Important notice - please read:

This firmware was based on code originally written by Jacques for the Montreal Doppler II.   Many new features have been added and most of the code has been extensively rewritten without Jacques' direct involvemen:  Please read the disclaimer.  For more information, use the contact information at the bottom of this page if you have any questions.
 

Operational differences between the Montreal Doppler II and Doppler III hardware:

The Montreal Doppler II and Doppler III hardware are very similar to each other - but with a few important differences:

• The Doppler II uses 4 pushbuttons to navigate the menus and select various items whereas the Doppler III has a single pushbutton and a potentiometer to do the same functions.
• The Doppler III has a "Roanoake-type" switched capacitor filter for additional filtering.  The Doppler II does not have this filtering - but the addition of such a filter is described on the Switched Capacitor Filter page and is highly recommended!
• The Doppler III has audio level metering taken at a point prior to the narrowband switched capacitor filter.  This metering provides an immediate response to signal level changes and may be used to set the audio level on the unit in a repeatable manner.  The Doppler II has audio metering - but this is derived from the signal processing and may be slow to respond to changes, particularly if the integration level is set fairly high.  The addition of similar circuitry to the Doppler II is highly recommended!
• The Doppler III has a compass rose display driven by a synchronous serial output ("SPI" - type.)  This replaces the on-screen 36-point compass rose that was used on the Doppler II unit and makes for a much easier-to-read intuitive display.  This firmware allows the implementation of this display on a Doppler II unit after appropriate modification.  This firmware supports the use of this same display with the Doppler II.
• The original firmware for the Doppler II and Doppler III have slightly different rotation frequencies:  The Doppler II operates at 500.8 Hz whereas the Doppler III operates at approximately 499.3 Hz.  This alternate firmware operates at 500.8 Hz for both the Doppler II and III, but a version operating at 499.6 Hz is available by special request.

Firmware installation:

There are two versions of this firmware:  One for the Montreal Doppler II unit, and another for the Montreal Doppler III unit.  While it is theoretically possible to have one piece of firmware that will automatically detect which hardware is being used, program space limitations prevent this, so two different versions are available, with the hardware platform being selected at compile time.

Both versions have exactly the same features, with the sole exception of a larger number of choices for integration/average settings being available on the Doppler III:  This is possible only because the menu potentiometer makes it practical to choose amongst a larger number of options whereas on the Doppler II, that number of selections would involve tedious button-pressing to go through all of the possible choices.  Another difference is that on the Doppler II it is possible to perform a complete EEPROM wipe at any time, resetting all parameters to default.

Installation of the new firmware involves only the replacement of the 40-pin processor:  No other changes are made, and one may put the original firmware back at any time.  The only possible caveat to this is that on the Doppler II, if the modifications to add the audio level meter and/or compass rose are made and the old firmware is installed, these pins may be floating at an indeterminate input level and cause the processor to draw a few extra milliamps of current unless they are tied either to ground or +5.

Figure 1-  The screen showing the compile date of the firmware for versions older than 7A.  This may be used to determine which version of firmware is installed.  Starting with version 7A this screen is no longer used as the version number is always displayed on the startup screen.

Modification of the Doppler II hardware:

• One may simply plug the newer processor into the existing socket and expect improved performance and capabilities, making no modifications at all.
• Some of the more useful features do require simple board modifications and minor circuit additions.  Details on modifying the Doppler II to add audio level metering, a drive signal for the switched-capacitor filter, and the compass rose output may be found on the Doppler II Modification page.  Additionally, "Adaptive Filter" control of an outboard 8-capacitor "Roanoke" type filter and a "Stop Rotation" control may be added as well - see modifications noted on this page.
  

• For the most part, no modifications are necessary to use most of the features.
• If you wish to implement the "Adaptive Filter" control 8-capacitor "Roanoake" type filter or the "Stop Rotation" switch, minor circuit additions will be required.  Details of these modifications are found near the end of this page!

Power up:

The bootup cycle will display "Mont Doplr II-CT" or "Mont Doplr III-CT" (depending on hardware) followed by software credits, and then it will display the main screen.  This cycle takes about 2 seconds, allowing power supply voltages and filters to stabilize.

Note on firmware version 7A and newer:  On these newer versions of firmware, the firmware version is displayed on the main startup screen sequence and the compile date is no longer shown.  Those versions using the PIC18F4620 processor will have a "+" symbol after the version number.

Checking compile date (versions prior to 7A only):

On both the Doppler II and Doppler III units, the firmware compile date may be checked as follows :

• On the Doppler II, press and hold the CENTER TWO buttons while turning on the power.
• On the Doppler III, press and hold the button while turning on the power.
• Again, on newer versions, this information is displayed on power-up.
  

Figure 2 - Doppler II ONLY:  This display shows the detected status of the modifications for audio level metering ("LVL Mtr") the SPI output  ("ROSE") for the compass rose and a clock signal for an added switched-capacitor filter ("SCF")

During startup, the date of compilation will be shown briefly:  No saved settings are altered when this done except on the Doppler II when it detects a new modification (see next section.)

Doppler II only:

On the Doppler II, powering up with the center TWO buttons pressed will cause an additional screen to be displayed, (following the date on versions older than 7A) showing the detected status of the audio level metering, the drive for the switched-capacitor filter, and the SPI interface for the compass rose.

If the displayed status is "1" (as shown in Figure 2) that indicates that the firmware has detected that the modifications to add these options may have been completed, and a "0" of the circuit board traces appear to be in their original configuration. All that is done is to check to see if the traces on the circuit board have been cut and it cannot actually tell if the audio level, compass rose or switched-capacitor filter hardware is really present.

EEPROM clear (Doppler II only):  If all four buttons on the the Doppler II are held down during powerup, the EEPROM (containing user settings) is wiped.  This is followed by the display of the compile date (on versions older than 7A) and the display of the audio level metering.  This is NOT available on Doppler III.

Important note (Doppler II only):

• If a modification to use the audio level metering, the switched-capacitor filter, and/or SPI output (for the compass rose) is made, you must force a display of either the software date or do an EEPROM wipe before the modifications are detected!  It is only if the "detected hardware status" screen is displayed that an attempt is made to detect any modifications!

Figure 3 - Main display before any bearing data has been received.  The asterisks (***) will be replaced with bearing information and the signal quality information will appear below the bearing.

Main Display:

Initial display:

After powering up, the main display will appear see Figure 3.  If no bearing data has yet been received, asterisks may be displayed instead of numbers.  This can also occur when exiting the menu system and the display has not been updated (because of quality/audio level gating.)

Note:  If there is audio present on the input when the unit is powered up, random bearing data may be displayed and you may not see the asterisks.  A random bearing may also be displayed on powerup as the circuitry may not have fully stabilized by the time the powerup cycle is completed.

Explaining the display:

The display will simultaneously display the current bearing (on the left) AND the averaged bearing (on the right) and their respective signal qualities in both numerical and graphical format. 

Figure 4 - The arrangement of the main display showing the location of the various data/status indicators.  Note the "audio level" icons surrounding the quality reading of the current ("integrated only") bearing:  This means that the buttons/knob will adjust the integration level.  Note also the diagonal bars on the left and right, which are the compass rose displays for the "current" and "averaged" bearings, respectively. (see text)

Figure 4 shows a typical display:

• Both the "current" and "average" happen to be showing a bearing of "307" degrees and a quality factor of "6".  Often, the two displays will show different readings - particularly if either the transmitter or receiver is moving.
• The diagonal line is the 16-point graphical compass rose, showing the respective bearing, in graphical format, to the nearest 11.25 degrees. Note:  The cardinal points (North, South, East and West) are displayed using "double" lines (see an example of this in the AF GATE description, below.)
• The numerical "Audio Level" reading (top, center) may be configured to show a number from 0 to 99 that could be used to represent an S-meter reading, field strength, or even battery voltage.  If the audio level is too high, this will display "OL" to warn the user of possible audio input overload - even if set to display the S-meter reading.
• The center-bottom number show the current setting of the integration or averaging - see below.
• Surrounding the signal quality (on the left side, in this case) are icons that represent audio level.  As the audio level increases, these icons get taller and wider.  This same location can also indicate whether the "audio gating" or "average clear" (see below) is currently in effect.  Note:  This feature is available on the Doppler II only if the audio level metering modification has been made.

Comment:  On versions 7A and newer, the "Square" brackets ([]) around the quality reading will be replaced with a "greater-than" and "less-than" sign (<>) if  pin 5 of the main processor is disconnected from the +5 volt supply and is available for use with the adaptive filter - read about this under the "Average Clear" description.

The difference between the "current" and "averaged" bearings:

The "current" bearing:

On the left side of the display is the "current" or "instantaneous" reading.  This reading has had applied to it no averaging and in those cases where the bearing is being degraded due to heavy modulation and/or multipath, it will be seen to change seemingly randomly.  It is possible to "smooth" this reading somewhat by applying integration (described below) that accumulates data from several readings.

The "averaged" bearing:

On the right side of the display is the "averaged" bearing.  This bearing is based on the processing of the present and past "current" bearings and applying a sliding average to the result.  Because multiple past bearings are used and the result is "weighted" according to bearing quality, this portion of the display can be significantly "smoothed" and there is the increased possibility that even if individual "current" bearings are inconsistent, that the averaging will help identify trends.

The effects of the "Integration" and "Averaging" settings:

Integration:

Internally, 20 readings are taken each second.  If the integration setting is "1" (e.g. NO integration) then each bearing is displayed as it is received.

Adjusting the integration/averaging on the Doppler III using the knob. As mentioned, when on the main display, the knob adjusts the amount of integration (or averaging - depending on setting) to be used.  There is one potentially confusing aspect of which the user should beware, however: The current setting of the knob is saved when the button is pressed-and-held.  This "saved" setting is restored when the user exits the menu and on powerup. Here's the confusing part:  Because the user uses the knob to adjust various settings within the menus, the knob may not be in the same position that it was before entering the menu system - but the setting of the integration (or average) will still revert to the saved setting - at least until the user changes the knob setting again. What this means is that if you'd set the integration to 16 (a value roughly correlating with mid-rotation) and you entered the menu system, that value would be saved.  To exit the menu system, however, you must select the "exit" icon which involves turning the menu pot all of the way to the end of rotation.  When back in the main display, you'll see that the setting is still 16 - even though the pot may be rotated all of the way to the end, which would mean a far different value.  It will "revert" to the pot's actual setting when the user turns it again. (Did any of that make sense?)

Having a higher integration rate also means that the display update is slower:  If the integration rate is set to "8" then the new readings are displayed only after 8 readings have been collected, resulting in about 2.5 readings per second.

Integration is useful for "smoothing" out noisy signals and/or slowing the display update rate to a more manageable level.  Note that the effects of the integration aren't quite the same as averaging as the raw data is simply accumulated for a longer period and the effects of rapid changes during the integration period - including those that might skew the data - are simply lumped in.

The "Integration" ability in this firmware is intended mostly to alow the slown down the display update and to provide a modicum of "averaging" of noisy data.  Although the unit can be set for a large number of integrations, it is not recommended that a setting of more than 8 be used as this slows the unit's response to signal!  Typically, one would use an integration setting of 1 or 2.

Averaging:

This firmware supports a "sliding average" that takes from 2 to 32  of the most recent "current" readings and averages them together (one of these 2-32 readings being the current reading.)  This has much the same effect as integration, except that it is updated at the same rate as the integrated bearing - but "smoothed."  Because the newest reading is included and the oldest is discarded, increasing the averaging setting does not slow down the update of the display as the integration setting does, but it does slow the response of the display to changes in signals!

Also consider exactly what sort of information comprises the "averaged" bearings:  They take into account not only the phase of the bearing, but the quality as well.  When it comes down to it, an "averaged" bearing is a close approximation to a software simulation of the 8-capacitor "Roanoake" filter.  What this means is that even if you turn the "Damping" potentiometer (assuming that you have added it) to minimum damping (a "fast" response) you will still get much of the same effect of having a switched-capacitor filter.  With the "integration" turned off (set to "1") the effect of 32 points of averaging is less than that of the switched capacitor filter's damping set to "maximum" - but it does a very good job of "smoothing" out noise.

With the averaging being under software control we have additional flexibility when it comes to how to handle the data that we receive.  For example, with the "Average Clear" function you can have a mixture of some of the advantage of a very fast response to received bearings yet some of the "smoothing" and noise-reduction of a higher average setting.  As mentioned elsewhere, one can configure the "Average Clear" to erase the "average buffer" a specified time after the signal being received disappears.  When the signal reappears, the software does not require that, assuming that the "average" setting is 32, that 32 readings arrive before the average is displayed, but rather it will display the average for the number of bearings that have been received thusfar.  So, if you receive a signal that lasts only 200 milliseconds, you can expect to have up to 4 readings averaged together and the result of those averages will be displayed.

One obvious advantage of the averaging is that if the integration rate is set to 1 (e.g. no integration) you still get an update rate of  20 readings per second.  If the averaging is set to 32, you get the same "noise reduction" effect as having an integration setting of 32, but the readings appear continuously, rather than appearing in "chunks" at a rate of just one reading every 1.5 seconds as would be the case were the integration set to 32.

Keep in mind that the "Current" (integrated) bearing is the source of readings for the "Average Bearing."  What this means is that increasing the integration will not only slow down the update rate of the average bearing, but the final result will be that the number of readings taken will be multiplied.

For example:  Suppose that you set the integration rate to 8 (about 2.6 readings/second) and the integration to 32.  In reality, the averaged bearing/quality being displayed is based on the past 256 readings (8 * 32) and the averaged bearing includes bearings that were taken over 12 seconds ago (that is, at 2.5 readings per second and 32 readings being averaged, it takes 12.8 seconds for the "oldest" reading to be discarded.  Another example is that if the integration level is set to 64 and the averaging is set to 32, the displayed average is based on contributions of the most recent 2048 readings.  Note that this also means that the reading will include data received around 100 seconds ago (assuming continuous updating, of course.)

One final comment:

ONLY "good" readings (those above the pre-set quality threshold and those allowed by the AFGATE setting) actually contribute to the averaged reading:  "Bad" readings are completely ignored!

Note:  The newest firmware using the PIC18F4620 device allows up to 64 averages to be taken.

Adjusting the amount of integration or averaging:

It is possible to adjust both the amount of integration and averaging being applied - but not at the same time.  Experience and testing has shown that one would usually set the integration value to one's taste (usually a value of 1 or 2) and then adjust the averaging as the situation demands.

Integration or Averaging is adjusted by using the pushbuttons on the Doppler II or using the knob on the Doppler III - but which parameter (integration or averaging) is adjusted from the main display depends on a setting in the "AFGate Menu" (see below) but this information may be discerned from the main screen.

Figure 5 - This display shows the "audio level" icons around the quality level on the average display.  This indicates that the buttons on the Doppler II (or knob, in the case of the Doppler III) will adjust the amount of averaging being done:  32, in this example.

Taking a look at the display, notice that the quality factor of the averaged reading is surrounded by the audio level icon but the quality factor of the current reading is surrounded by square brackets.  It is the number that has the audio level icons that may be adjusted (using the buttons or knob) while in the main menu.

In the case of the picture, with the audio level icons surrounding the averaged quality, we know that pushing the buttons (or turning the knob, as in the case of the Doppler III) will adjust the averaging AND that the number shown in the center (on the bottom) is the current averaging setting.

If the audio level icons are surrounding the quality factor of the current bearing (as shown in the picture with the labels at the top of the page) then it is the integration that is adjusted by the buttons/knob.

Button operation in the main display:

Doppler II -  The buttons operate as follows:

• Button #1 (typically, the left-most button):
• If the audio level icons surround the quality reading of the current bearing and square brackets surround the quality reading of the averaged bearing, this button resets the integration to 1 (e.g. NO integration.)
• If the icon surrounds the quality reading of the averaged bearing and square brackets surround the quality reading of the current bearing, this button clears/resets the average.
• Button #2:
• This button decreases the amount of averaging or integration, depending on mode.  Note:  When the averaged bearing is changed, the entire history of the averaging buffer is cleared.
• Button #3:
• This button increases the amount of averaging or integration, depending on mode.  Note:  When the averaged bearing is changed, the entire history of the averaging buffer is cleared.
• Button #4 (typically, the right-most button):
• This button causes the calibration menu to be displayed.
• Pressing buttons 2 and 3 (the center 2 buttons) simultaneously:
• A bearing is sent out via the serial port.  Note that this must be configured in the APRS menu - see below.

Doppler III -  The button and knob operate as follows:

• The knob:
• If the icon surrounds the quality reading of the current bearing and square brackets surround the quality of the averaged bearing, the knob adjusts the amount of integration.
• If the icon surrounds the quality reading of the averaged bearing and square brackets surround the quality of the current bearing, the knob adjusts the amount of averaging.  Note:  Whenever the averaging setting is changed, the averaging buffer is cleared.

Figure 6 - This display shows various displays of the "audio level" icons.  From left to right:
1:  Gated audio (GATE level set to 1)  2:  Audio level of 0 (GATE level set to 0)  3:  Audio level of 1  4:  Audio level of 2  5:  Audio level of 3  6:  Audio level of 4  7:  Audio level of 5  8:  Audio level of 6  9:  Audio level of 7  10:  Audio level of 2, but a GATE setting of 3 or higher  11:  The icon displayed on a Doppler II without the added audio-level circuitry.  12:  The indicator (vertical line) that shows that the "average clear" is active.

• The button:
• A momentary press of the button sends a bearing out via the serial port - if appropriately configured in the APRS menu.
• Pressing and holding the button for more than approximately one second will enter the menu system, starting with the calibration menu.

Surrounding one of the quality level displays (the one that may be adjusted by the knob/buttons) is the audio level icon.  As may be seen in Figure 6, as the audio level increases, this icon gets both "taller" and "wider."

The left-most icon in the picture shows the "audio gated" condition by the presence of the horizontal line at the bottom of the icon.  In this example, the GATE level (read about the GATE setting below) is set to 1.  The second icon from the left shows an audio level of zero but without the gating active (e.g. a GATE level setting of 0.)

The audio level icons will display the audio level despite the GATE setting.  The second icon from the right shows an audio level of 2 but with the audio gating active, hence the line at the bottom - this situation could happen with a GATE setting of 3 or higher.

Note:  If you are using an unmodified Doppler II unit (no audio level metering added) neither the audio level icons or the AFGATE feature is available then icon #11 (next to the one on the far right) will be displayed instead.
 

"Why have both the audio level icons AND the audio level readout?" These two readings are from difference sources and can indicate different things: - The audio level icons are derived from the voltage from the diode/capacitor circuit connected to the output of the first filter (before any switched capacitor filter.) - The numerical reading is based on the audio that the main processor gets - audio which, in the case of the Doppler III (or a Doppler II with the added switched capacitor filter) may have passed through the switched capacitor filter. This means that the numerical indicator is less affected by noise and/or modulation on the signal and will probably read lower as the signal being received degrades.  As the signal continues to degrade the switching tone becomes more diluted with noise which means that the audio level icon may still read a high signal but the numerical value drops. Note also that as the integration rate is increased, the update rate for the numerical value slows down, too:  The update rate for the audio level icons is not similarly affected and this is where the "AFGATE" threshold is derived from. (Also, I couldn't think of anything else to put there...)

Also note that the icon at the far right shows the "active" indication of the "Average Clear" function see below.

The Menu system:

The major difference in operation of the software between the Doppler II and Doppler III has to do with the user interface:  The Doppler II has four buttons while the Doppler III has just one button and a potentiometer.

On the Doppler II the menu system may be entered by pressing button #4, while on the Doppler III one presses-and-holds the button for approx. one second.

Selecting menu items:

Doppler II:

• Buttons 1-4 are typically placed under the display and they (roughly) correlate with menu items on the display:  In all cases, pressing button #4 by itself moves to the next menu/display.  Note that some functions on the Doppler II require that more than one button be pressed at the same time.

• The potentiometer moves the blinking cursor to the desired item and pressing the button selects/changes that item.  In all cases, moving the cursor to the far right and then pressing the button moves to the next menu/display.

This menu is used to calibrate the direction shown on the display with respect to the actual bearing.  If you are in a vehicle, this is currently set (often using the "AUTO" function) such that straight ahead is "North" (0 degrees.)  If the unit is used in a fixed location, one would typically calibrate the unit such that the bearings correlate with true-north bearings on a map.
 

Figure 7 - This menu is used to calibrate the direction shown on the display with respect to the actual bearing.  This picture shows the display from the Doppler III, with the cursor (a line) under the arrow on the bottom-right.  (The Doppler II does not have a cursor.)

Provisions are provided to increment the bearing in one or ten degree increments.  The calibration value ranges from 0 through 359, "wrapping around" to 0 at 360.

Doppler II:

• Button #1:  Return to the main display:  This button is typically located to the left of the display, above or below.
• Button #2:  Decreases the calibration direction by one.  This button is typically located toward the middle-left of the display.
• Button #3:  Increases the calibration direction by one.  This button is typically located toward the middle-right of the display.
• Button #4:  Moves to the next menu (e.g. the APRS menu.)  This button is typically located to the right of the display, above or below.
• Buttons 2 and 3 simultaneously:  This is "AUTO" where the most current direction is assumed to be where "zero" degrees ("due North") should be.  This is useful for calibrating "straight ahead" in a vehicle.  (These two buttons would typically be located above or below the word "AUTO.")
• Buttons 1 and 2 simultaneously:  This decreases the calibration direction by 10 degrees.  (These two buttons would typically be located above or below the "10" on the left side of the display.)
• Buttons 3 and 4 simultaneously:  This increases the calibration direction by 10 degrees.  (These two buttons would typically be located above or below the "10" on the right side of the display.)

• The menu potentiometer is used to place the cursor on the items mentioned above, and pressing the button changes/selects that item.
• Note that the arrow on the far left is selected to return to the main display while the arrow on the far right (above the cursor) is used to move to the next menu (e.g. the APRS menu.)

Important note about the AUTO function:

• For software versions dated prior to August 28, 2005 (Version 6H and earlier):  The AUTO function may be used ONLY when a signal is present!  This function takes the CURRENT reading and sets "Due North" (zero degrees) to that bearing.  Please note that if there is NO signal present, the result will be a random bearing being used for the AUTO setting.
• For software versions August 28, 2005 and later (Version 6I and later):  The code has been changed to use the averaged bearing for the AUTO setting.  This allows for averaging of the "calibration" bearing as well as permitting the most recently obtained averaged bearing to be used - even if the signal has disappeared.

Figure 8 - This menu is used to configure and test the antennas.

The antenna menu:

NUMBER - Number of antennas:

This "NUMBER" menu item is used to configure the type of antenna system being used.  This firmware supports the following types of antenna switching:

• 4, 6, or 8 antennas may be switched.
• Positive or negative switching.  If positive switching is selected, the antenna is "selected" of the voltage is high (5 volts.)  If negative switching is selected, the antenna is "selected" if the voltage is at ground.
• NOTE:  If the setting of 4 antennas is selected, differential antenna drive is available.  In this mode, outputs 1-4 have signals with the displayed polarity (+ or -) while outputs 5-8 always have the opposite polarity signal of outputs 1-4.
  

The "ROT" item selects the direction of antenna "rotation" - either clockwise (CW) or counter-clockwise (CCW.)  What your particular antenna requires depends on which order the individual antennas were connected.  If this is backwards, bearings that are left or right of the calibrated direction will be swapped right/left.

TEST - Antenna being tested:

The "TEST" item is used to select one single antenna to be turned on or, if no antenna is to be tested, one would chose NONE.  This facility is useful in testing a DF array to identify a particular antenna and/or troubleshoot it.

NOTE:  The antenna selected with the TEST parameter is always selected when displaying any menu.  When in the menu system, antenna rotation is stopped and the TEST antenna (which can include "NONE") is the one that is activated.  There are several situations were it may be useful to stop rotation:

• Listening for weak signals.  Often, the antenna rotation's switching noise will mask or degrade weak signals.  Stopping it will can help the user determine if the signal is there or not.
• Listen to a signal with low-level modulation.  Sometimes, as in the case of a stuck microphone, one can hear in the background a radio, TV, or some sort of distinctive noise.  Being able to mute the switching tone by stopping rotation can help one hear sounds in the background.
• Preventing the switching tone from being transmitted.  As any user of this type of DF system can attest, the "rotating" antennas impart a tone on any signal being received using nearby antennas.  Likewise, if one transmits from a vehicle using this sort of system, some of the switching tone will be "space-modulated" by the switching antennas (even if they are several feet apart!) on that transmission.  There are situations where one may wish to avoid appearing on the air with this switching tone, as its presence may raise questions with other users and/or alert a savvy jammer that "the game is afoot!"
• Except for testing and troubleshooting, it is recommended that one always selects antenna 1 - which guarantees that when rotation is stopped, your DF receiver will always be connected to a working antenna!
  

• It is always recommended that, if transmitting from the same vehicle as the RDF gear, that the TX and RX antennas be separated as far as possible and that the lowest transmit power - preferably 5 watts or less - be used to prevent damage to the antenna switching system and/or the RDF receiver.
• Other antennas on the vehicle can skew the pattern and reduce accuracy of the bearing and that they should be removed, if possible.  If you must have another antenna, make sure that it is as far away from the DF antenna array as possible (in all cases, more than 1/4 wavelength!)  If you need to have another antenna for monitoring/communicationg with others while using the DF unit, a small magnet-mount antenna on the hood or trunk, mounted as far forward or back as possible is least likely to cause bearing errors.

Figure 9 - This menu is used to calibrate the direction shown on the display with respect to the actual bearing.  This picture shows the display from the Doppler III, with the cursor (a line) under the arrow on the bottom-right.  (The Doppler II does not use a cursor.)

The APRS Menu:

This menu is used to configure several parameters related to the serial port:

BAUD - Baud rate selection:

 The baud rate.  Valid baud rates are 1200 (as of version 7C and later,) 2400, 4800, 9600, 19200, 38400, 57600, and 115200.  Higher baud rates are displayed as 19k2, 38k4, 57k6, and 115k, respectively.

APRS - APRS Serial data mode selection:

The "APRS" mode allows the sending of bearing data and/or GPS data from an attached receiver..  The available selections are as follows:

• NO - The serial port is disabled:  No data will ever be sent - even if the button(s) is/are pressed from the main display.
• D - The current direction only will be output on the serial port according to the TIME setting.
• A - The averaged direction only will be output on the serial port according to the TIME setting.
• G - Only GPS data will transmitted on the serial port. (See below.)
• DG - Both the current direction and GPS data will be transmitted on the serial port according to the TIME setting (other than "ON".)
• AG - Both the averaged direction and GPS data will be transmitted on the serial port according to the TIME setting (other than "ON".).
• Gd - GPS data is sent according to the TIME setting, but the current direction is sent ONLY if the button(s) is/are pressed to send a bearing manually.
• Ga - GPS data is sent according to the TIME setting, but the averaged direction is sent ONLY if the button(s) is/are pressed to send a bearing manually.
• dG - GPS data is sent according to the TIME setting, but current direction will be transmitted continuously (as if in the "ON" setting.)
• aG - GPS data is sent according to the TIME setting, but averaged direction will be transmitted continuously (as if in the "ON" setting.)

The time interval at which the data is to be sent.  Selections are:

• OFF - No automatically-timed bearing/gps is ever sent.  One may manually send bearings using the button(s) from the main display.
• ON - Data is continuously sent.  If bearing data only is to be sent (e.g. "D" or "A" above) then a continuous stream of bearing/quality information is sent.  This rate depends upon the integration setting:  If the integration is 1 (no integration) then 20 readings are sent each second, 10 readings/second if integration is set to 2 (e.g. 20 / 2) and so on.
• On versions 6H and earlier (dated prior to  August 28, 2005):  If this (the "ON" setting) is selected in conjunction with  the "DG", "AG", "Gd",  "Ga", "dG" or "aG" the TIME settings the TIME setting will be forced to "05s" when this menu is exited to prevent timing conflicts.
• On versions 6I and later (dated August 28, 2005 and later): On these later versions, the forcing to "05s" for the "DG", "AG", "Gd",  "Ga", "dG" or "aG" settings has been removed.  PLEASE NOTE that there is a chance that if you select "ON" with these settings - particularly at a baud rate under 19200 - the multiplexing may not work properly.  This often happens in conjunction with the GPS multiplexing where, at low baud rates, the sending of the NMEA sentances may occur too closely together for the software to detect the pauses between them, preventing the bearing from ever being sent.
• On versions 7O and newer:  A time setting of "01s" (one second) has been added.  Important note:  This setting is intentended only for those situations where GPS data is NOT being sent on the same serial port as the bearings, or is not being sent at all!  If a setting of 1 second is selected, it should be noted that this is more frequent than data is typically sent by most GPS receivers and the unit may not reliably send bearing data either at the specified interval, or in response to the press of a button.
  
• 05s to 16m - Various timing intervals (5 seconds to 16 minutes) at which the data is be sent may be selected.

Notes:

• Both the GPS receiver AND the DF unit must be set to the same baud rate!
• GPS data (typically, NMEA sentences from a receiver) may be input via the receive data pin on the serial port.  If "G" is selected, this data is "gated" (passed through) according to the timed setting.  Note that when sending  both bearings and GPS data, the firmware waits for the pause after the end of the current string of NMEA data before allowing the GPS receiver's data to pass through.
• WARNING ABOUT OPERATION AT LOWER BAUD RATES:  At lower baud rates (1200, 2400 and 4800) some NMEA sentences may take longer to transmit than the time allotted, resulting in the firmware NEVER sending a bearing - or frequently "skipping" a bearing to be sent.  For example, if the setting is for 5 seconds at 1200, 2400 or 4800 baud, it may take more than 5 seconds for all for the GPS data to be sent (depending on the NMEA transmit mode of the receiver, exactly how much data the receiver sends and how often it sends it out.)  In this case, it is possible that NO bearing data will ever appear:  The firmware detects the pauses between NMEA transmissions from the GPS receiver and if there is too much data, the GPS receiver may never pause long enough to send a bearing.
• For this reason, at least 9600 baud is recommended when sending both DF bearings and GPS data and it is NOT RECOMMENDED that NMEA data be transmitted at all at 1200 or 2400 baud!
• Please be aware that at 1200 baud, even when sending only bearing data, a bearing may occasionally be "skipped" owing the inherent limitations of how much data may be sent at 1200 baud.
  
• It is not recommended that the "ON" mode be used while passing GPS data and  transmitting bearing data.  If you wish to do this, look at the results using a dumb terminal program before proceeding further to see if you get the desired result.  If this combination is selected, GPS and/or Bearing data may appear erratically - or not at all.
• When a timing interval is used (e.g. 5 seconds to 16 minutes) only one DF bearing and/or packet of GPS data is sent at each interval when in the D, A, G, DG, or AG modes.  Additionally, a DF bearing is sent only if there is a bearing of adequate quality (meeting the threshold) available!
• The "Gd" and "Ga" settings send GPS data but do NOT send bearing data unless the user does so manually (e.g. press the center 2 buttons on the Doppler II, or the button on the Doppler III is pressed momentarily) with current data being sent if "Gd" is selected and averaged data sent if "Ga" is selected.  This facility exist to allow GPS data to be automatically sent out the serial port, but allow the user to manually send bearing data.
• This mode is very useful when using a program such as GPSS because this program requires a constant stream of GPS NMEA data to know the receiver's location as well as the heading of the vehicle.
• If tracking a signal it may be desirable to allow the user to send a bearing to GPSS only if he/she deems that the bearing is of good quality and/or, in the case of multiple signals, if the desired signal is present.  In other words, it allows the user to selectively (and manually) "DF" one particular transmitter.
• If the "dG" or "aG" mode is selected, bearing data is sent continuously - regardless of the timing interval setting - at a rate related to the integration setting.  If the integration is set to 1 (e.g. no integration) then over 20 readings will be sent per second. If an integration rate of 2 is selected, the rate would be half that, etc.  The GPS data is sent at the selected TIME interval.  Note that the bearing data is halted briefly while the GPS data is detected/sent.
• If it is desired that a bearing be manually sent from the main display (by pressing the two center buttons on the Doppler II or pressing the button on the Doppler III) then the APRS setting must be something other than NO or G.  Note that the button pressing does not allow or cause any GPS data to be transmitted - use the timer setting for that.
• You can still manually send the bearing to the computer (by pressing the button) even if no signal is currently being received.  There are a few things to remember:
• If you have configured to send the "current" bearing manually (e.g. the "Gd" mode) what will be sent is the current calculated bearing.  If you are receiving NO signal or just noise, the bearing sent will likely be random!  The "quality" setting does not affect the "current" bearing being manually sent - that is, a bearing will be send no matter what it's quality might be!
• If you have configured to send the "averaged" bearing manually (e.g. the "Ga" mode) the most recent averaged bearing will be sent.  Because the average bearing is calculated only on those bearings above the set quality threshold and are not being gated, when the signal disappears, the most recent averaged bearing is retained.
• This feature can be handy if the signal you are tracking disappeared before you had a chance to hit the button - see sidebar.
• If you have configured to send any bearing (the D, A, DG, AG, Gd, Ga, dG, or aG) pressing the button at any time will cause a bearing to be sent immediately (or, at least as soon as GPS data has been sent) and doing so does not affect the timing of the automatically-sent bearing.  NOTE:  If the setting is for a "current" bearing (the "D" or "d" settings) the bearing sent may be garbage if there is no signal present at that instant. IF, however, the setting is for an "averaged" bearing (the "A" or "a" settings) then the most recent averaged bearing will be sent, as described above.  Because the "average bearing" is calculated only from "good quality" bearings, it isn't likely to send a "garbage" bearing if the "quality" and "gate" settings have been set up properly.
• If you press the button to cause a bearing to be sent (or the 2 center buttons on the Doppler 2) it may take a few seconds before the bearing is actually sent if the unit is configured for "Timed" sending.  For example, in the Ga mode (where GPS data is sent every 5 seconds and the averaged bearing is send ONLY in response to a buttonpress) the firmware may be waiting for GPS data to become available:  As soon as the GPS data is sent (or after the unit has timed out while waiting for GPS data that it hasn't found) the bearing will be sent - something that could take a few seconds.  Also note that a new "buttonpress" is ignored until after the bearing to be sent in response to the most recent buttonpress has been sent.
  

Figure 10 - This menu is used to select which "Radio Settings" are to be used as well as the "minimum" quality factor that causes readings to be updated, and whether or not the "audio level" numerical readout (the center, top on the main display) is to show an "S-meter" reading instead.

The "Radio" menu:

RAD - Radio preset:

This unit has the capability of storing settings for up to five different radios.  The settings "remembered" for each radio are:

• The calibration offset.  Every radio has slightly different audio characteristics and thus, the calibration may vary.
• The number of antennas being used (4, 6, or 8)
• The antenna switching polarity (+ or -)
• The rotation direction - Clockwise (CW) or CounterClockwise (CCW).

DISPL QF - Minimum quality factor to be displayed:

This item sets the minimum quality at which the display of the current reading will be updated.  ONLY those current readings with a quality equal to or higher than this setting will contribute to the average.

• Version 6H and earlier (dated before August 28, 2005):  This threshold was applied ONLY to the current bearing (e.g. the one on the left side.)  On the right side, you could see bearings with qualities worse than the threshold owing to the fact that the averaged quality is calculated by averaging each quality/bearing as if it were a vector:  If, for example, there were 4 bearings, one each to the North, South, East and West, the overall quality would be zero because there was no clear trend.
• Version 6J and later (dated September 9, 2005 and later):  An additional selection setting is possible.  Advancing the "DISPL QF" above 8 causes the "DISPL" to change to "BOTH" and the quality setting starts a 1 and goes through 8:  The next step after this is, again "DISPL" at a quality setting of 0 (zero, or off.)  The "BOTH" setting causes both the current and averaged bearings to be subject to the quality factor threshold setting defined.

SMTR - S-meter display enable:

Firmware versions older than 7A:

• OFF - The display shows the "Processed" audio level reading.
• ON - A number 0-99 is displayed that is proportional to the voltage on the "S-Meter" input instead of the "processed" audio level reading.
  

• OFF - The display shows the "Processed" audio level reading..
• LCD - S-meter displayed only on the LCD
• DIR - The display shows the "Processed" audio level reading, but S-meter info is sent with the direction on the serial port (ONLY if the serial port has been configured to send a bearing) in the format of "%bbb/q/ss" where "bbb" is the 3-digit bearing, "q" is the quality from 0-8, and "ss" is the 2-digit signal level from 00-99.  Note:  There is no overrage indication provided on the serial port, so assume that a reading of "99" is an overload condition.
  
• BTH - S-meter is displayed on the LCD and sent on the serial port.

This voltage, if used, would typically come from the AGC line of the receiver, but it could also come from a field-strength meter or even be used to measure battery voltage.  With  the "maximum" setting of the "S-meter" potentiometer, a reading of 99 corresponds to approximately 1.93 volts:  Values higher than this will cause ">>" to be displayed on the LCD.

NOTE:  Regardless of the setting of this parameter, an "OL" indicator will appear on the LCD if the audio level is too high.

 
 

Figure 11 - This menu is used to select the "Gating" level, the source of the bearings for the compass rose display, and whether the integration or average is adjusted from the main display.
In this example, we see that "AVG 32" is displayed, indicating that the averaging setting is controlled from this menu and is currently set at 32.  This also means that it is the integration that is controlled from the main display.
Note also that the "Average Clear" setting is "NO" (turned off.)

The "GATE" menu:

This menu is an additional menu provided by this "alternate" firmware and it is used to adjust a number of parameters:

GATE - Audio gating enable:

With this setting one may halt readings (in the same way that the quality factor threshold does) when the audio level is too low.  An example of this would be to have the readings start/stop when the audio is squelched.

If the GATE threshold level is set to 0 (zero) this feature is disabled, while a setting of 7 (maximum) requires quite a bit of audio to be present.  For most applications, a setting of 1 works well - although some radios tend to have a bit of "squelch leakage" and may require a higher level.

A higher setting may also be used to better-reject signals with multipath distortion:  As signals degrade, the amplitude of the 500 Hz switching tone often decreases, replaced either with noise or with increased amplitude of the harmonics of the 500 Hz switching tone.  Because the harmonics cannot make it through the audio filtering, the "audio level" icon's reading will read even lower.  When it falls below the preset threshold, readings are ignored - and those readings are presumably poor-quality ones, anyway.

The main display also shows whether or not the audio gating is active, too:  If the audio input is below the set threshold, a "audio level" icon will have a horizontal line present at the bottom.  Shown in Figure 12 is an example of the audio gating being active - but the "Average Clear" not yet having cleared the average (if it is enabled):  In this state, all updates of bearing (both graphical and numerical, current and average) are frozen.

Figure 12 - An example of the "GATE" being active.
Note the line on either side of the Averaged quality (e.g. the "3" on the right side of the screen.)  If this is displayed, the reading update is halted due to the low audio level.
Also note that "double line" on the current direction graphical compass rose (the one on the left side):  This type of double line is shown when pointing North, South, East or West.

NOTE:  On the Doppler II only, if the audio metering is not present, "N/A" will appear below GATE to show that this feature is not available.  If you have made the modifications (see below) verify that you have done them properly and that the button(s) was/were held down during powerup to force detection.  This feature is always present on the Doppler III.

Versions 7O and newer using the PIC18F devices:

For these versions, there are the normal GATE threshold settings of 0-7, but when the button is pressed after a setting of 7, a plus sign (+) will appear, along with settings 1-7 (e.g. "2+).  When the plus sign (+) is present, only the first bad bearing (which is displayed in red) after the audio has dropped below the GATE threshold will be sent to the compass rose.

ROSE - Source of bearing data for the compass rose:

This setting determines exactly what is to be sent to the Compass Rose display as follows:

• DIR - The "current" (integrated) direction (that displayed on the left side of the LCD) is sent to the compass rose.  The current bearing is shown in green and an average, calculated by the compass rose display itself,  is shown in yellow, if enabled.
  
• AVG - The "averaged" direction (that displayed on the right side of the LCD) is sent to the compass rose.  The averaged bearing is shown in green and the yellow LED shows an "Average of the average" if averaging is enabled - see the note below explaining this.
  
• BTH - In this mode (present only on firmware version 7A and newer) the "current" (integrated) direction (on the left side of the LCD) is shown in green while the "averaged" direction (on the right side of the LCD) is shown in yellow.  In this mode, the fixed 16-point average built into the compass rose display itself is automatically disabled.  Note that this function is only supported on compass rose firmware version 2E or newer and is not compatible with older compass rose firmware.
  

If the current bearing is to be displayed, DIR will appear whereas if the average bearing is to be displayed on the compass rose, AVG will be displayed.

NOTE:  On the Doppler II only, if the SPI modification is not present (see below) "N/A" will appear below ROSE to show that this feature is not available.

AVG or INT - Setting for Averaging or Integration level:
 

Figure 13 - In this example, we see that "INT 01" is displayed, indicating that the integration setting is controlled from this menu and is currently set at 1 (no integration.)  This also means that it is the averaging that is controlled from the main display.  Also, the "Average Clear" setting is set to clear the average 5 seconds after the audio is gated or the quality falls below the preset threshold.

The menu item on 2nd from the far right of the display allows selection of the amount of averaging or integration and which one is adjusted by the buttons/knob from the main display.  This selection works as follows:

• If this menu item shows INT the number below it shows the amount of integration that will be used.  Pressing the button will increase the amount of integration - and when it reaches the maximum value (64) it will switch to displaying AVG instead.
• If this menu item shows AVG the number below it shows the amount of averaging that will be used.  Pressing the button will increase the amount of averaging - and when it reaches the maximum value (32 or 64, depending on the firmware version) it will switch to displaying INT instead.

Conversely, if AVG is displayed on this menu, the amount of averaging is selected from this menu and it is integration that is controlled from the main display.

AC - Average Clear:

On the far right edge of the display (see Figure 13) is the "Average Clear" (AC) parameter.  This is a very useful feature in that it can be used to automatically "throw away" the averaged bearings from previous transmissions that might "contaminate" new bearings and is, in fact, a form of "Adaptive Filtering."  This parameter has these available settings:

• NO - This feature is turned off:  No automatic clearing of the average occurs.
• 0 - The average is cleared the instant that the audio is gated (e.g. after "zero" seconds) - that is, audio falls below the "GATE" setting.
  
• 1/4, 1/2, 1, 5, or 10 - The average is cleared 0.25, 0.5, 1, 5, or 10 seconds after the audio is gated.  If the audio returns and is above the audio threshold level before this amount of time, the timer is reset and the average is not cleared.  Once the timer has expired, the "average clear" icon will appear see below.

This feature is used to automatically purge the averaging buffer after a transmission stops.  If multiple transmissions occur in sequence, the average buffer may contain bearings from a previous transmission from a different direction, causing subsequent averaged readings to be "contaminated" with the bearings from the previous transmission(s).

Notes:

• Clearing the buffer does not change the display:  The last reading stays on the screen until a signal of sufficient quality/audio level reappears to allow the user to see the "last reading" in case he/she missed it.
• Remember that the most recent averaged bearing may still be sent manually on the serial port (if the "APRS" settings are appropriate) even after the signal has disappeared - as long as a new signal hasn't appeared to replace it.
• Warning:  If the Integration setting is such that it takes longer to update the "current" bearing than the "Average Clear" setting, this function may not work properly.  For example:  Assume that AC is set for 1/2 (0.5 seconds.)  If the integration is set higher to 11 or higher, it will take longer to integrate a new reading (11/10.833 readings sec. = one reading every 0.528 seconds) than the AC period is set for.  The reason for this is that the AC period may "time out" between readings, causing the Average Clear function to seem to activate randomly.  When using the Average Clear function, an integration setting from 1 to 4 is recommended, with smoothing being accomplished by appropriate setting of the Averaging value.

Hint:  If the user has included the "damping" control on the switched capacitor filter (see here for the Doppler II, or here for the Doppler III) very brief transmissions may be detected as follows:
 

Figure 14 - This "vertical bar" icon is displayed when the average has been cleared by the "Average Clear" function.

• Set the damping control (the added potentiometer) to minimum.  This will assure a fast response to the signal.
• Set the integration to minimum (1).  This will process readings quickly.
• Set the averaging to maximum (32).  Because the average is cleared automatically, it is "built up" one reading at a time when the new signal appears.  After 32 readings (about 1.4 seconds, if the integration is set to 1) the averaged bearing consists of 32 readings, averaged together.
• Note:  If you add the "Adaptive Audio Filtering" (see below) then the "damping" setting is automatically adjusted by software when the "average clear" is active.
  

If the "Average Clear" function is enabled, the "Audio Level" icons will display a vertical line (see Figure 14) when the time set for the average clear has expired and the contents of the averaging buffer have been cleared.

Note that of this indicator can only appear once the audio has been gated (according to the AFGATE setting.)  It does not appear if the average is cleared any other way.

Comments:

• If you turn the "Average Clear" function off (change it to the "NO" setting) the "average clear" indicator bar may remain on the screen until the next time that audio appears of sufficient level to exceed the gating threshold, or until the unit is power-cycled.
• This function is available on the Doppler II only if the audio metering circuitry has been added.

Figure 15:
Diagrams showing the components added for the "Adaptive Filter" modifications - see text. 
Note that the TOP diagram assumes that R55 has been replaced with a 1 Meg front-panel potentiometer for "Q" adjustment.
The pin numbers in the above diagrams apply to the Doppler III ONLY.

• The analog "adaptive" audio filtering applies to the 8-capacitor filter built into the Doppler III unit.  Its function is to allow the fast detection of even very brief signals.  Note that if you have an older Doppler II unit and you have added to it a switched-capacitor filter (see the paragraph below) this firmware adds the feature to that unit as well.
• Even if you do not modify the 8-capacitor, switched-capacitor filter on your Doppler III (or, if you have the older Doppler II without a switched-capacitor filter) the "weighted averaging" built into the firmware - which is, in effect, a software-based simulation of the filter - can be operated in an adaptive manner (e.g. the "average clear" function mentioned above.)  In the case of the Doppler III without the modifications described in figure 15, the fast-response time will be limited by "default" parameters of its built-in filter - a parameter set by the value of R55.
  

Suppose that you are trying to locate a transmitter that only appears briefly - but there are several other transmissions occuring in sequence on the same frequency - as might be the case if someone were interfering with an ongoing QSO.  If the "damping" control is set to a fairly high value (as it might be if some of the signals are weak - or if you don't have a damping control) the Roanoake filter will still contain a "memory" of the last signal's bearing for a second or two after the signal disappears.  If the new signal appears before the Roanoake filter has "lost" its memory (e.g. the capacitors have discharged) it will "contaminate" the new bearing with some information from the previous bearing until all of the old bearing's signal has been "flushed out" - a process that may take a second or two!  If the new transmission is quite short, it may disappear before all remnants of the old transmission have been cleared from the Roanoake filter, resulting in a useless bearing.

• During antenna rotation, receiver sensitivity is reduced due to noise caused by the PIN diode switching.  Stopping rotation, while causing loss of bearing information, can allow a weaker signal to be detected.
• Antenna rotation produces a tone in the receiver - which is what is used to determine the bearing of the signal.  This tone can mask what is being modulated on the carrier being received, however.  An example of this would be a "stuck microphone" in which background noises (a TV, radio, or people speaking) may give additional clues as to the source of the signal and whether it is accidental or intentional.  Note:  The Comb Filter can mitigate this to some extent as well.
• Because of "Space Modulation" the antenna rotation can also cause the switching tone to appear on signals transmitted from nearby antennas as well - even if a different band is used or if the antennas are widely separated on the vehicle or dwelling.  In extreme cases, this modulation can hinder intelligibility of this transmitted signal, but it is more likely to just be annoying.  Another potential problem with having a tone imparted on the transmitted signal is that it may arouse the suspicions of a jammer if he/she hears a peculiar-sounding signal on the air, become wary, and stop transmitting before being located:  While cessation of jamming isn't necessarily an undesired effect, it is usually preferred that the identity of the jammer be known in order to provide future disincentive toward further such activities!

Figure 16:  This circuit could be used to detect RF from the transmitter to automatically stop antenna rotation.  The values of C2 and C3 could be from 0.01uF to 0.1uF, D1 and D2 are practically any silicon diode, like a 1N914, and Q1 could be about any small-signal silicon transistor, like a 2N3904 and resistors R2 and R3 could be 22k-100k and R4 would be 470 ohms to 1k.  The values of R1 and C1 would depend on the amount of RF available for sensing - see text.

• Place the "other" antenna as far away as possible from the DF array.  If you had to remove your "normal" antenna to mount the DF array, there's a good chance that you are using a magnetic-mount antenna instead.  Placing this antenna on the fender, the hood (or "bonnet") or the trunk (the "boot") may be the best bet, as it locates this antenna as far away - and below the DF array - as possible.
• Avoid large, high-gain antennas.  A lower-profile antenna array may have less of an effect on the pattern than a large antenna that sticks up into the air, even with or above the DF antenna.
• If the antenna seems to cause some pattern distortion, vary the distance between the antenna array and the transmit antenna.  This may only be practical if one is using a magnetic-mount antenna.  Another way to minimize the effects of antenna pattern distortion is to place the transmit antenna in a location that is symmetrical to the DF antenna array - that is, equidistant to several antennas, or inline with the array.  In this case, even though pattern distortion may be present, its effect may be less objectionable if it is symmetrical rather than lopsided.
• Do the transmitting on a higher band than the DF activity.  An example of this would be to use 70cm for coordination if you are looking for a 2 meter signal.  It is more likely that a 70cm-only antenna will not cause significant pattern distortion as it isn't as likely to be resonant on 2 meters (unless, of course, it's a dual-band antenna) and a 70cm quarter-wave antenna is quite short.
• Before you start moving antennas about, it would be a good idea to do some testing with and without the antenna just to see if there will be a problem - and if there is, how much of a problem it might be.  Who knows:  You might get lucky and find that the pattern distortion is acceptable!
  

There are a few things that the user should know about how this software operates.

• When changing the number of averages from the main display, the averaging buffer is cleared/reset.  If, for example, the number of averages is set to 32, the next reading after the averaging buffer is cleared is based on this single reading.  After the next reading, the average is based on two readings, and so on, up to the number of bearings set by the averaging setting.
• The averaged direction is based on both the direction and quality of the bearings on which it was based.  What this means is that good quality bearings contribute more toward the direction of the averaged bearing the poor quality bearings do.  For example, if you had one bearing toward the West that had a quality factor if 8 and 3 toward the East that had a quality factor of 1, the resulting bearing would be toward the West.
• The averaged quality is based on both the direction and quality of the bearings on which it was based.  For example, if you had one bearing of quality 8 in each of the directions North, South, East and West, the resulting averaged bearing's direction would have a quality of zero as there was no clear "winner."  Also note that, in this case, the direction of that bearing would be indeterminate since the "average" of all for directions is no direction at all.
• NOTE:  One may select whether the bearings sent to the compass rose are the integrated bearings or the averaged bearings. It is important to note that selecting these two sources causes the compass rose to behave differently, however:
• When integrated bearings are being sent to the compass rose, good bearings will display as green while bad bearings will display as red.  The threshold between "good" and "bad" is determined by the "QF" setting.  What this means is that even when there is no signal, or the signal is bad, you will still get direction indications.
• Averaged bearings, on the other hand, consist only of "good" integrated bearings.  For this reason, when the "good" integrated bearings stop, the average output stops being updated as well.  For this reason the compass rose will stop being updated when the input signal is of poor quality and/or if the signal goes away.  Note, however, that several "good" bearings together can comprise one "bad" averaged bearing:  An example of this would be if you got four "good" bearings - one each from the North, South, East and West.  Clearly, the average of these bearings is no direction at all and this result is "bad."
• When the integration is set to 1 (e.g. NO integration) only half of the main display is updated at the time of each reading (that is, the current bearing is updated one time, and the averaged bearing is updated the next time.)  What this means is that the display's readings are completely updated at just over 10 times per second.  This was done because it was found that updating the display 20 times per second made it flicker badly enough that it became difficult to read.  Note that this does not affect the update rate of the Compass Rose output (and the serial output, if so configured) as they are still updated at the full 20+ readings per second rate (if the integration is set to 1, that is...)
• In many situations, the user may find that it is best to operate with the integration set to 1 or 2 and have the averaging be adjusted from the main display.  This allows for the quickest update/acquisition of a signal and the averaging tends to offer good reduction of noisy or random variations in the bearing.
• On the Doppler III firmware, the use of the potentiometer allows the easy selection of more items.  For this reason, from the main display, a wider selection of integration (an integration level of up to 96) or averaging settings is available.  As mentioned before, adding this number of selections to the Doppler II would require tedious button-pushing on the part of the user to go through all possible menu selections.
• If you are trying to locate one particular signal, it is useful to prevent the "averaging buffer" from containing data from a previous transmission - possibly from another user.  To facilitate this, the "Average Clear" function (configured in the AFGATE menu) will automatically flush the buffer after transmissions.  Note:  Only the Doppler II version has provisions to manually clear the buffer as there simply aren't enough buttons on the Doppler III to do this.
• The original firmware for the Doppler II and Doppler III have slightly different rotation frequencies:  The Doppler II operates at 500.8 Hz whereas the Doppler III operates at approximately 499.3 Hz.  This alternate firmware operates at 500.8 Hz for both the Doppler II and III.  This fact is important if it is to be used with the comb filter.

Modification information for the Montreal Doppler II hardware:

Addition of the audio level detector circuitry to the Doppler II unit is highly recommended.  For more information on these modifications, go to the Doppler II Modification page.

Suggested initial settings:

The following settings are recommended the first time you begin using the Doppler unit.

The Calibration menu:

This menu is used for calibrating the unit's bearings. The Antenna settings (below) must be properly configured before an antenna can be calibrated.  The "Auto" selection calibrates the current bearing as being "zero" degrees (straight ahead.)

The Antenna menu:

- NUMBER:  Set this to the number of antennas that you use.  The polarity (+ or -) will depend on how your drive circuit works:  If set to +, the antenna is selected when +5 volts appears on the doppler unit's antenna drive terminal and a setting of - means that the output is at ground (0 volts.)

- ROT:  This is whether your antenna system should be "rotated" clockwise or counter-clockwise.  This setting is based on the order that the cables are connected to the antenna and/or antenna switch.  If set incorrectly, "left" will be "right" and vice-versa.

- TEST:  This is the antenna selected when the menu system is being displayed and the antenna rotation is stopped.  This value defaults to 1 and would normally be changed only when testing the antennas.

The APRS menu:

- BAUD:  If you are connecting this unit to a computer, this is the baud rate at which the DF bearing will be sent AND it is the baud rate to which the GPS must also be set.  If BOTH GPS and a bearing data is to be sent, it is strongly recommended that a minimum baud rate of 9600 be selected!

- APRS:  This determines which data should be sent and how it is to be handled - see the text, above.  If you have a GPS connected, I would recommend the "Ga" setting (with a TIME setting of 5 seconds) with the bearing being sent manually, by the user's pushing the button.

- TIME:  This determines the timing related to how the data should be sent - see the text, above.

The Radio menu:

- RAD:  There are five presets available for storing configuation such as bearing calibration, antenna number and polarity, and direction.  This is useful if the same unit is used with several different radios and/or antenna configurations.

- DISPL QF:  This is the setting is the minimum quality that should be considered as a valid reading.

A recommended starting value is 3.

- SMTR:  Please refer to the section above about the S-Meter.  Depending on the version of firmware, this can be set to display a number related to the voltage on the "S-meter" input terminal, send this value via the serial port, or both.

The Gate menu:

- GATE:  This is the minimum threshold at which the audio input level will trigger calculation of a bearing.

A recommended starting value is 1.  If your radio is "hissy" or has a bit of audio leakage even with the squelch closed, you may want to set this slightly higher - but you will probably never use a setting higher than 3 or 4.  Note that higher threshold levels may also be somewhat useful in weeding out severe multipath as poor-quality bearings are often accompanied by a drop in the level of the 500 Hz  switching tone due to increased energy in its harmonics:  Because these harmonics can't make it through the bandpass filter, the detected level will also drop.  The key here is to experiment and gain experience!

- ROSE:  This determines whether the bearing displayed on the compass rose is the "current" bearing (from the left side of the LCD) or the "averaged" bearing (from the right side of the LCD.) Once you get used to the system, you will probably set this to "AVG" and leave it there.

- INT or AVG:  If set to "INT" the number below it indicates the current integration setting (and indicates that the main display will be showing the average setting) and if set to "AVG" the number indicates the current average setting (and indicates that the main display will be showing the integration setting.)

For typical use, the setting will be "INT" and 01 or 02.

- AC:  This is the "Average Clear" setting.

A typical value is "1/2" where the average is reset 0.5 seconds after the signal disappears.

The main display:

- I recommend setting the "Average" to 32 (with the INT setting at 01 or 02.)  Because both the "current" (un-averaged) and the averaged bearings are displayed simultaneously, there is no reason not to provide maximum filtering in most cases.

Setting the audio level:

Second to calibrating the direction, setting the proper audio level is very important - and it should be done BEFORE calibrating the direction as excess audio input can skew the bearing.

Here is the recommended procedure for setting the audio level:

• Tune in a clean, full-quieting, unmodulated signal from a transmitter.
• Adjust the audio level up to where the quality reading just reaches 8.

• If you send the "current" bearing to the compass rose (the "INT" setting on the "ROSE" parameter) the averaged displayed on the compass rose in yellow is not the averaged bearing that you see on the LCD, but rather an averaged bearing calculated by the compass rose display itself based on the bearings that it receives.
• If you send the "averaged" bearing to the compass rose (the "AVG" setting on the "ROSE" parameter) the bearing displayed in green is the averaged bearing as shown on the LCD (and the "current" integrated-only bearing is not displayed on the LED compass rose at all) and that the "averaged" display on the LED compass rose (in yellow) is really an "average of the average."

Several different firmware dates, each newer fixing minor bugs or adding features as follows.  Skipped version numbers were never released:

• 20050323 (March 23, 2005, Version "6A"):
• First version released
• 20050619 (June 23, 2005, version "6C") - Items fixed since "6A":
• Added indicator for "Average Clear" condition (e.g. the vertical bar)
• Fixed bug in timing for "Average Clear" function where time was really 0 seconds no matter the setting.
• Fixed bug where the "Current" reading (on the left side of the display) wasn't being properly validated according to the quality threshold setting.
• 20050630 (June 30, 2005, version "6D"):
• Doppler II only:  Disabled "Average Clear" function if no audio metering present.
• 20050703 (July 3, 2005, version "6F") - Items fixed since "6D":
• Will now overlay "audio level icon" with the "Average Cleared" symbol (the vertical line.)  Prior to this, the audio level icon did not indicate while the "Average Clear" indicator was showing.
• Doppler II only:  Slightly changed scaling of numerical audio level metering to read 12.5% higher.
• 20050828 (August 28, 2005, version "6H") - Items fixed/changed since "6F":
• Doppler II only:  Increased wait times in LCD update.  Some LCD modules apparently take longer to process commands than others.  Since the Doppler II cannot poll the LCD module to see if it is "busy" it must wait for a time at which it is guaranteed to become "un-busy."  The symptom of this problem was that the display would become corrupted randomly.
• Doppler II only:  In the TEST menu, the antenna selection of "NONE" could not be saved when the menu was exited.
• In the CALIBRATION menu, the AUTO function had previously used the "current" direction.  This meant that the AUTO function worked properly ONLY while the signal being used for calibration was present. The firmware now uses the "averaged" direction:  This allows a more stable, filtered reading to be used, and the AUTO function may be performed even after the signal has disappeared if the quality/audio level thresholds have been set up appropriately.
• Ver. 7A  (November 28, 2005) - Items fixed/canged since "6H":
• Under the RADIO menu, a BOTH selection has been added to the "Quality threshold" selection.  The "DISPL QF" parameter can be set from 0 through 8, but the next selection is "BOTH QF" with a setting of 1 through 8 - and then back to "DISPL QF" with a setting of 0 (off.)  The "BOTH" mode causes both the current and averaged bearing display to be subject to the quality threshold.  Note that the "BOTH" setting does not apply to either the serial port or the LED compass rose display.
• Under the APRS menu, there is no longer the restriction against setting the TIME to ON when certain configurations were selected.  Please note that a setting of ON along with some GPS multiplexing may fail to work properly, particularly at baud rates below 19200.  If this happens, either the GPS or bearing (depending on the setting) will never be sent, so it is best to try it at a setting of "05s" first and then try it with the "ON" setting to see if it works.
  
• Added display of version number on startup splash screen (e.g. "7A")
  
• Fixed a minor bug that may occasionally have caused a bearing to be improperly calculated.  This was discovered when testing firmware for the Doppler I - but I had never seen it happen on the Doppler II/III.
• Made provisions to allow sending of both the "current" (integrated-only) and the averaged bearing to the compass rose.  When the "BTH" (for "both") setting under the "Rose" parameter is selected, the LED compass rose will now show the average direction that appears on the LCD (which is based on quality of each of the readings that constitute it) rather than an fixed 16-point average calculated within the compass rose display itself.   NOTE:  The use of this added mode is not fully compatible with the older compass rose display code and requires newer compass rose firmware.
  
• "S-meter" data (or, at least the number shown on when the "S-meter" function is enabled) can now be sent via the serial port (if it is enabled) in the format:  %bbb/q/ss where "bbb" is the 3-digit bearing, "q" is the quality, and "ss" is the S-meter reading.  It should be possible to integrate the "S-meter" reading into an existing receiver or field-strength meter and log received signal strength in addition to bearing and quality.  Under the S-meter menu setting one may select:
• Off - No S-meter function at all:  A number showing the audio level is displayed.
• LCD - S-meter displayed only on the LCD
• DIR - S-meter not displayed on the LCD, but is sent with the direction on the serial port
• BTH - S-meter is displayed on the LCD and sent on the serial port.
• Added the use of the PIC's hardware watchdog timer.  (It wasn't used in Jacques' original code and I'm not sure why I didn't use it before...)  If this unit is left unattended, the watchdog timer should increase the probability that the unit will restart the software if it "freezes" due to a power glitch or static - or even a bug.
• Added Adaptive Filtering for the 8-capacitor "Roanoake-type" switched capacitor filter.  When the "Average Clear" is enabled (the "AC" setting on the "Afgate" menu) one may optionally use Pin 5 of the main processor (IC3 on the Doppler II, IC70 on the Doppler III) to "dump" the Roanoake filter and shorten its time constant, allowing it to respond very quickly to signals that appear (even briefly) even when the "damping" is at maximum.  See above for more details.
  
• The software date (obtained by holding button(s) down during powerup) is no longer used as the version number is always displayed on the startup screen.
  
• Doppler II only:  Removed "EEPROM INIT" screen.
• Ver. 7B  (February 9, 2006) - Items changed since "7A":
• Only one minor change:  When sending "averaged" bearings via the serial port, data will not be sent unless:
• A bearing is manually sent using the pushbutton(s.)  This causes the most recent bearing to be sent - even if it is "stale"
  
• When in a mode where bearings are sent automatically, a new bearing will be sent only when it is available.  That is, averaged bearing data is sent only if there is new, updated information to send.  In previous versions, old averaged data would continue to be sent - even if the signal had gone away - but now, it stops.  Note that this has no effect whatsoever when "current" bearing data is being sent via the serial port:  Even with no signal, bearings will continue to be sent, but with a quality of "zero."
• Ver. 7C (June 13, 2006) - Items changed since "7B":
• Only one minor change:  The addition of 1200 baud.  This addition was done to allow the modulation of bearing data atop Bell 202 FSK to allow recording of bearing data on an audio logger.  For example, in using a program like "Xcorder" or "ScanRec" (do a web search for these and other similar programs) can allow a VOX-based automatic recording of audio to a computer hard drive.  In this instance, the audio could be recorded on one channel while the FSK data containing the bearing could be recorded on the other:  This scheme is simple to execute and has the advantage that the bearing data is always in perfect sync with the audio data - but it does require that another program and/or hardware be available to decode the bearing data.  Of course, another option would be to have two programs running:  One recording the audio, and the other recording the serial data either to another file or embedding it into a .WAV file for later extraction.  Note:  When recording FSK data, one of the "Linear" encoding schemes - such as PCM or ADPCM - must be used rather than an a "lossy" audio compression method such as MPEG, GSM, or CEP compression to avoid loss of data.
• Important note about operation at 1200, 2400 or 4800 baud:  At these lower baud rates (particularly at 1200 and 2400 baud) it is likely that NMEA data from a GPS receiver cannot be multiplexed into the bearing stream.  The reason for this is that there is just too much NMEA data to be sent at the lower baud rates.  The result of this is that while the NMEA data from the GPS receiver is being sent, one may NEVER be able to send a bearing until the GPS receiver stops - which may never happen!  For this reason, it is recommended that  you do NOT attempt to send GPS data with the unit set at 1200 or 2400 baud - or even at 4800 baud - if you have your GPS receiver configured to send very long NMEA sentences!
• Ver. 7G (June 26, 2006) - Items changed since "7C":
• On previous versions, when a "manual bearing send" was done (pressing the button on the Doppler III or pressing the two center buttons on the Doppler II) if the unit was waiting for or passing GPS data, the button press was queued and the bearing would be sent after it was done waiting for/passing GPS data:  At that point, the bearing and quality present at the moment that the bearing was finally sent is what appeared on the serial port.  Because the delay between pressing the button and the time that the bearing was sent out through the serial port could be a couple seconds (if GPS data was being sent) it is possible that the bearing could have changed during that time.  On this version, the bearing/quality is stored at the instant that the button is pressed and when the data is finally sent out the serial port, the bearing/quality at the moment of the button being pressed is what is sent instead of what is present at the moment that the bearing was sent.  Furthermore, once the button is pressed, further "buttonpresses" are ignored until the bearing stored at the moment of the first press is sent out.
• This code has been modified to allow the use of the newer PIC18F series of processors.  These processors can have much more program memory, RAM, and data memory allowing the addition of more features.  For the moment, only two minor changes have been made on versions using the PIC18F processor:
• The maximum size of the averaging buffer has been increased to 64.  It was noted that in experimentation, this size of averaging is approximately equivalent in its filtering capability to the 8-capacitor "Roanoake-type" filter (when the "integration" is set to 1) - a potential benefit to those using Doppler II units without the added Switched-capacitor filter.  In the future, testing will be done with larger buffer sizes.
  
• Versions using the PIC18F processors have a "+" following the version number (e.g. "7g+")
• Ver. 7O  (May 2, 2007) - Items changed since "7G":
• All versions:
• Minor change was made that allows the last averaged bearing to be sent automatically (using the TIME setting) even after the signal goes away.  Previously, if the signal went away before it was time to send a bearing, it would not be sent.
• Added a "1 second" option to the TIME setting.  Important note:  This option may not be usable if you are multiplexing both the GPS data and bearing data on the same serial port as it will probably take longer than 1 second to send all of the GPS data.  If you use this option, it is strongly recommended that GPS data be sent on a separate serial port, or not at all.
• When ROSE is set to AVG mode, the unit now sends the current bearing (e.g. integrated) as "bad" bearings when the signal disappears.  Previously, bearings would stop being sent to the compass rose when the signal was gone, but this prevented any "bad" bearings from being sent to the rose and the center LED would not display red to indicate bad bearings.
• A minor quirk was fixed having to do with manual buttonpresses:  Previously, pressing a button would also cause a GPS bearing to be sent in addition to a bearing if a button was pressed to cause the sending of a bearing - if the sending of GPS data was enabled.  Now, only the bearing is sent in response to a buttonpress.
• Doppler III versions only:
• When entering Calibrate menu from the main screen, the cursor will now always appear at the far right position on the screen - that is, the position that will advance to the next menu (the Antenna menu) when the button is pressed - unless the menu potentiometer is adjusted, at which point the cursor will jump to the position dictated by the setting of the menu potentiometer.  This was done to prevent the accidental modification of the calibration setting when trying to get into another menu, or if the button was accidentally pressed.
  
• Versions using PIC18F devices only:
  
• Pin 1 (RE3, or the !MCLR pin) was redefined as an input.  This pin has a pullup resistors and capacitor to aid in resetting the device, but because the PIC18F4620 has a fairly robust set of peripherals to assure proper startup/resetting, this pin was redefined as an input.  When this pin is grounded all antenna rotation stops and the display is frozen with the antenna selected in the "test" menu (if any) being activated.  This provides an easy method of stopping rotation (to silence the tone) with just an SPST switch.  It is recommended that a 470 ohm resistor be placed in series with the switch to provide some ESD protection to the processor.  Note that this feature is not available with the PIC16F877 version because pin 1 cannot be defined as an input - sorry.
• A new option has been added to the GATE setting:  Original settings 0-7 (audio thresholds) are still present, but "above" setting 7, a plus (+) appears with selectable settings of 1-7.  When the plus sign (+) is present, only the very first "bad" bearing (displayed in red) that is calculated after the audio drops below the threshold  will appear on the compass rose, preventing a red "bad" bearing from constantly appearing on the display.  There wasn't enough code space to add this feature to the PIC16F877 - sorry.
  

If you have an older version and would like to upgrade, please contact me.  Also, if you spot a bug, find something in its operation that seems awkward, or would like to suggest a feature, please let me know!

AVAILABILITY:

Disclaimer:

This code was originally based on that of the original Montreal Doppler II DF unit by the late Jacques Brodeur, VE2EMM, and full credit is given to him for this fine work.  Because the additional modifications are my own and were not done with his involvement, Jacques cannot reasonably be asked to offer any support or assistance regarding them!

Although good faith efforts have been made to make certain that the operation of the hardware/firmware is as described, it is possible that "undocumented features" (bugs) may be present:  It is through testing, use, and feedback from the users that projects such as this may be improved, and the user is asked to be understanding of this fact.  Further developments/bug fixes may be made and such changes will be noted on this page and it is up to the user to determine if he/she wishes to receive an update.  This firmware is strictly intended only for non-commerical amateur-radio use and any other use is in violation of applicable laws.

Additional note:  Neither the author or UARC officially endorse any vendors mentioned above or assume any responsibility for the use of the devices/products described herein.  The level and satisfaction of performance of any of the above is largely based on the skill and experience of the operator and no guarantee of suitability is to be implied.  Your mileage may vary.

Do you have any questions on this or other DF-related topics?  Go here.

This page last updated on 20120316