Radiacode 102 ¹³⁷Cs Dose Response
I've long wanted to get my hands on a Radiacode so I could check it against some known sources to see how it performs. Last night my fianceé gave me a Radiacode 102 as a gift. I'll do more tests as time goes on but today I ran by the lab for a few minutes and grabbed some quick dose rates vs a NIST-traceable ¹³⁷Cs calibration source.
It seems to be the prevailing opinion of radiation professionals on Reddit that the Radiacode line is a wonderful addition to hobbyist equipment and, for gamma spectroscopy, the value is unbeatable -- but that it is not a professional instrument, nor should it be used as one. This gets pushback from some hobbyists and Radiacode fans but perhaps this sort of test will help explain why we say it shouldn't be used as a professional instrument. All I've tested here is its reported dose rate against ¹³⁷Cs over the ranges a typical professional might encounter, excluding quite low rates one might encounter for environmental testing and higher rates above the range the Radiacode allows by default. In other words, I tested it from around 1.5 to 840 μSv/hr.
That the instrument doesn't measure above 1 mSv/hr (by default, perhaps I can change this in the settings once I explore the app more) is not reason to say it shouldn't be used in a professional setting. I've used plenty of professional instruments with much lower upper-ranges. It's hard to think of anything more important than knowing the capabilities of your instrument, so you don't bring those instruments to do jobs with higher dose rates. That said, 840 μSv/hr was right at the limit of the instrument, with its reading clipping the 1 mSv/hr limit more often than giving a reading within its range. As such, what I write below was done so ignoring the 840 μSv/hr table row.
Actual (μSv): True dose rate
Read (μSv): Radiacode 102 reading
Error: Percent confidence given by the instrument, rounded to nearest half-percent.
Diff: Difference between Actual and Read.
| Actual | Read | Error | Diff |
|---|---|---|---|
| 840 | 998^^* | ±3% | +18.8% |
| 681 | 965 | ±3% | +41.7% |
| 589 | 697 | ±3.5% | +18.4% |
| 473 | 640 | ±3.5% | +35.4% |
| 210 | 270 | ±5.5% | +28.6% |
| 189 | 222 | ±6% | +17.5% |
| 84.0 | 104 | ±8.5% | +23.8% |
| 68.1 | 98 | ±8.5% | +43.9% |
| 47.3 | 65 | ±10% | +37.6% |
| 21.0 | 29 | ±16% | +38.1% |
| 18.9 | 26 | ±16% | +37.6% |
| 14.7 | 19.0 | ±17% | +29.1% |
| 8.40 | 12.3 | ±17.5% | +46.4% |
| 6.81 | 11.0 | ±16% | +61.5% |
| 4.73 | 8.5 | ±15.5% | +79.9% |
| 2.10 | 3.8 | ±16% | +81.0% |
| 1.47 | 2.8 | ±15.5% | +86.8% |
Depending on the context, the most-common professional accuracy requirements I've seen outside of medical & research uses are either ±10% or ±20%, with the occasional ±15% tossed in as exceptions for instruments which specifically state they're made for a ±15% accuracy. The accuracy requirements for medial & research are far more strict, but I don't think I've seen any sub members suggesting the Radiacode is a medical or research-grade instrument.
Accuracy and precision are two different things. You need an instrument which can be measurably both. At every dose rate, the Radiacode over-responded. That's not an indication of a non-professional instrument. Nearly every non-energy-compensated Ludlum instrument you'll ever come across will be calibrated against ¹³⁷Cs and will over-respond at lower energies. Again, you should know your instrument. I don't know what isotope the Radiacode is calibrated against for dose rate,* (I believe it's calibrate against ⁴⁴Ti for energy discrimination) but I don't really need to know since I have a calibration source.
Accuracy. Looking at the table, I see that I can address accuracy pretty well with a correction factor. I like correction factors divisible by 5, and I find that if I reduce the readings by 30% across the board (again, ignoring the first row that clipped max) that I can get 9 and 13 readings within 10% and 20%, respectively. It should be able get all readings within 20% across its entire range, but 13/16 isn't too bad. For the 20% allowance, the 3 lowest readings were the ones falling outside the acceptable range. If I had a professional instrument come back for calibration and it had this issue I wouldn't toss the instrument or take it out of service unless the owner needed accuracy in that range. Instead, I'd mark on the calibration certificate and clearly on the instrument that it should not be used for dose rates below some nearby round number. In this case, below 5 μSv/hr. A new professional instrument, though? This particular Radiacode would not make the cut as a dose-measuring instrument. A new professional instrument should be accurate within 20% of true across its entire advertised range of measurement, 10% is what I (and many regulations) expect. Sure, someone could sell an instrument for measuring dose rates, say it's a professional instrument, and say that its tolerances at some dose rates in its range are outside of 20%... but I'm not buying that instrument.
Precision. This one really confounds me. I definitely could have spent more time taking these measurements, but with each of the unexpected results I did go back and take measurements again to be sure I wasn't in error. Take a look in the table from one dose rate to the next and see how its difference from true changes. From any given row to the next, will it go up or down? A little or a huge amount? There's no easy correction factor to tame this, and one correction factor (for accuracy) is as much as you want in a professional dose rate meter. It's precision more than accuracy that keeps this Radiacode from being a profession-grade dose rate meter, but either alone are enough to fail muster.
Don't get me wrong, I love the Radiacode. For the reasons I wrote above, it's an absolutely wonderful instrument. It has unmatched bang for your buck and brings gamma spec into the price range of the hobbyist -- or, at least, the hobbyist who was previously unwilling to use an attenuator set for do gamma spec by hand. Professional instrument manufacturers could learn a lot from Radiacode in terms of interface, specifically by recognizing that nearly every user carries a computer in their pocket everywhere they go nowadays. The form factor is great but, also, I'm certain a big source of its limitations. Smaller is easier to carry and use, and I love that, but the physics is against the instrument in that aspect. Regardless, I'll be carrying the instrument with me to many places and getting tons of value from it. It just won't be recording any official measurement.