Claire,

Duke validates the effectiveness of our passive monitoring program by measuring 
the sensitivity of the passive monitors.  We think that our method for 
characterizing the sensitivity of passive monitors described below is solid and 
we don't think anything would be gained by performing random WBCs.  Considering 
how little internal exposure we get, we would have to count the entire 
workforce to approach meaningful statistics.  We certainly do perform WBC for a 
variety of reasons like facial contamination or a monitor alarm.

We use Co-57, Ba-133. Cs-137 and Co-60.  For each radionuclide we have a set of 
sources of 1.00, 2.00, 4.00, 8.00 etc. nCi (for Co-57 50.0, 100, 200 etc. nCi)  
Combinations of these sources can be used to give any desired total activity in 
steps of 1 nCi (50 nCi for Co-57).  For each radionuclide, we place a phantom 
in the monitor and try an activity.  If we get 10 out of 10 alarms, we consider 
that a success and try a lower activity.  If that is successful, we try a still 
lower activity.  If we get one or more failures to alarm, we move to an 
activity between the tested activity and the most recent success.  This quickly 
converges.  We consider the sensitivity of the monitor for the radionuclide 
being tested to be the lowest activity tested that gave 10 out of 10 alarms.  
We repeat for the other three radionuclides.  Note that the activities are in 
integer values for the reference date.  We then decay correct each sensitivity 
for the actual activity on the date of testing.  We convert these sensitivities 
to sensitivities normalized to an imaginary radionuclide of the actual energy 
but a single 100% abundant photon.  We plot these normalized sensitivities 
versus energy.  We then look at our station radionuclide distribution, 
including alpha and other difficult to detect.  We compute the number of nCi of 
that distribution that would be required for an alarm (10 out of 10) using the 
normalized sensitivity versus energy plot every line of every radionuclide with 
appropriate weightings applied.  (much easier than it sounds using a 
spreadsheet)  Finally, we calculate the internal dose corresponding to this 
total activity of our station mix, again considering alpha and other difficult 
to detect.  We consider the GEM-5s at our protected area exits to be our final 
monitors and so characterize every one of them.  We have found by experience 
that monitors that are calibrated to perform identically at Co-60 or Cs-137 can 
have very different sensitivities for lower energy radionuclides like I-131.  
Per ANI, we also characterize one each of all the other types of fixed monitors 
that could be of any use as passive monitors.

Bob Sorber
Corporate Radiation Protection  704-382-7259  
[email protected]<mailto:[email protected]>

From: Claire C Fage [mailto:[email protected]]
Sent: Thursday, December 29, 2016 8:06 AM
To: '[email protected]'
Subject: [powernet] Passive Monitoring Program



*** Exercise caution. This is an EXTERNAL email. DO NOT open attachments or 
click links from unknown senders or unexpected email. ***
Recently at Cook we have instituted the practice of exclusively passive 
monitoring every worker in and out processing. My question is, what does your 
plant do to validate the effectiveness of your passive monitoring program?  Do 
you perform random whole body counts? If so, with what frequency?  Does your 
frequency increase during outages?

Thanks for your time,

Claire Fage
Associate Health Physicist
Donald C. Cook Nuclear Power Plant
(269) 465-5901 x1117

Reply via email to