frequency counter function

[ge]

I read in a 2006 topic... In the old days, digital counters would not work at microwave frequencies. So you bought a prescalar divider from a microwave house, and a PLL/counter from someone like Motorola, and married the two.

What is used now for frequencies up to 20GHz?

tnx,
GE

 

[A.Anand Srinivasan]

i think even now 20GHz is too high for frequency counters... this is due to the response time of the counter.... i'm not sure about this but i think upto 2GHz can be measured directly but that too with a lot of error...

 

[radiohead]

You can buy counters that go up to several GHz-es today, but in fact it's nothing more than a prescaler and low-frequency digital counter combines in one instrument. Whether that is the way to go for you depends on your situation. In lab environments the old way still works fine.

 

[rfmw]

"i think even now 20GHz is too high for frequency counters..."

I do not agree. Nowadays there is very cheap logic available with toggle rates well beyond 20GHz, so counters with input range >20GHz and resolution of 1Hz are easily possible...

 

[A.Anand Srinivasan]

"i think even now 20GHz is too high for frequency counters..."

I do not agree. Nowadays there is very cheap logic available with toggle rates well beyond 20GHz, so counters with input range >20GHz and resolution of 1Hz are easily possible...

i meant in terms of direct counting.... many counters which can count in GHz are available but they mostly using downconversion followed by counting...

 

[sergio mariotti]

For my knowledge , prescalers (frequency dividers) are still used.
But dividing meas loose resolution or increase gate time.

Hi end instruments such Agilent in both many modern and old instruments are (and was) able to read 1 Hz resolution every second up to X or Ka band.
This meas that other tequiques are (and was) used.
What I know is that for may years, hp counters used a comb generator wich was beating with the unknow signal, the IF was everytime in the IF range ( i.e from DC to 1 GHz) then counted whit no information loss ( 10Hz-0.1sec, 1Hz-1sec, 0.1Hz-10sec). The MSB was not measured but created by the counter and it was related to the "used" harmonic of the comb.

Also. in order to not loose any significative digit, the counter may choose automatically if perform a frequency measurement or a period measurement ( and the reciprocate the number).

many other tequinique exist, I think datailed principle of operation may be found into Agilent AN.

But rem. the most important, 30 yaers ago and nowadays is the stability and accuracy ot the timebase.

 

[raghuram_msc]

Hi All,

I quite agree with sergio mariotti. These days the state-of-the-art is computer-automated complete test-setup, many already commercially available, wherein, Integrated with LabVeiw the Controller-PC gives a virtual oscilloscope (for Ex) at as high as Ku. The vendors claim they can custom make at W-Band as well....!!!!

I personally never had chance to verify this (W-band issue !!), though, I had personally tested such a system at C&X-Bands. There it was perfectly OK.

Sai.