Radio amateurs knows the problem: To choose a suitable cable or calculate the effective radiation power of our station, it’s very usefull to calculate the attenuation of the desired cable. And here is a small tool for this task.
The motivation to create this tool was that with many cables only attenuation values of some single frequencies are available. To calculate the attenuation on an arbitrary frequency we must interpolate. To be to exact as possible and to allow also an extrapolation (outside the given frequencies) the tool first calculates the cable equation (see below) from the given frequency-attenuation pairs.
And this is the main advantage of this tool: To calculate the cable equation for the attentuation as a function from the frequency.
Little theory to the cable attenuation
To understand the program and to be able to recognize any malfunction (hey, it’s always good to know what happens) I tell something about the theory behind.
The attenuation of the cable as a function of th efrequency can be expressed as shown in the following equation:
D is the attenuation per meter in dB and f of course the frequency . k1 is the attenuation at DC caused by the DC resistance. k2 is an expression for the dielectric loss. k3 determines the additional attenuation as consequence of the Skin-effect.
For every cable this three coefficients must be calculated. To do it, we enter the known attenuation-frequency pairs from the data sheet (or the dealers cataloque) into the cable dialog and press “Calculate”. The program uses the least mean square method to fit the equation into the given supporting points.
It’s important to check the calculated coefficients k1, k2, k3 and the calculated attenuation values at the known frequencies because some published attenuation values of cables are terrible inexact.
First tests has shown that sometimes negative coefficients are calculated. Of course this is physically nonsense. One case is that dielectric losses of good cables are so low that the measurement failures of the entered attenuation-frequency pairs lead to a slighly negative k2. That’s not dramatic and the coefficients can normally be used.
On an other case the calculated attenuation at 160m was -2dB. Of course a negative attenuation is nonsense: The reason was that there was no supporting value (attenuation- frequency pair entered in the cable dialog) inside the short wave frequency range. An extrapolation is always problematic and only possible if the known values are very exact.
To make long things short: Be always critical and look for better attenuation-frequency pairs if the calculated coefficients or calculated values are not realistic.
A good way to verify the results is to calculate the attenuation at the frequencies of the known frequency-attenuation pairs.
Installation of the program
I thought long time about making an installation program. At the end I decided that the program is small enough and the user technically educated that a simple ZIP file will be OK. The program does not change anything in the registry and I think it will only be used from time to time.
So, please unzip the attcalc.zip file into an empty directory and launch the coaxatt.exe file. Hint for Windows7 user: Do not use a directory under the \programs directory because it will be write protected after installation and new coax cables cannot be saved.
For deinstallation, simply delete the directory with the program.
Hints for the usage
The usage is mostly self explaining: Select coax cable, enter frequency and length and read the calculatd attenuation.
The combo box “band” is filled with typical ham radion bands and is intended to set the frequency box.
Is the desired cable missing it must be createdby clicking “New”. Look for the data sheet and enter the given frequency-attenuation pairs and a clear name. At least three frequancy-attenuation pairs are required but as more as better.
After pressing the “Calculate” button the coefficient are calculated and the residual error is shown. This residual error is the remaining error after fitting the coefficients for the given frequency-attenuation pairs.
- With new cables test the quality of the calculation by calculating the attenuation at the known frequencies.
- Please check carefully what cable you have. There are often cables with similar names but different properties (for example RG-213U vs. RG-213UBX)
- Some given attenuation-frequency pairs are terrible inexact. If the above tests criteria results in a too inexact result look for better values. Often it can be found in the internet.
- If you create a new cable with the “New” button, please send me the created .cbl file (unter cables in the program directory). I can add this cable to the distribution.
– Bitte die Nutzungsbedingungen zur Kenntnis nehmen –
Unzip the attcalc1.zip into an empty directory and launch coaxatt.exe.
The program is compiled for Windows. I can create a Linux version on request. Please ask.
The implementation of the least mean square method is from Joachim Wuttke.