Hanbury Brown--The Intensity Interferometer

Didn't he write the book "Boffin"? He talks about the fact that lots of people thought his interferometer couldn't work.

John

On May 12, 12:31=A0pm, Phil Hobbs wrote:

anbury/hanbury2.pdf

Thanks Phil! The previous discussion sparked me to review HBT. (OK, I was really learning some of it for the first time.) So I had the following ideas. You can do HBT by a correlation technique, two fast detectors followed by a multiplier. This is the traditional method and if your detectors are slower than the correlation time of the light source you lose signal by the ratio of the detector response time divided by the correlation time of your light source. Since you start with an HBT signal less than the shot noise, every halving of the signal means four times longer averaging time to recover it. It quickly starts to look hopeless for slow photodiode detectors (PDs). Not wanting to give up on cheap PDs, I thought one might instead add the two signals from separate PD's and then look at the noise in this combined signal. The correlation lives up at some high frequency, but the excess noise from HBT is present all the way down to DC. (I think?) Now if I had some method of modulating the HBT noise signal I might be able to find it amongst the larger shot noise. I first thought of simply separating the detectors as was done by HBT. Then I thought you could put a polarizer over one of the detectors and rotate a polarizer in front of the second. This would modulate the HBT noise but not the shot noise. Now all I needed was a good light source. Not so easy. I thought our Rubidium discharge lamp would work. It meets all the HBT criteria, lots of photons, a long correlation time, but the photons current (10=92s of pA) was down below the dark current of the PD. Then a colleague suggested looking at a diode laser run below threshold. This looks great! It=92s got a nice small source area, lots of light and a 10-20 nm bandwidth. What=92s more you should be able to run it above threshold and see the HBT effect go away! So a long post for a simple question. Will a laser diode run as a LED below the laser threshold current show HBT noise?

George Herold

On a sunny day (Tue, 12 May 2009 12:31:07 -0400) it happened Phil Hobbs wrote in :

Phil thank you for making this publication available. I am happy to see the old fight waves versus photon as particle here. Did you see this?:

The subject is very dear to me, and I have for example posted about explaining the photoelectric effect from the wave perspective in sci.physics repeatedly. Will not go into that here, but to add the Einstein ... well... sucks.

LOL

Yes. This book explains why.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

To be fair Sir Bernard Lovell thought it stood a chance and made it possible for the early experiments at Jodrell Bank. They started out at

125MHz on the sun and then moved upwards a long way in frequency!

Most radio astronomers and physicists thought that it might be possible to get intensity only fringes. Most optical astronomers didn't. Despite the scepticism they got fringes and measured the apparent diameters of a bunch of stars in ~1958. They went on to build a much bigger intensity interferometer setup at Narrabi(sp?) which appears on the dust jacket. They could detect limb darkening on a good day. It required exceptional experimental skill to get this stuff to work.

To get an idea of how hard it is to get fringes. The last experimentalist able to get optical interferometry to work in astronomy was Michelson who published a paper with Pease in 1920. After Michelson died noone was able to ever make the kit work properly again.

It is a good book and also marks the beginnings of the mathematics for what would eventually be used for VLBI in radio astronomy. There are now optical interferometers being built that use both intensity and closure phase measurements as good observables. This is an incredible stunt in terms of mechanical tolerances in path compensation to get it to work.

BTW Isn't the book still in copyright from 1973 ?

Regards, Martin Brown

It's been out of print for about 20 years, and had a tiny print run back then. Second-hand copies are $400. If the publisher still cares, he can make print-on-demand copies--I'd buy one, and I'd also take down the PDFs if asked. For liability, you have to show actual harm, which would be a bit difficult in this case. Once my book has been OP for 20 years, I _hope_ somebody posts it!

Stellar interferometry like Michelson, VLBI, and modern adaptive optics works on quite different physics from HBT. The issue was an old misunderstanding of quantum mechanics--for a long time, a lot of people thought that light from two different atomic sources couldn't interfere.

Hanbury Brown was a radar pioneer, who knew about local oscillators and advanced stuff like that, so he quite rightly didn't believe it. However, he needed Twiss (who was a theoretical physicist) to show from QM that it could indeed work.

Cheers

Phil Hobbs

Thanks for posting Phil! much appreciated. Jure Z.

BTW the third file is now fixed. The second file is missing P.

111--I'll see if it's in my hard copy.

Cheers

Phil Hobbs

I was about to say I can't see much wrong with PDF part 3 apart from the last few missing pages 174-180 which includes a few more references and the index. p111 is the opening of chapter 9 in my 1974 copy.

I am astonished that it commands $400 second hand now. I paid £6 for mine new in Manchester. I expect it was a fairly small print run as it was never destined to be a best seller - too many equations in it ;-)

Regards, Martin Brown

[Originally from sci.electronics.design. For the sci.optics and alt.lasers folk: I crossposted this due to commonality of interest. Hanbury Brown was a pioneer of radar and quantum optics, and is one of my technical heroes. This is a long-out-of-print technical book about an astrometric telescope based on the intensity interferometer, which works by cross-correlating the *noise* rather than the E field as in a normal interferometer. Really a beautiful device.] [Re copyright: the book has been OP for 20 years, and used copies are fetching $200-$400. Once my book has been OP that long, I hope somebody posts it someplace too!]

Okay, I finally located the hard copy, and all the pages are now there. Due to using a hand scanner, there are some page magnification and registration issues, but it's all there now.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs@electrooptical.net
http://electrooptical.net

[...]

Thanks. I put the pdf file through the djvu converter at

This OCR'D the file and made it text-searchable. The size dropped from 8097508 bytes down to 3111953 bytes. I don't know how long they keep the file on their server, but you can get it now at

If you don't have a viewer, you can get WinDjView at

Mike

Thanks, Mike, that's great. I put that on my website as well.

Cheers

Phil

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net

Hmm.. I can't seem to find it on your home page. Can you tell us where you put it??

OK, I finally found it. It's at

Ha! XNews doesn't recognize the djvu extension, so you can't simply double- click to download. But it downloads and renders fine.

Thanks,

Mike