W8WWV - N2PK VNA - cialog Modified for the VNA

Greg Ordy


Several years ago, I wrote a Windows program that works with the AEA CIA-HF. This program, cialog (complex impedance analyzer logger), uploads data from the CIA-HF, and displays it as a graph of data versus frequency using a rectangular coordinate system. At version 1.08 of the program, I enhanced it to optionally work with the N2PK VNA.  This page describes the changes made to cialog so that it will work with the VNA. Ultimately, to use the program, you have to go to the cialog download page, and download the software from that page. The important point to remember is that there is one single program that works with both the CIA-HF and the VNA.

Due to the capabilities of cialog, its use with the VNA will be most useful when performing reflection measurements made with a bridge, such as the T1-6T. Further restrictions imposed by the CIA-HF (and, therefore, cialog), limit the data range to values typically found on antennas, and transmission lines connected to antennas. Resistance and reactance are limited to a range of plus and minus 999.9 Ohms. The frequency sampling widths are also adopted from the CIA-HF. The VNA has a  frequency resolution of less then 1 Hz. The CIA-HF has a minimum resolution of 1 KHz, and a set of fixed frequency steps. The CIA-HF frequency resolution is used by cialog, and, therefore, constrains the VNA in this program. For antenna-related work, however, this constraint is not unreasonable.

VNAccess and VNAgra

The extensions made to cialog, in order to support the VNA as an input device, are built upon the VNAccess and VNAgra programming libraries. Beginning with version 1.08 of cialog, the required files will be part of the download, and are located in the same folder as the cialog executable file. It is also required that the program have access to the parallel I/O port on the computer. This may require a special driver on some versions of Windows. We have been using UserPort 1.0.

Data Flow

The CIA-HF measures and uploads SWR, resistance, reactance, and impedance. The VNA measures the complex reflection coefficient, which is then turned into SWR, resistance, and reactance, by formula. Impedance is computed to be the square root of the sum of the squares of resistance and reactance.


In order to keep the program changes simple and small, the cialog internal data representation is used to describe readings uploaded from the VNA. The does the VNA a real disservice, since it can produce very accurate results, over a very wide range, with a great deal of relative resolution. The internal representation of resistance, reactance, and impedance, is plus or minus 999.9 Ohms, with a resolution of 0.1 Ohms. All VNA data will be rounded or truncated to conform to that representation. SWR is represented as a value with a maximum of 99.99, and a resolution of 0.01 units.

OSL Calibration

The VNA requires OSL (open, short , load) calibration. Calibration must be performed each time the capture specification changes. If the capture specification does not change, and if calibration data has been uploaded, then the second and subsequent captures do not upload calibration data.

Feed System Models

cialog supports a feed system modeling facility which transforms measurements made at the input of a transmission line to the load end.  In the case of the CIA-HF, this is the only technique available to compensate for the impedance transformation caused by the transmission line.

The VNA is far more flexible. In addition to the feed system transformation capability used by the CIA-HF, the VNA supports remote OSL calibration.  In this use model, a transmission line is connected to the VNA, and the OSL calibration is performed at the load end of the transmission line. No characterization of the line is needed, although it should be a 50 Ohm line of the best possible quality. This approach moves the reference plane from the VNA to the load end of the cable.

To summarize, there are two different techniques which can be used to make measurements at the end of the transmission line connected to the VNA.

In the first technique, the transmission line is specified as a feed system in cialog. The length and type of cable must be known. cialog will use lossy transmission line equations to transform the data measured at the VNA to the load end of the cable. In the second technique, the cable is connected to the VNA, and a feed system is not selected in cialog. The OSL calibration and device under test measurements are made at the load end of the cable.

Each approach has advantages and disadvantages. Here are a few of the points to keep in mind.

  1. Remote OSL calibration should provide the most accurate results.

  2. In order to use remote OSL calibration, it is necessary to physically reach the load end of the cable. If the cable is connected to an antenna on top of a tower, for example, it may be very difficult to perform OSL calibration where the antenna connects to the transmission line.

  3. The loss on the cable can only be computed when it is treated as an explicit feed system.

  4. The length of the transmission line must be known in order to use the feed system models built into cialog. Errors in the length  of even a few inches can shift the data by many Ohms. Even if the length is known precisely, variations in the cable characteristics may change the computed transformation on the cable.

It is also possible to utilize both techniques. For example, OSL could be used to determine the value of some typical load at the load side of a transmission line. The remote OSL calibration is then replaced with a specific feed system model which describes that line. The length of the line in cialog is adjusted until the computed results agree with the remote OSL results. This approach uses one technique to help configure the other.


Here is an example graph. The subject is the same length of 50 Ohm transmission line that I have used on a number of web pages. The cable is terminated in a 25 Ohm load, which creates a range of complex impedance values as a function of frequency. In addition, the SWR will be very close to 2:1.

cialog Example: Data from the VNA

Back to my Software Contributions Page

Back to my Experimentation Page

Back to my Amateur Radio Page

Last update: Wednesday, October 13, 2004 03:16:15 PM
Back to my Home Page