surface waves
It really stinks that I have to use the VECS (very expensive commercial software) to do my research. Not only does it take a lot of money to pay for it, but there is a decent amount of training that's required to use it effectively. Unfortunately, the policy where I work is that they don't like to pay for training for students. I personally think this is a bit of stupid policy (as I think a better policy is that the grad student going to training should write up a manual to be used after this student has left).
Anyway, the problem is that, as I mentioned, I'm putting my antennas in dielectrics. There is cheaper software we can use that is "2 1/2 D" software...short for 2 1/2 dimensional. What that means is you can create a structure that has any shape in two dimensions. You can then put that antenna into a dielectric. The problem is that you can't specify the shape of the dielectric. It's assumed to be an "infinite plane" or a big, flat slice of your dielectric.
I saw someone design an antenna using this software. When the antenna was encapsulated in something with a different shape, it didn't work as well.
The problem is surface waves. You know how when you shine light into a diamond, it bounces around (refracts)? Electromagnetic energy can do the same thing in a lot of substances, including plastics, glass, even wood. How it bounces around depends on the frequency of your EM wave, a property of your dielectric called the refractive index (or your permittivity-they're related), and the shape of your dielectric.
The energy bounces off the edges and faces of the dielectric, and it can actually interfere with itself. This can mess with how well an antenna works, which is why I decided to use the VECS rather than our 2 1/2 D software. I wanted to design something I knew would work with this particular dielectric and shape. Without training, however, I spent a lot of time spinning my wheels.
The other problem is that the VECS is incredibly computationally intensive. I probably wouldn't be finishing up my work in time if it weren't for the fact that our group recently got what I've affectionately named "Computerzilla". It's amazing what you can do with 32 GB of RAM.
Anyway, the problem is that, as I mentioned, I'm putting my antennas in dielectrics. There is cheaper software we can use that is "2 1/2 D" software...short for 2 1/2 dimensional. What that means is you can create a structure that has any shape in two dimensions. You can then put that antenna into a dielectric. The problem is that you can't specify the shape of the dielectric. It's assumed to be an "infinite plane" or a big, flat slice of your dielectric.
I saw someone design an antenna using this software. When the antenna was encapsulated in something with a different shape, it didn't work as well.
The problem is surface waves. You know how when you shine light into a diamond, it bounces around (refracts)? Electromagnetic energy can do the same thing in a lot of substances, including plastics, glass, even wood. How it bounces around depends on the frequency of your EM wave, a property of your dielectric called the refractive index (or your permittivity-they're related), and the shape of your dielectric.
The energy bounces off the edges and faces of the dielectric, and it can actually interfere with itself. This can mess with how well an antenna works, which is why I decided to use the VECS rather than our 2 1/2 D software. I wanted to design something I knew would work with this particular dielectric and shape. Without training, however, I spent a lot of time spinning my wheels.
The other problem is that the VECS is incredibly computationally intensive. I probably wouldn't be finishing up my work in time if it weren't for the fact that our group recently got what I've affectionately named "Computerzilla". It's amazing what you can do with 32 GB of RAM.