| T O P I C R E V I E W |
| shell |
Posted - 06/28/2023 : 03:27:30
Hello!
I have got Geiger counter (CAJOE clone) with J305 tube. Works fine.
I have a question about interpretation of CPM value. Tube has 25 pulses/minute at the background level and 12 pulses/minute of internal background noise. As I see, everyone just count the pulses and using tube conversion factor convert CPM into #956;Sv/h.
But when I count 25 pulses per minute (25CPM), should convert it into #956;Sv/h, or subtract noise? That means (25CPM - 12CPM) - counted pulses minus tube noise.
Which way is proper?
Regards!
|
| 2 L A T E S T R E P L I E S (Newest First) |
| ullix |
Posted - 06/29/2023 : 04:32:59
Well, it sure does matter.
Whatever the parameter you are trying to measure, you first try to determine any bias in the measurement, like reading a temperature being 0.35°C too high, or a weight being 7.5 g to heavy. You'd subtract that bias from the measured value, before you'd report that corrected value. That is standard practice.
You'd do the same when you report CPM from radioactive counts. But: the problem here is that you don't know the bias. The "internal background noise" is reported as "**up** to XYZ counts per minute". So this leaves a bias of something from 0 (zero) CPM up to 12 CPM. Which one to choose?
Choosing 0 CPM has the advantage that you get an error that is easy to convey and reproduce. Then a real measurement can hopefully be made on a source which gives so much higher CPM that the bias doesn't really matter.
You might be tempted to measure the bias by building a lead-castle (like this one: https://sourceforge.net/p/geigerlog/discussion/features/thread/e79e9bc71a/?page=2#2260) to define the internal background. But even this isn't guaranteed to work. Reason is that the lead itself is on the path of the decay chain of the heavy nuclides, and may contain other radioactive compounds, which may create more background than what you get without the lead-castle. Unless you purchase ultra-pure lead, certified for "low" radioactivity, which will cost you a fortune.
But what if your source isn't high enough in counts - like measuring the famous banana radiation (see e.g. my article "Going Banana" from: https://sourceforge.net/projects/geigerlog/files/Articles/) - then there is no option but to measure twice: once with the item (e.g. banana) and once without, with otherwise exactly the same setup. The assumption is that the bias is the same in both setups, so taking the difference will eliminate the bias. This is the only option I see for any low-count-rate measurement.
|
| EmfDev |
Posted - 06/28/2023 : 10:40:57
Does not matter as long as you calibrate it to the correct readings. |
|
|
