# network analyzer non-50ohm measurement

Status
Not open for further replies.

#### antenna_abc

##### Member level 3
citifile to matlab

Hi, everyone. I have a simple question about using the HP8753 VNA to do the S-parameter measurements.

It seems the reference impedance is 50 ohm, how can I change it to 100ohm reference system and get the S-parameters based on 100 ohm?

If yes, can I just use the same calibration kits (50ohm load) to get that?

#### Lance_RFdude.com

##### Member level 3
citifile touchstone conversion

For things I've done that weren't 50 ohms, I just manipulate it in a simulator.

If you have a good calibrated measurement at 50 ohms and can get the connectors close to the circuit, you can do an OK job using this technique.

In Agilent/Eagleware Genesys, you can just change the port impedance and interface to a 50 ohm s2p block.

The downside is that you can't see the "proper" results on the VNA screen, but you can get the data on your PC once transferred (using their TESTLINK interface or other utilities).

This is just my way of doing it, I'm not sure there is any way to do it in the VNA itself. This has worked "ok" at sub 2 GHz for the things I've tried to use it for.

Lance

### antenna_abc

Points: 2

#### Darren_pan

##### Member level 2
.s2p file in genesys

I suggest you use the smith chart to tune,if you really knew the VSWR meaning of 50 ohm system,it's not big problem for you.

#### antenna_abc

##### Member level 3
convert 50 ohm s11 to 100 ohm s11

Thanks, Lance. So do you means that even though the VNA itself cannot be changed to non-50ohm system, we can still export the measurement date and use PC to get that conversion?

If so, what kind of software you are using to finish this job, besides Agilent/Eagleware Genesys? I remember that the VNA HP8753 exports the "CITI" file but not "TOUCH STONE" file, is it still OK for the conversion?

Thanks.

#### Lance_RFdude.com

##### Member level 3
50ohm sample for s parameter mesurements

antenna_abc said:
Thanks, Lance. So do you means that even though the VNA itself cannot be changed to non-50ohm system, we can still export the measurement date and use PC to get that conversion?

If so, what kind of software you are using to finish this job, besides Agilent/Eagleware Genesys? I remember that the VNA HP8753 exports the "CITI" file but not "TOUCH STONE" file, is it still OK for the conversion?

Thanks.

You could do the conversions in various ways -- I'm not 100% sure of the capabilities of other simulators (you could put transformers at the ports, etc).

on my website (rfdude.com) you can find a perl script to convert CITI to s2p format. I've also recently uploaded a windows program that will interface to an 8753 via GPIB to suck out an s2p file (my personal 8753 is older and does not have an integral disk drive). The program is intended for national instruments GPIB interfaces (not sure if there is any chance of it working with something else), and is very crude (there are some known bugs that may cause it to crash)... but I use it quite a lot. I know of at least a couple of other people who have used it successfully.

Lance

#### zhonghua_li

##### Junior Member level 2
convert citifile to touchstone

1.get the complex impedance of ZL,by the equation of S11 = (ZL - Z0)/(ZL + Z0);
2. calculate the S11 with other Z0 (eg. 100ohm) by the equation of S11 = (ZL - Z0)/(ZL + Z0) too.

it's easy to be calculated by matlab's complex operation

##### Full Member level 6
renormalized 50 ohm s-parameters

If you have access to Microwave Office I wrote a VNA2MWO program that will acquire data from the VNA in real time. Everytime you push the simulate button it will fetch new data, update the 2-port circuit model, and show the results. Its good for real-time de-embedding and tuning in non-50 ohm environments. The READ_VNA is a sub-program that grabs the two-port data and optionally dumps it to a *.s2p or *.s1p file.

Series IV (parent of ADS) used to have this functionality however last time I used ADS (a long time ago) they got rid of it.

Some of the newer Agilent PNAs allow you to embed s-parameter blocks into measurement. I suppose you could embed a 2:1 transformer.

If you are a real masochist you can read the VNA correction terms, cascaded on you transformer t-parameters, then write them back into the VNA; I have never tried this. This will do it in real time similar to the PNA. The Maury Microwave FIXTURE software will do all this but it's . If you actually have a transformer you can measure, the same thing can be accomplished by tricking the VNA using it's adaptor removal algorithm. Again, I have not tried this. A guru from Maury Microwave told me about it. It's one of those gems of knowledge that has never been written down, only passed around verbally.

We'll shucks! I just tried the VNA2MWO with MWO 7.5 and it does not seem to work any longer. Back to the drawing board.

The READVNA still seems to work.

#### hititman

##### Newbie level 6
touchstone insertion loss calculation in excel

zhonghua_li said:
1.get the complex impedance of ZL,by the equation of S11 = (ZL - Z0)/(ZL + Z0);
2. calculate the S11 with other Z0 (eg. 100ohm) by the equation of S11 = (ZL - Z0)/(ZL + Z0) too.

it's easy to be calculated by matlab's complex operation

In 1-port measurement, you can do that. However, with 2-port measurement, the measured S11 in 50Ohm system is affected by the load on port2.

For a certain network, a unique Z, Y, or T matrix can be used to fully describe the network. While the Zo needs to be defined for the S matrix, which means S matrix can be different with different Zo. Usually, the S parameters measured by a VNA are normalized to 50Ohm. Several ways to renormalize the data to 100Ohm.

1. Use CAD tools. ADS, Ansoft Designer, Microwave Office, and EldoRF are all able to do the renormalization easily. In ADS, the S2P component can have CITI file as input. Let me know what software is available for you, I can show you how to do this.

2. Mathematic tool (matlab, or even in Excel). Transfer the S parameter matrix to either Z, Y, or T matrix. Then transfer them back to S matrix in reference to the new Z0. You may need a reference book on microwave network for the equations, eg Pozar's Microwave Engineering.

Regards,

hititman

westwind

### westwind

Points: 2

#### zhonghua_li

##### Junior Member level 2
cit files measurements

of course,when you measurement the s11, you can connect a load of 100ohm in port2

#### hititman

##### Newbie level 6
citifile to touchstone

You are right, assuming you have a good 100Ohm load directly attached to the output port. For a two-port network, you may get S11 or S22 data by switching port1 and por2 , but how can you get the insertion loss in reference to 100Ohm?

zhonghua_li said:
of course,when you measurement the s11, you can connect a load of 100ohm in port2

Status
Not open for further replies.