Really bright yellow LEDs?

[Martin Brown]

Just be glad you can GET the colors you want in LED form; I'd have been
happy to pay kilobucks for a dim blue LED, thirty years ago. We used a
mechanical chopper and HeNe-Cd blue laser instead.

There were some very dim blue LEDs available 30 years ago. Mostly used
on high end audiophool amplifiers and the like as indicators and just
barely visible when lit. ISTR they were Russian silicon carbide based.
Not mentioned on Wiki (except as available from 1990's).

Pretty sure they were on some high end kit in the late 70's early 80's.

 

[Ecnerwal]

I did notice that a couple of places (Ididex, Coherent, among others)
make a 577nm diode-pumped laser, but seem to keep it under tight wraps
and presumed high expense as a "medical" laser, so it seems unlikely to
be suitable.

 

[Phil Hobbs]

Probably wouldn't work for you, but I experimented some with UV LEDs
and fluorescent paints. Fun.

https://dl.dropboxusercontent.com/u/53724080/Optos/PH+Card.JPG

Yeah, I'm fighting a worst-case path loss of 80 dB optical, so I really
need a lot of photons. Turns out I can get quite a few just by
filtering a 4000K white LED, because 578 nm is pretty near the peak.

You know your LEDs are pathetic when a blue one, driving a phosphor,
filtered down to 5 nm wide, is still 100x brighter than yours and has
10x better wall plug efficiency.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

 

[[email protected]]

Hi, all,

Have you considered using a low pressure sodium vapour lamp for the yellow light?

Lots of problems in that approach,but you do get a lot of light per watt. From Wiki

"Another unique property of LPS lamps is that, unlike other lamp types, they do not decline in lumen output with age. As an example, mercury vapor HIDlamps become very dull towards the end of their lives, to the point of being ineffective, while continuing to consume full rated electrical use. LPS lamps, however, do increase energy usage slightly (about 10%) towards theirend of life, which is generally around 18,000 hours for modern lamps."

Dan

 

[Phil Hobbs]

Have you considered using a low pressure sodium vapour lamp for the
yellow light?

Lots of problems in that approach,but you do get a lot of light per
watt. From Wiki

"Another unique property of LPS lamps is that, unlike other lamp
types, they do not decline in lumen output with age. As an example,
mercury vapor HID lamps become very dull towards the end of their
lives, to the point of being ineffective, while continuing to consume
full rated electrical use. LPS lamps, however, do increase energy
usage slightly (about 10%) towards their end of life, which is
generally around 18,000 hours for modern lamps."

Dan

Thanks. The sodium lines are at 589 nm, which won't work in my
application. I really need 578 and 598 nm. If I could do it at 590,
there are a ton of bright LEDs there as well.

Cheers

Phil Hobbs



--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

 

[josephkk]

If you modulate the LED light and use synchronous detection, you can just
use more gain (without getting too badly clobbered by ambient light or noise).
That's a cheaper way to go, when one considers (for instance) the unknowns
like source aging/dimming.
Just be glad you can GET the colors you want in LED form; I'd have been
happy to pay kilobucks for a dim blue LED, thirty years ago. We used a
mechanical chopper and HeNe-Cd blue laser instead.

Cree had blue LEDs (SiC) over 30 years ago, but they had lousy spectral
purity and were expensive (>3$ ea. at 1 mW).

?-)

 

[Tim Williams]

Thanks. The sodium lines are at 589 nm, which won't work in my
application. I really need 578 and 598 nm. If I could do it at 590,
there are a ton of bright LEDs there as well.

So you need just 10nm either side of it?

How about... erm, hetrodyning what, a CO2 laser with it?

Better yet, a comb generator, and bandpass whatever you like... make an
optical Wadley Loop and cover the whole range!

Okay, maybe a *little* crazier than an LED array, certain notches (such as
the present problem) aside.

Tim

 

[Phil Hobbs]

Unfortunately both wavelength distances are beyond convenient thermal
pulling distance if i understand correctly. Plus, it could be hard to
regulate target wavelength well. Maybe a diffraction grating trick would
help.

Keep on thimkimg. >?-)

The current best guess is either a good white LED or the very greenest
among the high-efficiency yellow ones, shining through a 3-5 nm wide
interference filter.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

 

[josephkk]

Thanks. The sodium lines are at 589 nm, which won't work in my
application. I really need 578 and 598 nm. If I could do it at 590,
there are a ton of bright LEDs there as well.

Cheers

Phil Hobbs

Unfortunately both wavelength distances are beyond convenient thermal
pulling distance if i understand correctly. Plus, it could be hard to
regulate target wavelength well. Maybe a diffraction grating trick would
help.

Keep on thimkimg. >?-)

 

[Klaus Kragelund]

<snip>












This is a technological application--it's an on-line sensor for food

quality.

Interesting. Is the theory about the spectrum of different object describedanywhere?

Would for example an animal have a different spectrum than a human? Does the bad quality of the food show in a shift of the spectrum? (hope the question is not showing too much I haven't a faintest idea about optical theory apart from introductionary course)

Cheers

Klaus

 

[Klaus Kragelund]

On 12/21/2013 2:38 PM, Salmon Egg wrote:

Interesting. Is the theory about the spectrum of different object described anywhere?



Would for example an animal have a different spectrum than a human? Does the bad quality of the food show in a shift of the spectrum? (hope the question is not showing too much I haven't a faintest idea about optical theoryapart from introductionary course)

A link describes the different spectrum of normal tissue and fat:

http://www.opticsinfobase.org/oe/fulltext.cfm?uri=oe-17-15-12627&id=183577

So guess, that is what you are doing, refined using a LED instead of a laser/strope

Another picture:

www.electronicsdesign.dk/tmp/reflectance-chart.png

A plot of different species of plants with simelar shape, but visible gain difference. But how is the gain calibrated so it's possible to tell the difference? (a sensor could be sensible to different angle of attack etc)

Cheers

Klaus

 

[Phil Hobbs]

Interesting. Is the theory about the spectrum of different object described anywhere?

Would for example an animal have a different spectrum than a human? Does the bad quality of the food show in a shift of the spectrum? (hope the question is not showing too much I haven't a faintest idea about optical theory apart from introductionary course)

Cheers

Klaus

The basic idea is ratiometric multispectral measurement. The basic
idea is in Gielen, Robert M A M; De Jong, L.P.; Kerkvliet, Harry M M,
"Electrooptical Blood-Spot Detection in Intact Eggs," IEEE Trans.
Instrum. Meas., vol.28, no.3, pp.177,183, Sept. 1979.

The parlour trick is seeing the 578 nm haemoglobin absorption peak on a
baseline that goes up and down by 3 orders of magnitude in absorption,
due to the broad 598-nm protoporphyrin absorption peak. (Protoporphyrin
is what makes brown eggs brown.)

Cheers

Phil Hobbs


--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net