CST - BLE antenna with wood

Hi,

We have simulated a ceramic BLE antenna in two scenarios: with and without wood nearby. In the simulation, we observed that the presence of wood causes a shift in the S-parameter and reduces the antenna efficiency by about 30%. Without wood, the efficiency is around 70%.

However, in our real lab tests, the S-parameter only shifts when the PCB is in direct contact with the wood. If there is any separation, the effect is minimal or nonexistent.

Is it normal to see such a discrepancy between simulation and empirical results? Does it make sense that wood has such a strong impact in simulation but only affects performance in practice when there's direct contact? What could be causing this difference?

-Cases in simulation:

[1] Antenna + Full Wood Block

[2] Antenna + Wood with Cutout

[3] Wood with Cutout + Raised PCB (2cm)

[4] Bigger bounding box

CST Studio Suite