We have a few relatively wimpy CONCAM1s still lying around in the CONCAM lab. What to do with them? One idea is to place them near CONCAM3s. The advantage would be that meteors seen by the sensitive CONCAM3s might then be found also on the simultaneously operating CONCAM1s. From these two images of the same meteor, a 3D path for the meteor might then be constructed.
The problem is that CONCAM1s, ST-237AS cameras from SBIG, are perhaps 10 times less sensitive than CONCAM3s (a rough guess). So they are also less good at picking up meteors. The thinking is that if the CONCAM3 sees a meteor, it will be easier to find in the higher background noise of the CONCAM1 data. But this means that the CONCAM1 cannot be "too far" from the CONCAM3 or the meteor would appear so far away that it would be impossible to locate. On the other hand, if the CONCAM1 is too close, then the tracks would effectively coincide and no 3D information on the meteor track would be obtainable.
So my questions are first -- is this a good idea? And second, what would be an optimal distance to place the CONCAM1 from the CONCAM3 to optimize meteor recovery and 3D information simultaneously?
- RJN
Placing CONCAM1s near CONCAM3s for meteor studies
Lior,
The problem with placing the CONCAM1 that far from the CONCAM3 is that the likely difference angle between meteor streaks on the two images would be on the order of 90 degrees. So it would be very hard to find the corresponding faint image streak on the CONCAM1.
The absolute minimum would be about 10 arcminutes, the angular size of a single pixel in the 1024x1024 array. Assuming a meteor height of 100 km, then to get an angular distance of 10 arcminutes you would have to place the CONCAM1 a distance of (100 km) (sin 10 arcmintes) = 0.3 kilometers away. In other words, at 0.3 kilometers, the CONCAM1 would be expected to see the meteor one pixel seperated, on the average, from the CONCAM3 image of the same meteor. Clearly this is too small to get a 3D image.
Perhaps 10 times this amount -- 3 kilometers would be good? Searching the surrounding 10 pixels of the brightest CONCAM3 meteor pixel for the brightest CONCAM1 meteor pixel -- and its surrounding trail -- seems doable. I am unsure the accuracy of the determination of the 3D trajectory in this case, though. Perhaps it is enough to do as good as we could do even at 100 km seperation between the CONCAMs?
- RJN
The problem with placing the CONCAM1 that far from the CONCAM3 is that the likely difference angle between meteor streaks on the two images would be on the order of 90 degrees. So it would be very hard to find the corresponding faint image streak on the CONCAM1.
The absolute minimum would be about 10 arcminutes, the angular size of a single pixel in the 1024x1024 array. Assuming a meteor height of 100 km, then to get an angular distance of 10 arcminutes you would have to place the CONCAM1 a distance of (100 km) (sin 10 arcmintes) = 0.3 kilometers away. In other words, at 0.3 kilometers, the CONCAM1 would be expected to see the meteor one pixel seperated, on the average, from the CONCAM3 image of the same meteor. Clearly this is too small to get a 3D image.
Perhaps 10 times this amount -- 3 kilometers would be good? Searching the surrounding 10 pixels of the brightest CONCAM3 meteor pixel for the brightest CONCAM1 meteor pixel -- and its surrounding trail -- seems doable. I am unsure the accuracy of the determination of the 3D trajectory in this case, though. Perhaps it is enough to do as good as we could do even at 100 km seperation between the CONCAMs?
- RJN
-
nbrosch
- Ensign
- Posts: 78
- Joined: Tue Jul 27, 2004 1:45 pm
- Location: Back at Tel Aviv University after a sabbatical
Two CONCAMs
We have, as you remember, one CONCAM-1 that we have not yet put into operation. I could check whether it would be possible to mount it on top of our site manager's house in the town of Mizpe Ramon. The location is about five km away from the WiseObs, near the edge of the town and at a higher elevation than most of the town. Thus the city lights should only be a minor disturbance. The house is equipped with an ADSL connection, not REAL internet as we have at the WiseObs and downloading the images might be a drag. We do, however, worry that the camera would be an item of interest to any "souvenir" collector...
Another alternative would be to install the CONCAM-1 at the Desert Research Institute in Sde Boker. This is the location we identified once we would have had that CONCAM set up with a chopper, as we originally intended. However, the student that was tasked with that evaporated. At Sde Boker we would have a protected location within the institute, the computer would be housed in healthy surroundings, and the camera would be installed on top of a ~4-meter tall pole.
The only question if this alternative is chosen would be the running software. We would use the tiny old laptop we had in our CONCAM-2 to which we fitted a new disk. It can be connected to the web, as far as I remember.
Any comments?
Noah
Another alternative would be to install the CONCAM-1 at the Desert Research Institute in Sde Boker. This is the location we identified once we would have had that CONCAM set up with a chopper, as we originally intended. However, the student that was tasked with that evaporated. At Sde Boker we would have a protected location within the institute, the computer would be housed in healthy surroundings, and the camera would be installed on top of a ~4-meter tall pole.
The only question if this alternative is chosen would be the running software. We would use the tiny old laptop we had in our CONCAM-2 to which we fitted a new disk. It can be connected to the web, as far as I remember.
Any comments?
Noah
Noah,
One idea is to use your CONCAM1 as an "offline" CONCAM. Hook it up to a big disk and just let it run. Create a "ring buffer" that deletes files older than a certain amount. Then, if a bright meteor is seen by the WO CONCAM2, you can ask your site manager to download that specific FITS file for comparison and 3D meteor path reconstruction.
With hard disks always getting bigger, i just saw a 300 Gig hard drive for sale for $300. on pricegrabber.com. Now at under 0.5 Gigs of data a night, the "ring buffer" of the CONCAM1 can now go for over a year! In fact, now that I think about it, with smaller files it should record perhaps only 0.1 Gigs a night, yielding a ring buffer of about 8 (eight) years! At that rate, some hardware will likely fail before the hard disk fills.
So my advice is to just set it up with a large hard disk and let it run off-line. You might have your site manager check it every month or so to make sure it is still taking usable images. Then go in and manually get only the rare files that you really need.
Come to think of it, we should do this for as many stations as possible -- there is no need for them to be on-line.
One question -- can these CONCAM1s be fit with a light chopper?
- RJN
One idea is to use your CONCAM1 as an "offline" CONCAM. Hook it up to a big disk and just let it run. Create a "ring buffer" that deletes files older than a certain amount. Then, if a bright meteor is seen by the WO CONCAM2, you can ask your site manager to download that specific FITS file for comparison and 3D meteor path reconstruction.
With hard disks always getting bigger, i just saw a 300 Gig hard drive for sale for $300. on pricegrabber.com. Now at under 0.5 Gigs of data a night, the "ring buffer" of the CONCAM1 can now go for over a year! In fact, now that I think about it, with smaller files it should record perhaps only 0.1 Gigs a night, yielding a ring buffer of about 8 (eight) years! At that rate, some hardware will likely fail before the hard disk fills.
So my advice is to just set it up with a large hard disk and let it run off-line. You might have your site manager check it every month or so to make sure it is still taking usable images. Then go in and manually get only the rare files that you really need.
Come to think of it, we should do this for as many stations as possible -- there is no need for them to be on-line.
One question -- can these CONCAM1s be fit with a light chopper?
- RJN
Dr. Nemiroff,
Your idea souds good, but there are several issues that should be taken into consideration:
First, the 300GB disk needs to be supported by the old laptop that is going to be used.
Second, when the system is off-line, there is no way to know whether it is working or not. According to my experience, in case of some interesting discovery we might find out that the system has stopped running 2 weeks earlier.
Third, from our experience with MK/HL, many transients seem like meteors, but only when you compare two frames (from different CONCAMs) you see that they can't be meteors. The ratio of meteor/non-meteor in MK/HL was something like 1:10. That means that the site manager will need to spend a lot of time getting the files for us.
Getting the files from the laptop is also not trivial.
Supporting this off-line system in case of a failure might also be a problem.
Your idea souds good, but there are several issues that should be taken into consideration:
First, the 300GB disk needs to be supported by the old laptop that is going to be used.
Second, when the system is off-line, there is no way to know whether it is working or not. According to my experience, in case of some interesting discovery we might find out that the system has stopped running 2 weeks earlier.
Third, from our experience with MK/HL, many transients seem like meteors, but only when you compare two frames (from different CONCAMs) you see that they can't be meteors. The ratio of meteor/non-meteor in MK/HL was something like 1:10. That means that the site manager will need to spend a lot of time getting the files for us.
Getting the files from the laptop is also not trivial.
Supporting this off-line system in case of a failure might also be a problem.
Hi Lior,
All good points. I don't have the answers but I can try to make up things that sound believable. Here goes.
About the 300 GB drive -- the CONCAM1s don't run from our standard Pelican case, so there is no room to put a computer at all. So a nearby external computer can run it from indoors. This computer can be a old used inexpensive 1 GHz that can just barely support the 300 GB drive.
Next, about finding if the system is working. Yes, if it is offline, there is no networked way to know if it is working. Still, the unit can be causually checked perhaps monthly or so to see that the hard drive is still recording data every 4 minutes at night like it should. The system should be configured to reboot itself and start back up if the power goes out.
Next, the meteor - non-meteor rate is of course a problem. Still, the longest and brighest meteors are less ambiguous, and those are the ones the CONCAM1s are likely to record. Also those streaks that occur during a meteor shower and point back to the radiant are quite clearly meteors. These meteors will be the ones that can be checked on the CONCAM1.
- RJN
All good points. I don't have the answers but I can try to make up things that sound believable. Here goes.
About the 300 GB drive -- the CONCAM1s don't run from our standard Pelican case, so there is no room to put a computer at all. So a nearby external computer can run it from indoors. This computer can be a old used inexpensive 1 GHz that can just barely support the 300 GB drive.
Next, about finding if the system is working. Yes, if it is offline, there is no networked way to know if it is working. Still, the unit can be causually checked perhaps monthly or so to see that the hard drive is still recording data every 4 minutes at night like it should. The system should be configured to reboot itself and start back up if the power goes out.
Next, the meteor - non-meteor rate is of course a problem. Still, the longest and brighest meteors are less ambiguous, and those are the ones the CONCAM1s are likely to record. Also those streaks that occur during a meteor shower and point back to the radiant are quite clearly meteors. These meteors will be the ones that can be checked on the CONCAM1.
- RJN
-
nbrosch
- Ensign
- Posts: 78
- Joined: Tue Jul 27, 2004 1:45 pm
- Location: Back at Tel Aviv University after a sabbatical
CONCAM1 for meteors
That was the project I entrusted the student with. I'll give here a short description of the thoughts, but first the problems:RJN wrote:Noah,
One question -- can these CONCAM1s be fit with a light chopper?
- RJN
1. The camera does not have a shutter thus there is no way to obtain a dark.
2. Any shutter that must be installed should be before the fisheye lens, because there is no space and no access between the lens and the CCD. In this case, the shutter needs protection from the elements.
To deal with these points we planned to build a two-blade chopper that would have been rotated by a synchronous motor taking its frequency from the power grid. The blades should have been wide enough to completely cover the fisheye lens and would have passed rather close to this lens. To protect the chopper, we planned to buy an Edmund Scientific acrylic display dome; these are a few tens of $$ and should survive at least one season.
The dark exposure problem was a bit more tricky to solve. We planned to put a magnet on one blade of the chopper and install a solenoid next to the fisheye lens. Upon command, when switching off the power to the synchronous motor the power to the solenoid would come on. The blade with the magnet would, in this case, have stopped so that it would have covered the lens. Restricting the dark exposures to night-time would have ensured that the dark exposures would be as dark as possible.
We reached a stage of prototyping this device with a wooden chopper and a motor salvaged from a turntable (I think) when the student disappeared. He is still claiming in Emails that he'd like to complete this project, but I have my misgivings.
Cheers,
Noah