Global Radiation monitoring system / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

Global Radiation monitoring system

first previous 2 3 4 5 next last

Move Thread

LAN_403

Linas

Tue Jul 16 2013, 03:29PM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

so my original idea, is to show ticks per second, or if you want to standardize this go for ticks/(s*cross_section_of_detector) or something like that

Steve Conner

Tue Jul 16 2013, 05:08PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

That's how Radiationnetwork.com does it, they simply report the raw counts per minute. (the American equivalent of Radu's project)

Proud Mary

Tue Jul 16 2013, 05:12PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Steve Conner wrote ...

I think this device is so inaccurate that it'll only be good for scaring people.

However I've seen Russian Geiger counters that read out in uSv/hr probably using a similar setup and calibration, so maybe there is a precedent for it.

SBM20 calibration data gives gamma sensitivity Ra-226: 29 cps/mR/hr, and Co-60: 22 cps/mR/hr.

Ra-226 has many radioactive progeny, resulting in gamma emission across widely varying yields and energies between about 50 keV and 2.5 MeV. According to Calibration of Survey Instruments Used in Radiation Protection for the Assessment of Ionizing Radiation Fields and Radioactive Surface Contamination, NRCP Report 112, 1991, p. 54, Ra-226 has an effective gamma energy of 800 keV.

Co-60, by contrast, has an effective gamma energy around 1.3 MeV.

I would guess that SBM-20's higher counting rate with Ra-226 is due to significant emission <100 keV where the energy response curve of uncompensated GM tubes is at its most non-linear.

Because of Ra-226's wide spread of gamma energies compared with Co-60, I would suggest using the Ra-226 gamma sensitivity figure in Radhu's project i.e. 29 cps/mR/hr.

Finally, it is always a good idea to check a GM tube's plateau slope gradient at 5V intervals and compare the results with the data sheet figures before using any 'new old stock' parts - but so long as the slope is OK, moderate accuracy is possible with careful design.

radhoo

Tue Jul 16 2013, 08:37PM

Registered Member #1938 Joined: Sun Jan 25 2009, 12:44PM
Location: Romania
Posts: 699

Hi all,
As Stella kindly indicated, the geiger counters cannot distinguish between levels of energy. A GM tube will record a simple pulse as the radiation quanta intersects with the tube's volume space, resulting in emissions of electrons, captured between the internal electrodes, but will provide no info on what the energy level was.

Steve Conner wrote ...

I agree, I think this device is so inaccurate that it'll only be good for scaring people.

However I've seen Russian Geiger counters that read out in uSv/hr probably using a similar setup and calibration, so maybe there is a precedent for it.

The only thing people should be scared of is a rising graph. The current setup shows good stability, and is running since October 2012. I gathered the entire radiation data from October 21 to date, and plotted it to a graph, see it attached. The peaks are most likely to avalanche discharges in the tube. I don't know the exact cause of these events, but I suspect these to be misreadings, as on a few ocasions I used a secondary detector to rule out these events. But the idea is the background stayed quite constant, as expected. We would have seen a completely different graphic if something like Fukushima event would have happened near my device. And we all know how much disinformation is associated with such events.

Last year when I completed my first uRADMonitor station, I got some feedback from dr. Nicolae Victor Zamfir, director of the Horia Hulubei National Institute of Physics and Nuclear Engineering, pointing out that any attempt of guessing uSV/h readings out of CPM would be â€œan approximationâ€ in the best case. Read it if you like here:

, using an automated translator.

For compensating the lack of response towards the true energy levels, we might use a different type of detector (like scintillation detectors) , but there are not many true alternatives given the costs. PIN photodiode might be a choice, rather poor I'd say, due to their relative insensitivity, and lacking the capability of absorbing the entire radiation energy, giving out false readings. Sure they can be compensated using statistical methods, I'm not really interested in doing that at this point.

So this takes us back to the geiger tubes, indeed almost ALL detectors available on the market (commercial or professional) are showing the equivalent dose (measured in Sv/h). They do that with a given tolerance. When doing tests like in the video posted above, I was happy to see that Radex RD1706 and Terra-P are showing similar readings to my device, when tested against the 1B24 tube. So let's not make things more complicated than required. To move from cpm to Sv/h I used the CPM readings for : 1)background 2)an Am241 source 3) a Cs137 source. The data was then scaled for arbitrary CPM readings, resulting what you see on screen. If my device works at least as good as the Terra-P, I would say I did a decent job.

But here is another answer I gave to somebody concerned about the quality of uRADmonitor's calibration :
"Regarding the â€œcomparable readingsâ€, your point makes sense.

But .. Iâ€™m not sure how relevant it is. Be it placed in various ways outside (closer to -or- far away from the ground), inside (ventilated or not, etc), weâ€™ll get very much the same readings:
- a poorly ventilated cellar shows 0.17uSv/h
- a concrete made storage room shows 0.15uSv/h
- the room Iâ€™m living in shows 0.13uSv/h
- a salt mine shows 0.06uSv/h
- the inside of a house made of bricks 0.12uSv/h
- etc
Does it really matter?

No.

From a scientific point of view, the tiny differences would probably make sense: why is the 0.0X difference there and what are the phenomenons implicated, but for the end user is very much just the same. We will probably wonâ€™t get more than 0.10uSv/h in differences attributed to user misuse, misplace or other errors. We are still in the safe zone.

But things change when the monitor placed outside your house shows increasing readings, that quickly jump from 0.13uSv/h to more than 1uSv/h like it happened in Japan.
It sure means a lot when you live in a country close to another where a plant has gone critical, and you worry about the wind direction, and then you see you got lucky, but your friends somewhere far away didnâ€™t get that lucky as monitors close to their location register increasing readings.

IMO, Itâ€™s not the small differences caused by mounting the device differently that matter. But the significant changes, that when plotted on a map, can provide some great indications on the cause of the incident: The variations will be there, regardless of how well the monitors are installed."

Looking forward to your feedback . Big radiation chart attached, see it at full size:

Proud Mary

Tue Jul 16 2013, 11:03PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

It seems to me that the exact correlation of c.p.m. with Sv/hr in this experiment is of secondary importance compared with establishing a Law of Large Numbers base line against which the significance of any deviations can be evaluated.

Achieving this stable base-line needs good long-term mechanical, electrical, and thermal stability. It also needs freedom from moisture, vibration, EMI, gross contamination of the apparatus by, for example, roosting birds, and not forgetting cable damage by rats and squirrels, and infestation of the apparatus by insects.

Steve Conner

Wed Jul 17 2013, 09:10AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Well, surely the idea of a network of radiation monitors is that they won't all get infested with insects simultaneously.

The idea of ignoring the absolute value and looking only at changes is called "trending". It works well and is used in industrial condition monitoring. If a machine suddenly starts vibrating more than it used to, or its oil pressure starts to fall, you can probably guess something bad is about to happen.

However, it's not the change in radiation level that damages health, it's the absolute level. Even 1uSv/h is probably nothing to worry about. How were the "Low", "Normal", "High" and "Danger" levels chosen?

Proud Mary

Wed Jul 17 2013, 10:40AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

I think Steve's 'trending' concept is absolutely right for this project.

By removing the need for absolute dosimetric values - Gy/hr - we also remove the need for detector standardisation, so that a variety of GM tubes could be used in the same circuit.

I have seen ancient military dosimeters having meter scales marked with green ('safe') yellow ('safe' for short term exposure unless you are a coward) and red (perhaps Jesus does exist after all) but today the idea of there being any 'safe' level is questioned by many authors, because even natural background causes damage and very small numbers of people have their lives shortened because of it.

I think we should look at the way the idea of 'danger' is being used in this thread. As I write, I am being exposed to about 0.3 ÂµSv/hr background. If this climbs by an order of magnitude to 3 ÂµSv/hr, I am not in any sort of immediate danger, and few of the popular ex-millitary dosimeters will indicate anything at all despite this ten-fold increase. But while in health physics terms a rise in background to 3 ÂµSv/hr may only show up as excess mortality measured over years, the rise in background almost certainly indicates that something very dangerous requiring urgent investigation is taking place.

So if we drop the language of 'safety' and 'danger' - and use instead 'normal' and 'abnormal' - we shall enjoy less criticism and controversy.

radhoo

Wed Jul 17 2013, 10:43AM

Registered Member #1938 Joined: Sun Jan 25 2009, 12:44PM
Location: Romania
Posts: 699

Steve Conner wrote ...

However, it's not the change in radiation level that damages health, it's the absolute level. Even 1uSv/h is probably nothing to worry about. How were the "Low", "Normal", "High" and "Danger" levels chosen?

For the moment arbitrarily. <0.10uSv/h ; <0.2uSv/h; <0.3uSv/h and >0.3uSv/h

Sure, exposure to 1uSv/h is not a big concern, especially when the exposure is short. And pushing that to extreme, not to mention people that helped removing some of the dangerous debris at chernobyl , still alive today.

But we'll need some given thresholds in place, so the user knows that something has changed. Any suggestions are welcome!

BTW, I got the first two cases in post today, I'll post a few pictures later on; they look great, however I might not be able to fit everything inside.

Proud Mary wrote ...

I think Steve's 'trending' concept is absolutely right for this project.

Indeed, this is my feeling too. A constant value, regardless of its magnitude can be assuring that life goes on like it did yesterday (not necessarily perfect, but at least it's what we are used to), while a change , especially a significant one (relative term again :) ), should trigger at least some curiosity is not concern, on what was the cause of the event.

Proud Mary wrote ...

By removing the need for absolute dosimetric values - Gy/hr - we also remove the need for detector standardisation, so that a variety of GM tubes could be used in the same circuit.
..
So if we drop the language of 'safety' and 'danger' - and use instead 'normal' and 'abnormal' - we shall enjoy less criticism and controversy.

True, would be a lot safer, including from the criticism point of view which was highly concentrated on the conversion itself and less on the other technical details, but I will be unable to do that. People will still ask questions regarding the numbers, and as the majority of the dosimeters on the market have chosen to show the readings as the equivalent dose, I have to follow the trend, or risk a project failure.

radhoo

Wed Jul 17 2013, 10:51AM

Registered Member #1938 Joined: Sun Jan 25 2009, 12:44PM
Location: Romania
Posts: 699

Sorry PM, we wrote the answers simultaneously, I just added a few more comments in my previous post.

Proud Mary

Wed Jul 17 2013, 11:13AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

It would be interesting to monitor activity in the Bega Canal.

first previous 2 3 4 5 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob