The two rods have points at the end for penetrating the soil. The 1⁄4” drive rod is slightly longer than the probe, so that the point extends beyond the probe when the cup is bearing against the collar. The probe has a bevelled end to facilitate soil penetration. Similarly the pilot rod has a point at one end. A self-grip wrench, a 5/8" open or adjustable wrench, hammer and teflon tape are necessary tools.

2. Probe placement:

Choose a spot well above the water table preferably with few stones in the soil. Hammer in the pilot rod most of the way to the depth required. Remove the pilot rod (with the self-grip wrench if necessary). Place the drive rod inside the probe and hammer them both down to the required depth. Remove the drive rod.

There are two alternative methods of connecting the probe to the RAD7. Either push in the plug-in hose connector into the probe or screw on the NPT hose connector. With the plug-in connector make sure the O-rings are clean and not damaged and that it is a tight, push fit into the probe. For the NPT connector, tightly wrap the NPT thread on the probe with two or three layers of teflon tape. Apply the tape clockwise looking from above - that is in the same direction as the adapter when it is screwed on. Screw on the hose adapter (or the water shut-off valve). Tighten it down by holding the collar with the self-grip wrench and the hose adapter with the ?” open or adjustable wrench. Make sure it is good and tight. If there is a leak at this connection, ambient air will be sucked into the tubing to dilute the soil gas, thus reducing the radon concentration in the air sample delivered to the RAD7.

Tamp down the soil around the probe. Leakage of fresh air into the sample acquisition path or down the outside of the probe to the sampling point can be a major source of error in the measurement.

3. Connection

Attach the supplied plastic tubing to the hose connector. The other end of the tubing is then connected to the laboratory drying unit. (Note that this connection should go to the upstream connection on the drying unit. That is the one closest to the screw cap.) If a DRYSTIK is used, it should be inserted between the probe and the drying unit.

4. Measurement

Before starting any measurement, make sure the RAD7 memory and run number are not full. (Data, Free [ENTER]) and read the number of free cycles - should be more than 100, (Data, Read [ENTER]) and see the run number of the last run - should be less than 90. If the space or available run numbers are too few, download the data from the RAD7 to a PC and erase the RAD7 memory (Data, Erase, Yes [ENTER]).

There are three modes of measurement. One is by grab sample, another by continous monitoring in standard protocol and the third in THORON mode, with the pump running continuously.

4.1 For GRAB protocol, it is necessary first to purge the RAD7 for ten minutes or more with dry, fresh air, before connecting the probe.

With the probe disconnected but the drying unit hooked up to the RAD7 start purging (Test, Purge, [ENTER]). After at least five minutes push ‘Menu’ (to stop the purge’. With the printer switched off, put the RAD7 in SNIFF protocol (Setup, Protocol, Sniff [ENTER]) and start a reading (Test, Start). Go to the third status window (Menu, Enter, Enter, Rt. Arrow, Rt. Arrow) and observe the relative humidity in the top righthand corner. Let the RAD7 continue sniffing until the relative humidity drops below 6%.

If the RAD7 was previously used to measure a high radon concentration, it would be prudent also to measure the count rate in window A while the Sniff reading was continuing. Go to the fifth status window (from the third, just Rt. Arrow, Rt. Arrow). The lefthand number is the count rate. Typically, 0.25 cpm would be equivalent to a radon concentration of 1 pCi/L, or about 40 Bq/m3. Soil gas is seldom less than 100 pCi/L, so if the count rate in window A has dropped to below 0.5 cpm and the humidity to below 6%, the RAD7 is ready to make the next GRAB protocol reading.

Set the protocol to “Grab” (Setup, Protocol, Grab [ENTER]) then turn off the RAD7. Connect the tubing to the probe. Switch on the printer, switch on the RAD7 and let the printer print out a header for this measurement. Check the header to make sure the setup is what is required. Go to Test Start and push [ENTER] to start the measurement. The RAD7 pump will run for five minutes. The instrument will wait another five minutes and then count for four five-minute cycles. At the end of the half-hour period, the RAD7 will print out a summary of the measurement, including an average radon concentration in the soil gas from the four 5-minute cycle measurements. This method gives a quick (half-hour) reading and uses the least amount of soil gas. The accuracy will depend on the radon concentration, and would typically be better than +/- 10%.

If necessary, the soil gas sample can be pumped out of the ground by a sampling pump, and fed to a Tedlar sample bag for later analysis by the RAD7. With that methodology, care should be taken to ensure that the sample is truly soil gas from the sampling point, and that there is at least 5 litres of soil gas sample in the bag. When analysis is made later, at the RAD7, care should be taken that it is first purged thoroughly, so that the relative humidity inside the instrument drops below 6%. Then the bag is connected to the drying unit and a grab sample protocol measurement started. The result should be corrected for the decay of radon in the sample during the period from acquisition to analysis.

4.2 Continuous monitoring is a simpler method of measuring the soil gas radon concentration and provides time resolution in the event that the weather is changing. The setup is as above, but the RAD7 preset protocol should be set to 1-day. It is still desirable to purge the RAD7 before starting the measurement, but the continuous monitoring process will itself serve to purge the instrument and dry it out after a while. In this protocol, the RAD7 will pump for the first five minutes of every half-hour cycle, and then for only one minute in every five, once the relative humidity drops below 10%. There will be a reading after every half-hour cycle, though the first reading will be low because the 218-Po decay rate in the detector takes more than 10 minutes to reach equilibrium with the radon concentration in the measurement chamber. So there will be a reading printed out (and stored in the RAD7 memory) every half hour. The accuracy of readings will be +/-5% for the typical high radon concentrations found in soil gas. This method will draw from the sampling point a volume of soil gas equal to the flow rate of the pump (L/min) times 10 minutes, in every half-hour cycle. In a typical, porous soil, the pump flow rate may be around 0.5 L/min. So the soil gas extraction rate will be around 10 litres per hour. After two hours of continuous monitoring, this would amount to 20 L, which would come from a sphere around the sampling point of around 6" (15 cm) radius. Less porous soils would have a slower flow rate, so the volume of soil gas sampled would be less.

4.3 Thoron protocol uses 5-minute cycles and prints out both the radon and thoron concentrations at the end of every cycle. However, thoron has a short half-life (one minute) so that the pump has to run continuously. The first two cycles should be ignored, while the radon reading equilibrates. Thereafter, there will be a reading every five minutes. The print format should be set to SHORT to save paper.

Because of the short thoron half life, some estimate of sample acquisition time is needed if the thoron readings are to be properly interpreted. During acquisition, a flow meter may be connected to the RAD7 outlet. This will show how fast the air is flowing from the sampling point. An estimate of the sample acquisition volume will then allow a calculation of the time delay between sampling and measuring. Call the RAD7 one litre, and the laboratory drying unit one litre. The soil gas probe will only have a volume of around 50 ml, and can be ignored in comparison with the 2 litres for the RAD7 and the drying unit. The total acquisition delay will be 2 litres divided by the flow rate. If this is 0.5 L/min, the delay will be 4 minutes. The thoron sensitivity stored in the RAD7 anticipates an acquisition delay of 1.4 minutes, so the extra delay in this application will be 2.6 minutes. During that time, the thoron will decay to 0.165 of its original concentration, so the thoron reading should be multiplied by 6 to give the thoron concentration at the sampling point.

The final value of the thoron concentration has large uncertainties and the absolute accuracy is probably no better than +100/-50 %. For a soil gas radon concentration of 200 pCi/L (8,000 Bq/m3) the 5-min radon readings however, after the second 5-minute cycle, will be accurate to +/-10 % (95% confidence interval). Higher radon concentrations in the soil will improve the accuracy of the short term readings. Averaging several readings will also give a more precise measurement. One disadvantage of the thoron and sniff protocols is that the pump runs continuously, so that much soil gas is drawn from the sampling point. Compared with 1-day protocol, sniff protocol will draw three times as much gas. This means that the sphere of gas taken from the sampling point will be 1.4 times the radius for 1-day protocol in the same total time.

5. Continuous Monitoring

It has been proposed that continuous monitoring of soil gas for radon may be a useful technique for predicting earthquakes. For this purpose, an array of soil gas probes covering half a square kilometer, say, may be more informative than a single sampling point. Such an array can be monitored with a single RAD7. Fine bore plastic tubes may be run for hundreds of metres from each probe to a centrally located valve manifold system. The RAD7 may be switched from one tube to another under the control of a PC. Each tube is purged prior to the RAD7 hook-up, so that the soil gas sampled is fresh. With GRAB protocol, one measurement can be made every hour. If each sampling point had to be read six times a day, say, there could be four sampling points monitored by one RAD7.
With a wireless modem connected to the RAD7, the controlling PC can be located anywhere. The system can be controlled by the remote PC. Real-time data is collected by the PC while a back-up set of data is stored in the RAD7.