Answer to Question #12462 Submitted to "Ask the Experts"

What was the expected accuracy of a Landsverk-64 Roentgen Meter? I have some historical exposure rate measurements (without a stated uncertainty) made for a known activity ¹³⁷Cs source and would like to be able to estimate the uncertainty contribution that may be attributed to this instrument.

The Landsverk, Model L-64, Roentgen meter to which you refer was a high-quality instrument, an electrometer, intended for measurements of the quantity exposure. The ionization chambers available were of different sizes and could be fitted with caps to yield thicker walls necessary for measuring exposure from high-energy photons. In use, the selected chamber was charged on the electrometer and the quartz fiber contained in separate optics in the charging socket was set to zero on the scale visible through the lens system. After use the chamber was reinserted in the socket and the residual charge read as an upscale displacement of the quartz fiber.

I do not have the specifications available for the instrument but, under ideal conditions, when the appropriate chamber and equilibrium cap are used and if chamber leakage is not significant, I would expect to be able to obtain accuracies within perhaps +2% of the true exposure values. Such an accuracy might be attainable if the chamber/instrument had been calibrated by a primary standards laboratory (such as the National Institute for Standards and Technology [NIST] in the United States). If calibrations had been carried out at a secondary standards laboratory, the expected accuracy, under careful measurement conditions, might be on the order of +2% to +5%. Naturally these accuracies depend on the user applying all the proper corrections and techniques for field measurements—e.g., temperature and pressure corrections to correct air density to what was used at calibration, uniform photon field over the chamber volume, appropriate chamber wall (cap) thickness, restriction of field intensity to minimize ion recombination, environmental conditions acceptable to avoid excessive charge leakage, etc.

Since you appear to be attempting to evaluate measurements from the possibly distant past, you may have to make judgments about the quality of the measurements. While the instrument is capable of making high-accuracy measurements, if proper procedures were not followed significant errors could result and accuracies could be much worse than ideal. I cannot comment on the likely quality of the measurements. Of course, it would be a great advantage if any individuals who had participated in the use of the instrument for the said measurements were still available, and you could speak with them to attempt to gather appropriate information that might assist you in determining the reliability/accuracy of the measurements. I wish you well.

George Chabot, CHP, PhD