| T O P I C R E V I E W |
| Stargazer 40 |
Posted - 07/30/2019 : 11:23:14
When I schedule a Timed Count I set a minute or two thinking that for a ten minute count that the CPM would have time to build to the approximate average CPM and when the count starts it would begin with a number that is close to what it would calculate as the average. Instead the CPM starts counting at a very low number even though the led has been flashing at a high rate during the entire one to two minutes I'm waiting for the timed period to start. Hence the final count has a period that is lower due to the included initial run up.
Also I see the CPM climb to well above the average CPM I expect and then drop well below it and climb again over and over during the timed period. Just wonder how this work in the averaging world. At the end the total counts are ten times the average CPM for the ten minute timed period. What are you displaying for average CPM during the period?
Wondering if we could improve this by not starting timed period till counts stabilize somehow as an option. |
| 7 L A T E S T R E P L I E S (Newest First) |
| EmfDev |
Posted - 07/31/2019 : 15:09:07
Yeah using the Timed Count is better. |
| Stargazer 40 |
Posted - 07/31/2019 : 12:13:05
No, I just looked for counts to stabilize. It's one of the reasons I'm redoing a lot of my numbers to get the average over 10 minutes. I think it will improve the calibration numbers. |
| EmfDev |
Posted - 07/31/2019 : 08:36:28
Hmm did you measure the 10min counts from the old samples when you got then last September? |
| Stargazer 40 |
Posted - 07/30/2019 : 16:19:23
Certainly the difference is more than I would like it to be for both activity levels. It's different enough that I need to specify which of the sources I am using in any analysis or tube or meter comparison. I don't think V2 would give any meaningful difference in counts. I am in discussion with the maker of these to get their opinion if this is too much (I think they will say within tolerance for a non-calibrated source). |
| EmfDev |
Posted - 07/30/2019 : 15:58:48
Looks like the difference is too high. I wonder what would the V2 would read. |
| Stargazer 40 |
Posted - 07/30/2019 : 13:58:42
So total counts from zero divided by minutes is good for me. The ACPM was a little disconcerting.
Today I received another 10 uCi source and another 5 uCi source. The first two I got back in September of 2018. I did Timed Counts at 10 min. For the raw disks into my LND 712 wand grafted into the 500+ V1 I got the following table. To the right are the gamma only CPMs with 3mm of lead plate and 6mm of plexiglass in between.
Image Insert:
26984 bytes
There is a large +/- associated with these uncalibrated sources so the expected difference due to the ten months difference in manufacturing dates of -2.6% is lost in the noise. Still the old ones are up to 9% lower than the new ones. These are Cs137 so HL is 30.8 years. |
| EmfDev |
Posted - 07/30/2019 : 13:11:20
I think the CPM/Average CPM starts at 0 when time count is started because all counts before timed count are discarded. The final count is what was detected since the start-finish. I think it's correct. However, the ACPM calculation is not accurate. I will update it on the next firmware update. Should be the same as the normal ACPM.
'Total counts are ten times the average CPM for the ten minute timed period' that is correct.
lets say cpm = 100 throughout the 10min period. Then total = 1000 counts and ACPM = 100.
the calculation should be:
ACPM = Total Counts / (minute + (second/60)).
This is the calculation used for the normal CPM however it's not used in timed count.
That is maybe why there is a drop and climb every 1 minute or 30 seconds in the data. |
