Interesting product. Is the product you linked the specific option you recommend? I see there are other options on the page.
My application is home lighting. Specifically in dining and kitchen areas. I bought some of the 219c 4000/4500k cri90s thinking they were going to look like the 219b 9080 4500k. But as has been said there is a bit too much yellow for me. Could this film help my cause? And would the item linked be best or would one of the others listed with it better suit my application?
Thanks for the search suggestion.
I run them at 1600mA each. The tint is near perfect as long as I don’t get into the 35 degree zone. However, since I may build an even longer bar, there will be a full overlap.
I use a thin aluminum to mount all my leds so it may be hard to attach TIR optics/ reflectors. Do you have any suggestions of what I can use? Thank you for all the help!
My fav is Ledil. Can be installed in strip or individual:
Vanessa
Shelly
Then Luminit for it’s superior hiding power and simple application:
Very even uniform lights using the VLSD
so does this mean the consensus is that 219C 9080 aka 90cri 4500k are not going to happen? or do we just wait nichia out?
219Cs in R9050 exist in 4500K. That’s 90 CRI minimum with R9 (deep red) of 50 minimum. R9 usually exceeds the minimum by a significant amount with these, but less in very warm tints.
219Cs do not come in R9080, and might never.
I used this for all my latest tint shots. They’re too cheap to not to try: http://www.ebay.com/itm/10PCS-x-60-Degree-21mm-frosted-Lens-Cree-XPE-XPG-XTE-LED-/261515828211?hash=item3ce38edbf3
Just use clear silicone glue for easy removal (NEVER use any cyanoacrilate or superglue!). Fits most standard 16 and 20mm boards
- Clemence
A word of warning about silicone glue: make sure you use the good stuff for electronics. The cheap stuff produces acetic acid (vinegar) vapour as it cures, which won’t do electronics or reflectors any good. I’m not sure if it will fog plastic optics, though. If you’re not sure, try some on a piece of paper; you’ll smell the vinegar smell if it’s that kind.
I use the non acid type from Permatex, red automotive cylinder head gasket maker. Just for temporary fix. Thin double tape also do the job fair enough.
For permanent installation, transparent epoxy still the best so far.
Cyanoacrilate is a big no for any PMMA material, it will create thick haze on them.
- Clemence
Ouch. I didn’t know that. Thanks for mentioning it :+1:
After my first experience with these filters from a sample booklet I made a report, see here.
I am fortunate to have a Color Analyzer so I can measure the color points (xc,yc) before and after correction. That is useful, being a tint snob.
My first aim was to correct the awfully green Xiaomi lamps to have a tint on the BBL locus (arrow d in the drawing). This is about 20/1000 or 0.020 on the yc axis (In the drawing I interchanged yc and xc axis, sorry).
An offset on yc of 0.005 is easy to see as a green tint so here we have 4 times as much.
The best job would be done by a 279 plus a 003 filter but that would cost 35% in efficiency so I decided for only one layers of filter 279, that worked fine.
My final choice was to buy three filters:
- nr.279 (minus 1/8 green). Used it a lot to remove moderate greenish tint. Also makes the CCT a little warmer. Also to correct the slight greenish 219C (in my opinion) towards the magenta side.
- nr.278 (plus 1/8 green). Aimed to correct LEDs that is too magenta. Used only a few times.
I would advice you to ask for a (free) sample booklet from a vendor as I did. Then you can see the effect of a filter by shining a flashligt through the sample sheet (they are small, appr. 1.5 x 4 inch). Then decide which to buy. My local vendor did not have the half sheet size (cheapest) so I had to select carefully to save money.
Has anybody ever played with one of these cheap see through spectroscop?Amazon.com (only available in the U.S.)
to see if there’s a difference between different CRIs
Has anybody ever played with one of these cheap see through spectroscop?Amazon.com (only available in the U.S.)
to see if there’s a difference between different CRIs
Yes, you will see a spectrum and with some experience you can make out if the source is incandescent, led, sunlight etc. The difference between different colour temperature leds is already a challenge and you will be my hero if you can judge CRI from seeing a spectrum by eye. For that you need to record the spectrum, then calibrate the wavelengths and relative powers of them, and then analyse the data.
I hate you guys, you know that? I saw that, didn’t know such a thing even existed for so cheap, and had to order one (well, two, in case one breaks or gets lost), like, right now.
Will let youse know when I get it how it fares. Reviews look pretty good, so…
I always try to help.
Revenge for the two Boroit headlamps I bought? 
But lucky you, you can order one, I am in Germany and can’t.
Only 2? Damn, I got like 6 of ’em. Gave away a couple, still got the rest. Figure even if “disposable” I’ll have a long enough supply to keep me happy. Carry one in my bag every day, just as a spare light of whatever kind (wear it, remove from the band and “palm” it, etc.).
Oh it has a few issues:
- pwm
- a linear instead a cubic ramp
Yep.
I’ve gotten decent at catching it at/near the lowest point for regular day-to-day stuff, but would definitely prefer a log ramp.
Don’t really mind the full-power wave-thing as I don’t use it that often (I’m good with clicking it on/off as needed).
And for whatever reason pwm doesn’t seem to bug me as much as it does other people (’though tube-type fluorescent lights flickering at 120Hz does bug me after a short while).
For 10bux, though, I’ll live with it. 
Fixing everything but pwm could be as simple as reflashing the µC that’s controlling it.
A ~10µF chip-cap right across the LED might smooth the output. Output would still be jiggly, but not a hard on/off.
Then again, PWMing is a linear function, filtering/smoothing the voltage accordingly to an LED with a ~3V “knee” not so much. Ie, 50% might filter down the 3.8V from the cell down to a spiky 1.9V, giving a nice moonlight mode as the LED barely glows at all, if at all.
’Though if it’s not discharged when the voltage is that low, charge would simply build up in the next round of pulses. Never analysed that particular function, maybe someone else could try it in the Real World™.
Fact is, pwm is the simplest and cheapest(!) way to modulate a high power load like an LED when you have limited voltage headroom.
If Motorola still made their hexfets, you could get a sampling of the current throughput from a sense-out pin, without needing a series resistor to sniff the current through the load. Output from the sense-out pin would be a percentage of the total current, eg, 1%, so a 1A load would give a 10mA output. Sniff that and modulate accordingly.
Alas, I don’t think they’re made anymore…
Update 170601:
I finally finished sending all the papers needed to the last (hopefully) department. The license should appears within 5 - 30 working days (they say). I went to the warehouse to check if those Nichias were still in good conditions or not….they’re moved deeper to another warehouse because nobody claimed for the last 30 days (WTF!). I checked again, well they’re still in good (dusty) condition. Fortunately Nichia overdo the packing so they’re safe. I will have to pay the warehouse rent later, it’s USD 0,25/kg/day (6,4kg total).
Then I saw an approx. 10 meters high aluminum boxes piling there, multiply that with the rental cost……
- Clemence
Good luck with the fuss Clemence.