Band Pass Filter with very low IL, sharp rejection and small size.

[ali_cmi]

Requirement: [h=3]Band Pass Filters (BPFs) with very low Insertion Loss (IL) (Less than 0.9 dB practically), sharp rejection (Greater than 45 dB in stopband), wide stopabnd and small size (Not greater than 3 by 3 in.)[/h]
Hello,

I hope everyone is doing well. I am kind of stuck in a task given to me. I have to design a very fine tuned filter with above mentioned specs. Can you suggest any filter topology/structure (uStrip, Stripline etc.) which would fulfill above mentioned requirements? It would be much appreciated. I have searched a lot of IEEE papers and implemented a few of them on ADS but it does not fulfill my required specifications.

Currently, my Center Frequency is 1.09 GHz, less than 1 dB Passband 270 MHz and StopBand 565 MHz.

 

[BigBoss]

I think you can playaround this filter to optimize in according with your specs.
Dimensions are 24.37mm x 34.23mm on a substrate Alumina er=9.8, H=0.5mm, Microstripline configuration with 50 Ohm termination impedances.
If you have Sonnet Lite use the attached Sonnet project file, you can also optimize it by yourself.

 

[ali_cmi]

@BigBoss

Thank you for your reply, sir.

I have never used Sonnet. ADS fan. But will see to it.

Would you please tell me how come the size of this Tapped Hair-Pin filter reduced so much at this frequency? An insight would be great.

 

[BigBoss]

@BigBoss

Thank you for your reply, sir.

I have never used Sonnet. ADS fan. But will see to it.

Would you please tell me how come the size of this Tapped Hair-Pin filter reduced so much at this frequency? An insight would be great.

The high di-electric coefficient substrate is the key.If you move it up, dimensions go down..

 

[ali_cmi]

BigBoss, do you have ADS file of this? or in any other form which can be imported to gerber? I ahve sonnet free version which does not analyze this file due to limitations in free version.

- - - Updated - - -

Also, are conductor losses included in the above scenario?

- - - Updated - - -

BigBoss: I simulated your filter in ADS including conductor losses. It fails the insertion loss requriement.

 

[biff44]

most would do this with ceramic resonators. 0.9 dB IL depends on the specific requirements for bandwidth and beginning of the stop band.

 

[BigBoss]

BigBoss, do you have ADS file of this? or in any other form which can be imported to gerber? I ahve sonnet free version which does not analyze this file due to limitations in free version.

- - - Updated - - -

Also, are conductor losses included in the above scenario?

- - - Updated - - -

BigBoss: I simulated your filter in ADS including conductor losses. It fails the insertion loss requriement.

I have attached it as DXF file.I will simulate it today in ADS with the same condition.I can make it as zapped project file.
Give me some time..

 

[BigBoss]

Final simulation in ADS..
Response is not too bad, I have also attached project zapped file.
Good luck

 

[ali_cmi]

Thank you Bigboss for giving me the ADS project. However, it is unfortunate that I cannot use that filter because of its insertion loss. Insertion Loss being less than 1 dB is a very strict requirement even 0.1 dB greater than it would cause me to reconsider my options.

 

[chandu4kalyan]

View attachment 1.09ghz.pdf see this sir it will be help full to you, it's done by using ads

 

[andy_larkeen]

May I ask for what you need such a filter?
Honestly it can be a challange to design a filter with your specification and it won't take a day or two to do it.
For such low insertion loss I recommended a Substrate Integrated Waveguide Technology (SIW) with quarter miniaturization. To manufacture a filter only Rogers (expensive but with low loss) substrate should be considered.

 

[BigBoss]

View attachment 96151 see this sir it will be help full to you, it's done by using ads

Replace "real components or models" with ideal synthesized components in your filter and see what'll happen..