Neutral density filters and phase shift

[g***@gmail.com]

So we have been using these 0.3 ND (1/2 power) inconel
filters as beam splitters and 90 degree phase shifters
in a michelson interferometer... we also pick off the
reflected beam to get the other phase (sine and cosine).
Sometime several years ago someone-sigma changed the coating
to chromium, but didn't tell us. We can still get inconel
ND from elsewhere. So that OK

There is some uncertainty about where the
phase shift is, at the inconel/ air(glass) interface,
or is it the absorption in the inconel itself.
The fact that a 90 degree shift happens at 0.3ND
seems to point directly at absorption to me....
Is there something I might be missing?
0.3 chormium has less absorption...
hah! use some numbers to test theory,
what a concept.

George H.

[Phil Hobbs]

Post by g***@gmail.com
So we have been using these 0.3 ND (1/2 power) inconel
filters as beam splitters and 90 degree phase shifters
in a michelson interferometer... we also pick off the
reflected beam to get the other phase (sine and cosine).
Sometime several years ago someone-sigma changed the coating
to chromium, but didn't tell us. We can still get inconel
ND from elsewhere. So that OK
There is some uncertainty about where the
phase shift is, at the inconel/ air(glass) interface,
or is it the absorption in the inconel itself.
The fact that a 90 degree shift happens at 0.3ND
seems to point directly at absorption to me....
Is there something I might be missing?
0.3 chormium has less absorption...
hah! use some numbers to test theory,
what a concept.
George H.

We talked about this a year ago. I still think it's cool--lossless
beamsplitters are linear systems, but metal ones aren't, because the
total loss depends on the sum of the E fields in the two arms. (Of
course this works best in the S polarization, where the two arms have
the same direction of E.)

If you write the Fresnel formulae in terms of *k* instead of the
incident and refracted angles, lossy media and TIR are a lot easier to
account for. (The matching of transverse *k* at the boundary is the
physics behind the formulae, of course.) That's on P. 187 of my second
edition or P. 167 of the first.

I haven't gone through the math for this particular case though.

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

[g***@gmail.com]

Post by Phil Hobbs

Post by g***@gmail.com
So we have been using these 0.3 ND (1/2 power) inconel
filters as beam splitters and 90 degree phase shifters
in a michelson interferometer... we also pick off the
reflected beam to get the other phase (sine and cosine).
Sometime several years ago someone-sigma changed the coating
to chromium, but didn't tell us. We can still get inconel
ND from elsewhere. So that OK
There is some uncertainty about where the
phase shift is, at the inconel/ air(glass) interface,
or is it the absorption in the inconel itself.
The fact that a 90 degree shift happens at 0.3ND
seems to point directly at absorption to me....
Is there something I might be missing?
0.3 chormium has less absorption...
hah! use some numbers to test theory,
what a concept.
George H.

We talked about this a year ago. I still think it's cool--lossless
beamsplitters are linear systems, but metal ones aren't, because the
total loss depends on the sum of the E fields in the two arms. (Of
course this works best in the S polarization, where the two arms have
the same direction of E.)
If you write the Fresnel formulae in terms of *k* instead of the
incident and refracted angles, lossy media and TIR are a lot easier to
account for. (The matching of transverse *k* at the boundary is the
physics behind the formulae, of course.) That's on P. 187 of my second
edition or P. 167 of the first.
I haven't gone through the math for this particular case though.
Cheers
Phil Hobbs

Thanks Phil, I'd forgotten we talked about it already.
I was hoping for some quick and dirty result from the Kramers-Kronig
relations. I guess to really understand it I've got to work through
all the boundary conditions. (I haven't cracked open Jackson in a while)
For inconel maybe I can use the complex index of refraction for nickel?

George H.

Post by 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

[Phil Hobbs]

Post by g***@gmail.com

Post by Phil Hobbs

Post by g***@gmail.com
So we have been using these 0.3 ND (1/2 power) inconel
filters as beam splitters and 90 degree phase shifters
in a michelson interferometer... we also pick off the
reflected beam to get the other phase (sine and cosine).
Sometime several years ago someone-sigma changed the coating
to chromium, but didn't tell us. We can still get inconel
ND from elsewhere. So that OK
There is some uncertainty about where the
phase shift is, at the inconel/ air(glass) interface,
or is it the absorption in the inconel itself.
The fact that a 90 degree shift happens at 0.3ND
seems to point directly at absorption to me....
Is there something I might be missing?
0.3 chormium has less absorption...
hah! use some numbers to test theory,
what a concept.
George H.

We talked about this a year ago. I still think it's cool--lossless
beamsplitters are linear systems, but metal ones aren't, because the
total loss depends on the sum of the E fields in the two arms. (Of
course this works best in the S polarization, where the two arms have
the same direction of E.)
If you write the Fresnel formulae in terms of *k* instead of the
incident and refracted angles, lossy media and TIR are a lot easier to
account for. (The matching of transverse *k* at the boundary is the
physics behind the formulae, of course.) That's on P. 187 of my second
edition or P. 167 of the first.
I haven't gone through the math for this particular case though.
Cheers
Phil Hobbs

Thanks Phil, I'd forgotten we talked about it already.
I was hoping for some quick and dirty result from the Kramers-Kronig
relations. I guess to really understand it I've got to work through
all the boundary conditions. (I haven't cracked open Jackson in a while)
For inconel maybe I can use the complex index of refraction for nickel?

It's pretty flat with wavelength, so I'd probably just use a
normal-conductor model

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

[g***@gmail.com]

Post by Phil Hobbs

Post by g***@gmail.com

Post by Phil Hobbs

Post by g***@gmail.com
So we have been using these 0.3 ND (1/2 power) inconel
filters as beam splitters and 90 degree phase shifters
in a michelson interferometer... we also pick off the
reflected beam to get the other phase (sine and cosine).
Sometime several years ago someone-sigma changed the coating
to chromium, but didn't tell us. We can still get inconel
ND from elsewhere. So that OK
There is some uncertainty about where the
phase shift is, at the inconel/ air(glass) interface,
or is it the absorption in the inconel itself.
The fact that a 90 degree shift happens at 0.3ND
seems to point directly at absorption to me....
Is there something I might be missing?
0.3 chormium has less absorption...
hah! use some numbers to test theory,
what a concept.
George H.

We talked about this a year ago. I still think it's cool--lossless
beamsplitters are linear systems, but metal ones aren't, because the
total loss depends on the sum of the E fields in the two arms. (Of
course this works best in the S polarization, where the two arms have
the same direction of E.)
If you write the Fresnel formulae in terms of *k* instead of the
incident and refracted angles, lossy media and TIR are a lot easier to
account for. (The matching of transverse *k* at the boundary is the
physics behind the formulae, of course.) That's on P. 187 of my second
edition or P. 167 of the first.
I haven't gone through the math for this particular case though.
Cheers
Phil Hobbs

Thanks Phil, I'd forgotten we talked about it already.
I was hoping for some quick and dirty result from the Kramers-Kronig
relations. I guess to really understand it I've got to work through
all the boundary conditions. (I haven't cracked open Jackson in a while)
For inconel maybe I can use the complex index of refraction for nickel?

It's pretty flat with wavelength, so I'd probably just use a
normal-conductor model

OK.. thanks.. I was poking around at books, and then it struck me,
certainly Feynman did it.
I'm reading here...
http://www.feynmanlectures.caltech.edu/II_32.html

I'll hit you up for help right around section 32-7,
if I need it.

George H.

Post by Phil Hobbs
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