Author Topic: Options to measure frequencies beyond 6GHZ - LMX2594 15GHZ signal generator  (Read 8250 times)

0 Members and 1 Guest are viewing this topic.

Offline drew23

  • Contributor
  • Posts: 39
  • Country: au
Hi All.

It may be useful to read the Microchip AN3899 app note on measuring phase noise from microwave sources. They discuss ways of downconverting signals with minimum added phase noise. The same downconverter could be used with other instruments.

The second system uses conventional mixers and gives reasonable results.

The third system they detail, fig 19,  uses image-reject mixers to clean the signal more than a conventional mixer. This allows the phase noise analyser to work faster.

I am exploring the use of these downconverters with a low-cost (<US$1000) PN2060C phase noise analyser from China. He is also investigating microwave sources following AN3899.

I am trying to use cheap boards from China to do similar. I have started a (very) draft blog post on it.

Regards Drew VK4ZXI

https://ww1.microchip.com/downloads/en/Appnotes/AN3899-UHF-and-Microwave-Measurements-with-the-53100A-Phase-Noise-Analyzer-DS00003899A.pdf
https://qsl.net/bg6khc/pn2060c_phase_noise_analyzer.htm
https://vk4zxi.blogspot.com/2023/07/an-economical-way-to-measure-phase.html
https://vk4zxi.blogspot.com/2024/01/measuring-ghz-phase-noise-using-dual.html


« Last Edit: January 09, 2024, 02:06:53 am by drew23 »
 

Offline Gerhard_dk4xp

  • Frequent Contributor
  • **
  • Posts: 353
  • Country: de
@ Verticon

What programming and loop filter did you arrive at with the 4002 that
controlls the 100 MHz VCXO?

In my first try I divided the 100 MHz XO to 10 MHz with a 74LVC161
and the PLL was a 4046 then. The LVC161 seemed to create spurious
that got modulated on the clock, so for a redesign I did choose the 4002,
also to be more flexible with reference clocks.

I designed a new down converter board that has 2 ranges: 5MHz... 3GHz
and 3 GHz ... 12 or 20 GHz, depending on available mixers (Hittite or LT)

1st IF is 900..928 MHz with SAW filters, 1st LO is LMX..
2nd LO is 100 * 3 * 3 = 900 MHz, then SAW, 2nd IF is shortwave for the Timepod.

The timepod can do cross correlation via 2 independent down converters, so the
down converter should not be limited by the LMX to measure LMX2595 class
devices. No need to swim faster than the shark, it's enough to swim faster than
the guy next to you.

I think the board was mentioned somewhere else on EEVblog. 
There will be some boards left over and there will probably a redesign.

regards, Gerhard
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf