Are these radon results fubar?

Taken with a sun nuclear model 1027.


The data is read from left to right and each data point represents one hour of testing.


These were taken 6 minutes apart and the second tape was taken just before the last hour data point was recorded. Note the very different averages. I think the machine started to average the results in the electronics just before the last hour was integrated and recorded.


The graph immediately got my attention. I have not seen this shape of graph with such distinct spikes. When the homeowner was asked about laundry, the time frames that washing was done coincides with these spikes.


These results were printed within minutes of each other after hour #48. Note the very different averages and also note the last hour with a radon measurement of 500 pCi/L.


this shows the data points before and after hour #48. Note the different values recorded for the hours shown.





Has anyone else ever seen something like this?

Thanks, Bill

William -

Can you post associated graphs for tapes #3 and #4??

That appears to be an invalid test. Minimum test time is 48 hours. I only see 47 hours recorded. In regards to the spikes. I use SUN monitors and have never had such discrepancies. Has the machine been calibrated properly? Have cross checks and duplicate tests been done?

I would not use those test results.

get your monitors checked

Thanks David,

Tape 3 shows the results before and after hour #48. The first tape was printed just a few minutes before the last hour was of data was recorded. This test was performed using the same machine. It was calibrated last July and had a cross check performed last month. It was spot on.

I don’t look at the spikes as a discrepancy. (except for perhaps hour #48 with a level of 500 pCi/L. I think it is related to doing laundry. The laundry room was adjacent to the basement club room where the detector was located. The gas water heater and furnace are also in the laundry room.

The home has an active mitigation system, installed less than 2 years ago. It was pulling 3/4" of water. The post mitigation test was less than 2.0 pCi/L but it was performed prior to the owners occupying the home.

I did not certify these results. I commenced a retest with a different CRM. I am picking that monitor up in a few hours and will update you with those results.


I think I’ll be sending it to the manufacturer for repairs and recalibration.

Thanks Frank,


Here is the graph for the tape3.jpg Scott.

For tape4.jpg, this is the graph for the right side. tape2.jpg, posted above, is the graph for the left side.


The retest performed with a different detector shows the expected results which are very much different than the previous 48 hours.

(white out for client confidentiality)

I contacted my radon guru and his assessment is as follows:

“If you can rule out customer interference, and instrument noise, then you may have an extremely tender house. By that I mean a house were small changes in differential pressure greatly affects radon transport.”

“If the same machine for both tapes, and if the weather could not be the problem for the times indicated, you may have a bad machine.”

Frank was looking for a used 1027…:twisted:

“A reading of 500 pCi/L would require water radon levels of probably more than 5 million pCi/L of water---- I don’t think so.”

“Also, I don’t think the radon water concentration would vary that much. As you know, the weather was very cold with high winds.”

“The last reading should not have changed the other hourly readings. This is not true for the averages. You can now see why I always expose a charcoal canister along with EVERY machine test.”

“I would repeat the test with duplicated detectors.”

It is easy for him to use charcoal canisters. He manufacturers and analyzes them. :p:roll::stuck_out_tongue:

The last reading said 500 not 500.00 it looks fishy to me. I have never ran into this either. Also notice it wasn’t averaged in the test the first and last results are tossed if I am not mistaken to get the EPA average. Although it should have been addded into the first average. Maybe the unit was unplugged from the computer to early or there was a glitch.

I would use a different monitor and chect it again or better yet use two 1027’s to retest.

Hahaha don’t get any ideas William. I am interested to see what you come up with with this.

I’ll have to find the exact language but PA - DEP gives 2-3 hours leeway between placement/pickup for ease of scheduling. I typically leave my monitors an extra 10-12 hrs longer if I can not pick up close to the drop off time (rural area) … but some CAN pick up sooner.

Hi Chris,

For a 48 hour test, you have 48 + 2 hours to pick it up. After that for each additional 24 hours, you have plus or minus 2 hours.

Picking up a detector 58 to 60 hours later is not protocol. Do you certify a 60 hour test?


Dang it…:twisted:

I am conducting a co-located test right now with the suspect CRM and another CRM. So far they are in good agreement.

The unit drops the decimal once radon levels go triple digit. The results were puzzling indeed though Rick.

I will also from time to time leave a monitor for an extra 12 if I am not sure closed conditions are satisfied when I get there. These monitors only take the last 48 measurements under EPA avg. A short term test is defined as 2-90 days I believe. Don’t quote me on that.

If the windows are open, I’ll close the house and leave the CRM for 72 hrs + or - 2 hours. I will mathematically remove the first 24 hours of data and average the remaining.

By protocol to minimize the effects of the diurnal changes in radon level, the detector is supposed to be exposed for durations in increments of 24 hours plus or minus 2 hours from 48 hours on.

How do you add 12 hours to your tests? Are you retrieving them at 8:00pm to 8:00am? (assuming you might work from 8:00am to 8:00pm in the field). How do you account for the extra data? Do you remove data? Does your QAP address this situation?

The CRMs will record 90 intervals. The average and current values continue to update. On a loss of power, the partial hour value is removed from the average and the current value is lost.

The machine will issue an average based on all hourly results and the EPA average which is with the first four hours removed. The detector must be exposed for 48 hours in order to have the minimum 44 contiguous hours to produce a valid test.

FYI - The co-located test with the machine produced results with a RPD of 1.4%.

What were the readings for the co-located test?

The machines were in an undisclosed location deep inside a Hershey chocolate mine.:stuck_out_tongue:

The readings were 29.0 pCi/L versus 28.6 pCi/L for the suspect CRM

RPD = |(29.0 - 28.6)| ÷ (29.0 + 28.6) ÷ 2] x 100% = 1.4%

No anomalous data was observed. A difference of 0.4 pCi/L seems fairly insignificant, especially given the accuracy of machine is 25% or 1.0 pCi/L, whichever is higher after at least 24 hours of exposure.

Hi Bill… That’s not what PA DEP states, you have 48 -2 hours to pick it up. you must document why a 46 hour test was performed but you can do it. Anything over 48 hours is fine it does not have to be in 12 hr increments. Wayne Gemmill of Gemmill Associates, Kelly O. at the PA DEP office and a local mitigation specialist confirmed this… again documentation is key.