9 ms·

How to find a tiny radioactive source while doing 70 kph

134 points by DrZootron 4y ago

sundvor 4y ago
Back story: Rio Tinto apologises for losing radioactive capsule in Australia https://www.bbc.com/news/business-64448879 https://www.bbc.com/news/business-64448879
- bpbp-mango 4y ago
  private company fucks up and government cleans up the mess :)
  - account42 4y ago
    Still better than private company fucks up and hides their mess, continues polluting environment for decades. The funds for this kind of search and similar emergency cleanups really need to be collected ahead of time (e.g. via taxes or required insurance) as making the company pay for it on the spot only encourages coverups.
    - Symbiote 4y ago
      In one of the articles yesterday, Rio Tinto said they'll pay the full cost of the recovery if the government sends them a bill.
  - dsfyu404ed 4y ago
    What's the other option? Let the company search for it? You'd be complaining that the government wasn't involved enough if that's what they did. The company offered to pay for the search before the search was even concluded. Not sure what more you can ask for.
    - bpbp-mango 4y ago
      fines and potentially losing their license for repeated mistakes
dmurray 4y ago
Good job, but a lot of the reports really exaggerate the size of the search area. > in the formidable conditions of the outback... > It’s fair to say finding a piece of equipment the size of a button in a vast desert was no easy task Let's not go too far here. It was two metres from the road and they knew exactly which 1400km of roads to check.
- gggggg5 4y ago
  I was rather baffled by so many sources claiming that this would be super difficult to find. It seemed like such a simple task to drive up and down roads with some radiation detection equipment, log the data and investigate any hotspots. Well, I guess it was? Or maybe I'm severely underestimating the complexity of this?
  - mschuster91 4y ago
    The problem was the unclear scope - it was not sure if the capsule didn't get stuck into someone's tire, an animal ate or played with it, or if it got damaged by a car running over it.
  - curiousgal 4y ago
    Well imagine if the pellet fell on the road, another car drove by and the pellet got stuck in one of the car's tire grooves. The car eventually makes it to a parking lot, or to the back of someone's ranch. Good luck finding it, short of monitoring hospitals for radiation poisoning patients.
    - maven29 4y ago
      Perhaps a nation-wide campaign to cover your phone camera lens with tape and then look for artifacts from gamma rays hitting the CMOS sensor?
      - hilbert42 4y ago
        Well, that's one way. Monitoring/surveillance cameras in hot spots in nuclear facilities are often known to have horrible fuzzy images (sometimes it's difficult to see what's being monitored). It's an interesting question how close one would have to get for the CMOS sensor to register notably in this instance. Given the stated radiation signature of this sample it's likely one would have to get pretty close, so it may not be a useful technique. I can't recall having read anything about smartphone CMOS sensors and radiation thresholds so I'd be interested in hearing from anyone whose knowledgeable about the matter.
defrost 4y ago
This is more of an ANSTO PR piece to promote a new piece of kit [1] based on a recent paper [2]. I'm guessing they loved the opportunity to field test this in a sea of global public interest. What they're not actually saying is that the source in question could have just as easily been found using 1990s technology in a low flying aircraft at 70m/sec ( ~250 km/hr ) - once you get the target peaks (which were bright against the background) in a 70m square it's a simple matter to wave handheld detectors about to zero in on the capsule by the road. Also of note "How to find .." is somewhat misleading is there is great deal of gloss here but relatively little substance. [1] https://www.ansto.gov.au/products/detection-and-imaging-coris360 https://www.ansto.gov.au/products/detection-and-imaging-cori... [2] https://iopscience.iop.org/article/10.1088/1748-0221/15/04/P04014 https://iopscience.iop.org/article/10.1088/1748-0221/15/04/P... [2a] https://sci-hub.ru/10.1088/1748-0221/15/04/P04014 https://sci-hub.ru/10.1088/1748-0221/15/04/P04014
- SideburnsOfDoom 4y ago
  > What they're not actually saying is that the source in question could have just as easily been found using 1990s technology in a low flying aircraft at 70m/sec Well, yes. But it turned out, to everyone's relief, that the optimistic assumption held: the capsule was in fact still lying near the side of the expected road. It could have gone another way - it could have been picked up and moved elsewhere by vehicle tyre, small animal or a human, in which case things would be much harder.
  - defrost 4y ago
    In which case an aircraft gives a better wide area search frame. Also, things that are possible are frequently unlikely - bits of metal fall from road trains all the time, the near road side is littered with bolts bits of tyre, etc. They largely don't travel far for the most part. Ergo: flying a plane on day one of known missing object would have been faster and cheap in the long run, provided more information, and been immediate leaving less time for the chance of a pick up and move elsewhere. Waiting several days for the ANSTO group to roll up increased the chances of movement but provided a chance for them to take the glory and road test their new toy in the field. It's a mixed bag of possibilities, and it's found now, so <shrug>, not much to see here.
cm2187 4y ago
Do gamma rays bounce back on a rock or do you need line of sight to capture the source? If it is the latter they were very lucky, effectively one small stone away from never being able to find it.
- closewith 4y ago
  Most rock is semi-transparent to gamma rays, which is how the lost source is intended to be used, to determine density of rock.
- sampo 4y ago
  If the gamma rays are at 662 keV [1], you'd need about 8cm of rock [2] to block 50%. For comparison, sunglasses typically block 70%-80% of visible light. [1] https://en.wikipedia.org/wiki/Caesium-137 https://en.wikipedia.org/wiki/Caesium-137 [2] https://pdfs.semanticscholar.org/7248/4e8995c811d31837199d7c629e51a882e3e0.pdf https://pdfs.semanticscholar.org/7248/4e8995c811d31837199d7c...
- amelius 4y ago
  If something is easy to block, then it's probably not very dangerous. So probably not the case.
- kennend3 4y ago
  When you think of things like Gamma or x-rays you cant relate them back to "light" because while there are similarities there are also a number of differences. Gamma penetrates rocks, these sensors are used to help determine the density of rock by measuring how much was stopped by the material. This is also why Gamma sources are so dangerous, because things like a coat, a car door are effectively "transparent" to them. If there was just "one small stone" in front of the source it would almost be like it wasn't present at all.
perihelions 4y ago
Here's a more technical/curiosity-satisfying answer to the question the title poses, https://inis.iaea.org/collection/NCLCollectionStore/_Public/22/072/22072114.pdf https://inis.iaea.org/collection/NCLCollectionStore/_Public/... ("Airborne Gamma Ray Spectrometer Surveying", 1991; 116p) There's some really interesting case studies in chapter 9 ("Searching for Radioactive Objects"). The first one was more or less identical to the highway incident in Australia: "On 21 June 1968, a 325 mCi 60Co source was lost in transit somewhere between Salt Lake City, Utah, and Kansas City, Missouri, a distance of 1800 km." The other two were an American test ICBM that crashed in remote desert in Mexico, and a Soviet orbital nuclear reactor that crashed in Canada.
collyw 4y ago
Wasn't this the opening scene to the Simpsons?
jacquesm 4y ago
The paranoid part of me has me thinking that all this sounds tailor made to showcase a product and a capability in a relatively safe setting. And you're really so hot on nuclear tech: don't lose your sources. That way you don't have to show off later on how good you are at finding them, aka the 'hero model' applied to nuclear physics.
aaron695 4y ago
[dead]
jonstewart 4y ago
In the US, the NNSA has helicopters with very sensitive radioactivity detectors. I live in DC and they‘ve flown straight rows over the city, back and forth, to create a background map of natural radiation. It’s not too surprising AU was able to find their missing capsule.
SeanLuke 4y ago
This is not the title of the article, and it should be changed. Besides, what's kph?
- bruce343434 4y ago
  km/h
- staringback 4y ago
  American moment
  - SeanLuke 4y ago
    Indeed. I use km/h
- IanCal 4y ago
  The given title here is much more informative than the article title of "WA outback proves no match for Aussie nuclear know-how". > Besides, what's kph? Kilometres per hour.
  - SeanLuke 4y ago
    > The given title here is much more informative than the article title of "WA outback proves no match for Aussie nuclear know-how". It's in violation of HN guidelines.
- angelbar 4y ago
  Not a imperial thing
nukenuke 4y ago
Cool use of gamma-ray imaging technology! The underlying technology is a coded aperture to give a random response on the detector that can be used to reconstruct the direction of the gamma-ray source. I think they are using a moving mask with a single detector in the middle. They probably didn't know if the source had broken apart and caused more contamination when they found the general area. So the imaging allows you to know from a distance if the source has been distributed without going into a potentially loose contamination area. Shameless plug: our company makes gamma-ray imaging systems and combines them with LiDAR mapping to make real-time 3D radiation maps: https://www.gammareality.com/ https://www.gammareality.com/ Basically we can make "nuclear street view" in real-time while driving around, or walking it around on Spot.
- thadt 4y ago
  Cool tech! Spent a chunk of a past life building radiation detection equipment and mapping it in various ways - it's neat to see the state of the art moving forward. Is your LAMP device using coded aperture (in that tiny form factor?) or some other technique for mapping directionality?
  - nukenuke 4y ago
    Awesome! we're using coded aperture and Compton imaging
webdoodle 4y ago
I was hoping it was going to be by using antineutrinos, something the U.S. and the U.K. have been developing behind the scenes: https://www.sheffield.ac.uk/physics/research/particle/neutrino/watchman https://www.sheffield.ac.uk/physics/research/particle/neutri...

134 points by DrZootron 4y ago

Back story: Rio Tinto apologises for losing radioactive capsule in Australia https://www.bbc.com/news/business-64448879 https://www.bbc.com/news/business-64448879

bpbp-mango 4y ago
private company fucks up and government cleans up the mess :)
- account42 4y ago
  Still better than private company fucks up and hides their mess, continues polluting environment for decades. The funds for this kind of search and similar emergency cleanups really need to be collected ahead of time (e.g. via taxes or required insurance) as making the company pay for it on the spot only encourages coverups.
  - Symbiote 4y ago
    In one of the articles yesterday, Rio Tinto said they'll pay the full cost of the recovery if the government sends them a bill.
- dsfyu404ed 4y ago
  What's the other option? Let the company search for it? You'd be complaining that the government wasn't involved enough if that's what they did. The company offered to pay for the search before the search was even concluded. Not sure what more you can ask for.
  - bpbp-mango 4y ago
    fines and potentially losing their license for repeated mistakes

dmurray 4y ago

Good job, but a lot of the reports really exaggerate the size of the search area. > in the formidable conditions of the outback... > It’s fair to say finding a piece of equipment the size of a button in a vast desert was no easy task Let's not go too far here. It was two metres from the road and they knew exactly which 1400km of roads to check.

gggggg5 4y ago
I was rather baffled by so many sources claiming that this would be super difficult to find. It seemed like such a simple task to drive up and down roads with some radiation detection equipment, log the data and investigate any hotspots. Well, I guess it was? Or maybe I'm severely underestimating the complexity of this?
- mschuster91 4y ago
  The problem was the unclear scope - it was not sure if the capsule didn't get stuck into someone's tire, an animal ate or played with it, or if it got damaged by a car running over it.
- curiousgal 4y ago
  Well imagine if the pellet fell on the road, another car drove by and the pellet got stuck in one of the car's tire grooves. The car eventually makes it to a parking lot, or to the back of someone's ranch. Good luck finding it, short of monitoring hospitals for radiation poisoning patients.
  - maven29 4y ago
    Perhaps a nation-wide campaign to cover your phone camera lens with tape and then look for artifacts from gamma rays hitting the CMOS sensor?
    - hilbert42 4y ago
      Well, that's one way. Monitoring/surveillance cameras in hot spots in nuclear facilities are often known to have horrible fuzzy images (sometimes it's difficult to see what's being monitored). It's an interesting question how close one would have to get for the CMOS sensor to register notably in this instance. Given the stated radiation signature of this sample it's likely one would have to get pretty close, so it may not be a useful technique. I can't recall having read anything about smartphone CMOS sensors and radiation thresholds so I'd be interested in hearing from anyone whose knowledgeable about the matter.

defrost 4y ago

This is more of an ANSTO PR piece to promote a new piece of kit [1] based on a recent paper [2]. I'm guessing they loved the opportunity to field test this in a sea of global public interest. What they're not actually saying is that the source in question could have just as easily been found using 1990s technology in a low flying aircraft at 70m/sec ( ~250 km/hr ) - once you get the target peaks (which were bright against the background) in a 70m square it's a simple matter to wave handheld detectors about to zero in on the capsule by the road. Also of note "How to find .." is somewhat misleading is there is great deal of gloss here but relatively little substance. [1] https://www.ansto.gov.au/products/detection-and-imaging-coris360 https://www.ansto.gov.au/products/detection-and-imaging-cori... [2] https://iopscience.iop.org/article/10.1088/1748-0221/15/04/P04014 https://iopscience.iop.org/article/10.1088/1748-0221/15/04/P... [2a] https://sci-hub.ru/10.1088/1748-0221/15/04/P04014 https://sci-hub.ru/10.1088/1748-0221/15/04/P04014

SideburnsOfDoom 4y ago
> What they're not actually saying is that the source in question could have just as easily been found using 1990s technology in a low flying aircraft at 70m/sec Well, yes. But it turned out, to everyone's relief, that the optimistic assumption held: the capsule was in fact still lying near the side of the expected road. It could have gone another way - it could have been picked up and moved elsewhere by vehicle tyre, small animal or a human, in which case things would be much harder.
- defrost 4y ago
  In which case an aircraft gives a better wide area search frame. Also, things that are possible are frequently unlikely - bits of metal fall from road trains all the time, the near road side is littered with bolts bits of tyre, etc. They largely don't travel far for the most part. Ergo: flying a plane on day one of known missing object would have been faster and cheap in the long run, provided more information, and been immediate leaving less time for the chance of a pick up and move elsewhere. Waiting several days for the ANSTO group to roll up increased the chances of movement but provided a chance for them to take the glory and road test their new toy in the field. It's a mixed bag of possibilities, and it's found now, so <shrug>, not much to see here.

cm2187 4y ago

Do gamma rays bounce back on a rock or do you need line of sight to capture the source? If it is the latter they were very lucky, effectively one small stone away from never being able to find it.

closewith 4y ago
Most rock is semi-transparent to gamma rays, which is how the lost source is intended to be used, to determine density of rock.
sampo 4y ago
If the gamma rays are at 662 keV [1], you'd need about 8cm of rock [2] to block 50%. For comparison, sunglasses typically block 70%-80% of visible light. [1] https://en.wikipedia.org/wiki/Caesium-137 https://en.wikipedia.org/wiki/Caesium-137 [2] https://pdfs.semanticscholar.org/7248/4e8995c811d31837199d7c629e51a882e3e0.pdf https://pdfs.semanticscholar.org/7248/4e8995c811d31837199d7c...
amelius 4y ago
If something is easy to block, then it's probably not very dangerous. So probably not the case.
kennend3 4y ago
When you think of things like Gamma or x-rays you cant relate them back to "light" because while there are similarities there are also a number of differences. Gamma penetrates rocks, these sensors are used to help determine the density of rock by measuring how much was stopped by the material. This is also why Gamma sources are so dangerous, because things like a coat, a car door are effectively "transparent" to them. If there was just "one small stone" in front of the source it would almost be like it wasn't present at all.

perihelions 4y ago

Here's a more technical/curiosity-satisfying answer to the question the title poses, https://inis.iaea.org/collection/NCLCollectionStore/_Public/22/072/22072114.pdf https://inis.iaea.org/collection/NCLCollectionStore/_Public/... ("Airborne Gamma Ray Spectrometer Surveying", 1991; 116p) There's some really interesting case studies in chapter 9 ("Searching for Radioactive Objects"). The first one was more or less identical to the highway incident in Australia: "On 21 June 1968, a 325 mCi 60Co source was lost in transit somewhere between Salt Lake City, Utah, and Kansas City, Missouri, a distance of 1800 km." The other two were an American test ICBM that crashed in remote desert in Mexico, and a Soviet orbital nuclear reactor that crashed in Canada.

collyw 4y ago

Wasn't this the opening scene to the Simpsons?

jacquesm 4y ago

The paranoid part of me has me thinking that all this sounds tailor made to showcase a product and a capability in a relatively safe setting. And you're really so hot on nuclear tech: don't lose your sources. That way you don't have to show off later on how good you are at finding them, aka the 'hero model' applied to nuclear physics.

aaron695 4y ago

[dead]

jonstewart 4y ago

In the US, the NNSA has helicopters with very sensitive radioactivity detectors. I live in DC and they‘ve flown straight rows over the city, back and forth, to create a background map of natural radiation. It’s not too surprising AU was able to find their missing capsule.

SeanLuke 4y ago

This is not the title of the article, and it should be changed. Besides, what's kph?

bruce343434 4y ago
km/h
staringback 4y ago
American moment
- SeanLuke 4y ago
  Indeed. I use km/h
IanCal 4y ago
The given title here is much more informative than the article title of "WA outback proves no match for Aussie nuclear know-how". > Besides, what's kph? Kilometres per hour.
- SeanLuke 4y ago
  > The given title here is much more informative than the article title of "WA outback proves no match for Aussie nuclear know-how". It's in violation of HN guidelines.
angelbar 4y ago
Not a imperial thing

nukenuke 4y ago

Cool use of gamma-ray imaging technology! The underlying technology is a coded aperture to give a random response on the detector that can be used to reconstruct the direction of the gamma-ray source. I think they are using a moving mask with a single detector in the middle. They probably didn't know if the source had broken apart and caused more contamination when they found the general area. So the imaging allows you to know from a distance if the source has been distributed without going into a potentially loose contamination area. Shameless plug: our company makes gamma-ray imaging systems and combines them with LiDAR mapping to make real-time 3D radiation maps: https://www.gammareality.com/ https://www.gammareality.com/ Basically we can make "nuclear street view" in real-time while driving around, or walking it around on Spot.

thadt 4y ago
Cool tech! Spent a chunk of a past life building radiation detection equipment and mapping it in various ways - it's neat to see the state of the art moving forward. Is your LAMP device using coded aperture (in that tiny form factor?) or some other technique for mapping directionality?
- nukenuke 4y ago
  Awesome! we're using coded aperture and Compton imaging

webdoodle 4y ago

I was hoping it was going to be by using antineutrinos, something the U.S. and the U.K. have been developing behind the scenes: https://www.sheffield.ac.uk/physics/research/particle/neutrino/watchman https://www.sheffield.ac.uk/physics/research/particle/neutri...