GPSDO Phase noise measurements

[Focus]

I see periodic jitter and a ~300KHz spur on phase noise plots of a BG7TBL GPSDO 10MHz sine wave output.  I would appreciate knowing if anyone else has seen the same behaviors.  Please let me know if you have any GPSDO phase noise plots to share. 

I attached phase noise plot images from my spectrum analyzer and oscilloscope.  Both show the same phase noise out of the GPSDO.  I also attached a phase noise plot image of a free running 10 MHz oscillator.  The free running oscillator doesn’t have any phase noise jitter. 

Thanks,

[drew23]

Hi.

Spurs can be an artifact of the measuring process, usually intermodulation. The analysers have spur removal algorithms to clean up the plot.

I have a blog post on phase noise measurement using an inexpensive analyser.

I would be curious to measure phase noise on the BG7TBL GPSDO if anyone has one in the western suburbs of Brisbane.

Regards Drew VK4ZXI

https://vk4zxi.blogspot.com/2023/07/an-economical-way-to-measure-phase.html

[Focus]

Thanks for the comment.  I’m aware of spur generation during the measurement process. That’s why I used two different types of instruments that use two different measurement methods.  And I cross checked that by measuring two different types of 10 MHz oscillators.  I’m confident the spur is generated by the GPSDO unit. 

I hope someone has a unit you could test.  I’d really like it if someone would test the new GB7TBL GNSS Disciplined Rubidium oscillator. 

Cheers,

[drew23]

Hi

I have an SRS PRS10 RUBIDIUM GPSDO on the way from eBay. The main advantage of them is that they use an ultra-low noise TXCO. The rubidium clock and the GPS both use a PLL on a TCXO. But many of the TCXOs don't have low-phase noise. The TCXO determines accuracy and phase noise in the short term. The GPS and rubidium are more for the long term.

The SRS PRS10 RUBIDIUM clock has an input for a 1 pps GPS correction signal, so it is easy to include the GPS correction.

The spur you are detecting is a long way down and I suspect within spec for spurious signals. Does it show up on a spectrum analyser?

Regards Drew VK4ZXI

https://www.ebay.com.au/itm/185788400189?var=693733706363

[Focus]

Thanks for the suggestion. Yes, the GPSDO spectrum shows the noise.  I attached the GPSDO spectrum plot and both plots of the OCXO.  So, the spectrum reinforces my belief the BG7TBL GPSDO is the source of phase noise. 

I’m looking forward to your Rubidium analysis.

Cheers,

[jayk]

That 300KHz peak could be from a DCDC converter in the unit.  Looking at the PCB, it definitely looks like there's some sort of switching power circuit on the back side of the board (see https://blog.febo.com/?p=636).

[Focus]

Thanks Jay.  That blog had the phase noise plot I was looking for. 

[Focus]

I finally got a rubidium reference.  The two plots below illustrate the difference between my GPSDO and the rubidium.  The two significant differences for me are the elimination of the GPSDO adjustment phase noise and the rubidium’s low close in phase noise. 
Cheers,