TM 5-6635-386-12&P
If the tester consistently exhibits a tendency towards a
low deviation, that is more towards the 0.75 end for the
calculator or more towards a 2 out or 1 out for the
manual method, then the tester is picking up some
periodic count rather than purely random count, tester
repair may be required.
This smooth tendency is probably due to a failing filter
network in some supply system which would allow a
steady noise to creep into the system, a noise not
subject to the variation of a random source.
Using either of these techniques, either calculator or
manual evaluation, the operator can maintain a close
watch on his tester and can be confident that his field
work is accurate without electronic error or decay drift.
It is not adequate to merely run the automatic standard
count and not run the periodic evaluation.
The daily automatic standard count will serve to verify
general good operation, but it will also mask minor
variations between counts that are indicators of potential
tester failure.
The standard count in the automatic mode can cover up
erratic tendencies such as a noisy detector, or other
problems. The tester may appear very normal under
such conditions.
We must maintain a professional attitude about our
equipment and its operation at all times.
This means frequent checking of standard count and
tester performance.
1-13. SITE PREPARATION
Tests will only be as good as the quality of site
preparation. The site must be flat, free of voids, with
surface irregularities filled with native fines or sand.
If transmission is used, the hole must be neat and
square to the surface, and protected from puffing or
movement during the drilling of the transmission hole or
retraction of the drill pin.
It is desirable to rotate the tester 900 or more on rocky
soils to obtain the largest sample size practicable. This
avoids the possibility of a single rock skewing the results.
Site Selection
CPN advocates statistical testing wherein the average of
several test sites is used to accept or disapprove the
project. The nuclear tester is much faster than the older,
conventional test methods and statistical programs may
be employed.
Using the CPN Random Selection Cards, (See Appendix
J) select five site locations and take one test on each
location. Average the sites for a statistically valid
evaluation of the project.
Each site must be free from major protuberances
reasonably smooth, and should be representative of the
overall project.
Preparation
Using the CPN Cast Aluminum Guideplate, smooth the
surface, removing large stones or surface debris which
would prevent the tester from sitting flat. Any technique
of smoothing, pounding, or other flattening will be
adequate. It is impossible for the operator to pound the
site so hard as to compact the site. He can only flatten
down the top 1/16" or so to provide a good surface.
Using a sieve, shake some native fines or sands over the
site to fill in any remaining air voids to remove possible
errors from surface roughness.
The tester must rest upon native promontories, however,
with only the voids filled in. The tester must not rest
upon a "cushion" of filler material.
If the test site is asphaltic concrete, (AC), a coffee can
"salt shaker" with fine sand may be used to fill in the
voids.
The tester has a reduced bottom surface to provide the
best possible seating under adverse conditions. This
facilitates seating the tester on AC where a rubber tired
roller has been used.
1-14. SEATING QUALITY EXPERIMENT
An experiment can be run to test the required effort
1-12
