The Montreal Doppler III is an update of its predecessor - the Montreal Doppler II.  It is recommended that you read about them on their respective official pages.  Please note that this was NOT  an official page of VE2EMM - please read the disclaimer at the bottom of this page.
 
Important notes:

• As of late 2006, the Doppler III circuit boards are available from FAR Circuits - see the Main ARDF page for more information.

• In 2006, Jacques produced the "Version 3.2" circuit boards that have some minor differences as noted below.  The pictures below show only the original version of the circuit board.

• The alternate firmware is still available - see the link at the bottom of this page.
  

My version of the Montreal Doppler III.  Note that this unit uses the alternate firmware for the main processor  and for the compass rose display.
Click on image for a larger version

Differences between the Doppler II and Doppler III:

The Doppler II and Doppler III units (running their respective original firmware) are very similar to each other - with the following differences:

• The Doppler II has only an on-screen LCD-based compass rose.  This compass rose is somewhat "squashed" and is somewhat difficult to read at a quick glance.  The Doppler III has a no on-screen LCD compass rose (only a numerical display) but it has a full-sized compass rose using 36 LEDs, resulting in an easy-to-see, intuitive display.
• The Doppler II uses a 2-stage op-amp based peaking filter:  Unless very temperature-stable components are used, this filter may be prone to drift, causing phase/direction errors at different temperatures. The Doppler III uses a 2-stage integrated switched-capacitor filter that is intrinsically stable in terms of temperature.
• The Doppler II does not have any audio filtering other than the 2-stage peaking filter.  The Doppler III includes a "Roanoake-style" 8-capacitor switched-capacitor filter with extremely narrow bandwidth.  This additional filter allows for better performance in the presence of noise and/or other modulation.
• The Doppler II does not have a built-in speaker amplifier, but the Doppler III does, allowing one to set the audio drive to the Doppler unit independently of the speaker volume.  This can be important - particularly on HTs - where the lack of audio power may make it difficult to hear the audio over high ambient noise.  Additionally, the change in volume from the radio can cause a shift in the calculated bearing and not having to adjust the radio's volume eliminates this potential source of error.
• The Doppler II uses 4 buttons to navigate menus and select items.  The Doppler III uses a single pushbutton and a potentiometer navigate menus and select items.
• The Doppler II has a numerical audio level indicator - but this is based on the integrated/processed audio.  If the integration is set such that it takes over a second to process the readings, the audio level detection operates only as fast as this.  The Doppler III has a bar graph audio level indicator using a separate circuit that detects the audio level and can respond immediately.

Filtering:

Owning both the Doppler II and Doppler III units, I can say that both work very well - but the Doppler III, having the 8-capacitor "Roanoke-type" switched-capacitor filter allows it work better in the presence of weak, noisy, and/or heavily-modulated signals.  The addition of this filter is a natural evolution, I hadn't had the Doppler II too long before I'd added an additional switched capacitor filter well before the Doppler III was available.  With the turn of a potentiometer (a recommended modification) this filter can be adjusted for a quick response (e.g. little or no filtering) able to react to a brief "kerchunk" or for fairly heavy damping and a slower response (a lot of filtering) that is appropriate for very weak, noisy, or heavily-modulated signals.

Note that the potential temperature drift problem of the Doppler II is minimized by using good-quality stable capacitors in the peaking filter:  I'd found that allowing the unit to warm up for 5 minutes or so before calibrating minimized this problem.  With the clock-based switched-capacitor filter on the Doppler III temperature-related drift of the bearings has not been noticed - only a slight "amplitude" drift (that is, a slight increase in the apparent gain of IC50 when it is cold) that does not appear to have any noticeable effect on the bearing.

Compass rose display:

Perhaps the best improvement of the Doppler III over the Doppler II is the built-in LED-based compass rose display:  I'd found that the original on-screen LCD "compass rose" on the Doppler II - while clever - to be nearly unusable - mostly because it was "squashed" (not round) and difficult to interpret.  For this reason I'd built my own LED-based Pelorus (compass rose) that was driven from the serial output data to work with the Doppler II.  Note that my alternate firmware allows the Doppler III compass rose hardware to work even with the older Doppler II.
 

The "guts" of my Doppler III
Click on image for a larger version.

The unmodified Doppler III works noticeably better on weak/noisy/modulated signals than the unmodified Doppler II unit, but with the added switched-capacitor filter, I would say that the Doppler II performs equally to the Doppler III - and the LED compass rose makes the Doppler III much easier to use - particularly if operating in a mobile environment.

My version of the Doppler III:

Being one to tinker, I made some modifications to the Doppler III right off:

• The addition of a "damping" control.  This allows adjustment of the "Roanoake-type" switched-capacitor filter from a fairly wide, fast-acting filter to a fairly narrow, slow filter - and anywhere in-between.  This allows the flexibility to determine the approximate direction of very brief signals (such as kerchunkers) and sweeping signals using a quick response, to very weak, noisy signals that require more filtering. I would highly recommend this modification - see below for details.
• The LCD display that I used has an LED-based backlight.  Having used the same LCD module in my Doppler II, I knew that this display would be unpleasantly bright at full intensity when used in the dark, so I added a "dim/bright" switch.  See below for details pertaining to this modification.
• A completely new version of the Doppler III Compass Rose Display software to display averaging, history, and to use dual-color LEDs to indicate signal quality.  For more information, see the New Doppler III Compass Rose firmware page.
• After getting the Doppler III operational, I rewrote the software for the Doppler II to include more features and, while I was at it, made a compile-time option to have this new software work with the Doppler III hardware as well.
• Later, I added provisions for, and constructed an "Outboard Compass Rose" - that is, a display that may be placed away from the main Doppler unit.

For a series of photographs visit the VE2UMS Doppler III web page (English version). There are a lot of pictures there with some good ideas!

Comment:  As of late 2006, the Doppler III circuit boards are available from FAR Circuits - see the Main ARDF page for more information.

When I originally built the Doppler III, the circuit boards (two of them) and pre-programmed pics (three of them) were available directly from Jacques.  These boards, while apparently home-made, were of reasonably good quality, being etched single-sided on 1/32" glass-epoxy circuit board.

There were only two minor difficulties that I'd encountered:  I found an open trace (on the "contrast" control line of the LCD - it caused the display to appear to be blank and non-working) and the fact that the green lacquer on the board had be be laboriously scraped from the solder pads.  While the green lacquer looked nice, it is probably unnecessary.  One builder mentioned to me that he'd simply stripped the lacquer completely off, but another mentioned that he was able to "solder through" it without difficulty.

View of the back side of the Montreal III circuit board.  Close inspection will reveal board mods near the upper-left corner and the bottom edge in the center where the additional components were added for the backlight control and negative supply voltage, respectively.
Click on image for a larger version.

This is not a project that should be undertaken by anyone unfamiliar with basic kit-building techniques.  Not only must the builder obtain most of the parts, but he/she must be able to install the components in their proper locations using only the parts-placement diagram on Jacque's web site and (in cases where the image is indistinct) by back-tracing the schematic diagram.

Finally, in the event that it doesn't work the very first time you power it up (mine didn't - not only did I have to locate the open trace, but I'd left an inadvertent solder bridge somewhere...) you must be prepared to do some simple troubleshooting.

It should not need to be said, but it is strongly recommended that the builder socket all integrated circuits.  This allows preliminary testing of the power supply/regulator as well as incremental testing/troubleshooting of various circuits, not to mention easy replacement of parts.

Important modifications:

Note:  The three modifications below have been incorporated into the more recent (Version 3.2) circuit boards.

Here are three modifications that I would recommend making to the Doppler III board no matter whose firmware you plan to use:

• When you are building the main board, do *not* install C64, a 0.1 uF capacitor connected to the "2.5 volt" line on the output of IC53A as its presence may cause IC53A to oscillate and cause erratic readings.  More recently, this component  has been removed from the schematic, but if you used a parts list/and or schematic dated prior to early 2005 when you built it you may have installed it, so it would be a good idea to check for it and remove it if it is present.
• It would be a good idea to add a 1k resistor in series with the wiper of P50 (the 10k pot used for the "S-Meter" input) and Pin 2 of IC50, the main processor - preferably cutting the trace to put C54 on the "Pin 2" side of the added resistor.  This provides additional RFI protection as well as keeping the PIC from being damaged should a full +12 volts be applied to the "S-meter" input with P50 turned all of the way up!
• It is strongly suggested that a 0.001 uF capacitor be connected across pins 2 and 3 of IC30, the LM386 audio amplifier, on the backside of the circuit board.  Without this modification, IC30 may be extremely sensitive to nearby RF, causing clicking or a loud switching tone to be present if a nearby transmitter is keyed.

• The use of a 3-lead LED is specified.  Originally, a 2-leaded bi-color LED was chosen for the "center" LED, but in version 3.2, this was changed to a 3-lead device, each color having its own current-limiting resistor.
• The wiring of the 3-lead LED is "upside-down" as compared with the original 2-led device.  This is only important if you decide to construct a compass rose using the version 3.2 board with a 2-lead LED, in which case the center LED must be installed "backwards" while skipping the center (ground) lead.
  

The enclosure:

I chose a fairly small, thin enclosure in which to mount the Doppler III's circuit boards and components (a Hammond 1599KBK) and I also wanted to make sure that the various switches and knobs were also low-profile, so I used a flat pushbutton switch for the "select" button, rocker switches for the power, backlight dim/bright, and compass rose control.

Mounting the boards:

The two boards (compass rose and main board) are intended to be mounted back-to-back (see picture) but depending upon exactly how you wish to mount these boards/display in your enclosure, you can change this as needed.  The holes drilled are intended to accommodate 4-40 hardware (or its metric equivalent) and an assortment of various standoffs is very helpful in spacing/mounting these boards.

The potentiometers:

Also present (in my version) are three potentiometers:  One for menu selection, another for the volume control, and a third to adjust the "damping" of the switched-capacitor filter.  If you look closely at the "inside" picture, you'll see that these controls actually mount on pieces of glass-epoxy circuit board material offset from the front panel rather than mounting directly to the front panel.  These pots are miniature units that use 1/8" shafts (instead of the more usual 1/4" variety) and the knobs themselves are recessed, protruding through holes in the enclosure.  The result is that only about 1/4" of the knobs themselves are revealed on top, maintaining the low profile.

Connectors:
 

Close-up showing the back-to-back circuit boards.
Click on image for a larger version.

On the right side of the enclosure may be seen two "D"-type connectors and a flat rocker switch.  The upper connector is for the serial port in/out:  Some users choose to put a separate "in from GPS" and "out to computer" connector, but I decided to use just one connector and construct an "octopus" cable consisting of 3 connectors - one for the DF unit, one for the GPS, and another for the computer.  Unless I need the GPS input, a standard 9-pin cable may be used to connect the DF unit to a computer.

Below the "data" connector is a 15-pin high-density "D" connector of the sort typically used to connect to a computer monitor.  This connector has connections to all eight antenna drive pins plus the unregulated V+ and a V- supply (extracted from the MAX232 RS-232 driver IC) used to power the antenna array switch plus the "S-meter" input.  While not all of these pins are used for all applications, putting them on this connector makes at the time of construction makes them more easily accessible for later use.

On the left side is the power connector - a coaxial-type - but in hindsight, I would, were I to construct the unit again, hardwire a power cable:  The power cable is one item that could get separated from the unit and lost, making it unusable.

Finally, located between the two switches on the left side (but not visible on the "front panel" view) is a 3.5mm audio jack:  This allows the user to reroute the speaker audio to headphones or to an external speaker.

The audio amplifier:

Because the Doppler III uses an on-board amplifier (an LM386-4 - be absolutely certain to order the "-4" version of the LM386 - only the "-4" version is rated for operation above 12 volts!) capable of about one watt of audio output, I chose a high-quality 4-ohm speaker - in this case Jameco Electronics P/N 99995.  This speaker is rated for several watts, has a large magnet, and is very efficient:  With just the single watt output of the LM386-4 it can be very loud with minimal distortion - sometimes a requirement in a noisy car. (No, Jameco doesn't sell the "-4" version of the LM386, but Digi-Key does...)

As mentioned above, it is recommended that a 0.001 uF capacitor be connected directly across pins 2 and 3 on the backside of the circuit board to prevent any problems with RF sensitivity of the audio amplifier.

The LCD module:

The sharp-eyed observer of the pictures will notice something else:  The LCD module.  Jacques, in his board layout, made provisions to mount the LED "piggyback" to the main display board.  I, on the other hand, got the largest 2-line by 16 character backlit LCD module that DigiKey Electronics had available at the time.  Unfortunately, this LCD module had its interface connector along the bottom (the conventional location, actually) rather than along the top.  This necessitated the use of the gray interconnect cable that may be seen in the picture.  I was able to find a 13 conductor pre-assembled cable on a defunct VCR - but this left me one wire short of the 14 conductors needed.  No problem:  The Doppler III uses the "4-bit transfer mode to send data to the LCD and thus only 10 wires were actually needed, so some of the extra were used to make up the deficit.

The switches:

To maintain a "low profile" I chose to use "snap-in" switches for the power, backlight, menu select, and compass rose control.  Each of these switches require that a fairly precise square hole be made in the enclosure, but this is easily done using the "drill 'n file" method in the plastic enclosure:  With a little bit of care, a suitable mounting hole can be made.

Not everyone stocks the variety of switches required, but I was able to find the necessary switches in the Mouser Electronics Catalog:

SPST/SPDT rocker switch:  CW Industries,  model:  GRS-4012-1600  (used for power on/off and backlight on/off)
Center-off SPDT momentary:  CW Industries, model:  GRS-4013C-0001  (used for mode control of the compass rose)
SPST momentary pushbutton:  Mountain Switch, model:  103-1211  (used for the "menu" button)

Details of the modification to add a "damping" control to the Doppler III.

Power supply fusing:

Another important feature is overcurrent protection on the power supply leads.  While fuses are suggested on the schematic, I used 0.6 amp self-resetting fuses (such as those sold by Jameco Electronics - click here for a datasheet) instead along with a reverse-biased diode across the power supply leads (placed after the fusing) for polarity protection.  It is strongly recommended that you put your chosen means of overcurrent protection in both the positive and the negative leads.

Damping control:

Once you have gotten the DF unit completely operational, a highly-recommended and useful feature to add is an adjustable "damping" control.  This replaces R55, a fixed 680k resistor, with a 1 meg potentiometer (an "audio taper" if is preferred, if available,) a 47k resistor, and a capacitor of "approximately" 180 pf - more on the "approximately" part in a moment.

It is recommended (but not required) that the metal body of the potentiometer be grounded:  At the "1 meg" end of the rotation (highest amount of damping) it is possible that a small amount of noise or some "hand capacity" effects may slightly skew the reading when the metal shaft is being touched if the potentiometer is not grounded and a metal knob is used.

When wiring the damping control, I used some "miniature 2-conductor plus shield" (with the shield grounded) wire extricated from a junked VCR.  If you do not have such wire, three conductors twisted tightly (or braided together)  will work just as well.  In this case, one wire would connect to each side of the (removed) R55 location and the third wire (or shield) should go to ground.

The 180pf capacitor is used to counteract various other reactances found in the circuit:  Without it, there would be a 10-15 degree shift in bearing when the control was rotated from minimum to maximum damping.  While 180 pf is a good "starting" value, it may not be the exact value required.  To determine the precise value, tune in an unmodulated, clean, multipath-free, stable signal and observe the bearing at all rotations of this potentiometer:  If it moves more than a degree or so, increase or decrease the capacitance slightly (by 5-10pf or so) until this variation is nulled out.  One possibility is to (initially) use a variable capacitor (5-50pf, for example) and use a fixed 150pf capacitor:  Adjust the variable until the phase variation is nulled out and either measure the variable/fixed capacitor combination and replace it with fixed capacitor (or combination of fixed capacitors) or just leave the variable in place.

Schematics showing the components added for the "Adaptive Filter" modifications - see text.

An "Adaptive" filter  (Version 7A and newer of the alternate firmware only):

One advantage of the filtering used in the Doppler III, if one has added the Damping Control is that it may be adjusted to provide a very fast response or a slow response.  A fast response (low damping or "Q") is often desirable when the signals being monitored are of fairly good quality while a slow response (high damping or "Q") provides better filtering of noise and modulation and can be helpful weak or noisy signals.

Occasionally, one may wish to use the "slow" filter but encounters difficulty when trying to determine the bearing of signals that last only a short time, such as a "kerchunk."  If a slow response is used, the "Q" of the filter may be such that it does not register the signal before it is gone.  Another problem is when a short-duration signal such as this immediately follows another signal:  In this case, the filter will still "contain" the bearing from the previous signal and may not be able to read a correct bearing until enough time has passed for the "old" bearing to be purged from the filter.

This "adaptive" filter option causes a "fast" mode to always be automatically selected when the signal goes away, reverting back to the selected response time shortly after a signal is detected.  In this way, the "old" signal is purged from the filter immediately and when the new signal appears, the filter is "charged" up for a brief moment (between 50 and 100 milliseconds) at the "fast" rate before the "slow" rate is re-enabled once again.

With this feature, it is more likely that even very brief transmissions (especially if they closely follow other transmssion) can be analyzed - even if the damping/Q of the switched-capacitor filter is set to a fairly high value (e.g. a "slow" filter.)  For more information, see the section on Adaptive Filtering in the New Firmware manual as well as this brief description of adaptive filtering.

This modification involves the isolating of pin 5 of the main processor on the Doppler II to disconnect it from the +5 volt supply.  Once this is done, the firmware detects that this modication is present and when the "Average Clear" indicator is active (see the "Adaptive Filtering" section in the New Firmware manual) this pin will be set at +5 volts.  This signal may be applied to a 4066 analog gate that, when the 5 volt signal is present, "shorts out" the "damping" control in the schematic above.  Note that once this is done, it is likely that the value of 180 pF capacitor will have to be readjusted to accommodate the added capacitance of the 4066.  When wiring in the 4066, be sure that the 47k resistor shown in the above diagram is still in series with the circuit when the 4066's gate is closed!

The schematics to the right show how this adaptive filter may be implemented.  For the analog gate, either a standard 4066 or a 74HC4066 may be used, powered from the +5 volt supply, and it is up to the user to decide whether to wire up just a single gate, or all four of the 4066's gates:  The resistance of the 4066 is not significant in this case, so a single gate will suffice.  It is strongly recommended that one tie all unused pins to either +5 volt supply or ground to keep the pins from floating at undefined voltages.  For the ultimate in simplicity and small size, it is also possible to use the Toshiba TS4S66 - a surface-mount chip that consists of a single 4066 gate.


Selection of the proper capacitor value:

One important aspect of this modification is the addition of the parallel capacitor.  As in the case of the damping control, a bearing shift will occur with different values of resistance.  Without the capacitor, the user would observe a 10-15 degree bearing shift immediately after acquisition of the signal that would, after a second or two, disappear.  The reason for this is that when the 4066 is closed, the phase shift across the filter is different that that which would occur with the 4066 open and the full value of R50 (or the damping control) in place.  To compensate for this, the parallel capacitor should be added.

The precise value of this parallel capacitor should be determined experimentally, but it will likely be in the range of 150 to 180pF.  If you have installed the damping control, determining the precise value is easy:  Just follow the procedure described above in the section describing the damping control.  If you have NOT installed the damping control, do the following:

• Tune in a stable, clean signal.
• Set the damping control (if used) to maximum resistance.
  
• Disconnect the "control" pin of the 4066 from the CPU and connect it to +5 volts to "close" the 4066 switch.  Note the bearing.
• Now, connect the "control" pin of the 4066 to Ground to "open" the 4066 switch and note the bearing.
• Repeat the previous two steps, adjusting the value of the capacitor for the minimal bearing shift when the 4066 switch is "opened" and "closed."
• When you are done, reconnect the "control" pin of the 4066 to the CPU.

Bright/Dim backlight control:

The LCD module that I chose uses green LEDs in series-parallel resulting in a voltage drop of approximately 4.2 volts.  As with my implementation of the Doppler II backlight, I decided to operate the LED using "free" current by putting in series with the Doppler III's logic supply.

A slight complication arose, however:  The LM386 audio amplifier and the compass rose display.

I did not want to run the current from either the audio amplifier or the compass rose display through the LCD's backlight:  The backlight's brightness would vary wildly with the speaker audio variations (not to mention the lowered voltage to the LM386 reducing it's capable power output) and a potentially similar effect with a varying load current by the compass rose.

Details showing the modifications for a "Dim/Bright" backlight switch
Click on image for a larger version.

The solution was simple board modification:

• The trace going from + side of C20 to the LM386 was severed and another capacitor added in its place.
• The addition of a 78L05 voltage regulator (its input connected to the + side of the new capacitor on the LM386) to power the compass rose display.
• It is strongly recommended that, on the main board, a 78L05 regulator (the low-current version, in the TO-92 "transistor" package) be used instead of the full-sized 7805 regulator with the metal tab.  The reason for this is that of one accidentally shorts out the +5 volt line on the main board while testing, only 200mA or so of extra current can flow through the LED's backlight - something that will not immediately damage the backlight's LEDs.  With the regular 7805, up to 2 amps can flow - which can instantly destroy the LED's backlights!  (This would mean that, if one does the modifications, that there will be two 78L05's - the one to run the main board, and the added one for the compass rose, as depicted in the diagram.)
  

The caveat to this method of powering the backlight is that there is a (nominal) 4.2 volt drop through the backlight.  In order for the 78L05 regulators to do their job properly, they need at least 1.75 volts higher than this (almost 7 volts) and this means that a minimum of 11 volts (I'd measured 10.75 volts) or so is required for proper operation.  If this amount of voltage drop bothers you, consider using a low-dropout voltage regulator - such as the LM2930 or LM2936 for VR20.  (The separate regulator is used for the compass rose, this need not be a low-dropout regulator as it is not in series with the LCD's backlight.)

AVAILABILITY of the "Alternate Firmware":

• For information about availability of this firmware, please go here.

Disclaimer:

The code for the alternate firmware 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!

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.  This firmware is strictly intended only for non-commercial amateur-radio use and any other use is in violation of applicable laws.

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

KA7OEI - 20120316