Accuracy issue regarding HFSS versus IE3D

I have a question for the simulation experts. I computed the radiation efficiency of an antenna in HFSS and plotted the results in the file attached. I also used IE3D to get the antenna pattern of the same antenna but when I plotted radiating efficiency versus frequecny in IE3D, the results are not similar to the HFSS results. IE3D also has another quantity it calls the antenna efficiency. I am not sure what the difference between the two types of efficiencies defined in IE3D is, and which one can be compared to the radiation efficiency but neither of them compares to the radiation efficiency obtained using HFSS in this case (see word doc file attached). What could be the problem? Anyone knows? I appreciate your response.
Thanks,
Waveplumber

Hi, Wavenumber:

1. What antenna is it? If it is a patch antenna with a reasonable size substrate, the radiation efficiency should normally between 50% to 90% in the resonance bandwidth. Surface wave is an important source of loss to efficiency. IE3D is normally quite accurate in predicting the radiaiton efficiency for such a case. If it is an antenna in free space or an antenna with the substrate with the same size or slighly bigger size than the patch, the radiation efficiency should be between 95% to 100%. For such a situation, you should use finite substrate modeling on IE3D in order to get accurate radiation efficiency and it is also quite accurate.

I looked at your data. It is changing much with frequency and there are many up and down's. I seldom see anything like this. I am not sure what antenna is or there are any problem in your model.

2. On IE3D, we defined two efficiency values: radiation efficiency and antenna efficiency. Radiation_Efficiency is defined as Radiated_Power / Net_Input_Power. Antenna_Efficiency is defined as Radiated_Power / Incident_Power while Net_Input_Power = Incident_Power - Reflected_Power. In some sense, the mismatch loss is included in the antenna efficiency.

Best regards.

Thanks for your reply, Jian. You are the man. Could you please explain the two scenarios in which radiation eficiency is between 50% and 90% and the one in which it is 95% to 100% again. My antenna has a PCB antenna on a 5cm by 5cm substrate, it is a microstrip-fed radiating patch. In HFSS, I could cut the substrate but I don't know how to model a finite size substrate in IE3D version XML that we have. However, the results still look too discrepant for the difference to be explained by the finite size substrate! I think I got my HFSS simulation right. Would like to point out that for the antenna in question, the S11 characteristic is similar in HFSS and IE3D, though.

Hi, Wavenumber: Please send me the description of your antenna and, possibly, the IE3D geometry files (jian@zeland.com).