Badge Sensitivity Topic

[Orville Cypret 441-6411 <CYPRETOW@lmus.com>]

A comment was made yesterday about "arbitrarily lowering" doses.  I would be 
surprised if anyone is "arbitrarily" lowering doses.  However, evaluating the 
TLD processor's report of measured doses relative to a worker's expected dose is 
a legitimate act of dose evaluation.  A case in point is one I had to deal with 
a few years ago.  At my plant, we only have uranium hexafluoride, a few button 
sources for instrument checks and calibration, and some laboratory standards.  
One of our employees came up with a quarterly whole body dose of more than 9 
rem.  I've forgotten what the skin dose was.  Review of the glow curves left no 
doubt the TLD had actually been exposed to something, but we have no sources 
onsite that could have caused this kind of dose in 3 months.  I think the TL 
chip readings inferred exposure to hard photons and low energy particles.  I 
tried every way I could think of to find something wrong with the TLD 
processor's system.  No luck--they were doing everything right as far as I could 
tell.  Based on this, we concluded the TLD (in this specific instance) did not 
accurately report the person's occupational dose and I assigned dose to the 
person based on his coworkers' dose for the quarter.  In the end, I wrote a 
report describing the investigation and documenting my reasons for not assigning 
the TLD processor's reported dose.  This has subsequently been reviewed by DoE 
and the NRC and they concurred with my findings.

A few other relevant facts about this incident:  people took their TLDs home 
with them each day when this happened; the person's work location was not in a 
radiation area, a contamination area, or a radioactive materials area.  We also 
surveyed the employee's house, houseboat, vehicles, etc.  Didn't find anything 
radioactive.  

I've been in the business for 24 years and I'm convinced that anyone who takes a 
number from a black box (e.g., a TLD processor) unquestioningly (as some replies 
in this thread have implied) is asking for trouble.  We question the 
reasonableness of the results from any TLD processor we've used over the years. 
Sometimes a processor problem can result in a faulty reading, sometimes there 
will be something structurally wrong with the TLD hanger or filters, sometimes 
someone will slip in an unauthorized source on you and you'll get an unexpected 
result.  I think a good dosimetry program evaluates TLD results against 
expectations and, if there is a disconnect, initiates an investigation to 
determine why there is an discrepancy.  We do very, very few of these 
investigations and we manually assign a dose different from the processor's 
reported dose even less frequently, but, given sound reasons for doing a manual 
assignment, you shouldn't hesitate to assign the dose you believe is most 
representative of the person's dose.  But---document, document, document!!  And 
if you have to do more than one or two a year, something may be wrong with your 
program.

Obviously, in a complicated situation like this one, it's hard to include all 
the detail.  I have left some facts unstated, but I think I've included most of 
the relevant details.

Orville Cypret CHP, PE
Radiation Protection Manager
Paducah Gaseous Diffusion Plant
cypretow@lmus.com

****************************
DISCLAIMER:  This material has not been reviewed, endorsed, or approved by my 
employer.  Responsibility for the content of this message is solely mine.
************************************************************************
The RADSAFE Frequently Asked Questions list, archives and subscription
information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html