PDA

View Full Version : Telescope interferometer in space



philippeb8
2017-Mar-03, 06:05 AM
Is it possible to have multiple Hubble-like telescopes in space in a way it can act as a giant interferometer?


Regards,
philippeb8

Shaula
2017-Mar-03, 06:32 AM
Probably just about technologically feasible - just very expensive and difficult to do bearing in mind the requirements for stability and timing precision.

antoniseb
2017-Mar-03, 12:43 PM
I'd go a step beyond Shaula and say not possible for the foreseeable future. We can make interferometers in space at longer wavelengths, and the event horizon telescope (ground based) is the current state of the art being able to do it with millimeter wavelengths. Maybe in the next decade it will be possible to do it in the mid-infrared range. I am hopeful that it will be possible in the optical range in my lifetime, but I'm not counting on it.

philippeb8
2017-Mar-03, 03:59 PM
Here's a more complicated setup which would require less telescopes and which I think would work:

- Suppose you have 5 Hubble-like telescopes all aligned on the same longitude but with the help of the Earth's rotation then the telescopes on that longitude will swipe the entire imaginary spherical surface. That means these 5 moving telescopes can simulate, within half a day, a giant interferometer.

- What about using the Earth's rotation around the Sun as well? We could have these 5 telescopes pointing exactly in the same direction, for entire half days, for half a year.

I think for the price of 5 Hubble-like telescopes it's quite a good deal.

Grey
2017-Mar-03, 04:10 PM
Just to make it clear what the issue is, in order to be able to serve as an interferometer, the separation of the two devices needs to be precise to better than roughly the size of the wavelength you're observing. Since radio waves have a much longer wavelength than visible light, it's much easier. For an optical interferometer, you have to be able to keep the telescopes steady to within a couple hundred nanometers.

philippeb8
2017-Mar-03, 04:14 PM
Just to make it clear what the issue is, in order to be able to serve as an interferometer, the separation of the two devices needs to be precise to better than roughly the size of the wavelength you're observing. Since radio waves have a much longer wavelength than visible light, it's much easier. For an optical interferometer, you have to be able to keep the telescopes steady to within a couple hundred nanometers.

With the technology we have these days, it's easy to recompute the measurements based on the exact location and orientation of the telescope (and the wavelength measured).

philippeb8
2017-Mar-03, 05:32 PM
With the technology we have these days, it's easy to recompute the measurements based on the exact location and orientation of the telescope (and the wavelength measured).

All you need is for each telescope to memorise the amplitude of the photons with some 3D matrix (len width * len height * wavelength measured) and you can properly superimpose this matrix with the ones from the other telescopes. Here we simulate the interferometry using computers.

Shaula
2017-Mar-03, 05:57 PM
With the technology we have these days, it's easy to recompute the measurements based on the exact location and orientation of the telescope (and the wavelength measured).
Not to the tolerances required. It is not easy at all to get the time and space measurements to that level of precision in a dynamic system. Places like CHARA (https://en.wikipedia.org/wiki/CHARA_array) spend a lot of time and money to do interferometry at near IR wavelengths. As Grey says, you need sub-micron accuracy which is not easy.


All you need is for each telescope to memorise the amplitude of the photons with some 3D matrix (len width * len height * wavelength measured) and you can properly superimpose this matrix with the ones from the other telescopes. Here we simulate the interferometry using computers.
It is about phase, though. Not wavelength. And to get the phase you need something much more complex to build. If it were as easy as you say ... we'd already be doing it more.

Shaula
2017-Mar-03, 05:59 PM
I'd go a step beyond Shaula and say not possible for the foreseeable future. We can make interferometers in space at longer wavelengths, and the event horizon telescope (ground based) is the current state of the art being able to do it with millimeter wavelengths. Maybe in the next decade it will be possible to do it in the mid-infrared range. I am hopeful that it will be possible in the optical range in my lifetime, but I'm not counting on it.
For smaller baselines (100s of metres) we can do it in the near IR - there are several running or planned systems for this, there used to be one outside LA which I had the pleasure of visiting. Which really rammed home the challenges they faced!

philippeb8
2017-Mar-03, 06:05 PM
Not to the tolerances required. It is not easy at all to get the time and space measurements to that level of precision in a dynamic system. Places like CHARA (https://en.wikipedia.org/wiki/CHARA_array) spend a lot of time and money to do interferometry at near IR wavelengths. As Grey says, you need sub-micron accuracy which is not easy.


It is about phase, though. Not wavelength. And to get the phase you need something much more complex to build. If it were as easy as you say ... we'd already be doing it more.

"phase", sorry.

(If we could move the funds for dark matter / dark energy into this then that could accelerate the process).

But thanks for the information.

antoniseb
2017-Mar-03, 06:08 PM
For smaller baselines (100s of metres) we can do it in the near IR - there are several running or planned systems for this, there used to be one outside LA which I had the pleasure of visiting. Which really rammed home the challenges they faced!
Cool! Outside LA? Palomar, Wilson, someplace else?

Shaula
2017-Mar-03, 06:14 PM
Cool! Outside LA? Palomar, Wilson, someplace else?
It was Wilson - the CHARA array. Seeing the little carts used for adjusting the light path ... it was amazing. A real feat of technology.

Darrell
2017-Mar-03, 07:05 PM
The ESO facility at Paranal has been doing interferometery in the near infrared for years with four 8.2 meter and four 1.8 meter telescopes, and some very complex systems to combine their light. If I recall correctly it took them much longer than originally planned to get the interferometery systems up and running.

Ara Pacis
2017-Mar-03, 07:19 PM
If -big if- you could do it with 5 telescopes, then you could do it with 1, allowing half an orbit to give you an earth-sized baseline, and half a year to give you a 2 AU baseline. But the technical issues remain.

philippeb8
2017-Mar-03, 07:27 PM
If -big if- you could do it with 5 telescopes, then you could do it with 1, allowing half an orbit to give you an earth-sized baseline, and half a year to give you a 2 AU baseline. But the technical issues remain.

Well with 1 telescope then it would be even easier. As Shaula said: all you need is 1 phase telescope and you build a matrix based on that.

philippeb8
2017-Mar-03, 07:40 PM
Well with 1 telescope then it would be even easier. As Shaula said: all you need is 1 phase telescope and you build a matrix based on that.

Can you imagine the benefits? We know we'll get something out of this experiment whereas dark matter / dark energy we don't know.

antoniseb
2017-Mar-03, 07:40 PM
... But the technical issues remain.
Like the technical issue of making two observations half a year apart be simultaneous.

philippeb8
2017-Mar-03, 07:52 PM
Like the technical issue of making two observations half a year apart be simultaneous.

You just need to know the exact position of the telescope within "that sea of electromagnetic waves". If the telescope is at the 104th wave or the 505th wave later during the day is the same thing, as long as you can locate the telescope within these waves.

ngc3314
2017-Mar-03, 08:23 PM
You just need to know the exact position of the telescope within "that sea of electromagnetic waves". If the telescope is at the 104th wave or the 505th wave later during the day is the same thing, as long as you can locate the telescope within these waves.

If you could do that with one telescope, you'd be able to move it ahead of the wavefront faster than c, and if that were possible we wouldn't need the interferometry so much. A key feature is that the wavefront from real objects is stochastic due to the random emission processes, so it must be sampled at exactly matching points on the wavefront; same phase many peaks apart won't work. It takes at least two telescopes (or two apertures in a giant optical component) for interferometry, even for Earth-rotation aperture synthesis. The optical case is more difficult because there is technology allowing us to measure the phase of a radio detection, but not in the optical (so we have to use phase differences between paths, which can be measured from photon rates after the beams interfere).

To see how the state of technology compares to the requirements of space-based optical interferometry, check out the documents from the (since-cancelled) Space Interferometry mission (SIM). SIM in turn was largely a pathfinder project for one concept for the Terrestrial Planet Finder, which would have put multiple cryogenically-cooled ~1.5m telescopes on a precision-metered optical bench at about Jupiter's distance to reduce the background from zodiacal light. The precision needed is mostly within the state of the laboratory art, "only" needing development to make it work 5 AU away at cryogenic temperatures.

philippeb8
2017-Mar-03, 08:25 PM
You just need to know the exact position of the telescope within "that sea of electromagnetic waves". If the telescope is at the 104th wave or the 505th wave later during the day is the same thing, as long as you can locate the telescope within these waves.

Once again you measure and record phase matrix "snapshots" based on the and position, direction and time, send that to the Earth's labs and process that data using the desired wavelength. Other than a highly precision location device, I don't see anything blocking us.

philippeb8
2017-Mar-03, 08:40 PM
same phase many peaks apart won't work

I see.



To see how the state of technology compares to the requirements of space-based optical interferometry, check out the documents from the (since-cancelled) Space Interferometry mission (SIM). SIM in turn was largely a pathfinder project for one concept for the Terrestrial Planet Finder, which would have put multiple cryogenically-cooled ~1.5m telescopes on a precision-metered optical bench at about Jupiter's distance to reduce the background from zodiacal light. The precision needed is mostly within the state of the laboratory art, "only" needing development to make it work 5 AU away at cryogenic temperatures.

Thanks for the information.

WaxRubiks
2017-Mar-03, 09:17 PM
What about two mirrors on opposite sides of the Earth, reflecting light back to a third satalite, that recombines the light?

StupendousMan
2017-Mar-04, 04:16 AM
One of the issues involved here is the requirement to measure the phase and amplitude of light waves detected by each antenna (or telescope) in the array. In the radio regime, engineers can easily build receivers which can measure both of these quantities, and record them as they arrive for future reference. So, in the radio case, we can simply write the measured phase and amplitude to a tape drive (or hard disk drive) at each telescope separately. Later, we bring the records to a single location, and perform the mathematical calculations on them there.

But in the optical, most detectors measure only the amplitude of the wave -- not its phase. There's no way one can recover the phase information after the fact. Thus, most optical interferometers are designed to bring the light waves gathered by two (or more) different telescopes to a single location, allow them to interfere, and then record the result. It's that step of "bringing two light waves to a single location" that's the hard part.