Combline Filter Using Ansoft HFSS

site:www.edaboard.com hfss bandwidth

Hi guys,

I am designing a bandpass combline filter using Ansoft HFSS. I am using rectangular bars and stripline for the implementation. My specifications are as follows:

Center Frequency = 4 GHz
Passband Bandwidth = 80 MHz
Passband Insertion Loss = 0.5 dB

I have used electric walls to reduce the coupling between the resonators but I'm not able to achieve the above specified bandwidth. I am attaching my HFSS program and my result. Can you please help out of this problem as I have only 4 days to submit it? I would appreciate and very grateful to the one who can help me.

combline filter

Here is my HFSS program!

hi
i have seen ur file.its very complex.first try to do for less no. of taps.have u calculated the dimensions urself or its u have just modeled it with arbitrary dimensions?
lets see wat can i do .u continue ur effort.
take care

Hi Anoz,

It's nice that u replied me n I appreciate it very much. I calculated all the values of lengths, widths n heights using Theory n Design of Microwave Filter by Ian Hunter and my supervisor for this project is Ian Hunter himself. He agreed that my design is good. It is an 8th order combline filter with 2 transformers at the start and end. I have used electric walls to reduce the coupling between the resonators so that the bandwidth will reduce. At start I was getting a bandwidth of around 500 MHz and now it's around 200 MHz. So still i need to make it around 80 MHz. I've tried to reduce the number of rectangular bars but I've reached a situation where I can't reduce the bandwidth anymor. Is there any other technique to reduce the bandwidth? I'm trying hard evryday with this program and will still try in the future.

I thank you for your email and I hope you'll get something for me.



Best Regards,

Vish2808

Re: Combline Filter in Ansoft HFSS

Hi,

Can someone tell me how to reduce the coupling between two resonators in a stripline implementation? I want to reduce the bandwidth of my filter.


Thanks!

Hi,

Can someone tell me how to reduce the coupling between two resonators in a stripline implementation? I want to reduce the bandwidth of my filter.


Thanks!

Hello Vish.
Just take a look is this discussion:


There are papers that I proposed in this subject. You will also find the famous basic program that can make this design much easier for you. At least as a starting hint.
There are also references for tutorials etc..

Hope it helps.
S.

Re: Combline Filter in Ansoft HFSS

Hello Vish.
Just take a look is this discussion:


There are papers that I proposed in this subject. You will also find the famous basic program that can make this design much easier for you. At least as a starting hint.
There are also references for tutorials etc..

Hope it helps.
S.

Try adjusting the distance between the bars and your "capacitor". This should have major impact on the coupling - bandwidth... Probably moving it closer to the bars should reduce the coupling. Probably making it wider should also reduce the coupling... Increasing your walls heights should further reduce the coupling...

Thanks a lot Sinatra!

Added after 3 minutes:

Hi rfmw,

Thanks for ur reply! Changing the distance between the resonators and capacitor changes the centre frequency and the bandwidth remains similar. also i tried increasing the wall's height, the electric coupling becomes greater than magnetic coupling. hence the bandwidth increases. So may be increasing the distance between the resonators will reduce the bandwidth. thanks again because your suggestions have made me think about it.

Yeah, sorry if my notes were in contrast with your simulations (i did use probably .

Your structure resembles the usual resonator filter (metalic rods in a rectangular metalic enclosure, but has unusual implementation, that is resonators are cubes, etc... At least i haven't seen implementation like this yet. Anyway placing resonators further apart should definitely reduce the bandwidth. I find it strange that increasing the wall's heights increase the bandwidth... the field distribution in your structure is not so obvious to me, so I admit that I'm missing something here Good luck with your project.

Hi rfmw,

Your notes did help me n thank u that. if i increase the height of the wall, the bandwidth will not become smaller because it's the optimum height, that is, the elctric coupling and magnetic coupling almost cancel each other. I've used an IEEE paper named "Full Wave Modeling of Conducting Posts in Rectangular Waveguides and Its Applications to slot Coupled Combline Filters". Thanks again for taking time to have a look at my design n reply me.

Hi Sinatra,

Can u please tell where can i get the following paper:

E.G. Cristal “Capacity Coupling Shortens Combline Filters” Microwaves, Dec 1967, p 44


Thanks in advance!

Hello Vish.
I have lots of papers in this field.
I may have it somewhere. I will take a look and tell u later if I find it.
Cheers
S.

Hello, I am designing a combline filter. The first resonator is fed with a coaxial cable. Where should i place the wave port? in one of the faces of the dielectric of the coax? the thing is that the results of the simulations are different depending on the length of this cable, which shouldnt be as far as i know. the input coupling is made by a probe coupling, its not tapped, so i dont know where shoud the inner condutor of the coax go to.

Could anybody help me please??