I'm completely out of my depths here but hear me out: could pulsars be observed with such a thing? That would definitely be the icing on the cake.
A way to precisely point the dish would be needed and an equatorial mount to compensate for Earth's rotation. If the signal was too faint, maybe some lock-in amplifier magic by creating a synthetic trigger signal with the expected rotational frequency of the targeted pulsar and then shifting the phase until the signal is maximized?
The "lock-in magic" you are describing is called "phase folding" in the pulsar community and can be done in software. The expected rotational frequency would be taken from an ephemeris file and there is tools to search for it, if unknown, as well. If the telescope is sensitive enough, is a good question. I don't see any results in the article that shows sensitivity numbers.
> I'm completely out of my depths here but hear me out: could pulsars be observed with such a thing? That would definitely be the icing on the cake.
In principle, yes. But I think in practice it would be difficult at best. Pulsars become fainter at higher frequencies, and this dish looks relatively small. So I suspect that combination of factors would mean that there aren't many (if any) pulsars that one could detect with this setup.
> expected rotational frequency of the targeted pulsar and then shifting the phase until the signal is maximized?
A technique like this is one way to search for pulsars. Though you don't use trigger signals since you don't a priori know the pulsar's spin period. So you record data for a period of time, then try many timesteps over which to fold the data and see if there's a pulse at that period. I'm sure there's better ways now, for it to be done.
Yes, ONE telescope may not be able to adequately resolve a pulsar at that frequency.
But a /network of/ these telescopes, appropriately coordinated, that's a different story. And seems like a logical next step. You could use GPS conditioned timekeeping and standardized directional setup to coordinate data from multiple telescopes. We do that in some distributed physics projects like cosmic ray studies.
I'm not sure what you mean by "resolve"; are you referring to resolving the pulsar's pulse in time?
I was speaking more about the flux detection limit of such a dish (implicitly assuming the receiver could make sufficiently short measurements to enable folding of the data to detect pulses in the time-folded dataset). One could try to coherently sum the measurements from a number of telescopes to increase the signal to noise.
Alternately, one could also try to detect the pulsar by averaging over the pulse profile, but that still requires that the telescope+receiver sensitivity is better than the period-averaged flux density. But then you're risking confusion of other, continuum, radio sources in the beam.
It's still the case that most pulsars are much fainter few GHz frequencies than they are 1 GHz (e.g., https://arxiv.org/abs/1302.2053). Though there are likely selection effects (discussed in the linked paper), most of the pulsars we know about can be expected to be ~40x fainter at 10 GHz than they are at 1 GHz. The linked paper cites a 6.5 GHz survey that identified 18 pulsars (compared to > 1000 detected in the ~1.4 GHz survey).
I suppose it depends on what one's aims are, though. Someone wanting to only detect pulsars in general, it'd be easier to do at lower frequencies. But there's certainly some science to be done by observing them at higher frequencies.
Well, I am a physicist, so this wasn't purely techno-babble but since I never had anything to do with astronomy or radio-observations a caveat seemed warranted.
From the linked article: "... With their system [...] they were able to estimate that the minimum dish aperture required to observe the Vela pulsar would be 6m, noting that the Vela pulsar is probably the strongest pulsar ever detected"
As a kid, I built a crystal radio. There it was the same: with a very long wire, even such a simple radio could receive far-away stations
An excellent next step would be to use a distributed mesh of these and do interferometry to resolve spatial details on the sky.
Observations could be distributed with bittorrent and coordinated over a blockchain. There exists off the shelf open source interferometry packages used by professional astronomers today that can be used for the analysis.
Accurate positioning is needed but can often be achieved to within a couple of meters using google maps and a sat image of your back yard.
EDIT: Do the downvoters hate interferometry? Oh no, it's just because I used the world blockchain. Chill people this is a thread about a blog post about turning a piece of trash into a fun day of amateur astronomy.
Accurate timing is also needed to shift and combine the signals from the various dishes.
> often be achieved to within a couple of meters using google maps and a sat image of your back yard.
Position accuracy requirements are fractions of the wavelength being observed. So for this 11.2 GHz dish (26.8 mm), the position of the dish needs to be known to much better than 2mm.
In practice this can be done by observing a bright source with a well-known sky position and then solving for the antenna positions. Aspects of this are discussed in Memo 503 from the Atacama Large Millimeter Array Memo series. http://legacy.nrao.edu/alma/memos/html-memos/alma503/memo503...
Yes, super-duper accurate timing. A nanosecond of timing error means your focal point moves by an angle equal to 1 foot divided by the distance between dishes.
You can calibrate out static timing errors, like due to different lengths of wire. But if the delay is varying for any reason your focal point will be slewing all over the heavens.
Absolute positioning is not that important as long as the relative positions are constant to within a small fraction of the wavelength, in this case to within a millimeter. Which is feasible if you mount it to the side of a building and there isn't too much wind.
The much bigger problem is that you need a very accurate clock. Much better than what you can get from a GPS receiver.
Are there any crystal or photonic oscillators that are up to the task for this time keeping? Or is RF interferometry bound to something like a chip atomic clock [1] (~$4k in low quantities)?
It's been almost ten years since I last done any of the back end math for this kind of thing, but you could probably get within the ballpark using COTS rubidium oscillators like those used for cell towers. They often show up on auction sites at affordable prices.
Use the pulse output to lock a station clock for your local time needs (like syncing your data recording to wall time) and the sine output to lock your mixing oscillators.
Except you need closer to 0.01ns for this use case instead of the 55ns you get out of the GPS receivers. And half your data would have more, possibly much more, deviation than that anyway. If you don't know which half, that will lead to loss of sensitivity and possibly artifacts in the reconstructed signal. Interferometry is actually hard.
GPS receiver accuracy relative to the constellation's time base is much much better than 55n. I'm working with a cheap (< 100 USD in prototype qty) OCXO that's good to a few parts in 10^-12 over 10 second periods. The GPS timebase varies from the TAI time base by up to a couple dozen nanoseconds, but it is very slowly varying.
A good GPS-disciplined OCXO can hold that accuracy over much longer intervals.
To take advantage of that accuracy you have to share the master reference oscillator between the telescope's clock tree and GPS receiver's clock tree. A commercial chipset's PPS output isn't going to cut it, but a dedicated amateur could definitely build a GPS-tagged radio telescope off of a common crystal like that.
Something like that might work. I am surprise that the temperature stabilized local oscillator is so cheap. Care to share a link?
The other thing that can help is to put receivers not too far apart. Say 25 dishes in an area with 100 meters diameter. That improves sensitivity by an order or magnitude, resolution by two orders of magnitude and the station can either be connected with a clock network, or (if using GPS) will at least see correlated clock errors that don't matter much in the correlation. The 1 part in 10^-7 accuracy of the GPS receivers is plenty accurate for the phase folding.
We've measured Milliren 220 series to be about 5e-12 at 10 seconds, but that's not fair since it turns out that part is "call for quote", even if we only bought it in small qty.
Conner-Winfield OH320 advertises an Allan Deviation of 2.7e-12 at one second. Its a very similar part, without all the hirel and space qualifications. 98 USD/ea at Digi-Key.
I agree wholeheartedly on the low-baseline setup case. An offline carrier phase position-time solution could stretch you out to much wider baselines, though.
I'm completely out of my depths here but hear me out: could pulsars be observed with such a thing? That would definitely be the icing on the cake.
A way to precisely point the dish would be needed and an equatorial mount to compensate for Earth's rotation. If the signal was too faint, maybe some lock-in amplifier magic by creating a synthetic trigger signal with the expected rotational frequency of the targeted pulsar and then shifting the phase until the signal is maximized?
The "lock-in magic" you are describing is called "phase folding" in the pulsar community and can be done in software. The expected rotational frequency would be taken from an ephemeris file and there is tools to search for it, if unknown, as well. If the telescope is sensitive enough, is a good question. I don't see any results in the article that shows sensitivity numbers.
> I'm completely out of my depths here but hear me out: could pulsars be observed with such a thing? That would definitely be the icing on the cake.
In principle, yes. But I think in practice it would be difficult at best. Pulsars become fainter at higher frequencies, and this dish looks relatively small. So I suspect that combination of factors would mean that there aren't many (if any) pulsars that one could detect with this setup.
> expected rotational frequency of the targeted pulsar and then shifting the phase until the signal is maximized?
A technique like this is one way to search for pulsars. Though you don't use trigger signals since you don't a priori know the pulsar's spin period. So you record data for a period of time, then try many timesteps over which to fold the data and see if there's a pulse at that period. I'm sure there's better ways now, for it to be done.
PRESTO is one of the major pieces of software used to search for pulsars: https://www.cv.nrao.edu/~sransom/presto/
Yes, ONE telescope may not be able to adequately resolve a pulsar at that frequency.
But a /network of/ these telescopes, appropriately coordinated, that's a different story. And seems like a logical next step. You could use GPS conditioned timekeeping and standardized directional setup to coordinate data from multiple telescopes. We do that in some distributed physics projects like cosmic ray studies.
I'm not sure what you mean by "resolve"; are you referring to resolving the pulsar's pulse in time?
I was speaking more about the flux detection limit of such a dish (implicitly assuming the receiver could make sufficiently short measurements to enable folding of the data to detect pulses in the time-folded dataset). One could try to coherently sum the measurements from a number of telescopes to increase the signal to noise.
Alternately, one could also try to detect the pulsar by averaging over the pulse profile, but that still requires that the telescope+receiver sensitivity is better than the period-averaged flux density. But then you're risking confusion of other, continuum, radio sources in the beam.
It's still the case that most pulsars are much fainter few GHz frequencies than they are 1 GHz (e.g., https://arxiv.org/abs/1302.2053). Though there are likely selection effects (discussed in the linked paper), most of the pulsars we know about can be expected to be ~40x fainter at 10 GHz than they are at 1 GHz. The linked paper cites a 6.5 GHz survey that identified 18 pulsars (compared to > 1000 detected in the ~1.4 GHz survey).
I suppose it depends on what one's aims are, though. Someone wanting to only detect pulsars in general, it'd be easier to do at lower frequencies. But there's certainly some science to be done by observing them at higher frequencies.
If you want to do that at 140MHz, sure. But at 11GHz interferometry is -- despite being conceptually the same -- quite hard.
Doesn't sound like you're out of your depth. It's been done with a worse receiver - https://www.rtl-sdr.com/detecting-pulsars-rotating-neutron-s...
Well, I am a physicist, so this wasn't purely techno-babble but since I never had anything to do with astronomy or radio-observations a caveat seemed warranted.
Thanks for the link btw!
Worse receiver, but better antenna:
From the linked article: "... With their system [...] they were able to estimate that the minimum dish aperture required to observe the Vela pulsar would be 6m, noting that the Vela pulsar is probably the strongest pulsar ever detected"
As a kid, I built a crystal radio. There it was the same: with a very long wire, even such a simple radio could receive far-away stations
An excellent next step would be to use a distributed mesh of these and do interferometry to resolve spatial details on the sky.
Observations could be distributed with bittorrent and coordinated over a blockchain. There exists off the shelf open source interferometry packages used by professional astronomers today that can be used for the analysis.
Accurate positioning is needed but can often be achieved to within a couple of meters using google maps and a sat image of your back yard.
EDIT: Do the downvoters hate interferometry? Oh no, it's just because I used the world blockchain. Chill people this is a thread about a blog post about turning a piece of trash into a fun day of amateur astronomy.
> Accurate positioning is needed
Accurate timing is also needed to shift and combine the signals from the various dishes.
> often be achieved to within a couple of meters using google maps and a sat image of your back yard.
Position accuracy requirements are fractions of the wavelength being observed. So for this 11.2 GHz dish (26.8 mm), the position of the dish needs to be known to much better than 2mm.
In practice this can be done by observing a bright source with a well-known sky position and then solving for the antenna positions. Aspects of this are discussed in Memo 503 from the Atacama Large Millimeter Array Memo series. http://legacy.nrao.edu/alma/memos/html-memos/alma503/memo503...
Yes, super-duper accurate timing. A nanosecond of timing error means your focal point moves by an angle equal to 1 foot divided by the distance between dishes.
You can calibrate out static timing errors, like due to different lengths of wire. But if the delay is varying for any reason your focal point will be slewing all over the heavens.
Absolute positioning is not that important as long as the relative positions are constant to within a small fraction of the wavelength, in this case to within a millimeter. Which is feasible if you mount it to the side of a building and there isn't too much wind.
The much bigger problem is that you need a very accurate clock. Much better than what you can get from a GPS receiver.
And lose the blockchain bullshit.
Are there any crystal or photonic oscillators that are up to the task for this time keeping? Or is RF interferometry bound to something like a chip atomic clock [1] (~$4k in low quantities)?
[1] https://coverclock.blogspot.com/2017/05/my-stratum-0-atomic-...
It's been almost ten years since I last done any of the back end math for this kind of thing, but you could probably get within the ballpark using COTS rubidium oscillators like those used for cell towers. They often show up on auction sites at affordable prices.
Use the pulse output to lock a station clock for your local time needs (like syncing your data recording to wall time) and the sine output to lock your mixing oscillators.
I suspect that you can get a start by looking at how cosmic ray studies conducted at different sites are coordinated:
https://ieeexplore.ieee.org/document/1351816
Except you need closer to 0.01ns for this use case instead of the 55ns you get out of the GPS receivers. And half your data would have more, possibly much more, deviation than that anyway. If you don't know which half, that will lead to loss of sensitivity and possibly artifacts in the reconstructed signal. Interferometry is actually hard.
GPS receiver accuracy relative to the constellation's time base is much much better than 55n. I'm working with a cheap (< 100 USD in prototype qty) OCXO that's good to a few parts in 10^-12 over 10 second periods. The GPS timebase varies from the TAI time base by up to a couple dozen nanoseconds, but it is very slowly varying.
A good GPS-disciplined OCXO can hold that accuracy over much longer intervals.
To take advantage of that accuracy you have to share the master reference oscillator between the telescope's clock tree and GPS receiver's clock tree. A commercial chipset's PPS output isn't going to cut it, but a dedicated amateur could definitely build a GPS-tagged radio telescope off of a common crystal like that.
Something like that might work. I am surprise that the temperature stabilized local oscillator is so cheap. Care to share a link?
The other thing that can help is to put receivers not too far apart. Say 25 dishes in an area with 100 meters diameter. That improves sensitivity by an order or magnitude, resolution by two orders of magnitude and the station can either be connected with a clock network, or (if using GPS) will at least see correlated clock errors that don't matter much in the correlation. The 1 part in 10^-7 accuracy of the GPS receivers is plenty accurate for the phase folding.
We've measured Milliren 220 series to be about 5e-12 at 10 seconds, but that's not fair since it turns out that part is "call for quote", even if we only bought it in small qty.
Conner-Winfield OH320 advertises an Allan Deviation of 2.7e-12 at one second. Its a very similar part, without all the hirel and space qualifications. 98 USD/ea at Digi-Key.
I agree wholeheartedly on the low-baseline setup case. An offline carrier phase position-time solution could stretch you out to much wider baselines, though.