[time-nuts] Using GPSDO as a Refrence for Protable Amateur Radio Microwave Operations

Tom Curlee tcurlee at sbcglobal.net
Thu Dec 22 01:00:41 EST 2016

I used one of the Jupiter GPS receivers that has a 10 KHz output to control my 10 GHz LO to discipline one of the common "brick" type of microwave oscillators.  These oscillators have an internal crystal oscillator (106.6 MHz  for a 10.224 GHz LO) that is multiplied up to the needed microwave frequency.  I divided the 106.5 MHz oscillator frequency down to 10 KHz that was then compared to the 10 KHz output from the Jupiter.  The control loop is very simple: a single op amp, a resistor, and a large capacitor.  The response time is extremely slow - 5 to 10 seconds or more.  All I wanted was to nudge the crystal oscillator onto frequency.  

>From a warm GPS start, the LO is within 1 to 2 HZ at 10.224 GHZ within 45 to 60 seconds from power on.
The hardest part was designing the divider circuit to get 10 KHz from the 106.6 MHz oscillator.



      From: Eric Haskell <eric_haskell at hotmail.com>
 To: Time Nuts <time-nuts at febo.com> 
 Sent: Wednesday, December 21, 2016 11:06 AM
 Subject: [time-nuts] Using GPSDO as a Refrence for Protable Amateur Radio Microwave Operations
Hello Time Nuts,  I have been on the group and have promoted it to other folks for a while but this may be my first post here.

I am microwave amateur radio operator and I have question to pose relating to the use of GPSDO's with amateur radio for microwave communication.

First, the more generic question.  A friend was discussing using a eBay purchased Trimble 57963-D for providing a 10 MHz refrence for his portable microwave station  (primarly at 10GHz).  He wants a clean high stability 10 MHz refrence mainly to lock the station LO. First I think a GPSDO is overkill for this application and I am thinking that a good surplus ovenized crystal oscillator should get him to within a few Hz after warm up and a Rb could do better but may have short term stability that may degrade phase noise of the LO.  I am concerned that a GPSDO is not designed for portable operations.  Moving it should probably force a new site  survey which may take a day or more  to complete before it goes into disciplining mode so you would loose any potential benefit of a GPSDO by moving around frequently.  If he wants to do this I think he should leave it connected at his home location for an extended time (several days at least), then when he want to go portable (roving), he should
  disconnect the GPS antenna entirely to force the unit into holdover mode maintain continuous power with battery backup which should maintain the internal OCXO very close to the target frequence and allow the holdover algorithm to compensate for OCXO for aging and best it can.  I would guess that if he chooses to used the GPSDO with the antenna connected it would probably never exit the site survey mode and you would have the output default to the last known good DAC value when it was been disciplined so it would be operating as a OCXO only (although potentially starting from a very accurate starting point, if it had been in use at a fixed location for a good while) before going portable.  Is this a correct view of the situation?  Any recommendations?

I also know of a fellow who has developed some excellent open source Linux software to drive an Ettus Research USRP microwave SDR transceiver for amateur radio microwave applications.  His code also has features to calculate antenna baring and with other available code compensates for satellite Doppler shift and/or synchronize digital communication modes using the GPS coordinates and timing data.  He has a built in interface for a Trimble Thunderbolt for this purpose.  I think it also might be a better solution to use a OCXO for 10 MHz and a cheap USB GPS sensor for location?  Is there a cheep USB GPS that provides PPS?  Any recommendations?

I have seen simpler GPS controlled 10 MHz sources like the Miller design that divides down a 10 MHz ref and compares it to a 10 KHz output from a Jupiter T GPS to tweak the ref freq that may or may not be better suited to this application as it may add phase noise to the LO but would be more real time in it's GPS correction to the reference frequency.


Norman Eric Haskell


Keller, TX USA
time-nuts mailing list -- time-nuts at febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.


More information about the time-nuts mailing list