DIY reflector paint

[Martin Brown]

On 11/02/2020 02:30, Phil Hobbs wrote:

On 2020-02-10 04:17, Martin Brown wrote:

The OP's requirements are ultimately up against laws of thermodynamics.

Yes, but the arguments are a bit subtler there since a LED is not
exactly a thermal source.

It means you can get away with a simple aspheric cast lens though since
the wavelength range of a colour LED emitter is typically 50nm fwhm.

You would be crazy to use a white LED in a signalling application.

--
Regards,
Martin Brown

 

[Phil Hobbs]

On 2020-02-12 03:47, Martin Brown wrote:

On 11/02/2020 02:30, Phil Hobbs wrote:
On 2020-02-10 04:17, Martin Brown wrote:

The OP's requirements are ultimately up against laws of thermodynamics.

Yes, but the arguments are a bit subtler there since a LED is not
exactly a thermal source.

It means you can get away with a simple aspheric cast lens though since
the wavelength range of a colour LED emitter is typically 50nm fwhm.

Spatial coherence is a big issue. A LED is a diffuse source with light
coming from inside a material with n~=3.45, so if you got 100%
extraction efficiency you'd have a Lambertian source about 11 times the
area. That makes the solid angle/beam diameter tradeoff difficult.

You would be crazy to use a white LED in a signalling application.

Actually not. I've designed a couple of tactical optical transceivers
for the US military, both of which used white LEDs. You get a lot of
bang for the buck--much better than the corresponding blue LED dice.


Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:3dc0b0e9-5c74-0201-bbb5-f26329344340@electrooptical.net:

On 2020-02-12 03:47, Martin Brown wrote:
On 11/02/2020 02:30, Phil Hobbs wrote:
On 2020-02-10 04:17, Martin Brown wrote:

The OP's requirements are ultimately up against laws of
thermodynamics.

Yes, but the arguments are a bit subtler there since a LED is
not exactly a thermal source.

It means you can get away with a simple aspheric cast lens though
since the wavelength range of a colour LED emitter is typically
50nm fwhm.

Spatial coherence is a big issue. A LED is a diffuse source with
light coming from inside a material with n~=3.45, so if you got
100% extraction efficiency you'd have a Lambertian source about 11
times the area. That makes the solid angle/beam diameter tradeoff
difficult.


You would be crazy to use a white LED in a signalling
application.

Actually not. I've designed a couple of tactical optical
transceivers for the US military, both of which used white LEDs.
You get a lot of bang for the buck--much better than the
corresponding blue LED dice.


Cheers

Phil Hobbs

Can place at the back of a black tube for stopping the side reads.

Deep voice inflection...
(coat it with Vantablack)(Bou! bou!)(oh yeah!)(chic-a-chic-ahhh)

Seems though that lasers (diodes) are far more efficient for data
xcvrs.

Dice-a-roni...

Dice... Dice... very nice, so many people in the same deDice!

 

klaus.kragelund@gmail.com wrote in
news:ab8052ff-78a5-4ad8-91dd-bf63a368900d@googlegroups.com:

I have found a reflector I would try out

Another idea is to place an array of 20 of the LED, that should
give me much more combined intensity

COB has been ordered and will try it out next week

Cheers

Klaus

There is a street lamp up in a parking lot that has a VERY bright
LED flood lamp on it. During the day I went and looked at it. It is
about 14 (two rows of seven) 'pads' of ten LEDs each (2x5). So there
*must* be off the shelf parts out there in the form of high per LED
wattage arrays. Like ten 1 or 2 watt LEDs.

 

COB = PCB

Damn autocorrect function

 

[Phil Hobbs]

On 2020-02-12 10:01, DecadentLinuxUserNumeroUno@decadence.org wrote:

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:3dc0b0e9-5c74-0201-bbb5-f26329344340@electrooptical.net:

On 2020-02-12 03:47, Martin Brown wrote:
On 11/02/2020 02:30, Phil Hobbs wrote:
On 2020-02-10 04:17, Martin Brown wrote:

The OP's requirements are ultimately up against laws of
thermodynamics.

Yes, but the arguments are a bit subtler there since a LED is
not exactly a thermal source.

It means you can get away with a simple aspheric cast lens
though since the wavelength range of a colour LED emitter is
typically 50nm fwhm.

Spatial coherence is a big issue. A LED is a diffuse source with
light coming from inside a material with n~=3.45, so if you got
100% extraction efficiency you'd have a Lambertian source about 11
times the area. That makes the solid angle/beam diameter tradeoff
difficult.


You would be crazy to use a white LED in a signalling
application.

Actually not. I've designed a couple of tactical optical
transceivers for the US military, both of which used white LEDs.
You get a lot of bang for the buck--much better than the
corresponding blue LED dice.


Cheers

Phil Hobbs


Can place at the back of a black tube for stopping the side reads.

There were baffles and stuff for that, yeah.

Deep voice inflection... (coat it with Vantablack)(Bou! bou!)(oh
yeah!)(chic-a-chic-ahhh)

Just shiny black plastic. Shiny black baffles let you ray-trace the
reflections and make sure they hit black surfaces three or four times
before escaping. That gets you down into the PPMs.

Seems though that lasers (diodes) are far more efficient for data
xcvrs.

These were intended for handheld use on ships, e.g. the ramp rats
talking to the island on an aircraft carrier.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

I have found a reflector I would try out

Another idea is to place an array of 20 of the LED, that should give me much more combined intensity

COB has been ordered and will try it out next week

Cheers

Klaus

 

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:859c286a-3444-f3b4-cbf5-908b2113c7bc@electrooptical.net:

On 2020-02-12 10:01, DecadentLinuxUserNumeroUno@decadence.org
wrote:
Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:3dc0b0e9-5c74-0201-bbb5-f26329344340@electrooptical.net:

On 2020-02-12 03:47, Martin Brown wrote:
On 11/02/2020 02:30, Phil Hobbs wrote:
On 2020-02-10 04:17, Martin Brown wrote:

The OP's requirements are ultimately up against laws of
thermodynamics.

Yes, but the arguments are a bit subtler there since a LED is
not exactly a thermal source.

It means you can get away with a simple aspheric cast lens
though since the wavelength range of a colour LED emitter is
typically 50nm fwhm.

Spatial coherence is a big issue. A LED is a diffuse source
with light coming from inside a material with n~=3.45, so if you
got 100% extraction efficiency you'd have a Lambertian source
about 11 times the area. That makes the solid angle/beam
diameter tradeoff difficult.


You would be crazy to use a white LED in a signalling
application.

Actually not. I've designed a couple of tactical optical
transceivers for the US military, both of which used white LEDs.
You get a lot of bang for the buck--much better than the
corresponding blue LED dice.


Cheers

Phil Hobbs


Can place at the back of a black tube for stopping the side
reads.

There were baffles and stuff for that, yeah.

Deep voice inflection... (coat it with Vantablack)(Bou! bou!)(oh
yeah!)(chic-a-chic-ahhh)

Just shiny black plastic. Shiny black baffles let you ray-trace
the reflections and make sure they hit black surfaces three or
four times before escaping. That gets you down into the PPMs.

Seems though that lasers (diodes) are far more efficient for data
xcvrs.

These were intended for handheld use on ships, e.g. the ramp rats
talking to the island on an aircraft carrier.

Cheers

Phil Hobbs

A 'new' variant of the shutter windowed shipboard 'signal lamps'.

Hopefully at higher rate and better character set than morse code.

You should check that vantablack stuff out (for applicable uses of
course).

 

[Phil Hobbs]

On 2020-02-12 12:17, DecadentLinuxUserNumeroUno@decadence.org wrote:

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:859c286a-3444-f3b4-cbf5-908b2113c7bc@electrooptical.net:

On 2020-02-12 10:01, DecadentLinuxUserNumeroUno@decadence.org
wrote:
Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:3dc0b0e9-5c74-0201-bbb5-f26329344340@electrooptical.net:

On 2020-02-12 03:47, Martin Brown wrote:
On 11/02/2020 02:30, Phil Hobbs wrote:
On 2020-02-10 04:17, Martin Brown wrote:

The OP's requirements are ultimately up against laws of
thermodynamics.

Yes, but the arguments are a bit subtler there since a LED is
not exactly a thermal source.

It means you can get away with a simple aspheric cast lens
though since the wavelength range of a colour LED emitter is
typically 50nm fwhm.

Spatial coherence is a big issue. A LED is a diffuse source
with light coming from inside a material with n~=3.45, so if you
got 100% extraction efficiency you'd have a Lambertian source
about 11 times the area. That makes the solid angle/beam
diameter tradeoff difficult.


You would be crazy to use a white LED in a signalling
application.

Actually not. I've designed a couple of tactical optical
transceivers for the US military, both of which used white LEDs.
You get a lot of bang for the buck--much better than the
corresponding blue LED dice.


Cheers

Phil Hobbs


Can place at the back of a black tube for stopping the side
reads.

There were baffles and stuff for that, yeah.

Deep voice inflection... (coat it with Vantablack)(Bou! bou!)(oh
yeah!)(chic-a-chic-ahhh)

Just shiny black plastic. Shiny black baffles let you ray-trace
the reflections and make sure they hit black surfaces three or
four times before escaping. That gets you down into the PPMs.

Seems though that lasers (diodes) are far more efficient for data
xcvrs.

These were intended for handheld use on ships, e.g. the ramp rats
talking to the island on an aircraft carrier.

Cheers

Phil Hobbs


A 'new' variant of the shutter windowed shipboard 'signal lamps'.

Hopefully at higher rate and better character set than morse code.

You should check that vantablack stuff out (for applicable uses of
course).

Vantablack and other designer blacks aren't that useful in the general
case, because they scatter light into 2 pi steradians. Even if their
reflectance is very low, they aren't generally competitive compared with
properly-designed baffles. They're good for cleaning up the general
background scatter.

A lot of the time designer blacks get pulled in to fix a design problem.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

klaus.kragelund@gmail.com wrote in news:11c282f5-4ab5-48cb-9669-
c04f8766f952@googlegroups.com:

COB = PCB

Damn autocorrect function

Chip On Board.

At least it was not a complete morphing.

 

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:b1e5398c-c7f6-7016-6950-32d0a4bb920a@electrooptical.net:

On 2020-02-12 12:17, DecadentLinuxUserNumeroUno@decadence.org
wrote:
Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:859c286a-3444-f3b4-cbf5-908b2113c7bc@electrooptical.net:

On 2020-02-12 10:01, DecadentLinuxUserNumeroUno@decadence.org
wrote:
Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:3dc0b0e9-5c74-0201-bbb5-f26329344340@electrooptical.net:

On 2020-02-12 03:47, Martin Brown wrote:
On 11/02/2020 02:30, Phil Hobbs wrote:
On 2020-02-10 04:17, Martin Brown wrote:

The OP's requirements are ultimately up against laws of
thermodynamics.

Yes, but the arguments are a bit subtler there since a LED
is not exactly a thermal source.

It means you can get away with a simple aspheric cast lens
though since the wavelength range of a colour LED emitter is
typically 50nm fwhm.

Spatial coherence is a big issue. A LED is a diffuse source
with light coming from inside a material with n~=3.45, so if
you got 100% extraction efficiency you'd have a Lambertian
source about 11 times the area. That makes the solid
angle/beam diameter tradeoff difficult.


You would be crazy to use a white LED in a signalling
application.

Actually not. I've designed a couple of tactical optical
transceivers for the US military, both of which used white
LEDs. You get a lot of bang for the buck--much better than the
corresponding blue LED dice.


Cheers

Phil Hobbs


Can place at the back of a black tube for stopping the side
reads.

There were baffles and stuff for that, yeah.

Deep voice inflection... (coat it with Vantablack)(Bou!
bou!)(oh yeah!)(chic-a-chic-ahhh)

Just shiny black plastic. Shiny black baffles let you ray-trace
the reflections and make sure they hit black surfaces three or
four times before escaping. That gets you down into the PPMs.

Seems though that lasers (diodes) are far more efficient for
data xcvrs.

These were intended for handheld use on ships, e.g. the ramp
rats talking to the island on an aircraft carrier.

Cheers

Phil Hobbs


A 'new' variant of the shutter windowed shipboard 'signal
lamps'.

Hopefully at higher rate and better character set than morse
code.

You should check that vantablack stuff out (for applicable
uses of
course).


Vantablack and other designer blacks aren't that useful in the
general case, because they scatter light into 2 pi steradians.
Even if their reflectance is very low, they aren't generally
competitive compared with properly-designed baffles. They're good
for cleaning up the general background scatter.

A lot of the time designer blacks get pulled in to fix a design
problem.

Cheers

Phil Hobbs

Does the hubble tube have these baffles? I am unaware of the
configuration as I thought it was a smooth bore tube.

Anyway, I know that it does not have vantablack coating.

I want a car with several pieces done in it.

It is a forest of vertical carbon nanotubes and it lets light in,
but does not let it out.

 

[Phil Hobbs]

On 2020-02-12 12:55, DecadentLinuxUserNumeroUno@decadence.org wrote:

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:b1e5398c-c7f6-7016-6950-32d0a4bb920a@electrooptical.net:

On 2020-02-12 12:17, DecadentLinuxUserNumeroUno@decadence.org
wrote:
Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:859c286a-3444-f3b4-cbf5-908b2113c7bc@electrooptical.net:

On 2020-02-12 10:01, DecadentLinuxUserNumeroUno@decadence.org
wrote:
Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:3dc0b0e9-5c74-0201-bbb5-f26329344340@electrooptical.net:

On 2020-02-12 03:47, Martin Brown wrote:
On 11/02/2020 02:30, Phil Hobbs wrote:
On 2020-02-10 04:17, Martin Brown wrote:

The OP's requirements are ultimately up against laws of
thermodynamics.

Yes, but the arguments are a bit subtler there since a LED
is not exactly a thermal source.

It means you can get away with a simple aspheric cast lens
though since the wavelength range of a colour LED emitter is
typically 50nm fwhm.

Spatial coherence is a big issue. A LED is a diffuse source
with light coming from inside a material with n~=3.45, so if
you got 100% extraction efficiency you'd have a Lambertian
source about 11 times the area. That makes the solid
angle/beam diameter tradeoff difficult.


You would be crazy to use a white LED in a signalling
application.

Actually not. I've designed a couple of tactical optical
transceivers for the US military, both of which used white
LEDs. You get a lot of bang for the buck--much better than the
corresponding blue LED dice.


Cheers

Phil Hobbs


Can place at the back of a black tube for stopping the side
reads.

There were baffles and stuff for that, yeah.

Deep voice inflection... (coat it with Vantablack)(Bou!
bou!)(oh yeah!)(chic-a-chic-ahhh)

Just shiny black plastic. Shiny black baffles let you ray-trace
the reflections and make sure they hit black surfaces three or
four times before escaping. That gets you down into the PPMs.

Seems though that lasers (diodes) are far more efficient for
data xcvrs.

These were intended for handheld use on ships, e.g. the ramp
rats talking to the island on an aircraft carrier.

Cheers

Phil Hobbs


A 'new' variant of the shutter windowed shipboard 'signal
lamps'.

Hopefully at higher rate and better character set than morse
code.

You should check that vantablack stuff out (for applicable
uses of
course).


Vantablack and other designer blacks aren't that useful in the
general case, because they scatter light into 2 pi steradians.
Even if their reflectance is very low, they aren't generally
competitive compared with properly-designed baffles. They're good
for cleaning up the general background scatter.

A lot of the time designer blacks get pulled in to fix a design
problem.

Cheers

Phil Hobbs


Does the hubble tube have these baffles? I am unaware of the
configuration as I thought it was a smooth bore tube.

The one fatal mistake in baffle design is to allow any single-bounce
reflection to reach the detector. That is, as you go towards the
detector you shrink the baffle diameter so that the earlier stages are
out of the FOV of the later ones.

Telescopes are actually an easier case in general, because reflecting
the light back out the aperture is just as good as absorbing it.

The exoplanet detection folks are making the running in baffle design
these days, but I haven't looked very closely at what they're doing.
Soft rather than hard edges, I expect--that reduces the amplitude of the
diffraction rings just as windowing reduces leakage in spectral estimation.

Anyway, I know that it does not have vantablack coating.

I want a car with several pieces done in it.

It is a forest of vertical carbon nanotubes and it lets light in,
but does not let it out.

Plus or minus, depending on the angle of incidence. It turns out that
(if you can stand the outgassing) Edmund's black flocked paper is about
as good in the visible as any of the fancy designer blacks.

No use on spacecraft, of course.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com