Leaky Coax Simulation Power Loss Error
Altair EmployeeI'm currently modeling a segment of leaky coax using MoM. The segment has a waveguide source on one end and is terminated with a 50 Ohm structure on the other. I keep receiving the following warning: 'WARNING 3397: The power loss is larger than the active power'. The warning makes me question the validity of the results (which are not exactly what I expected numerically).
In the past I have found this warning has occurred with this model with lossless dielectrics and has been fixed by making the loss tangent of the dielectrics a small nonzero number. In this case however, I am using copper as the only lossy medium (as a friend did in a simulation of the same structure with COMSOL). Why are the losses exceeding the power inputs and does this incorrect result have significant physical meaning? Obviously the model has nonphysical results, but does that invalidate the results entirely? What are the unrealistic features of the model causing this incorrect calculation? Can this situation be resolved without making the dielectrics lossy (to maintain equivalence with my friend's COMSOL model)?
Even with lossless dielectrics, how can the solution of the wave equations result in the power loss being noticeably (~1%) greater than the active power?
Answers
-
Dear riiily
Your solution shows a total source power of 10.091 mW being applied. With 37.653 uW of losses in the copper and only 1.020 uW being radiated, we would therefore expect most of the power to be dissipated in the 52 - j24 ohm load you have applied at the end of your coax.
(If we assume these values to be correct) the expected power dissipated in the load would be (P_source - P_copper_losses - P_radiated = ) 10.052 mW. FEKO reports power dissipated in the load as 10.134 mW. As this is higher than the source power, you receive this warning.
It should be noted that the value reported for the power dissipated in the load is very close to the expected value (assuming the copper losses and radiated power is correct). The reported value is only 0.8% higher than expected. It should be noted that the power in the load is directly related to the current flowing through your wire section. As such you should consider modeling this more accurately.
I would suggest that you manually reduce the mesh size in the areas of the waveguide source and your wire section that has the load applied. You should also consider adding a plane of magnetic symmetry for the Y=0 plane. All of this should improve the accuracy.
0 -
Thanks Johan,
I've tried your suggestions and still receive the same warning stating the power losses exceed the active power. Does the warning invalidate my results? How close do the active and dissipated powers have to be for FEKO to be satisfied (and to not return this error)? I mainly wonder because the ~0.8% discrepancy between the active power and the total power loss is greater than the total power radiated. I am interested in simulating far field gain of the cable segment.
I've included an updated model. The model has your suggestions included. I also included an error estimation request. The error is highest on the end segments of the line containing the wire port load. I decreased the wire segment length and the global wire radius which had some effect on the wire error but not much. Are there other methods of decreasing the model errors?
0 -
riiily, in this case the warning does not invalidate your results. It only points to the fact that the results are numerically very sensitive.
With about 10mW being applied and only about 1uW (0.1%) being radiated through the aperture, the source and load (loss) powers should be virtually identical. A very high level of accuracy is required to get these two numbers to be within the expected 0.1% of each other. Practically speaking, I would be surprised if it was possible to measure the source and load powers of such a physical model to such accuracy.
Another option would be to replace your wire section that represents the load also with a waveguide port. This might be a more accurate representation of the load that would actually be seen by the end of this line section.
0 -
Hi I have a problem.I am manufacturing a coax cable rg6 but it came a error at the analyzer.It have a pick and do not pass the test.The cable is showing from 13 to 19 dB but I need 20 min.I have tested all of the cable but the problem is there.We checked all the lines for some mechanical problems but no.If somebody can help me write me a mail at otchev.radoslav@gmail.com Thanks
0