Re: Personnel Monitoring Data Needed

[Bob Flood <bflood@SLAC.Stanford.EDU>]

While at TVA, I participated in the development of the multibadging system,
which developed much like that described by Sandy - a feeding frenzy in
response to an NRC I&E Notice. In the tradition of Federal agencies,
anything worth doing is worth overdoing. Multibadging at Browns Ferry in
the early to mid 80's got so intense that offsite processing could no
longer keep up with the demand, and an onsite TLD lab had to be set up to
handle all the badges (maybe that was the program goal to begin with).

Like Sandy said, analysis later found the overwhelming majority of the
multibadging to be completely unnecessary.

Eventually a reasonable approach was developed:

Under normal conditions, a single dosimeter worn on the front of the upper
torso would be adequate to monitor deep dose (whole body dose), dose to the
skin of the whole body, to the skin of the extremities, and to the lens of
the eye. This assumes uniform irradiation of the body including extremities.

When the radiation fields are not uniform, the single dosimeter may be
moved to the location on the body that will receive the highest dose, and
this measurement, too, would be an acceptable measure of all of the dose
quantities  mentioned above. This is best illustrated by the case of
working under headers where the source of exposure is overhead - moving the
dosimeter to the worker's head (hardhat) avoids a non conservative
measurement.

However, in cases where the location of the highest dose cannot be
predicted, a set of badges will be used. Multibadging criteria were set,
and if the criteria were met, up to eight dosimeters could be worn. The
body locations monitored were the head, chest (normal badge location), left
and right upper arms, left and right thigh (femur), gonads, and back. If
the conditions permitted, some badges could be omitted (e.g., reaching into
a steam generator, where the lower half of the body would receive
relatively little dose compared to the upper body). Such decision could be
made only based on previous multibadging measurements for the same job or
location in the facility.

The criteria were (at the time I left TVA):

Whole body dose rate of at least 100 mrem/h (don't bother multibadging
outside High Radiation Areas)

An expected whole body dose of at least 300 mrem

The gradient between the reference dosimeter location (the chest) and any
part of the body receiving a higher dose is at least 50 percent.
-----------------------------------------------------------------------------

Now it's my turn on the soapbox. I got to review and comment on the
multibadging standard while was being drafted and didn't like the 30
percent gradient criterion then, and I don't now. I find it unnecessarily
restriction. The standard combines this with a threshold dose of 10 percent
of the person's limit, without qualifier. In the NRC world, that means
those people who do not require monitoring have a de facto limit of 500
mrem/y, and this standard would require multiple dosimetry if the person
may encounter fields where a 150 mrem/y gradient exists. People classified
at DOE facilities as not meeting Radiation Worker criteria are limited to
100 mrem/y, making the threshold gradient 30 mrem/y! Surveys can't even
measure such a gradient.

Better the threshold should be expressed as a percentage of the REGULATORY
limit. But, too late.

Also, have you seen Dan Reece's calculations of the angular dependence of
EDE for photons (published in HPS Journal)? He also calculated deep dose
for each geometry, and then calculated EDE using the ANSI standard way and
a few techniques of his own. What he found is far simpler than the ANSI
standard: if a person will be working in non-uniform fields, supplementing
the chest badge with a second one on the worker's back was sufficient, with
EDE calculated from a specific formula using the dose measured by the front
and back badges. The correlation with EDE over the range of geometries was
superior to the standard. So how will regulators (and maybe lawyers) react
to using Dr. Reece's more reasonable approach?


Bob Flood
Stanford Linear Accelerator Center
(415) 926-3793     bflood@slac.stanford.edu
Unless otherwise noted, all opinions are mine alone.