DURRIDGE RAD AQUA User manual

RAD AQUA
Continuous Radon in Water Accessory for the RAD7!
User Manual!
!
!
!
!

TABLE OF CONTENTS!
!
INTRODUCTION 5
1 RAD AQUA SETUP 6
1.1 Exchanger Assembly 6
1.1.1 Nozzles 6
1.1.2 Temperature Probe 6
1.1.3 Air return 6
Fig. 1 RAD AQUA with Temperature Probe 6
1.1.4 Tie rod 6
1.2 Connections 7
1.2.1 Air loop 7
Fig. 2 RAD7 AQUA Standard Setup 7
1.2.2 DRYSTIK 7
1.2.3 RAD7 and Exchanger Location 7
1.2.4 Water Supply 8
1.2.5 Temperature Probe 8
1.3 Water Flow 8
1.3.1 Water Source 8
1.3.2. Water Level 8
1.4 Air Flow 8
1.4.1 Continuous Pumping 8
1.4.2 Pump on “Auto” 8
1.4.3 Mode set to “Auto” 8
1.5. Protocol 9
1.5.1 RAD7 Protocol 9
1.5.2 RAD AQUA Protocol 9
1.5.3 User Protocol 9
2 MEASUREMENT PROCEDURE 10
2.1 Start Up 10
2.1.1 Temperature Probe 10
2.1.2 Start Measurement 10
2.2 Speed of Response 10
2.2.1 Measurement in Progress 10
2.2.2 Inuencing Factors 10
2.2.3 Water ow rate 10
2.2.4 Air ow rate 10
2.2.5 RAD7 Mode 11
2.3 Long Term Measurement 11
Table of Contents
2

2.3.1 Desiccant 11
2.3.2 Memory 11
3 DATA 12
3.1 Data Handling 12
3.1.1 Printer 12
3.1.2 Memory 12
3.1.3 RADLINK 12
3.1.4 CAPTURE for Windows or Mac 12
3.1.5 Temperature Data 12
3.1.6 Time Relationship 12
3.2 Data Conversion 12
3.2.1 FRITZ WEIGEL 12
3.2.2 CALCULATION 13
4 THORON IN WATER 14
4.1 Why Thoron? 14
4.2 Measurement in Water 14
4.3 Thoron Sensitivity 14
4.3.1 Source to Exchanger 14
4.3.2 Exchanger to RAD7 method 1 14
4.3.3 Exchanger to RAD7 method 2 14
4.4 Speed of Response 15
5 DRYSTIK 16
5.1 Passive DRYSTIK 16
5.2 Active DRYSTIK 16
5.3 Effect on Response Time 16
5.4 Custom designed Active DRYSTIK 16
6 BOATING 18
6.1 Response time 18
6.1.1 Minimizing T1 with increased water ow rate 18
6.1.2 Minimizing T1 with reduced air volume 18
6.1.3 Minimizing T2 with a higher sensitivity RAD7 18
6.1.4 Minimizing T2 with multiple RAD7s and one RAD AQUA 19
6.1.5 Minimizing T2 with multiple RAD7s and multiple RAD AQUAs 19
6.2 Pump position 19
6.2.1 Positioning Equipment 19
7 CARE, MAINTENANCE, AND TROUBLESHOOTING 20
7.1 Water Catastrophe 20
Table of Contents
3

7.2 RAD7 Care 20
7.3 Exchanger Care 20
7.4 Desiccant Regeneration 20
7.5 RAD AQUA Troubleshooting 20
7.5.1 Rising Water Level 20
7.5.2 Spray Chamber Fills With Water 21
7.5.3 Air Path Integrity 21
7.5.4 Poor Spray Formation 21
7.5.5 Water Overowing From Base 21
REFERENCES 22
Table of Contents
4

INTRODUCTION
!
e RAD AQUA is an accessory for the DURRIDGE RAD7 Electronic Radon Detector. It is a device to bring the
radon concentration in a closed air loop into equilibrium with the radon concentration in a ow-through water
supply. It consists of a spray chamber, called an “exchanger”, that brings the air and water into equilibrium. e
radon in the air is monitored continuously by the RAD7.
e partition coefficient, the ratio of radon concentration in the water to that in the air at equilibrium, is
determined by the temperature at the air/water interface. is temperature is measured with a temperature
probe inserted into the exchanger. At typical room temperature the coefficient is about 0.25. at means there is
four times higher concentration of radon in the air than in the water, so there is, in effect, a gain of four times in
the sensitivity of the system to radon in water, compared to radon in air.
It takes time for the water to deliver radon to the air loop and for the RAD7 to respond to the changed radon
concentration. With optimum conguration the response time of the system may be reduced to less than half an
hour.
Introduction
5
CAUTION
Tap water and typical ocean water have sufficient dissolved gases to maintain the water
level in the exchanger at an acceptable level. However, should the water level in the
exchanger start to rise to an unacceptable height an air bleed may be added as
described in Section 7. This will prevent water from being drawn into the desiccant and
hence into the RAD7.
!

1 RAD AQUA SETUP
1.1 Exchanger Assembly
e RAD AQUA exchanger is supplied semi-
assembled. First the chosen nozzles and other
components should be installed in the head assembly
then the base, trivet, cylinder, rod and head should be
assembled together and held in place with the brass
thumb screw.!
1.1.1 Nozzles
e RAD AQUA is supplied with three pairs of
alternative nozzles. ese are the WL4, WL1 and
WL0.25. We install one and include the other two in
the accessories. At 20psi (138 Pascal) water pressure
the published ow rates for each are:
WL4: !10.98 L/min.
WL1: !4.1L/min
WL0.25: 0.68L/min.
!
WL0.25 is intended for slow continuous monitoring
where rapid changes are not expected and where
conservation of water is a consideration. WL4 is
intended for those applications where speed of
response is a major goal. WL1 is a compromise
between the two.
!
1.1.2 Temperature Probe
e probe is inserted through the stem adapter. A
little petroleum jelly may help it to slide into position.!
1.1.3 Air return
e air return is sent, via the check valve, to the
internal tubing that terminates halfway down the
cylinder. e actual length of that internal tubing is
not critical. If it is very short air would be able to
short-circuit the exchanger without passing through
the spray. If it were too long it may terminate
beneath the internal water surface and may lose air
through the water outlet. It should be about halfway
down the cylinder."
!
Fig. 1 RAD AQUA with Temperature Probe
!
1.1.4 Tie rod
Insert one end of the tie rod into the thread in the
trivet. Place the cylinder in the trivet slots. Push the
head assembly onto the rod. Attach and tighten the
thumb screw to draw and hold the assembly together."
Section 1 RAD AQUA Setup
6

1.2 Connections
!
1.2.1 Air loop
Two pieces of tubing connect the RAD7 and drying
unit to the RAD AQUA air/water exchanger, as
shown in Fig. 2. ese two pieces of tubing can be
several tens of meters long. e standard tubing
supplied with the RAD7/RAD AQUA is sufficient for
a connection up to ve feet between the exchanger
and the RAD7.
Connect the OUTLET of the RAD7 to the check-
valve connected to the head assembly. For this, the
5long, 3/16" ID tubing, with a 1/8" ID section at
one end, may be used. e 1/8" end ts on the RAD7
outlet, and the 3/16" end ts the check valve.
Connect the other 3/16” hose connector on the head
assembly to the laboratory drying unit, at the
SCREW CAP end, with the tubing and sleeve
provided.
Connect the other end of the laboratory drying unit
(there should be at least one inch of blue desiccant
leat this end) to the air inlet lter (with 1/8" ID
tubing at the lter end), which is then placed on the
RAD7 INLET. e Luer taper ensures an airtight
connection."
Fig. 2 RAD7 AQUA Standard Setup
1.2.2 DRYSTIK
Please note that the above instructions are for use of
the RAD AQUA without a DRYSTIK humidity
exchanger. A DRYSTIK, if available, should be placed
between the exchanger and the drying unit, and the
outer sheath should be purged with dry air from the
RAD7 outlet. See Fig. 3 in Section 5. More details are
provided in the DRYSTIK user’s manual.
1.2.3 RAD7 and Exchanger Location
Place the RAD7 on a clean, dry surface, preferably
inside a laboratory. If it has to be located in a harsh
environment, then it should be protected from the
elements (especially water). A simple way to do this
is to place the RAD7 inside a large transparent plastic
bag, such as the one in which it was originally
shipped. e bag opening should be gathered around
the inlet and outlet tubes, so that the instrument is
Section 1 RAD AQUA Setup
7

inside a closed space, completely protected from the
elements, while still allowing observation of the LCD
and print-out, and operation of the key pad.
e exchanger unit may be placed upright in a sink,
or higher than a boat’s gunwales. Water will ow
from the hose outlet, in the base of the exchanger. A
garden hose may be connected to the exchanger base,
to take the outow, provided it runs downhill from
the exchanger.
1.2.4 Water Supply
e water supply should be connected to the two
large hose connectors in the head assembly. If the
supply will be at high pressure, then clamps may be
necessary around the water tubing, to hold the tubes
on.
For a slow-response application, where water
conservation is important, one nozzle can be sealed
with a cap over the hose connector and water
supplied only to the other one.!
1.2.5 Temperature Probe
e temperature probe should be inserted down the
stem adapter as far as it will go. A little Vaseline
smeared on the shawill help it go in more easily,
and will also ensure an air-tight tting. e probe
should be plugged into the temperature data logger,
which should be put in its own plastic bag to protect
it, once it has been launched.!
1.3 Water Flow
1.3.1 Water Source
e water entering the instrument should come
direct from the sampling point, below the surface,
and should not have been exposed to any air-water
interface en route. e water should be clean and
free from debris. If necessary, it should be ltered
(but not with charcoal) before entering the exchanger.
e preferred delivery system is a submerged pump,
delivering the water at fairly low pressure, straight
from the sampling point to the exchanger. ree
sizes of spray nozzle are supplied. e choice will
depend on the pump performance and the speed of
response required.
1.3.2. Water Level
With the water owing, a spray will be observed
inside the body of the exchanger. Water will
accumulate inside the base and overow out through
the hose connector(s). If the ow rate is very high it
may be necessary to utilize both hose connectors.
e level may also start to rise too high inside the
body of the exchanger. To correct, either reduce the
ow rate or purchase the optional 12” cylinder. A
high water level inside the exchanger is no problem
provided it is no more than 1/3 of the way up the
exchanger and is stable.
If the water level rises slowly but continuously up the
cylinder it will be due to the water source being
completely without dissolved air. To correct this a
bleed may be added to the return air path from the
RAD7. It should be a long piece of tubing open at
one end and connected with a T-connector to the
tubing between the head assembly and the check
valve.!
1.4 Air Flow
1.4.1 Continuous Pumping
If you choose “SETUP, PUMP, ON [ENTER]” and
“SETUP, MODE, SNIFF [ENTER]”, then the pump
will run continuously, regardless of the length of the
cycles, or status of the RAD7, and the RAD7 will
count only 218-Po decays. In this operational mode,
the system will have the fastest response time, but
desiccant is quickly hydrated.
1.4.2 Pump on “Auto”
e pump, in auto operation, pumps for ve minutes
at the beginning of every cycle and then for one
minute in every ve. e length of each cycle is
chosen by the user. Short cycle times will involve
more pumping and speed up the response of the
instrument, but will consume more desiccant. Longer
cycle times will give better statistical precision to the
individual readings, and will conserve desiccant and
memory space.
For rapidly changing concentrations of radon in the
water, 10 minute cycles, with the PUMP on AUTO
and MODE on SNIFF, may be a suitable compromise,
though not giving fastest response.
1.4.3 Mode set to “Auto”
For very low concentrations, or when long-term
monitoring is desired, long cycle times of an hour or
longer, with the PUMP and MODE set to AUTO,
may make good sense.
Section 1 RAD AQUA Setup
8

1.5. Protocol
1.5.1 RAD7 Protocol
First, please read the RAD7 manual and learn how to
use the instrument for measurement of radon in air.
e RAD7 should normally be operated with AC (or,
with 12V option, 12V) power applied, to keep the
batteries in a fully charged condition.!
1.5.2 RAD AQUA Protocol
Switch on RAD7. Push the MENU key. Go to:
SETUP, CYCLE, push [ENTER]. Set the cycle time
required (as discussed above).
SETUP RECYCLE to 00, for continuous operation.
SETUP MODE: For fast response, with moderate or
high radon concentrations, choose SNIFF. For low
concentrations, to gain better statistics, choose AUTO.
SETUP THORON: Choose OFF
SETUP PUMP: Choose ON or AUTO, depending on
choice, see above"
SETUP TONE: Choose what you like
SETUP FORMAT: Choose what you like, but LONG
format with short cycle times uses a lot of paper. You
will probably not need to use the printer at all, in the
eld.
SETUP UNITS: Your choice
SETUP SAVUSER: Push [ENTER]. When it says “Are
you sure?” use arrow keys to change response to “Yes”
and push [ENTER].
1.5.3 User Protocol
You now have your personalized USER protocol
saved. To recall your settings, go to SETUP,
PROTOCOL, USER and push [ENTER]. To make a
change, simply display the parameters to be changed,
make your changes then, once more, go to SETUP
SAVUSER and save them."
Section 1 RAD AQUA Setup
9

2 MEASUREMENT PROCEDURE
2.1 Start Up
2.1.1 Temperature Probe
Load the temperature logger soware and connect
the temperature logger to the PC using the serial
cable provided. Congure the logger to take
temperature readings at frequent intervals (these may
be far more frequent that the RAD7 test cycles.)
Choose the second (external) temperature sensor.
Connect the temperature probe to the logger and
note that when you hold the probe the indicated
temperature rises.
When everything is set, begin recording. e logger’s
LED may ash periodically. Once the logger has
begun running, you may remove the serial cord from
both the logger and the PC.
Warning! Make sure that previous temperature data
has been downloaded before launching the logger.
e launching process may erase previous data.
If not already done so, insert the probe into the RAD
AQUA.
2.1.2 Start Measurement
Start the water owing. Note that, aer a few seconds,
water starts to ow out of the outlet hose. Switch on
the RAD7 (have the printer switched on if you are
using it. e RAD7 will then print a header for the
data printout, including a review of the setup, before
it gives you a ‘Test’ prompt.)
Provided that the RAD7 has been set up properly, see
above, at the ‘Test’ prompt, push [ENTER] then the
right arrow, to see ‘Test Start’ on the LCD, then
push [ENTER] to start the test.!
2.2 Speed of Response
2.2.1 Measurement in Progress
e instrument is now measuring the radon in the
water. With high concentrations and short cycle
times, and depending on the air and water ow rates,
it will take half an hour or more before there is much
of a reading, and maybe y minutes before you can
rely on the count rate being close to the equilibrium
value. Aer that you need to accumulate sufficient
counts for the precision desired. For example, 100
counts would give a reading with a standard
deviation of 10%. At very low concentrations, it may
take hours, and averaging over many cycles, to reach
a sufficiently precise value.
2.2.2 Influencing Factors
ere are two processes requiring time. One is for
the air in the closed loop to approach equilibrium
with the water and the other is for the RAD7 to
respond to the changed radon concentration in the
air loop. e rst is primarily controlled by the water
ow rate and the second is determined by the half life
of the rst daughter of radon, namely 218-polonium.!
2.2.3 Water flow rate
At typical room temperature, the equilibrium
coefficient for radon in air and water is about 4:1.
at is the concentration of radon in air at
equilibrium with water will be four times higher than
the concentration in the water. If the water were able
to give up all its radon to the air it would take four
times the air volume just to deliver the radon. In
practice the transfer is not complete so we may
estimate that ten times the air volume is required.
Considering the volume of the RAD7, the drying unit
and the RAD AQUA, we can conservatively estimate
the volume of the air loop to be of the order 4 litres.
erefore about 40 litres of water is needed to deliver
the radon to the air loop before it can reach
equilibrium. A water ow rate of V L/min will take at
least 40/V minutes to deliver the radon.
2.2.4 Air flow rate
ough important for thoron, see below, the air ow
rate is less critical to the radon response time. For
maximum speed of response, the air should keep
circulating around the loop so that the air in the
exchanger is continually being replenished with air
from the measurement chamber of the RAD7. us
the shortfall from equilibrium and hence the
efficiency of transfer is maximized. To achieve this
the pump may be set to ON, see 1.4.1 and 1.5.2
above.
For a more relaxed operation, the pump may be set to
AUTO, which will preserve the desiccant and
Section 2 Measurement Procedure
10

increase the life expectancy of the pump. Air will
remain stationary in the RAD AQUA for four
minutes before moving to the desiccant where it waits
another four minutes before entering the RAD7.
While in the RAD AQUA, the stationary air may
approach equilibrium with the water thus inhibiting
further radon transfer from the water to the air. It
will be about 15 to 20 minutes before that parcel of
air returns to the RAD AQUA. We can therefore
estimate that the response time of the system will be
increased by about 20 minutes if the pump is set to
AUTO
Having the pump on AUTO would normally be
associated with having the RAD7 cycle time of 30
minutes or longer. So an extra 15 to 20 minutes on
the response time will not be excessive.
2.2.5 RAD7 Mode
In AUTO (the default) mode, the RAD7 will
automatically switch from SNIFF mode to NORMAL
mode aer three hours into the run. is is to take
advantage of the additional counts provided by the
214-Po decays that will, by then, have approached
equilibrium with the (steady) radon concentration in
the measurement chamber.
For slow, long-term measurements with long cycle
times AUTO mode for the RAD7 is appropriate. e
RAD7’s response time will be a couple of hours or so.
e RAD AQUA can be running with a low water
ow rate and the RAD7 pump can be on AUTO also.
For fast response, however, it is essential to force the
RAD7 to stay in SNIFF mode (Setup, Mode,
Sniff [ENTER]). It will then always count only the
218-Po decays, giving it a 13-minute, 95% response
time.
2.3 Long Term Measurement
2.3.1 Desiccant
As set up, above, the system will continue making
measurements indenitely. ere are, however,
various resources that are being used up in the
process, and which must be replenished. e most
obvious is the desiccant. A new, or regenerated,
laboratory drying unit will normally last for about ten
days of continuous use in a temperate climate. In this
application, however, it is receiving saturated air and,
therefore, will be hydrated more quickly. When the
remaining length of blue (dry) desiccant is less than
one inch, the desiccant should be replaced. Please see
the RAD7 manual on desiccant regeneration. If the
desiccant is not replaced, and the humidity in the
instrument rises above about 20%, then the
sensitivity drops offand the reading is lower than the
true value.
2.3.2 Memory
e capacity of the internal memory of the RAD7 is
1,000 records or cycles. If each cycle is half an hour,
that would be data for 500 hours, or just over 20 days.
Every time the desiccant is changed, therefore, all the
stored data should be downloaded to a PC, backed
up, and erased from the RAD7 memory."
Section 2 Measurement Procedure
11

3 DATA
3.1 Data Handling
3.1.1 Printer
e IR printer will print out data in short, medium or
long format - see RAD7 manual. In long format,
there will be a spectrum printed at the end of every
cycle.
3.1.2 Memory
e internal memory of the RAD7 stores the date and
time, the radon concentration, live time, total count
and percentage in each of the main energy windows,
as well as a host of other parameters, for every cycle -
see RAD7 manual. ese data can be downloaded to
a PC at any time, during or aer a run, with
CAPTURE soware, supplied.
3.1.3 RADLINK
e RADLINK remote control soware installed in
the RAD7 enables a PC to control the RAD7
remotely, to grab any or all of the data, and to divert
what would have been sent to the IR printer to the PC
instead - see RAD7 manual. ese, as with all other
commands, may be invoked from DURRIDGE’s
CAPTURE soware, as explained below, or using a
terminal program that can communicate through the
computer’s serial port.
e PC must be connected to the RAD7 with a null-
modem cable, as supplied with the RAD7. If the PC
does not have an RS232 serial port the Keyspan USB/
RS232 adaptor, as supplied, should be used. It is
advisable to go to the Keyspan web site
(<www.keyspan.com>) to download the latest driver
for the adaptor and the operating system of the PC.
3.1.4 CAPTURE for Windows or Mac
Data recorded to the RAD7 may be easily download
and graphed DURRIDGE’s CAPTURE soware.
CAPTURE is available from the DURRIDGE web site
(<www.durridge.com>).
RAD7 data that has been collected using the RAD
AQUA should be downloaded using CAPTURE, then
graphed using CAPTURE’s RAD AQUA radon
source option, which uses water temperature data to
calculate radon concentrations. is water
temperature data is obtained by exporting it from the
temperature logger soware, (see Section 3.1.5,
below). For specic instructions on downloading and
graphing RAD7 data, please refer to the CAPTURE
user’s manual, which is available from the
DURRIDGE web site.
Once the data has been downloaded and backed up
securely, you should erase the data on the RAD7, to
prevent it from accumulating and lling the device’s
memory.
3.1.5 Temperature Data
To obtain the temperature data, hook up the
temperature logger to the PC and run its soware to
download the data. e program will take a moment
to download the entire memory, and then display it
as a graph. You should save it to your hard drive
before doing anything else. You can also export it to
a comma-delineated .TXT le for use with
CAPTURE, or for incorporating into a spreadsheet or
database program.
Warning! Make sure that the data is properly
downloaded and backed up before launching the
logger again. Certain operations may erase previous
data.
3.1.6 Time Relationship
A temperature reading is made at the moment in
time indicated with the reading. A radon reading, in
contrast, is the average value taken over the cycle
whose end occurred at the time indicated. More
precisely, in SNIFF mode, taking into account the
218-Po half life, a cycle whose end occurred about 5
minutes before the time indicated. For constant
radon and temperature values this is of no
consequence but if the temperature was changing
quickly, then the temperature readings during the
course of the radon cycle, and for ve minutes before,
should be averaged to give the average temperature at
the air water interface when the radon being
measured was leaving the water.
3.2 Data Conversion
3.2.1 FRITZ WEIGEL
e RAD7 gives an accurate reading of the radon
concentration in the air. With the RAD AQUA, this
Section 3 Data
12

air reaches equilibrium with the water in the
exchanger. To convert the air concentration to water
concentration, the air concentration must be
multiplied by the partition, or equilibrium,
coefficient, given by the Fritz Weigel equation
(Weigel, 1978):
a = 0.105 + 0.405 * exp(-0.0502*T)
where T is the temperature in deg C.
At room temperature, a is around 0.25, giving, at
equilibrium, a four-to-one ratio of radon in air to
water."
3.2.2 CALCULATION
e RAD7 radon data is stored in the RAD7 and the
water temperature data is stored in the temperature
logger. It is necessary to download the RAD7 data to
a PC, preferably using DURRIDGE’s CAPTURE
soware, and to download the temperature logger
data using the soware provided by the logger’s
manufacturer. e water radon concentration may
be calculated using CAPTURE (see the RAD AQUA
information in the CAPTURE user’s manual) or it
may be calculated manually, according to the Fritz
Weigel equation. CAPTURE automatically gives the
average radon concentration of the selected data
between the cursor lines, but the temperature needs
to be assessed separately and entered into the
equation above to determine the partition ratio."
Section 3 Data
13

4 THORON IN WATER
4.1 Why Thoron?
oron, 220-Rn, an isotope of 222-Rn radon, has a
55.6 second half life. As a result, almost everywhere,
it is not to be found. Close to a thoron source,
however, the water will still have measurable thoron
as it will be still be young and not have had time for it
all to have decayed away.
oron coexists with radon in the soil. Ground water
entering the ocean will therefore bring thoron as well
as radon with it. Around submarine springs,
therefore, there may be thoron in detectable amounts
and this may be used to locate the springs (Burnett et
al, 2007).
4.2 Measurement in Water
Because of its short half-life, the measurement of
thoron in water is fraught with difficulties.
First, the concentration in the water will vary
signicantly from point to point and from time to
time depending on the position of the sampling point
relative to the position of the source and the water
ow between the source and the sampling point.
Second, during the process of getting the thoron
atoms into the measuring device, the thoron will be
decaying thus reducing the size of the sample. An
estimate of the time taken for this process is required
in order to apply a correction to the reading.
However, if using thoron as a tracer, a knowledge of
the absolute sensitivity is not so important as
minimizing the lower limit of detection. is may be
done by making the transfer of thoron atoms into the
RAD7 as quick as possible.
4.3 Thoron Sensitivity
Sample is lost by decay while in the water en route to
the exchanger and then again in the air en route to
the RAD7.
4.3.1 Source to Exchanger
e time delay in the water can be made small by
having a high water ow rate and a short, small
diameter hose. If the hose is no more than 3m long,
with an internal diameter of no more than 8mm, say,
the hose volume will not exceed 0.15 litre and a water
ow rate of, say, 4 L/min will mean a time delay for
the thoron to reach the exchanger of no more than
2.5 seconds.!
4.3.2 Exchanger to RAD7 method 1
e RAD7 pump will typically generate an air ow
rate of around 0.9 L/min. e volume of the air
above the water spray in the chamber will be about
0.5L, the laboratory drying unit is about 1L as is also
the RAD7 itself. An estimate, therefore of about 2.5L
of air is required to be pumped for a thoron atom in
the exchanger to reach the RAD7. At 900ml/min this
will take a little less than 3 minutes. e thoron will
have decayed through about three half lives between
leaving the sampling point and being detected in the
RAD7. In addition, the transfer from the water to the
air will not be complete and the returning air from
the RAD7 will have lost almost all its thoron, so there
is another factor to be multiplied in. All in all, we
may estimate that we see no more than 10% of what
would have been seen had there been no thoron
decay during acquisition.!
4.3.3 Exchanger to RAD7 method 2
Instead of using the RAD7 pump, a separate pump
may be used to circulate air round the loop much
faster than the RAD7 pump. However, the RAD7
cannot tolerate an air ow rate higher than 3L per
minute. e RAD7, therefore, should be connected
to tap into the fast recirculating air loop, using its
own pump to do so, with the main air ow bypassing
the RAD7.
With a circulating air ow rate of, say, 10L/min the
delay from the exchanger to the RAD7 tap will be
about 0.2 minutes and from the tap to the RAD7
chamber about 1 minute. So now, the total time delay
from sampling point to entry into the RAD7 will be
no more than 1.5 minutes so reducing the attenuation
due to radioactive decay of the thoron to no more
than 70%.
Furthermore, with the external pump, the time for air
to circulate once round the loop will be reduced to a
fraction of a minute. e thoron concentration in the
recirculated air will now be signicant, giving a better
chance for the air leaving the exchanger to be closer
to equilibrium with the water.!
Section 4 Thoron In Water
14

4.4 Speed of Response
ere is another advantage to using thoron as a tracer
to locate submarine springs. at is the almost
instantaneous response of the RAD7 to thoron.
e rst daughter of thoron, 216-Po, has a half life of
just 150mS. us in 0.5 seconds the RAD7 window
B, 216-Po, count rate will have nearly reached
equilibrium with the thoron in the chamber. So the
speed of response of the RAD7 to thoron is limited
not be the half life of the polonium daughter but by
the time it takes to get the sample into the
measurement chamber.
We have seen that with the highest sensitivity
conguration, using a separate pump to circulate air
round the loop, the total time for thoron to go from
the sampling point to the RAD7 is only about 1.5
minutes.
A boat carrying the system, therefore, moving slowly,
will see the thoron count rate increase within a
minute or two of passing over a submarine spring
and drop again shortly thereaer."
Section 4 Thoron In Water
15

5 DRYSTIK
5.1 Passive DRYSTIK
A passive DRYSTIK may be installed in the RAD
AQUA system without modifying any other part of
the system or the operating conditions. e inner
membrane tube goes between the exchanger and the
desiccant while the outer sheath is purged by dry air
from the RAD7 outlet. e two ows should be in
opposite directions along the DRYSTIK. A 12”
DRYSTIK will increase the life of the desiccant by a
factor of about ve. A 48” DRYSTIK will increase the
life by about 10 times.
5.2 Active DRYSTIK
In active conguration, there is a pump upstream and
a needle valve downstream of the inner membrane
tube. is increases the pressure inside the tube
which increases its efficiency.
A typical setup has the RAD7 pump set to OFF
(Setup, Pump, Off [ENTER]), the DRYSTIK
pump running continuously and the needle valve
adjusted to give a ow rate of about 0.2 L/min.
When using an active DRYSTIK, a small drying tube,
inserted between the DRYSTIK needle valve outlet
and the inlet lter on the RAD7, will keep the air
sample in the RAD7 below 7% RH. It will also last
for a very long time and add only a tiny volume to the
air loop, maintaining a fast response. See Fig. 3."
5.3 Effect on Response Time
With a ow rate of only 0.2 L/min it will take about
20 minutes for the air in the loop to go round once.
is will make thoron detection impossible and also
add an extra 10 or 15 minutes to the response time
for radon. For long term studies the slower response
is generally not important, whereas the frequency of
replacing the desiccant may be. So an active
DRYSTIK may be of considerable benet.
5.4 Custom designed Active DRYSTIK
e 0.2 L/min ow rate of the DURRIDGE-supplied
active DRYSTIK arises because it matches the average
ow rate of a RAD7 in AUTO mode and also matches
the performance of the installed pump at a pressure
of 44 PSI (3 atmospheres). With the appropriate
choice of pump and needle valve, an active DRYSTIK
can be produced that will maintain a 44 PSI pressure
inside the inner membrane tubing and a ow rate of
1L/min or even more, to restore the speed of
response of the system while virtually eliminating the
need to replace the desiccant periodically."
Section 5 DRYSTIK
16

Fig. 3 RAD AQUA configuration with Active DRYSTIK and Water Switch
Section 5 DRYSTIK
17

6 BOATING
e RAD AQUA was originally designed to serve the
needs of oceanographers making surveys, by boat, of
coastal zones and lakes. In this application there are
a number of considerations.
6.1 Response time
If the boat is moving, which it usually is, the speed of
response of the system translates to the spatial
discrimination of the data. Roughly speaking, the
minimum distance that can be separated in terms of
radon concentration is the 95% response time, to a
step change in radon, times the boat velocity. It
behoves one, therefore, to minimize the response
time in order to have reasonable spatial resolution at
a boat speed that is not too slow.
By going through the spray chamber the water is, in
effect, delivering radon to the air loop. As discussed
earlier, if V is the total volume of the air loop, it needs
10 to 20 times that volume of water to pass through
the exchanger to deliver sufficient radon to the air for
the reading to approach close to equilibrium. Aer
equilibrium is reached, sufficient counting time then
has to pass for the reading to achieve the desired
precision. Call T1 the time to reach equilibrium and
T2 the counting time needed to reach the desired
precision. For a given setup and ow rates, T1 will be
a constant. T2 will be a function of the radon
concentration.
6.1.1 Minimizing T1 with increased water flow
rate
Use the fastest pump that is consistent with power
availability. A Rule360 bilge pump draws less than
2.2A at 12V. It delivers 4.8 L/min, unobstructed ow,
to 2above the water surface, and 2.4 L/min through
two W1 nozzles mounted in a RAD AQUA. For
many purposes, this pump rate is adequate and the
power requirement is quite modest. A 45 AH car
battery can run the pump for two 8-hour days of
surveying and have energy to spare.
However, for fastest possible response time, the RAD
AQUA can handle ow rates up to 10 L/min or more,
through two W4 nozzles. But the power needed
increases signicantly. e Rule 3700 delivers 6 L/
min through a W4 nozzle, but it takes over 15A. One
8-hour run would completely discharge a 120 AH
battery.
6.1.2 Minimizing T1 with reduced air volume
A typical RAD AQUA setup has a RAD7, a spray
chamber and a drying unit. e spray chamber and
the RAD7 are each around 1L in volume. e
Laboratory Drying Unit is about 0.5L. Tubing
volume can be minimized by avoiding unnecessarily
long lengths of tubing and using small ID tubing
where possible. is setup would have an air volume,
V, of about 2.5L.
An Active DRYSTIK, that efficiently removes
humidity from the air even when there is no
desiccant in the circuit, has only a small air path
volume. It can replace the Laboratory Drying Unit.
For optimum drying efficiency a small drying tube
can be inserted between the DRYSTIK and the
RAD7. is setup would have an air volume, V, of
about 2.2L.
With a Rule360, 10 times V of water would be
delivered in 11 or 9 minutes, respectively.
Conservatively, the response time to reach
equilibrium may be taken to be 20 minutes in either
case.!
6.1.3 Minimizing T2 with a higher sensitivity
RAD7
A standard RAD7 has a typical sensitivity of 0.25
cpm/(pCi/L) (0.007 cpm/(Bq/m3)) in Sniffmode
(counting only 218Po decays) and double that in
Normal mode (counting both 218Po and 214Po
decays). Water at typical room temperature with 2.5
pCi/L radon concentration will reach equilibrium
with radon in air at around 10 pCi/L. is will
produce around 2.5 cpm in Sniffmode and 5 cpm in
Normal mode, in a standard RAD7.
For 10% standard deviation, or 20% two-sigma
uncertainty, we need, according to Poisson statistics,
100 counts. In Sniffmode that would take 40
minutes. us, at 2.5 pCi/L radon in the water, with a
Rule 360 pumping the water and a single, standard
RAD7, the total time to reach a reading with better
than 20% two-sigma uncertainty would be about one
hour. A much faster pump could perhaps reduce that
time by about 10 minutes.
A high-gain RAD7 generally has an oversized dome
and and oversized alpha detector. e gain is
increased by better than 50%, over a standard RAD7,
Section 6 Boating
18

and the volume increased by about 0.2L. us the
equilibrium time, T1, will be increased by a minute
or two while the counting time will be reduced by
more than 33% which, in this case, will be around 14
minutes. So, for this radon concentration, there will
be a net reduction of more than ten minutes in the
response time.
6.1.4 Minimizing T2 with multiple RAD7s and one
RAD AQUA
Several RAD7's may be deployed in parallel, all
accessing one RAD AQUA. In analyzing the data
uncertainty, the counts of all the RAD7s are summed.
us three RAD7s will reduce the time to reach a
given precision by a factor of three. On the other
hand, each RAD7 adds about 1L to the air-loop
volume, thus potentially increasing T1.
For example, with the radon concentration as above,
three high-gain RAD7s with one Laboratory Drying
Unit, one RAD AQUA and a Rule 360 pump, the
counting time to reach about 20% two-sigma
uncertainty will be about 10 minutes (reduced from
30 minutes with only one RAD7). On the other
hand, there is an extra 2L in the air loop, requiring an
extra 20L - 40L of water. With the Rule 360 that
would take an extra 10 - 15 minutes, reducing the
advantage of the three pumps to only 5 to 10 minutes.
But with the Rule 3700, the extra 20L to 40L of water
would take only 3 or 4 minutes to deliver, thus
preserving almost all the 20 minute gain of the
multiple counting technique."
6.1.5 Minimizing T2 with multiple RAD7s and
multiple RAD AQUAs
If each RAD7 has its own RAD AQUA and bilge
pump, adding another complete system will not
change T1, but will reduce T2. A compromise of
adding another bilge pump and RAD AQUA for
every two RAD7s may best meet a user's need for
fastest response on the one hand and minimum
power requirements on the other.!
6.2 Pump position
In a stationary boat it is simple to hang a bilge pump
over the side to the depth of interest. But when the
boat is moving there is a lateral force on the pump
that may force it towards the stern and consequently
bring it closer to the surface.!
6.2.1 Positioning Equipment
A long tether, running from the near the bow, will
stop the pump from swinging to the stern, when the
boat is under way. If the pump is heavy enough, or if
a weight is added, this tether may be enough.
Another tether, from the near the stern, will support
the pump when the boat is stationary.
A stiffpole could be used, with the tether, to keep the
pump at a xed depth, regardless of the boat speed."
Section 6 Boating
19

7 CARE, MAINTENANCE, AND TROUBLESHOOTING
7.1 Water Catastrophe
If water ever enters the RAD7, or if the RAD7 ever
goes swimming in the water, it will probably cease to
operate and immediate steps should be taken to
minimize the impact on the instrument.
Keep the RAD7 upright. is will prevent water from
touching the detector, which is close to the face plate
at the top of the dome. Put a piece of tubing on the
RAD7 outlet with the other end in a sink. Use the
RAD7 pump if it still works or, otherwise, an external
pump into the inlet, to blow air through the
instrument. When water ceases to be blown out of
the outlet, put desiccant upstream of the RAD7 to dry
out the air path. When the air path is fully dry (aer
dry air has been blown through it for approximately
one hour), remove the face plate from the case, empty
the water out of the case and blow dry the case and
the RAD7 electronics.
Once there is no visible water in or on the
instrument, it can be put in an oven at 50°C for a few
hours to dry out completely. Additionally, desiccated
air can be passed through the air path until the air
leaving the RAD7 drops below 10% RH. Aer this
treatment further corrosion will be prevented, and
the RAD7 will boot once more and you can use the
internal RH sensor to measure how dry the air path
is. At this point the instrument should be returned to
DURRIDGE for service.
7.2 RAD7 Care
Water, particularly salt water, is hostile to electronic
instruments. Please keep the RAD7 in a relatively
clean and dry environment. One way is to enclose
the instrument in a large, transparent plastic bag, see
Section 1.2.3 above. Should it ever be seriously
splashed with salt water, please follow the
instructions in Section 7.1, above.
As a preventive measure, plastic cling wrap can be
placed over the RAD7 face plate and down the sides
of the RAD7. Push it down around the hose
connections, push the power and RS232 plugs into
their sockets, and push the lid onto its hinges. e
wrap will make the RAD7 almost watertight. If it
tears it can be easily replaced at any time.
e instrument should, in any case, be returned every
year for recalibration.
It is useful to look at a cumulative spectrum
periodically. is may be obtained by having the
printer on and allowing the RAD7 to complete a run.
e “Recycle” number may be set to the current cycle
number (Setup, Recycle, NN [ENTER]). When
the RAD7 reaches the end of the current cycle it will
then print out the end of run summary including the
cumulative spectrum. Look to see that the peaks are
clean and in the normal position.
7.3 Exchanger Care
e exchanger should be kept as clean as possible in
the circumstances. Sea water, if carrying any solid
matter, should be ltered. e spray nozzle should be
examined for build-up of deposits, and cleaned if
necessary.
7.4 Desiccant Regeneration
Please see the RAD7 manual for information on the
care and regeneration of the desiccant. Regenerated
desiccant, aer a few regenerations, loses most of its
indicating ability (due, we believe, to migration of the
cobalt chloride to the interior of the calcium sulphate
crystals). One way to ‘indicate’ the status is, every
time you rell the laboratory drying unit with
regenerated desiccant, you rst add half an inch or so
of new, blue desiccant, out of the jar. is way, you
can always tell if the unit is still working, as the new
desiccant will only turn pink when the rest of the
desiccant, upstream, has become hydrated.!
7.5 RAD AQUA Troubleshooting
!
7.5.1 Rising Water Level
Should water rise inside the exchanger there is a
danger that it may be sucked out into the desiccant
and RAD7. is may occur because of the water
supply having no dissolved gas and absorbing air
from the exchanger. To prevent this, a bleed
consisting of a long, small bore piece of tubing may
be connected, with a T-connector, to the return air
supply downstream of the check valve.
Rising foam, due to some forms of pollution in the
water, may also be treated with a bleed as in the above
Section 7 Care, Maintenance, and Troubleshooting
20