Calculating Instrument Efficiency

[Wojo]

This question goes out to everybody that is involved in site closure activities.  MARSSIM section 6.5.4 states the following, “The instrument efficiency is determined during calibration by obtaining a static count with the detector over a calibration source that has a traceable activity or surface emission rate”.  Now how I read this is that when calculating total efficiency (εt) in accordance with ISO-7503, the value that you would use for (εi) is the calibrated efficiency from the instrument’s calibration certificate.  However, there is a large contingent of people who calculate instrument efficiency daily as part of the daily set-up of the instrument.  Subsequently, the (εt) value they use fluctuates daily.  This would be expected over a normal count distribution.  I am struggling with the technical wisdom of determining efficiency in this way.  Aside from not being strictly in compliance with the guidance, anybody else have any thoughts, pro or con?

[GLW]

Unless you are more conservative than the ISO,...

Stick to the ISO,...

You do not have to defend the ISO,...

Anything different you can be called on to defend,...

end result is dueling CHP's,...

[Laning]

I think your thoughts are correct. The daily performance tests otherwise known as source checks are not intended to take the place of the calibrated efficiency. They're conducted to confirm the calibrated efficiency, and shouldn't be used for the purpose of calculating activity. If the daily count falls within your acceptance range (+/- 20% or 2 sigma), then that tells you the instrument is working properly and the calibrated efficiency is good. There's more to calibrating an instrument than just doing a source check so using a daily source check efficency to quantify activity is not a very "efficient" thing to do.

[Wojo]

GLW

I acknowledge and agree with everything you said.  However, you didn't answer or speak to the question.  And of all the people out on this board, I am really interested in hearing what you have to say about this

[OldHP]

As Lanning said there is a big difference between 'periodic calibration' and 'daily performance checks', and this applies to operational facilities as well as site closure activities.  The calibration establishes the efficiency for a period of time, i.e., until the next 'periodic calibration' or until calibration is required because the instrument fails the daily checks.

Anyone who is trying to change the established efficiency on a daily basis does not understand the process!

That being said, there are certain instruments, that are 'calibrated' with each use, and thus an efficiency and 'conversion factor' are established for the particular routine!

[GLW]

GLW

I acknowledge and agree with everything you said.  However, you didn't answer or speak to the question.  And of all the people out on this board, I am really interested in hearing what you have to say about this

I'm vegetating at DCA, check your voice mail,.....

[peteshonkwiler]

Having operated in both listed venues, I can say without hesitation that I prefer to use an instrument efficiency calculated during calibration in accordance with ISO-7503.  The times when I had a governmental inspector on-site who queried me about instruments' efficiencies was most always on a site which did daily efficiency calculations and utilized those efficiencies on the surveys and reports.  I say most always as a weasel wording as I have been questioned about efficiencies of instruments during inspections, however that line of questioning drops when I state and show that the efficiency(s) in usage are from the calibration sheet(s).  When it's done as a daily thing, the line of questioning continues.

[fiveeleven]

Usually not being one to deviate too far away from simple approach mindset, I would point out that on any given day we in the HP/RP arena release to the enviroment hundreds or thousands of pieces of equipment, cranes,trucks, hammers, pens, hardhats, etc.,declared to be free of detectable radioactive contamination.  We use different protocols, various instrumentation including scintillators, semi-conductors and what not. But the single instrument that has far and away been used to declare more stuff free of detectable radioactive contamination - is typically calibrated yearly, has a daily efficiency verification check performed on it, and its efficiency is calculated by cpm/dpm = 0.10 or thereabouts.

[GLW]

Usually not being one to deviate too far away from simple approach mindset, I would point out that on any given day we in the HP/RP arena release to the enviroment hundreds or thousands of pieces of equipment, cranes,trucks, hammers, pens, hardhats, etc.,declared to be free of detectable radioactive contamination.  We use different protocols, various instrumentation including scintillators, semi-conductors and what not. But the single instrument that has far and away been used to declare more stuff free of detectable radioactive contamination - is typically calibrated yearly, has a daily efficiency verification check performed on it, and its efficiency is calculated by cpm/dpm = 0.10 or thereabouts.

That is so Somers 1975,...

[AARPSHP]

The answer to this problem of free release and detector efficiency is a easy one.  If one is concerned enough to do a daily calculated efficiency than do the following.  Use the calibration efficiency and a pass/fail of 20% on the daily checks, then follow it up with calculating a new detector efficiency if you notice a trend.  Setting up a spreadsheet to do this work for you is very easy.  Trend graphs can tell you many things and are a good troubleshooting tool.  When using a instrument like a Ludlum 2350 with many different detectors calculating a daily efficiency would mean it could take over a hour to do each instrument.  Sometimes people think to hard and make the job of the Technician in the field much to difficult.   People like this need to get away from their desks and look at what goes on in the field once in awhile.  The best Rad Engineers started out as field techs, need I say anything more.

[GLW]

The answer to this problem of free release and detector efficiency is a easy one.  If one is concerned enough to do a daily calculated efficiency than do the following.  Use the calibration efficiency and a pass/fail of 20% on the daily checks, then follow it up with calculating a new detector efficiency if you notice a trend.  Setting up a spreadsheet to do this work for you is very easy.  Trend graphs can tell you many things and are a good troubleshooting tool.  When using a instrument like a Ludlum 2350 with many different detectors calculating a daily efficiency would mean it could take over a hour to do each instrument.  Sometimes people think to hard and make the job of the Technician in the field much to difficult.   People like this need to get away from their desks and look at what goes on in the field once in awhile.  The best Rad Engineers started out as field techs, need I say anything more.

The OP couched the scenario in MARSSIM, MARSSIM is not synonymous with free release.

Calculating a new efficiency based on trend observation is wrong.

Troubleshooting, repair as necessary and re-calibration as necessary is right.

The meter-detector combination is either "As Left" within prescribed, accepted quality control parameters or it is not.

You can adjust for trends to your hearts content, my intervener is going to tag you with fudging numbers to get your data where you want it to get the license terminated.

Stick to the standards, empire building (hours and hours to get meters in the field) helps almost nobody in the long term.

(sic)

[OldHP]

The OP couched the scenario in MARSSIM, MARSSIM is not synonymous with free release.
Calculating a new efficiency based on trend observation is wrong. 

The 'daily difference' might mean a difference of 0.001 or 0.002 perecent in the calibrated efficiency!  Is the time do do a full efficiency calibration for the difference - NO!  The background changes, in a lot of cases, between readings, unless there is an obvious (increase/decrease) do you change it with each reading, and if so, how do you justify that it didn't change before you took the next reading?

[GLW]

The 'daily difference' might mean a difference of 0.001 or 0.002 perecent in the calibrated efficiency!  Is the time do do a full efficiency calibration for the difference - NO!  The background changes, in a lot of cases, between readings, unless there is an obvious (increase/decrease) do you change it with each reading, and if so, how do you justify that it didn't change before you took the next reading?

Good comments and illustrative of why I dislike attemting to condense standards and TBD's into breakroom threads.

[Wojo]

My opinion, efficiency is an empirically derived conversion factor to convert a reading in observed units to a measurement in reported units.  When you perform a daily or pre-use/post-use performance check of an instrument, the purpose is not to recalculate that conversion factor but rather, to ensure that the instrument is operating within expected parameters.  When you receive a newly calibrated instrument, you assess the instrument’s response to a check source through a series of measurements sufficient to provide a mean response with a robust standard deviation.  Instrument efficiency is irrelevant to this process.

[GLW]

My opinion, efficiency is an empirically derived conversion factor to convert a reading in observed units to a measurement in reported units.  When you perform a daily or pre-use/post-use performance check of an instrument, the purpose is not to recalculate that conversion factor but rather, to ensure that the instrument is operating within expected parameters.  When you receive a newly calibrated instrument, you assess the instrument’s response to a check source through a series of measurements sufficient to provide a mean response with a robust standard deviation.  Instrument efficiency is irrelevant to this process.

bingo,....with the caveat that the instrument efficiency may dictate what strength of source you may require