[GB ended,discussion only] CRI > 80+ NICHIA 2000K-6500K [E21A/219B/219C/319A/144A/757GT-F1(Optisolis)]

1608 posts / 0 new
Last post
zak.wilson
Offline
Last seen: 3 hours 19 min ago
Joined: 09/29/2014 - 14:27
Posts: 888

I put one of the R9080s in a light and now everything else is green and horrible, including the sun. Help?

Ceilingbounce – flashlight testing and runtime graphs for Android

clemence
clemence's picture
Online
Last seen: 9 min 9 sec ago
Joined: 07/12/2015 - 02:58
Posts: 2338
Location: Bali - Indonesia

zak.wilson wrote:
I put one of the R9080s in a light and now everything else is green and horrible, including the sun. Help?
LOL LOL
Texas_Ace
Texas_Ace's picture
Offline
Last seen: 5 hours 27 min ago
Joined: 03/24/2016 - 07:44
Posts: 8535
Location: Everything is brighter in Texas

I just put 7 of the 219B 9080 LED’s into a SRK with the stock non-DTP mcpcb.

I can legitimately say that I am blown away by the tint of these! I have seen quite a few LED’s and this by FAR blows them all away.

These are the whitest, purest, most gorgeous LED’s I have ever seen! Keep in mind that over half of my lights are 90+ CRI.

Even compared to all my 219B 9050’s I have, these have no noticeable tint shift either green or red. Just pure amazing white. All my other 219b’s have a slight shift to either the green or rosy side.

I was worried when I saw these outside of a reflector when on the bench that they would be too rosy but seems it was just an effect of having tested the 319A right before it and that screwing my frame of reference.

Now I will see what it looks like tonight to get a better idea but at this point I am in love, I just wish that they had higher output.

A 219C 9080 would be perfect, if only they could put the Phosphor from the 219B 9080 on the 219C die it would be the perfect LED for all your CRI needs. Hmmm, maybe they would be open to a special run of these clemence?

Only downside is the low output but in triples and quads it is perfect. An M34 or M43 would be amazing with these.

Now on the down side, DO NOT use these with non-DTP mcpcb’s. I killed 3 of the 7 LED’s I installed in the SRK. I am not sure if it is a mismatch in the Vf or just not being able to get rid of the heat. Either way 4 are nearly dead and released magic smoke. I now have to run the light with only 1 laptop pull cell installed to limit current.

Later I might indeed build a M34 with these.

emarkd
emarkd's picture
Offline
Last seen: 1 day 8 hours ago
Joined: 04/14/2015 - 22:04
Posts: 1601
Location: Georgia, USA

Oh wow, sorry you smoked some. I’ve only used one of mine and I put it in a pretty weak light (a 1 amp HDS), but I absolutely love it. TA isn’t exaggerating folks, they’re that good. I’ve gotta figure out what I’m doing with the rest of mine.

clemence
clemence's picture
Online
Last seen: 9 min 9 sec ago
Joined: 07/12/2015 - 02:58
Posts: 2338
Location: Bali - Indonesia

Texas_Ace wrote:
I just put 7 of the 219B 9080 LED’s into a SRK with the stock non-DTP mcpcb.

I can legitimately say that I am blown away by the tint of these! I have seen quite a few LED’s and this by FAR blows them all away.

These are the whitest, purest, most gorgeous LED’s I have ever seen! Keep in mind that over half of my lights are 90+ CRI.

Even compared to all my 219B 9050’s I have, these have no noticeable tint shift either green or red. Just pure amazing white. All my other 219b’s have a slight shift to either the green or rosy side.

I was worried when I saw these outside of a reflector when on the bench that they would be too rosy but seems it was just an effect of having tested the 319A right before it and that screwing my frame of reference.

Now I will see what it looks like tonight to get a better idea but at this point I am in love, I just wish that they had higher output.

A 219C 9080 would be perfect, if only they could put the Phosphor from the 219B 9080 on the 219C die it would be the perfect LED for all your CRI needs. Hmmm, maybe they would be open to a special run of these clemence?

Only downside is the low output but in triples and quads it is perfect. An M34 or M43 would be amazing with these.

Now on the down side, DO NOT use these with non-DTP mcpcb’s. I killed 3 of the 7 LED’s I installed in the SRK. I am not sure if it is a mismatch in the Vf or just not being able to get rid of the heat. Either way 4 are nearly dead and released magic smoke. I now have to run the light with only 1 laptop pull cell installed to limit current.

Later I might indeed build a M34 with these.

There are some sm503 9080 D220 “domeless” 219C on my desk right now Wink .
Unfortunately no sw45k available at the moment. And they’re the hardest LED to solder I’ve ever had. No thermal pads, even the super tiny cramped solder points are well hidden under the die completely.

Texas_Ace
Texas_Ace's picture
Offline
Last seen: 5 hours 27 min ago
Joined: 03/24/2016 - 07:44
Posts: 8535
Location: Everything is brighter in Texas
clemence wrote:
There are some sm503 9080 D220 “domeless” 219C on my desk right now Wink . Unfortunately no sw45k available at the moment. And they’re the hardest LED to solder I’ve ever had. No thermal pads, even the super tiny cramped solder points are well hidden under the die completely.

Do explain, how does a 219C not have a thermal pad? Don’t they call the version without a thermal pad the 229C or something like that? Without the thermal pad it is basically useless for a flashlight sadly.

Although I would gladly take a 219C 9080 in 5000k, 4500k would be my preference but not going to argue over 500k!

Also what do you mean by domeless?

clemence
clemence's picture
Online
Last seen: 9 min 9 sec ago
Joined: 07/12/2015 - 02:58
Posts: 2338
Location: Bali - Indonesia
Texas_Ace wrote:
clemence wrote:
There are some sm503 9080 D220 “domeless” 219C on my desk right now Wink . Unfortunately no sw45k available at the moment. And they’re the hardest LED to solder I’ve ever had. No thermal pads, even the super tiny cramped solder points are well hidden under the die completely.

Do explain, how does a 219C not have a thermal pad? Don’t they call the version without a thermal pad the 229C or something like that? Without the thermal pad it is basically useless for a flashlight sadly.

Although I would gladly take a 219C 9080 in 5000k, 4500k would be my preference but not going to argue over 500k!

Also what do you mean by domeless?

OK, give me a moment. I’ll show you.
PS: don’t take it to literally Silly

zak.wilson
Offline
Last seen: 3 hours 19 min ago
Joined: 09/29/2014 - 14:27
Posts: 888

I think they’re NVSWE21As, not 219Cs.

I find the sw45k significantly rosy, but shined on anything but a white wall that just makes all the colors look extra rich. After my eyes adjust all other light sources look greenish and pale.

The sm503 is a little cooler than I expected. It’s unambiguously cooler next to other 5000K 219Cs and an XHP35 HI in 3C. sm503 left, 3C right

Ceilingbounce – flashlight testing and runtime graphs for Android

Texas_Ace
Texas_Ace's picture
Offline
Last seen: 5 hours 27 min ago
Joined: 03/24/2016 - 07:44
Posts: 8535
Location: Everything is brighter in Texas
zak.wilson wrote:
The sm503 is a little cooler than I expected. It’s unambiguously cooler next to other 5000K 219Cs and an XHP35 HI in 3C. sm503 left, 3C right

The SM503 looks more like what a 5000k should look like to my eye, the 3C is too warm / green the way I see it.

5000k should be basically pure white with the smallest hint of blue in the beam if you look close.

zak.wilson
Offline
Last seen: 3 hours 19 min ago
Joined: 09/29/2014 - 14:27
Posts: 888

The 3C definitely has a touch of green in it. sm503 does not, and what I may be seeing here is the lack of off-colors more than anything. Also, Cree hotspots tend to be warmer than spec while the spill is cooler. Regardless, these are nice, and do, indeed look like a very pure white.

Ceilingbounce – flashlight testing and runtime graphs for Android

clemence
clemence's picture
Online
Last seen: 9 min 9 sec ago
Joined: 07/12/2015 - 02:58
Posts: 2338
Location: Bali - Indonesia

Sorry dudes…. edited the pics and then fell asleep. I was just teasing Tex Big Smile
Yes, correct it’s E21A! Congrats Zak Grad

gchart
gchart's picture
Offline
Last seen: 1 hour 41 min ago
Joined: 03/19/2016 - 11:57
Posts: 1728
Location: Central IL

I’d love to see some more about the E21A. I wonder how they would fare on djozz’s custom copper boards for LEDs such as these (moderate power but no thermal pad).

clemence
clemence's picture
Online
Last seen: 9 min 9 sec ago
Joined: 07/12/2015 - 02:58
Posts: 2338
Location: Bali - Indonesia

Finally an official Nichia 219 "HI".

Finally, a true official 219C "HI"

 

Reflowing this one is PITA. Without precise SMD, you have to use a jig/centering tool. Not auto centering effect to be expected. even the excess solder flux will ruin it.

To accurately reflow this one without precise SMD is PITA. Not much auto centering effect to be expected

 

Cheers,

Clemence

Texas_Ace
Texas_Ace's picture
Offline
Last seen: 5 hours 27 min ago
Joined: 03/24/2016 - 07:44
Posts: 8535
Location: Everything is brighter in Texas

Interesting LED for ultra low current applications.

clemence
clemence's picture
Online
Last seen: 9 min 9 sec ago
Joined: 07/12/2015 - 02:58
Posts: 2338
Location: Bali - Indonesia
Texas_Ace wrote:
Interesting LED for ultra low current applications.

Yes, not so interesting stuff for you guys. Will requires specially made PCB. The anode – cathode separation is only 0,2mm! Just a LITTLE too much solder and they will bridged. Keeping 2 or more emitters perfectly parallel and flat is another challenge. I didn’t use tweezer as it’s only 0,2mm thick, and the coating is very fragile. The soft silicone protection coating has phosphors mixed in it (warmer CCT/higher CRI has thicker coatings). I just dragged it here and there using fine feeler gauge blade.

Imagine (manually) creating a light panel consists of 100’s of these… Sick Ughh

Pöbel
Pöbel's picture
Offline
Last seen: 2 hours 51 min ago
Joined: 02/14/2013 - 08:27
Posts: 234
Location: Germany

and then imagine Emitters somewhere in between having to be replaced Big Smile

tuelleric
Offline
Last seen: 3 months 1 week ago
Joined: 12/05/2016 - 12:17
Posts: 109
Location: Germany

Mine just arrived! Thanks again to clemens, chouster and TA, who made this possible!

Not to mess up with all those tiny LEDs when repacking must be a real challenge! Really hard to imagine how much effort went into this!

clemence
clemence's picture
Online
Last seen: 9 min 9 sec ago
Joined: 07/12/2015 - 02:58
Posts: 2338
Location: Bali - Indonesia

tuelleric wrote:
Mine just arrived! Thanks again to clemens, chouster and TA, who made this possible!

Not to mess up with all those tiny LEDs when repacking must be a real challenge! Really hard to imagine how much effort went into this!

Ladies and Gentlemen…. the awards goes to Chouster and Texas_Ace….
I only did two big bulk packings Silly

Guys, please let me know about the colour of 219C 9050 sm503. Mustsimon said it’s pretty green (even though he compared it to some of his 219B’s collection). I don’t have any yellow/green or even yelloish/greenish problem with them all.
Those sm503 9050 219C and sm573 9050 144A are by far the most neutral white I have.
The more current, the (slightly) rosier 9050 219C becomes

- Clemence

tuelleric
Offline
Last seen: 3 months 1 week ago
Joined: 12/05/2016 - 12:17
Posts: 109
Location: Germany

Anyway, clemens, you were the one who initiated the whole thing!

I only tested one 219C sm503 with my DMM so far and did not notice any green. But I should test it properly to really state the tint.

Now I need boards, many boards… Any quick suggestion where I can buy SinkPad or Noctigon for a reasonable price (shipped internationally)? Thanks!

clemence
clemence's picture
Online
Last seen: 9 min 9 sec ago
Joined: 07/12/2015 - 02:58
Posts: 2338
Location: Bali - Indonesia

tuelleric wrote:
Anyway, clemens, you were the one who initiated the whole thing!

I only tested one 219C sm503 with my DMM so far and did not notice any green. But I should test it properly to really state the tint.

Now I need boards, many boards… Any quick suggestion where I can buy SinkPad or Noctigon for a reasonable price (shipped internationally)? Thanks!

Try Vestureofblood or Hank. They’re both member of BLF and CPF

Jerommel
Jerommel's picture
Offline
Last seen: 5 months 1 week ago
Joined: 01/04/2014 - 13:18
Posts: 5721
Location: the Hague, Netherlands

Those “mini COBs” look interesting.

2Q19

chouster
Offline
Last seen: 3 hours 21 min ago
Joined: 02/20/2014 - 15:05
Posts: 688
Location: germany

Those E21A are very interesting. Imagine a 20mm triple under Carclo optics. Unfortunately this would need a special MCPCB. For good heat transfer a special AlN Board would be nice, or a direct cathode MCPCB combined with a special driver that switches battery+, so that the board wouldn’t have to be isolated.

3x E21A and Carclo 10511 should make a very nice beam, smooth but with some throw.

That beeing said, I would expect the NCSL219B-V1 R9080 to be quite similar concerning beam characteristics under those 105xx Carclos, but much easier to implement.

gchart
gchart's picture
Offline
Last seen: 1 hour 41 min ago
Joined: 03/19/2016 - 11:57
Posts: 1728
Location: Central IL

I just built up a Convoy S2 using one of the sm503 9050 219C. Looks pretty nice to my eyes, but it’s hard for me to tell as I’m red/green colorblind. I wouldn’t mess with high CRI for myself, this is for a friend who will enjoy the CRI. It does appear to be a wonderfully neutral white though.

clemence
clemence's picture
Online
Last seen: 9 min 9 sec ago
Joined: 07/12/2015 - 02:58
Posts: 2338
Location: Bali - Indonesia

chouster wrote:
Those E21A are very interesting. Imagine a 20mm triple under Carclo optics. Unfortunately this would need a special MCPCB. For good heat transfer a special AlN Board would be nice, or a direct cathode MCPCB combined with a special driver that switches battery+, so that the board wouldn’t have to be isolated.

3x E21A and Carclo 10511 should make a very nice beam, smooth but with some throw.

That beeing said, I would expect the NCSL219B-V1 R9080 to be quite similar concerning beam characteristics under those 105xx Carclos, but much easier to implement.

Thx for the AlN idea Chouster, will think about it.
Carclo 10511 designed favourably towards Cree XP-G in the first place (0,73mm focal plane). The rest of the emitters will need some adjustment in optics’s peg height. E21A paired with 10511 isn’t as tight as it potentially could. The problem lies in much lower focal plane in E21A 0,27 – 0,3mm +/- 0,05mm (height vary with different CRI). Compare this to 0,4mm in NVSx219C/NCSx219B-V1/NVSx219B-V1.
We all know that domeless emitter with the same die shape should have much tighter beam. But from this beamshots, the difference is only marginal. Approx. 1 degree tighter compared to 219C (the same die as E21A).
Cree XP-L HI usually have ~20% tighter beam compared to XP-L HD in any TIR optics.

NVSW219B

NVSWE21A

NVSW219C

Just my Rp 2

clemence
clemence's picture
Online
Last seen: 9 min 9 sec ago
Joined: 07/12/2015 - 02:58
Posts: 2338
Location: Bali - Indonesia

Just for fun....

 

chouster
Offline
Last seen: 3 hours 21 min ago
Joined: 02/20/2014 - 15:05
Posts: 688
Location: germany

Hi Clemence,

I appreciate your Rp 2 very much. Smile

I read your thoughts about AlN in your VirEnce CDTP MCPCB thread, so you already mentioned it.

The beamshots and FWHM-graphs you posted are from Carclo, right?

I think their numbers are very much like I would have expected, with one big exception…

… the NVSL219B. I really think they mixed something up there. In fact, I have the feeling they tested a NCSL219B…

All numbers would make perfect sense to me than.

EDIT: I just wrote them an e-mail asking for clarification.

clemence
clemence's picture
Online
Last seen: 9 min 9 sec ago
Joined: 07/12/2015 - 02:58
Posts: 2338
Location: Bali - Indonesia

chouster wrote:
Hi Clemence,

I appreciate your Rp 2 very much. Smile

I read your thoughts about AlN in you VirEnce CDTP MCPCB thread, so you already mentioned it.

The beamshots and FWHM-graphs you posted are from Carco, right?

I think their numbers are very much like I would have expected, with one big exception…

… the NVSL219B. I really think they mixed something up there. In fact, I have the feeling they tested a NCSL219B…

All numbers would make perfect sense to me than.

Yes, I got them pics from Carclo’s website.
I don’t think they mixed it with NCSL, it’s what it is.
Beam angle order in 10511: 219B – E17A – E21A – 219C – 319A (weird huh?)
If we follow common sense based on die area geometries it supposed to be like this: E17A – E21A – 219B – 219C – 319A
So far, the only explanation logcal to me is in their focal plane (die height) differences. Anything lower than ~ 0,6mm – 0,73mm (XPG) will off focus. 219B, 219C, and 319A are all have 0,4mm focal plane. That’s why their sequence looks OK.

chouster
Offline
Last seen: 3 hours 21 min ago
Joined: 02/20/2014 - 15:05
Posts: 688
Location: germany

Hi,

please take a look at the Cd/lm numbers of the 10193 optic.

  • NVSW 319A Cd/lm: 9.5
  • NVSW 219C Cd/lm: 12.5
  • NVSW E21A Cd/lm: 20.66
  • NVSL 219B Cd/lm: 23.1 (NCSL219B!?)
  • NCSW E17A Cd/lm: 34

I really do think they mixed that up.

clemence
clemence's picture
Online
Last seen: 9 min 9 sec ago
Joined: 07/12/2015 - 02:58
Posts: 2338
Location: Bali - Indonesia

chouster wrote:
Hi,

please take a look at the Cd/lm numbers of the 10193 optic.

  • NVSW 319A Cd/lm: 9.5
  • NVSW 219C Cd/lm: 12.5
  • NVSW E21A Cd/lm: 20.66
  • NVSL 219B Cd/lm: 23.1 (NCSL219B!?)
  • NCSW E17A Cd/lm: 34

I really do think they mixed that up.

I don’t know. But that looks OK to me since 219B has smaller die than 219C. E21A and E17A could be close to 30 and 40 cd/lm if the height adjusted.

chouster
Offline
Last seen: 3 hours 21 min ago
Joined: 02/20/2014 - 15:05
Posts: 688
Location: germany

Also, why would the NVSL219B have so much higher Cd/lm values than the NVSL219C, when they have about the same die-height and die-size?

Pages