Page 1
DIGITAL R.F. PROBE INSTRUCTIONS
INDEX
Page
1. GENERAL DESCRIPTION ........................................................... 2
2. PRINCIPLE OF OPERATION ....................................................... 2
3. SPECIFICATION ........................................................................... 3
3.1. Digitrans probe 3
3.2 Digitrans controller - model C2C 4
4. INSTALLATION ........................................................................... 4
4.1 Probe installation 4
4.2 Controller installation 4
4.3 Controller terminations 5
5. OPERATION OF PROBE AND CONTROLLER ........................ 6
5.1 The keypad 6
5.2 Operation 6
5.3 View or set parameters 7
5.4 Calibration options 8
5.5 Display mode 8
5.6 Switch 4 / 20 mA 8
5.7 Set Access code 8
5.8 Test Watch dog 8
6. COMMISSIONING ..................................................................... 9
6.1 Power up 9
6.2 Check probe operation 9
6.3 Loop testing 10
7. CALIBRATION ........................................................................ 10
7.1 Simple 2 point calibration 10
7.2 Automatic calibration using regression 11
7.3 Calibrator 14
8. FAULT DIAGNOSIS ................................................................. 16
Page 2
1. GENERAL DESCRIPTION
The DIGITRANS DIGITAL R.F. Probe consist of a probe (electrode) fitted with a
head containing the measurement electronics which is connected to a second box
(the controller), via a single co-axial cable.
The controller contains a power supply, a microprocessor, keypad and display,
A/D and D/A converters, and is also used to supply power to the probe, control
the probe, receive signals from the probe, calculate brix from these
measurements, and provide digital indication plus 4-20 mA output signals for
connection into PID controllers or DCS.
One control box can operate two probes in multiplex. In addition, for each probe
there are two analog outputs available, and an analog input suitable for
connecting a 2 wire temperature transmitter (one for each probe).
R.F. readings may therefore be corrected for temperature if required.
2. PRINCIPLE OF OPERATION
The instrument consists basically of a variable frequency oscillator built into the
probe head which is coupled to the probe electrodes. The oscillator is switched to
operate at two different frequencies separated by about 10 MHz and the probe is
also switched in and out of circuit by a second switch to provide two measurement
and two reference frequencies.
The probe is controlled by a microprocessor controller unit which operates the
switches in the probe head. Four output frequencies (two measurement and two
reference), are then measured in sequence over about a 3,8 second period.
These frequencies are designated Fa, Fb, Fc, and Fd. Fa and Fd are reference
frequencies with the probe disconnected from the oscillator, and Fb and Fc
measurement frequencies with the probe in circuit. The differences between the
two measurement and reference frequencies Fb-Fa and Fc-Fd are calculated and
scaled and designated F1 and F2. These differences are then used for calculating
brix.
F1 and F2 are each affected differently by massecuite resistance and capacitance
and these signals are combined in a mathematical formula to derive an optimum
measurement signal calibrated in brix units.
In order to calibrate the instrument, a series of measurements of F1 and F2 are
taken for different known brixes and brix correlated against the measurements
using a regression technique. The constants derived from this correlation are then
used by the microprocessor to calculate brix from F1 and F2 when in the
measurement mode.
For each probe there are two 4-20 mA analog outputs. These outputs (1 & 2) may
each be used for different calibrations (eg one for brix and the other for ash), or
for calibrating a probe for use on two different products.
Page 3
The formula used to derive probe 1 output 1 is given below:-
Output = K3 + K4*F1 + K5*F2 + K6*F12 + K7*F23 + K8*F13 + K9*F23 + K10*T
K3 to K10 are constants stored in memory which are inputted during calibration.
T is temperature measured by a temperature probe (Range 0 - 100 C).
The above formula gives the output in engineering units. The 4-20 mA output may
be spanned to operate over any range by setting parameters K1 (zero) and K2
(full scale)
Parameters K1 to K10 can be entered via the keypad, but alternatively constants
K3 to K9 may be automatically calculated and stored by the microprocessor when
a calibration procedure is carried out.
For Probe 1 output 2, the corresponding constants are K11 to K20, and for probe
2 the constants for output 1 and 2 are K21 to K30, and K31 to K40 respectively.
3. SPECIFICATION :
3.1 Digitrans Probe
Probe length overall : Model T2 422 mm
: Model T4 507 mm
: Model T3 577 mm
Probe length into pan : Model T2 165 mm
: Model T4 250 mm
: Model T3 320 mm
Probe diameter : 26 mm
Probe body material : 316 stainless steel
Connection into pan : Adaptor flange/bushing
Probe body temperature : Max 120 deg.C
Probe head temperature : Max 70 deg.C
Connection : N type coaxial socket
Cable to controller : RG 58 C/U 50 ohm coaxial
Power consumption : Typically 55 ma 16-25 V DC.
Protection class : IP 65 enclosure.
PCB encapsulated in epoxy for added protection.
Measurement frequencies : approx. 18 and 27 MHz
Page 4
3.2 Digitrans Controller - Model C2C
Power requirement : 24 V DC 300 mA.
Probe inputs : 2 off N type coaxial sockets
Program : EPROM
Parameters : RAM battery back up
Display : 4 line alpha numeric illuminated.
Keypad : 16 keys - access through password.
Temperature : 50 deg. C max.
Outputs : 4 x 4-20 mA into 500 ohms
Inputs : 2 x 4-20 mA two wire self powered.
Isolation : Power supply and 4-20- mA signals galvanically
isolated from probes
Terminations : Terminals in separate compartment to electronics.
Protection class : IP 65
4. INSTALLATION
4.1 Probe installation
The probe is push fitted into a housing which is mounted in the bottom of a pan or
on a pipe where liqour brixes are measured. The housing consists ot two parts, an
adaptor flange which is welded into the vessel, and a bronze inner housing which
bolts onto the adaptor flange. The probe is fitted into the bronze housing and is
secured by a single locking bolt.
A dummy plug is recommended for use when a probe is removed for cleaning or
maintenace. Details of the probe housing and plug are given in the instruction
manual to allow fabrication by the user. Alternatively the housing and plug may be
ordered as optional extras.
4.2 Controller Installation
The controller box requires a 24 volt DC supply. The negative of the supply is
common with the 0 volts of the 4-20 mA outputs but is galvanically isolated from
the probe circuits.
Page 5
4.3 Controller Terminations
The terminals in the termination box are designated as follows:
Terminal Description
1 + 24 V supply
2 Negative supply
3 Screen
4 Screen
5 + Temp. transmitter probe 1
6 - Temp. transmitter probe 1
7 + Temp. transmitter probe 2
8 - Temp. transmitter probe 2
9 + 4-20 ma output 1 probe 1
10 - output
11 + 4-20 mA output 2 probe 1
12 - output
13 + 4-20 mA output 1 probe 2
14 - output
15 + 4-20 mA output 2 probe 2
16 - output
The coaxial fittings on the side of the terminal box are as follows:
Right Probe 1
Left Probe 2
Screened cable must be used to connect analog outputs to the controllers or
DCS. The screens may be connected to terminals 3, and / or 4.
If temperature transmitters are installed, these must be galvanically isolated from
earth. Screened cable must be used and the cable screens terminated only at the
controller (not at the transmitter). These screens should be connected to screens
of cables back to contol room.
The coaxial cables connecting the probes to the controller must be of good quality
with an impedance of 50 ohms (MIL spec. RG 58) and may be any length up to
70 m. Although cable lengths up to 100 m may be used, it is advisable to use a
coaxial cable with a lower loss and DC resistance (eg RG 8) in which case lengths
greater than 100 m are possible.
After installation of the coaxial fittings, the cables should be checked prior to
connection. Check for continuity (resistance) and for short circuits between centre
pin and screen. Resistance through the cable (with one end shorted) should be
less than 10 ohms.
Page 6
The probe is supplied with 16-25 volt DC through the cable. Current drawn is
about 55 mA. There is short circuit protection and current is limited to prevent
damage.
5. OPERATION OF PROBE AND CONTROLLER
On power up, the microprocessor displays "Warm start" followed by "on line"
Thereafter the pre-selected probe will be selected and the following displayed :
PROBE 1 TEMP. ...
OUTPUT 1 = ....
OUTPUT 2 = ....
This display is updated every 3,8 seconds. If there are no calibration constants, or
the probe is disconnected, then output 1 and 2 will be zero.
If there is no temperature transmitter connected, TEMP will also display zero.
5.1 The Keypad
The probe is calibrated and the mode of operation set via the keypad. Keys are
allocated as follows:
A: When in the run mode, pressing A allows access via a password to the
menu used for calibration or checking. The default password is 0 and
this may be changed by the user.
B: This is used to back track one digit when entering parameters, or to
backtrack parameters when viewing data.
ENT: This is the ENTER key and is pressed after entering data or to advance
to the next parameter.
F: This is used to return to the run mode and is active only when not in the
run mode.
In addition there are keys for digits 0-9, negative (-) and decimal point.
5.2 Operation
On power up, after initialising, the probe will go into the measurement mode and
the following will be displayed:-
PROBE 1 TEMP 0 C
OUTPUT 1 = ....
OUTPUT 2 = ....
The outputs will indicate zero unless there are calibration parameters in memory.
Page 7
The temperature will display 0 unless there is a temperature transmitter
connected.
On pressing key A, "ENTER ACCESS CODE" will be displayed.
Key in the acess code (default 0) followed by ENT. The following menu will be
displayed:
VIEW / SET VARS. 1
CALIBRATE 2
DISPLAY MODE 3
NEXT PAGE 4
On selecting 4 the following additional menu will be displayed:
SWITCH 4-20 mA 1
SET ACC. CODE 2
TEST W. DOG 3
FINISHED 4
5.3 View or set parameters
This option allows mode of operation to be changed or calibration parameters to
be entered or viewed.
On selecting this option the following is displayed:
PROBE 1
PROBE 2
BOTH 3
Select the option depending on which probe is required.
If 3 is selected the unit automatically returns to the measurement mode and both
probes are in operation in multiplex. If 1 or 2 is selected followed by F, the unit
also returns to the measurement mode with the selected probe in operation.
Alternatively, after probe 1 or 2 is selected and the following appears:
OUTPUT 1 OR 2 ?
When the required output is selected, calibration constants K .. are displayed. The
next parameter may be viewed by pressing ENT. If B is depressed, the previous
parameter is displayed.
New calibration parameters may be keyed in and entered if required.
Pressing F returns to measurement mode.
Page 8
5.4 Calibration options
On selecting this option, a further menu allows selection of the required probe and
output. The following new menu then appears:
NEW CALIBRATION 1
ADD TO CALIBRATION 2
ENTER ANALYSIS 3
REGRESSION 4
These will be used for automatic calibration and will be described later.
If option 1 is selected, all previous calibration data is erased but not calibration
constants (K values).
If option 1 or 2 is selected, then the frequency differences F1 and F2 will be
displayed. Pressing F returns to the measurement mode.
5.5 Display mode
When this option is selected display is changed to display the four frequencies. To
return to the normal display, this option must be again selected by pressing A etc.
5.6 Switch 4-20 mA
When this option is selected, all outputs cycle from 4 to 20 mA maintaining each
output for 5 seconds. This option may be used for loop testing.
Pressing F will terminate this option.
5.7 Set access code
When this option is selected, a new acess code may be entered. This may be any
number of up to 8 digits.
This code must be recorded since if forgotten access to the unit will only be
possible by using a master code.
Initially during calibration, it is advisable to use a simple number (eg 0) and only
enter the final code once calibration is complete.
5.8 Test watch dog
This option is used to check operation of the watchdog timer.
The function of the watchdog timer is to reset the microprocessor if there is a
failure due to a glitch in the power supply etc. This prevents the unit from "hanging
up".
Page 9
6. COMMISSIONING
6.1 Power up
After completing installation, when power is applied, the display will light up and
after a short while, the following will be displayed:
PROBE 1 TEMP 0 C
OUTPUT 1 = ....
OUTPUT 2 = ....
If the display does not light up, check for 24 volts across terminals 1 and 2 of the
controller terminations. Terminal 1 should be + 24 volts. If polarity is incorrect,
then it will be necessary to replace the fuse in the controller after polarity is
corrected. This is done by removing the controller cover. This should be replaced
by a 0,5 amp fuse.
If the display lights up but is blank, check supply voltage.
After sucessful power up, continue as follows:
6.2 Check probe operation
i) Press A, enter the acess code (default 0) , and select CALIBRATE.
The display will prompt :-
PROBE 1 OR 2 ?
ii) Press 1 to select probe 1. The following will now be displayed:-
OUTPUT 1 OR 2 ?
iii) The display will now give the following options:
NEW CALIBRATION 1
ADD TO CALIBRATION 2
ENTER ANALYSIS 3
REGRESSION 4
iv) Press 1. The display will now give:-
SAMPLE 1
F1 = ......
F2 = ......
Readings of F1 and F2 (between 1 and 20) will be shown if the probe is operating
correctly. If no readings are obtained, check connections to probe.
Page 10
If probe 1 is operating correctly, press F to return to the measurement mode and ,
if there are two probes installed, repeat selecting probe 2.
6.3 Loop testing
Press A, enter the acess code (default 0), and select NEXT PAGE ( Key 4) from
menu, then select SWITCH 4/20 mA (Key 1).
All outputs will switch alternatively between 4 and 20 mA. The signals can then
be tested at the control room or controller. When complete select FINISHED
(Key 4).
7. CALIBRATION
Two calibration procedures are available, a simple 2 point calibration based on F1
or F2 measurements which will provide a useful measurement signal (output
0-100%) without requiring laboratory analyses, and a more accurate regression
calibration requiring laboratory analyses which will provide optimum combination
of F1 and F2 signals, and provide a linearised readout in engineering units.
7.1 Simple 2 point calibration
The F1 or F2 signal will normally correlate to brix. Experience has shown that in
most applications the F2 signal provides a better correlation to brix than F1.
The method for setting up the probe is as follows:
Go into the calibration mode, select the required probe (PROBE1), select
“OUTPUT 1”, and then select “NEW CALIBRATION”.
F1 and F2 signals will then be displayed.
Reduce brix or solids content in the process to the lowest value required and
record F1 and F2 values. These values are designated F1L and F2L.
Increase brix or soloids content to the highest value required, and then record F1
and F2 values. These are designated F1H and F2H.
If the F2 signal is to be used, calculate the following :
K3 = 100 F2L / (F2L-F2H)
K5 = - 100 / (F2L-F2H)
These parameters are entered into the controller as follows:
Press A and select 1 (VIEW / SET VARS). The following will then be displayed:-
PROBE 1
PROBE 2
Page 11
BOTH 3
Press 1. The following is displayed :
OUTPUT 1 OR 2 ?
Press 1. Constant K1 = O is now displayed.
Press E. Constant K2 = 100 is now displayed.
These constants represent zero and span for the 4-20 mA output signal.
Press E and K3 is then displayed. Enter the value for K3 as calculated above and
then press ENT. K4 is then displayed. This should be zero. Press ENT again abd
K5 is displayed. Enter the value for K5 calculated above and press ENT. Check
that contants K6 to K10 are zero and then press F. The probe is now calibrated
and will provide a readout of 0 -100 corresponding to 4-20 mA output.
Alternatively if the F1 signal is to be used, calculate the following :
K3 = 100 F1L / (F1L-F1H)
K4 = - 100 / (F1L-F1H)
These values are then entered as described above.
Note the probe output is determined by:
OUTPUT = K3 + K4*F1 + K5*F2 + K6*F12 + K7*F22 + K8*F13 + K9*F23 + K10*T
7.2 Automatic calibration using regression
The probe can be calibrated against brix or solids content measured in the
laboratory of a number of samples taken from the process. It is important that
these samples cover as wide range of brix values as possible and are not all
around the same brix value.
The procedure is as follows:
7.2.1 Taking samples
Go into the calibration mode as described before.
After a short while the following is displayed:
CALIB. PROBE 1 OR 2 ?
Press 1 or 2 to select the probe to be calibrated and the next prompt is :
OUTPUT 1 OR 2.
Page 12
Each probe has two analog outputs. Normally only one needs to be used. Press
1 or 2 to select the output to be calibrated and then the following options will be
displayed:
NEW CALIBRATION 1
ADD TO CALIB. 2
ENTER ANALYSES 3
REGRESSION 4
Option 1 is for a complete new calibration where a new set of samples are to be
taken. Pressing 1 erases all previous calibration data from memory.
Option 2 allows additional calibration data to be added to the data bank.
Assuming a new calibration is to be done, enter "1".
The following is then displayed:
SAMPLE 1
FI = ...
F2 = ...
F1 and F2 are the two frequency differences. These frequency differences are
used in the multi- linear regression.
Sample 1 is taken from the pan and at the same time the E key pressed. This
then stores F1 and F2 against sample 1.
SAMPLE 2 is then displayed.
The pan conditions are changed to give a different brix and this will give different
values of F1 and F2. Take a second sample and press "E".
SAMPLE 3 is then displayed.
By pressing "B" one can go back and resample if a mistake has been made.
Pressing "A" moves forward without changing the previous F1 and F2 values
stored.
Between samples, it may be desirable to go back on line so that the pan can be
controlled and the set point changed to vary brix. This is accomplished by
pressing F.
When returning to calibrate to take a sample, select "ADD TO CALIB." which will
allow new data to be added to the stored data.
At least 8 samples (up to 20) must be taken to calibrate a probe. The
microprocessor will not carry out an automatic calibration unless there are 8 or
more samples entered.
Page 13
7.2.2 Entering of laboratory data
Send the samples to the laboratory for analysis and when the results are
obtained, go into CALIBRATE mode, select PROBE NO., and OUTPUT as
before, then select " ADD ANALYSES" by pressing 3.
The following is then displayed:
SAMPLE 1 0
F1 = ...
F2 = ...
Key in the laboratory brix for sample 1 and then press "E".
Pressing "B" will backtrack if an error is made.
The data will be stored and SAMPLE 2 ... then displayed.
When all analyses have been keyed in, press "B" to backtrack if the data is to be
checked, otherwise press "F" to go back on line.
It is desirable to write down the stored values of F1 and F2 and the analysis for
each sample for future reference.
If a set of data (for example sample 5) is wrong, it can either be overwritten by
going to add to calibration and taking a new sample and then keying in the
analysis, or in the ADD analysis mode, the data may be erased by pressing A. In
this case, sample 6 becomes sample 5 etc.
7.2.3 Regression
The microprocessor can now be used to automatically calibrate the probe in brix
units as follows:-
Go into the calibration mode, select probe and output as before and then select
"REGRESSION" by pressing 4.
The microprocessor then checks the data for completion and may come up with
the following messages:
ERROR IN DATA
or INSUFFICIENT DATA
There must be at least 8 sets of data to be accepted.
If ERROR IN DATA is displayed press "F" to return on line and then go to ADD
ANALYSES. Check the data by pressing "E" to go forward, or "B" to backtrack.
Correct the error. There must be an analysis for each set of frequencies and vice
versa.
Page 14
If the data was acceptable, the microprocessor then does two separate
regressions between F1 and F2, and the analyses.
During calculation the following messages are displayed:-
CALCULATING
REGR. No 1
MATRIX COMPLETE
REGR. No 2
MATRIX COMPLETE
When calculation is complete, the standard errors of the two regressions are
displayed and the regression chosen (one with the lowest standard error)
The standard error is a good indication of the accuracy that may be expected in
that application.
If the results are acceptable, press "E" and the calculated regression constants
(for example K3 to K9 for probe 1, output 1) go into memory and are then used to
calculate brix in the measurement mode. The probe then goes back on line.
If F is pressed, the results are ignored and the probe goes back on line.
The various regression constants are as follows:
Probe 1 output 1 ........... K3 to K9
Probe 1 output 2 ........... K13 to K19
Probe 2 output 1 ........... K23 to K29
Probe 2 output 2 ........... K33 to K39
On completion of calibration, the K constants should be recorded for future
reference. These constants are applicable for the probe calibrated (record probe
serial no.) and may be entered into another microprocessor to exactly duplicate
the calibration. Note that this calibration is only valid for the probe used during the
calibration.
7.3 Calibrator
A calibrator consisting of a number of resistors and capacitors which may be
switched in parallel across the probe electrodes to simulate product impedance is
provided. There are 4 various types of calibrator available, the CA1 which is used
for Raw Sugar massecuites, the CA2 calibrator used for Refinery White boilings,
the CA3 which is used for Syrup and Remelt, and the CAX universal calibrator
which covers all the ranges and can be used on all products.
The CAX calibrator has four binary switches. The first two set the resistance
value, and the second two the capacitance value. Each switch has 15 positions ie
Page 15
0, 1, 2, 3, 4, 5. 6, 7, 8, 9, 10, A, B, C, D, E, and F. As the swich is increased,
resistance or capacitance increases.
The first switch on the left gives high resistance values, going from 680 ohms for
position 1, reducing down to 42,8 ohms for position F. The second switch gives
resistance values from 39 ohms down to 3,6 ohms.
The third switch sets capacitance from 5,6 pF to 48,5 pF. The fourth switch is only
used from position 1 to 7 and provides capacitances from 100 pF to 680 pF.
By switching in various combinations, a large range of different resistance and
capacitance values can be simulated.
7.3.1 Using the calibrator
Once a probe calibration has been carried out using the regression, record the
values of F1 F2 and brix used in the regression. (These can be read in the
calibrate mode by going to " ADD ANALYSES ").
Plot F1 vs F2 on a graph.
This will give roughly a curve but with scatter.
Remove the probe from the pan, fit the calibrator, then go into the calibrate mode.
Go to "ADD TO CALIB." and F1 and F2 will be displayed
Now we require 10 calibration points with F1 and F2 values similar to those
obtained during calibration. These must be scattered around the calibration data
on the graph to cover a similar range of F1 and F2.
Adjust the switch positions to give a calibration point as above.
Note the switch position for this calibration point and mark F1 and F2
corresponding to this calibration point on the graph.
Adjust the switch positions to give another calibration point and repeat until 10
points have been obtained.
Now go back into the measurement mode and set the switch positions to each of
the calibration points chosen.
Write down the switch positions and the brix reading for each of the 10 calibrator
settings, for example using a CA1 calibrator, one of these may be as follows:
1 2 3 4 5 6 7 8 9 10 Brix
0 1 0 1 1 0 1 1 0 0 90.3
Alternatively, if the wide range calibrator type CAX is used, the readings may be
as follows:
1 2 3 4 Brix
5 A 3 1 90.3
Page 16
These 10 settings and corresponding brixes can now be used for recalibration of
any other probe using the regression procedure.
Note that this calibration is only valid in the pan in which original calibration is
carried out. If the probe is used in a different pan having a different physical
configuration (eg calandria closer to probe) then calibration may be different.
The calibration curve may be elevated or suppressed by adjusting parameter K3
(for probe 1 output 1) etc.
8. FAULT DIAGNOSIS
1. No power- Check for power at terminals 1 and 2 and if there is power check
polarity. Wrong polarity will blow a fuse on the main controller
PCB. To replace undo the screws on the main controller panel
and carefully remove the front panel. Check and replace with 0,5
amp fuse if blown.
2. Unit working but no reading or erratic readings from probe.
Go into calibration mode "ADD TO CALIB".
Check if FI and F2 displays a reading. If no response, unplug
the probe cable from the controller and measure volts across the
plug. The voltage should vary from 16 to 25 volts.
If the voltage is present, check the probe and cable.
4. Unit working but keyboard faulty.
Check cable from keyboard to plug in board.
5. Display faulty.
Check cable from display to plug in board.
6. Loss of parameters on power down.
This indicates failure of the RAM battery back up and may occur
after 5 to 10 years of operation. The battery is soldered onto the
mother board and may be checked with a voltmeter. The
voltage should be greater than 3 volts ( new battery 3,6V ) This
may be unsoldered from the top of the board if faulty and
replaced with a new unit by a qualified technician.
Revision 6 3- 7- 03
