- Community Home
- :
- User Groups
- :
- More Groups
- :
- Network Analysis
- :
- What is the Time Domain Resolution of the PXIe-5632 VNA?

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Printer Friendly Page

This is an open group. Sign in and click the "Join Group" button to become a group member and start posting.

Highlighted
##
##### What is the Time Domain Resolution of the PXIe-5632 VNA?

02-07-2014
01:31 PM

Options

- Mark as New
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report to a Moderator

The PXIe-5632 vector network analyzer (VNA) provides time domain analysis as an option that can be purchased to complement its normal frequency domain analysis functionality. Time domain analysis performs inverse FFTs upon the acquired VNA data to allow S parameter data to be viewed as a function of time/distance instead of frequency. It is often used as a debugging tool, and shares some similarities to time domain reflectometery (TDR), though time domain analysis offers some advantages over TDR. The attached presentation covers some time domain analysis basics.

The question posed above is what is the resolution (on the X axis) of the time domain plots shown. The answer depends on the VNA sweep span and the time domain processing mode.

The wider the frequency sweep span, the finer the resolution of the time domain plot. A sweep utilizing the full range of the VNA will provide finer resolution than a narrowband sweep.

Another factor is the processing mode - bandpass or lowpass mode. The attached presentation details the difference in these modes, but in regards to resolution, lowpass mode provides resolution that is twice as good as bandpass mode when using the same settings. However, lowpass mode has additional sweep requirements that must be met (e.g. f0 = df).

As an example, I setup a sweep from 1 MHz to 8500 MHz using 8500 points. This meets the requirement of f0 = df, so lowpass processing mode can be used. The resulting data is attached in Time Domain.txt. The data itself is meaningless, but it shows the time domain sample resolution to be ~0.588 ps. Converting this time value to a measure of distance depends on the dielectric constant of the DUT.

3 REPLIES

02-07-2014
02:01 PM

Options

- Mark as New
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report to a Moderator

To confirm, if you are measuring a connection that uses several different cable types, can our software accept multiple dielectric constants?

Does the PNA use the same technique?

02-07-2014
02:37 PM

Options

- Mark as New
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report to a Moderator

02-09-2014
07:16 AM

Options

- Mark as New
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report to a Moderator

Sweeping from 1 MHz -> 8.5 GHz with 8500 points (harmonically related grid k * 1 MHz with k=1.. 8500 -> lowpass mode is applicable) gives an effective time resolution of 1/{( 2 * 85000 +1) 1 MHz} = 58.8201 ps.

On can choose different windows which is equivalent on selecting different shapes of pulses that are used as stimulus.

Also the VNA's use interpolation techniques (e.g. chirp Z-transform) so one can have a lower time resolution but this is just interpolating between the time points which are limited by the effective time resolution (e.g. 58.8201 ps)

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Printer Friendly Page