
Here are a number of examples of using atlc . Each example can be found in the examples directory.
Cross section 
Properties 
Efield 
Ex field 
Eyfield 
Voltage 
Energy 
Permittivity 

multidielectric.bmp: C= 93.7871 pF/m L= 279.3847 nH/m Zo= 54.5795 Ohms 







ushape.bmp: C= 76.4283 L= 145.5809 nH/m Zo= 43.6441 Ohms 







coax2.bmp: C= 47.6374 pF/m L= 233.5667 nH/m Zo= 70.0215 Ohms 







veryodd.bmp: C=59.1756 pF/m L= 188.0251 nH/m Zo= 56.3685 Ohms 







twinwire.bmp: C= 63.2867 pF/m L= 175.8111 nH/m Zo= 52.7068 Ohms 






The full details of exactly what the files mean is given here.
In the case of the twinwire, and the mutidielectric, the electric field extends to infinity, as there is no surrounding conductor like in the case of the coax, ushaped conductor of the very odd shape. By putting the twinwire or multidielectric onto a finite grid, which does not extend to infintiy, we have introduced an error. Hopefully this error is very small, but in this case it is not, as we can see from the plots of electric field, that the Efield has not fallen to a negligible amount at the edges. In cases like this, the same bitmap should be drawn on a larger grid, as in the next example.


twinwire2.bmp:With the bitmap drawn on a larger (256x256) image like this, the error introduced by truncating the fields to zero at the edges of the image are smaller. For this simulation, atlc reports C= 57.1575 pF/m L= 194.6639 nH/m Zo= 58.3588 Ohms The results are clearly very different from what the first simulation showed. 
The results with the twinwire show the problem with simulations in which the fields theoretically stretch to infinity. The calculations were then repeated on a 512x512 and 1024x1024 grid. The results for all the calculations are shown below as well as the times taken on a dual processor Sun Ultra 60 (2 x 300 MHz CPUs).
128x128 (twinwire.bmp) 
Zo= 52.7068 Ohms 
run time = 1 s. 
256x256 (twinwire2.bmp) 
Zo= 58.3588 Ohms 
run time = 8 s 
512x512 (twinwire3.bmp) 
Zo= 60.7080 Ohms 
run time = 82 s. 
1024x1024 (twinwire4.bmp) 
Zo= 61.9047 Ohms 
run time = 834 s. 
2048x2048 (twinwire5.bmp) 
Zo= 62.3786 Ohms 
run time = 4300 s. 
Inspecting the results one can see that as the grid size is increased, so the results converge. and for all practical results, the last result should be sufficiently accurate. Unfortunately it was obtained at the expense of a lot of CPU time.
The problem with the fields being truncated does not exist with fully enclosed transmission lines. As is shown in the accuracy section, the results are very acurate on other such lines in a very short time.
atlc is written and supported by Dr. David Kirkby (G8WRB) It it issued under the GNU General Public License
Return to the atlc homepage
