Important Note:  Unless you are absolutely certain that you know what you are doing, do not perform this conversion.  
I cannot be - nor  will  be - responsible for any damages to your radio incurred as a result of your reading this website. 

A Dual Digital VFO for the Ten-Tec Omni A, B, C, D, Corsair, Century 21, Argosy, Triton, Delta, Drake TR7 & Similar Digital Display, 5.0 to 5.5 Mhz VFO / PTO Equipped Radios

omni picture
1. Introduction

    Although this site currently describes how this DDS VFO can improve the Ten-Tec Omni D series of PTO controlled radios, it may be readily installed in other Ten-Tec radios including the Corsair I and Corsair II, the Triton, the Delta, the Argosy II and the Digital Century 21 as these radios are all equipped with a factory stock digital display.

    However, if the user is willing to install a home brew digital display, other Ten-Tec radios like the analog Century 21, the Century 22 or the early Argonauts could likewise enjoy the benefits of this conversion, provided there's space within the radio to accommodate the DDS VFO circuit board which measures 2 and 3/4 by 3 and 1/4 inch.  Removing the stock PTO should provide an adequate mounting space.


2. The 'Contributors' - Giving Credit Where Credit Is Due

    Having built an AD9850 DDS signal generator using a modified version of George Heron's (N2NAP's) code, I decided to develop a quality DDS VFO using additional information published on the internet.  This website shares my results (and even the software) for anyone who might like to partake.

   Porting George Heron's DDS software to a Freescale MC908JL16 chip was easy.  I bought a Chinese AD9850 DDS chip boards on eBay for $6 (including shipping).  Since it used a 125 Mhz crystal, I changed the multiplication constants in George's software, and it worked right off.

   The output of the DDS chip needed amplification.  An internet search produced a 2 transistor amplifier designed by K8IQY.  Once breadboarded and operational, I had a nice 4 volt peak-to-peak output.  I sold off my old Heathkit SG-8 and began to experience the wonders of DDS.

     I then reworked the N2APB code into a dual 5 - 5.5 Mhz VFO with SPLIT operation, RIT and LOCK.  Also included was an interrupt driven tuning acceleration algorithm that significantly facilitates traversing from one band edge to the other, as was the ability to store the last used frequency so that it was available on power up.  This VFO tunes much like an analog VFO with flexibility depending upon the speed and duration of the optical encoder.

Note: a 128 step optical encoder will enable a frequency excuirstion of about 5 khz for each complete rotation.  A 256 step encoder will double this to 10 khz.  In my opinion, either is totally satisfactory and there are most often economical choices on eBay
Note: This VFO could also be housed in a metal enclosure and serve as a versatile dual VFO for those reluctant to modify their Omni, TR7, etc and / or don't want to take the time (and expense) to rebuild the mechanical PTO.  Here's just one example.

3. Parts - Itemized Component List

    You'll need about $60 - much less if you have a well stocked junkbox.  The most expensive part is the optical encoder.  I found a couple of nice ones on eBay for $20 each.  Truly clever folks may be able to construct their own encoder using photo-diodes and a home built optical interrupter with 'bearing type' parts from discarded potentiometers.  Surf the web for suggestions.

    Two P/C boards have been designed so that the Chinese AD9850 board will plug right into it.  Here are pictures of the prototype Version 1 board (no longer available), the schematic, the P/C board 1 artwork , the wire connection points and the first completed assembly.  

Note: The Version 1 board has been replaced by a slightly smaller (2.75 inch x 3.20 inch) and improved Version 2 board which will be available in early January, 2014. Here's a picture of the P/C board 2 artwork, the wire connection points and a completed assembly.

4. Conversion Suggestions - Ten-Tec Omni Radio

    If you plan to undertake this conversion, the first thing is determining how to mount your optical encoder.   To do this, you'll have to disassemble the radio to the point of removing the front panel.  You might want to get a small container for all the knobs, parts, screws, etc.

    a) Disassembling the Omni (as an example).
    Note: Be sure to save the 2 felt pads from the main tuning knob as they will come in handy later.
    b) Removing the PTO
     c) Installing the Optical Encoder

    Before installing the encoder, make sure your board is working properly.  Following the schematic, connect up the encoder, a frequency counter to the output and then 12 VDC and ground.  You should see a 5.000 Mhz signal (or something very close) which should change as the encoder is rotated.  If you don't have a counter, your station receiver will suffice.  You can even connect it directly to your Omni at the rear jack provided for a remote VFO.  
Important Note: The DDS VFO output level should be set as close to the output of the Omni's PTO as possible as the Omni's circuitry was designed around this value.  Setting the DDS VFO's output higher than this will generate some 'birdies' here and there and needlessly increase the receiver's background noise level.  The Ten-Tec's PTO's output is approximately .5 Volt peak to peak.  A resistor trimmer (R5) has been provided on the Version 2 circuit board for this purpose.  If you have a 'scope, you may use it to set the DDS VFO's board right on the money.  If not, just turn R5 slowly until the radio begins to receive properly on all bands, but no higher.
 
Also Note: If you monitor the signal on your receiver, you'll probably observe a rough, warbling note.  This is normal as the RIT connections have yet to be made and the processor's A/D converter (used for the RIT function) will be 'hunting' a bit.
    Connect 4 (2.5 foot strands) strands of 30 gauge wire to the encoder.  Tag the power and ground connections (I used knots in the wire).  The phase leads can be swapped later (if it tunes backwards).

    In this case, the size (length) of your encoder shaft matters.  By far, the easiest way is to place a small metal plate (drilled out for the outside diameter of your encoder and mounted in the existing PTO mounting holes) on the outside of the radio's sub-panel.  Temporarily install your encoder (finger tight) and then the front panel.  If you can satisfactorily attach the tuning knob of your choice, and if it spins properly - that's great.

   But if you need a bit more shaft length, either check around for another tuning knob whose set screw is closer to the back of the knob and / or attempt to drill out the old Ten-Tec tuning knob using a 1/4 inch bit and a drill press.  I tried to my old knob out with just a hand drill and muffed the job because the knob had a slight wobble when tuning.  

    To use the tuning knob of my choice whose set screw was too far back to securely grasp the optical encoder mounted on the sub-panel (as shown), I mounted the encoder on the front panel by securing it to the same piece of drilled P/C board material which itself was mounted on the rear of the front panel using the two existing screw holes.  Here's a picture of the panel.  That's a steel washer under the nut.  It fits perfectly, and it tunes like a dream.

    Since the optical encoder didn't fit into the PTO opening, I had to enlarge the sub-panel opening using a Greenlee chassis punch.  I mounted the encoder with the wiring pins downward and didn't obliterate the former PTO mounting holes. Once you are satisfied that the front panel can be properly installed, add about 2.5 feet of wire to the old SPOT switch, mount it on the front panel and temporarily set the whole thing aside.

d) RIT Functionality

    The goal here is to give the controller access to the Omni's offset control (RX-OT - 22K for my radio) and (optional) access to the existing offset switch.

    First, remove the 4 screws holding the preselector and turn the whole assembly backwards.  You'll note another example of  TEN-TEC's 'frugal' 'point-to-point' wiring technique.  In my case, I was able to flip it back just enough to remove the Offset control from which I clipped all 3 wires.  

    Next, solder 3 wires (about 2.5 feet long) to the control, tagging the center (wiper) lead, and set this control aside.  If you want to use the existing Offset control switch and the OT led to indicate that the offset mode (only on receive) has been activated, one more step is required (see below).  If you don't want RIT on/off functionality, then there is no need to modify the switch assembly.  If the RIT Disable lead is left open (high),  the Offset control will be active (except while transmitting). You'll have to set it properly when in the SSB mode (so that you're not off frequency).

    Skip ahead to this step to reinstall the Offset potentiometer, the preselector, the digital display and the notch filter.

    Remove the 4 nuts holding the switch assembly and push it back.  The mounting screws are inside the radio and they are equipped with vinyl spacers. The goal is to access to the portion of the OT (offset) DPDT switch that controls the tuning voltage being set to the PTO while the offset function is active.

    Looking down from the front of the radio, disconnect the two wires connected to the rightmost side of the offset switch.  While there, disconnect the two wires from the SPOT switch.  They can be tagged and securely wrapped.  Alternatively, they can be removed if you don't plan to revert to the PTO.  

    Two resistors (R2 and R4) on the switch board itself that need to be removed if you want to use the OT switch to activate / deactivate the controller's RIT function.  Here's a picture of one.  They have to be removed for the switch to activate / deactivate the RIT function and light / extinguish the OT LED.  

    I wiggled the board until I could see both 1/4 watt resistors through the hole and then clipped them both out with needle nose pliers. The ideal way to remove these parts would be to remove the 4 screws holding the board and then bring it up a bit.  However, Ten-Tec's stingy wiring practice struck again.  Needless to say, if you clip out these resistors it will be difficult restoring the rig to its PTO condition without a lot of additional work.

    Carefully solder wires (each about 2.5 feet long) to the two upper, rightmost portions of the offset switch.  Once done, use your voltmeter to verify that the switch shorts when the offset is off, and that this short is removed when the offset switch is in either of the 2 upper positions.  Run this wire as you've done with the previous ones.

    Using needle nosed pliers, carefully replace the vinyl spacers that may have fallen off their screws and bolt the switch assembly up.

    Reinstall the OT control on the front panel and run the wires beneath the chassis.  Before tightening nut on the OT control, connect a meter and set the control to its exact electrical center (11K from the wiper either leg).  You want to ensure that this position will be maintained when the OFFSET knob is installed.  It should point directly up - for a zero offset. Replace the preselector assembly followed by the digital display and the notch filter.

e) Reinstalling the Front Panel

    Ensure that the leads for the optical encoder, the SPOT switch, RIT on / off (if used) and the Offset control are all run properly.  Ensure that the three flat 'black donuts' are mounted over the toggle switches beneath the preselector and that no wiring is kinked / snagged.  Move the front panel close to the chassis and connect the 3 wire plug that operates the ALC and OT lamps and then place the panel AND the trim ring in place, securing it with the 4 corner screws.

  Remount the mike and headphones jacks, placing them in the proper holes along with the large lock washers.

    Place a 1/4 flat washer up against the optical encoder shaft followed by the 2 felt washers removed earlier, and then mount the tuning knob.  If your encoder has little to no torque resistance (as one of mine did), the felt washers will keep the knob from unduly rotating after it has been turned and released.

  f) Mounting the Controller

    I mounted my board in the right rear corner of the radio where the crystal calibrator mounts in the analog Omni A. This is a convenient spot as it has both a source of 12 VDC and existing coaxial cable access to the VFO amplifier board to which the existing PTO is connected.  This spot permitted last-minute refinements to the M/P code (during the debugging phase) without removing the front panel, digital display, etc, etc.  

   The ideal way to mount the controller would be to drill 4 holes in both the corners of the P/C board and then into the chassis, providing a physically secure grounding arrangement.  Being pressed for time, I mounted a piece of insulating perf board under my P/C board, and secured this with one of the existing screws used to hold one of the connectors on the other side of the board.  One of the 2 P/C board mounting screws is located in the corner of this board, while the voltage regulator grounding tab is used for the other.  I installed a longer screw in an existing hole (see picture).  Grounding wires at the other corners of the board are connected to convenient locations.  

    Note: Since the whole board draws close to 140 ma, the LM7805 needs a heat sink, and the chassis alone is perfect.

g) Connecting the Controller

    After the controller is mounted securely, it can be wired up:
Note: Be sure that you have the right pin by measuring its voltage in receive (12 VDC) and transmit (0 VDC).
5. Commands

    When powered up, both the A and the B VFO will be set to the lower band edge, that is, 7000, 3500, 1800, 28000 (etc).  VFO A will be enabled.  The user may then tune with VFO A in the normal manner, and VFO A will be used for transmitting.  If the RIT (OFFSET) switch is activated, the receive frequency will vary based upon its setting; the transmit frequency will not change.  When the OFFSET is turned off, the original frequency will be restored.

    To switch to VFO B, depress (tap) the function button (most often wired to replace the Omni SPOT button) briefly, and the system will be using VFO B.  The frequency previously stored in VFO A will not be changed.

        Note: if you have wired up the optional LEDS, the LED for either VFO A or VFO B will be illuminated.

   To enter the SPLIT mode, just tap the function / SPOT button twice (a short followed by a longer tap - like a ' A' in CW) and the radio will enter the SPLIT mode.  The on-line VFO will control reception, while the off-line VFO will control transmitting.

        Note: if you have wired up the optional LEDS, the SPLIT LED will be illuminated.

    To exit the split mode, tap the function / SPOT button twice (another short - long tap sequence) and the radio will revert to the normal mode.  The contents of the on-line VFO will be copied into the off-line VFO.

    To LOCK the system at any point, just hold the function / SPOT button down for 2 seconds and the system will be LOCKED, and cannot be changed until UNLOCKED.  To unlock the system, just tap the function / SPOT button - and that's it!  While LOCKED, the RIT control will work.

        Note: if you have wired up the optional LEDS, the LOCK LED will be illuminated.  

    An optional button has been added dedicated to the SPLIT function.  Tap it one time and the SPLIT function is active.  You can then use the main button to switch between the VFO's.  Tapping the SPLIT button again will disable the split function and map the on-line VFO into the standby unit.

    To store the last used frequency before powering down the radio, just operate the LOCK function, release the button and push it again within one second.  If you have equipped the LED's, they will all flash 3 times to indicate that the frequencies (VFO-A, VFO-B and the split function) have all been stored in flash memory and will be available whenever the radio is next powered up.  

Note: The instruction manual for this processor states that the flash memory can be updated just 10,000 times, so you might want to use this function frugally.  If you want to disable it, simply ground the FLASH INH lead (see the schematic).

7. On-the-Air Results
a) Receiving

    The P/C board is mounted in the rear of the chassis, with no shielding and with rather long leads for the Encoder, Offset and Push Button controls, and it works well for me.

Note: One way to cut down on spurious mixing products is to follow the instructions in the service manual to adjust both R23 and R2 on the oscillator / mixer board. Perform the adjustments shown in Step 3 and in Step 7 (Mixer Balance).

   With an antenna connected and the preselector peaked, there are a few detectable narrow banded (200 to 300 khz spurs).  The louder spurs are asterisked. 
Note: This 10 meter spur (28987 khz is quite loud) is a known Omni issue.  To tune this frequency, set the bandswitch to the 29 Mhz position and tune downward. Check the service manual for more information.
 b) Transmitting

    Several SSB QSO's were made on 40 meters, and the reports were comparable to what one would expect from a PTO equipped Omni - generally very good.  Both the SPLIT and QSK functions work properly on CW. 

8. Other Concerns / Considerations 

    If the user decides to tune up the antenna to make a QSO (say, answering a CQ), and if the antenna SWR is too high - the power supply circuit breaker will trip.  Since the DDS board is powered by the same supply, the desired frequency will be lost when the breaker is reset.  This is one of the drawbacks of using DDS in lieu of the analog PTO when the DDS is powered by the current sensing power supply.  

    Three solutions are possible. 
  1. The DDS board could be powered separately - say by a wallwart supply - and left on all the time.  This way, should the Omni's power supply trip out, the desired frequency information will be retained on power up.   Gauche? - yes, but workable.
  2. If the AIRPAX (or equiv) circuit breaker used to safeguard the radio's finals were to be installed in the Omni proper, then the DDS VFO could be powered on the 'line side'.  This way, the circuit breaker's tripping would not cut the power to the DDS module.
  3. Alternately, the operator may gradually increase output power (using the drive control) when tuning up to an antenna.
9. Using an External DDS VFO Controller - Front, Side, Rear Views

    Those reluctant about digging into their radio to mount the DDS VFO P/C board, encoder (etc), may opt to build the whole thing in a separate enclosure as shown in the above pictures.  As you'll note, the prototype unit (just sold) has 4 unmarked LED's across the top (VFO-A, VFO-B, SPLIT and LOCK), a red and a black push button, the tuning knob for the optical encoder itself, and the RIT control with an activation / deactivation switch.

    The rear panel shows the power connector, the VFO output and a third phono jack into which the transmit signal from the radio is to be plugged, as was done with the internal modification (R lead) that was just described.  

    The black button is the multi-function unit that lets the user switch VFO's, operate split and lock the dial.  The red button is just a one press access to the split function.

    Note: The unit shown here worked very well with both my Ten-Tec Omni and Drake TR7.  Another ham is using it with his Corsair 1.

Loose Chassis Grounding Screws

    Intermittent connections are very often frustrating and difficult to pinpoint / resolve.  One such example involves the chassis grounding method used by Ten-Tec on the RX Trimmer and bandpass boards on the radio's underside.  When the screws holding these grounding straps loosen, they can be very difficult to tighten as the PC boards themselves limit access, and narrow needle nose pliers are often required.  Even worse, when the chassis threading becomes stripped, there's no way that the screws can be secured.

    A simple solution is to insert a short screw in the chassis hole beneath.  Then, solder a short bare wire to the topside of the board and secure it to the new screw, as shown here.

Intermittent Heterodyne Oscillator Crystals - Check the Chassis Grounding Screw and the Crystal Pins

    I purchased my Omni used on eBay.  The seller mentioned that the 40 meter band was sometimes intermittent, and would usually work after the radio had warmed up for a bit.  He was right.  I spent a couple of hours on this problem noting that the 80 meter band would occasionally crap out.  Two bad crystals???

    I put a scope on the output of the oscillator / mixer board and watched.  When the 40 meter band was starting to fail, the output of this board would start to gradually diminish to the point where the received signal would weaken and then totally disappear.  Sometimes the display would remain on frequency, but most times it would just revert to 1.000 Mhz.

    I found that I could restore operation by tapping on the crystal band and / or just flexing the board.  However, one problem was positively traced to a loose grounding screw which is partially hidden by the oscillator / mixer board itself.  There was enough of the screw 'showing' so that I was able to tighten it with a small bladed screwdriver, and this solved my problem at least for a while.  While you're at, try tightening up all similar grounding screws on the bottom of the chassis.

    Here's a picture.  Poor quality? - yes, but it should point you in the right direction.

    After a week or so the problem reappeared.  So, I turned the rig off, let it cool and then removed the bottom cover.  Upon power up, 40 meters worked for a while and then began to fail.  I found that I could reproduce the problem by slightly tapping on the 11 Mhz crystal.

    So, I removed the crystal from its socket, cleaned the pins and then reinserted it.  No good.  There was either a poorly soldered connection on the bottom of the crystal socket, or the crystal was intermittent.  Then I noticed that the crystal worked perfectly well if it was only partly inserted into its socket.

    Next, using a toothpich I carefully narrowed the openings on the crystal socket.  I also gently spread the pins of each crystal so that a bit more force was required to insert them.  So far- so good.  No matter how I tap on the heterodyne oscillator board or the crystals themselves, I cannot repeat the failure.