: 01/30/2013 - 14:14
Posts: 4595
Location: 联合王国

Thanks for the colour chart pictures!

ImA4Wheelr
Offline
Last seen: 1 hour 53 min ago
Joined: 02/03/2013 - 14:51
Posts: 7909
Location: SC

I guess what I'm thinking is that the comparative 100 cri shot should not actually be one that is true to the real colors, but true to the same method the other shots were taken.  So the shots are apples to apples.  Meaning, the same camera settings etc.  The colors will be off, but they will be off in the same direction as the flashlight pictures.  Not asking you to do all that work.  Just saying that is what I think would be the best comparison.

Jalben
Jalben's picture
Offline
Last seen: 4 months 2 weeks ago
Joined: 12/12/2011 - 20:58
Posts: 452
Location: Mount Pearl, Newfoundland, Canada

I got a feeling that when djozz said it’s not the same as looking at it in person that the issue maybe with colour balancing in his camera. But I’m no camera guy either.

TODO List:

Acquire more parts.

Devise cheap way to make TIRs.

Rub some bacon on it.

Finish something that I 

lightme
lightme's picture
Offline
Last seen: 3 years 7 months ago
Joined: 11/12/2012 - 19:51
Posts: 615
Location: WV

You can get 100 CRI from both an incandescent bulb or the sun but that doesn’t mean objects can’t appear more blue or yellow. The way I understand it, higher CRI produces more visible colors so outside you get the full spectrum but it’s mostly blue because of our atmosphere. Inside under an incandescent lamp, you still have the full spectrum but less blue light is being produced. If you don’t have equal intensity throughout the spectrum for a comparison chart, the extra blue from the sky will make the yellowish charts look worse than they are and vice versa.

biggulp
Offline
Last seen: 3 weeks 23 hours ago
Joined: 12/21/2012 - 09:45
Posts: 125

CRI comparison is only useful when using the same color temperature.

djozz
djozz's picture
Offline
Last seen: 3 hours 23 min ago
Joined: 09/07/2012 - 17:04
Posts: 15781
Location: Amsterdam

I am still learning this colour reproducing trick, and I am not doing really well..

My holy grail was to get on my and your monitor exactly the colours that I see in real life. I knew that was difficult, but I thought it to be possible to get close. That appears to be not true, far from it. Several reasons. The main reason is that the dynamic range of the real life colours can not be reproduced on a monitor, You miss out at least the vibrant reds, yellows, blues. For that reason those charts above do not show the best performing leds very well, on screen the charts look closer to each other than they really are, in real life differences are further apart, some differences are even absent on screen.

Another reason is that a (at least my) camera is not so good in capturing bright red and bright yellow, bright blue is much better. That said, in the digital colour chart that lightme provided (thanks for that, I find it very informative Smile ) the bright yellow and bright red tiles are also not as bright as in real life.

Perhaps the best way, to at least get the fairest idea of the differences between leds, is not to try to adjust the images on screen to match the real life colours (as I have tried above), but just show them without colour correction. 

lightme
lightme's picture
Offline
Last seen: 3 years 7 months ago
Joined: 11/12/2012 - 19:51
Posts: 615
Location: WV

djozz wrote:
Another reason is that a (at least my) camera is not so good in capturing bright red and bright yellow, bright blue is much better. That said, in the digital colour chart that lightme provided (thanks for that, I find it very informative Smile ) the bright yellow and bright red tiles are also not as bright as in real life.

I just edited that post to indicate that. Every chart I find seems to be little different. Some have lighter shades but duller colors. Only you know what they’re supposed to look like.

Pages