Re: Efficiency Determination

[Judd Sills <sillsj@GAT.COM>]

At 11:22 AM 11/26/96 -0600, you wrote:
>Dear Radsafers,
>
>I have recently obtained a Ludlum Model 3/44-9 and am in the process of
>performing efficiency calculations for various isotopes in use at my
>facility, namely P-32, S-35, I-125, and C-14.  In determining efficiency, I
>plan to take a measured amount of each isotope, place it on a disc
>smear, determine the count rate, and perform the necessary
>calculations.
>
>Yesterday the above method gave me a 30.6% efficiency for P-32.
>
>I would like to know if this is within the expected range for this
>instrument and if this is an efficient (so to speak) method for determining
>efficiency.
>
>I would also appreciate a list of expected efficiency ranges for the
>Model 3/44-9 for the isotopes I have listed if anyone is using similar
>instrumentation and has these numbers handy.
>
>In Anticipation, 
>Julia Sober
>Acting RSO
>Northern Illinois University
>DeKalb, IL 60115
>
>jsober@niu.edu
>Phone (815) 753-1093
>Fax (815) 753-0461
>
Ludlum reports the probe to be 65% efficient for P-32.  This is a 2 Pi
number, so if the number that you obtained was 4 Pi, which it should be,
then you are "in the ballpark."  Your method of calibration is not, however,
a very good approach to calibrating the probe/instrument, or to empirically
determine probe efficiencies for various isotopes.  You should be starting
with NIST traceable sources and using either a scatterless mount type source,
or a planchette type source with known/reported backscatter factors.  In order
to get a good efficiency, the source should be designed to prevent
self-absorption.
Electroplated sources are good for this type of situation.  FYI, the Ludlum
reported efficiency specifications (2 Pi) for this beta-gamma pancake probe are:
C-14    10%
Sr-90/Y-90      45%
Tc-99   38%
P-32    65%
Pu-239  30%

The 4 Pi geometries are one half of the above values.

Bear in mind that the instrument response for specific activities of an isotope
will vary depending upon the degree of backscatter from the surface that is
being
monitored, and that this is directly related to the atomic number of that
surface.
In other words, low Z materials will have low backscatter, and high Z
materials will
experience a lot of backscatter.  The degree of backscatter is also a
function of
the E-Max of the beta in question.  This "should" be taken into account when
making assay of surface actities by direct frisk, though generally, it is
conservative
to correct for scatter at the time of efficiency calibration, and assume no
scatter when "frisking" an object/surface.

In order to determine probe efficiencies, you should establish a standardized
calibration procedure for the detector(s) and have that reviewed and approved
by a technical person that is qualified/experienced with instrument calibration.

The above are my opinions, not my employers.....

Hope that this helps.

Best regards
J. M. Sills, CHP             |   Office: (619)455-2049
General Atomics, Room 01-166C|      Fax: (619)455-3181
3550 General Atomics Court   |   E-Mail:  sillsj@gat.com
San Diego, CA  92121         |