Microstrip Filters - RF Cafe Forums

The original RF Cafe Forums were shut down in late 2012 due to maintenance issues - primarily having to spend time purging garbage posts from the board. At some point I might start the RF Cafe Forums again if the phpBB software gets better at filtering spam.

Below are the old forum threads, including responses to the original posts.

-- Amateur Radio
-- Anecdotes, Gripes & Humor
-- Antennas
-- CAE, CAD, & Software
-- Circuits & Components
-- Employment & Interviews

-- Miscellany
-- Swap Shop
-- Systems
-- Test & Measurement
-- Webmaster

Antonio
Post subject: Microstrip Filters Posted: Tue Jun 19, 2007 3:57 am

Captain

Joined: Sat Jun 16, 2007 12:26 am
Posts: 12
Hi All,

I am trying to desing a Microstrip bandpass filter using edge coupled line filter technique.

This filter is a narrow band filter. I did lot of trials to design it but the problem is the Insertion Loss in the band is too high. I am not able to reduce it.

I want to know that is it true that Edge coupled line filter technique is not suitable for a narrowband filter?

If it is not suitable then which technique will be suitable :
a) End coupled structure
b) Hairpin line structure
c) Comb line structure

Can anyone help me in this problem??

Is there any book or link in the web from which I can get useful information for designing a narrowband filter??

Top

nubbage
Post subject: Posted: Tue Jun 19, 2007 10:13 am

General

Joined: Fri Feb 17, 2006 12:07 pm
Posts: 218
Location: London UK
Hi Antonio
For a coupled multi-element filter of the Chebychev type, the in-band insertion loss is approximately

L = (4.34/(w*Qu)* SUM(g) ......dB

where w = bandwidth/centre freq
Qu is the un-loaded resonator Q
and SUM(g) is the sum of all the element values in the low-pass prototype.
Thus if w is small (say < 0.2) then for a given Qu the loss is high.
With a strip-line the Qu is always lower than say copper bars, due to film skin resistance and dliectric dissipation.
The best design for really narrow band filters will always be coupled cavities, such as used in narrow-band diplexers.

Top

IR
Post subject: Posted: Tue Jun 19, 2007 2:30 pm

Site Admin

Joined: Mon Jun 27, 2005 2:02 pm
Posts: 373
Location: Germany
Hello nubbage,

I have one question concerning the term Sum (g) in the equation. Can you please elaborate on that?

Thank you!

Top

Antonio
Post subject: Microstrip Filter ImplementationPosted: Tue Jun 19, 2007 10:21 pm

Captain

Joined: Sat Jun 16, 2007 12:26 am
Posts: 12
Hello nubbage,

Thanks for your reply. But my question is still unanswered. Due to resource constraint I am trying to desing it as edge coupled filter. I want to know that among the listed structure which is suitable for designing a narrow bandpass filter.
a) Edge coupled
b) End coupled
c) Hairpin line
d) Combline structure.

I am able to design this filter but how to calculate the length, width and spacing for different structures is still not clear.

Is there any book or website where I can get detail explanation on all these structures to calculate the length, width and spacing for diiferent structures.

Thanking you,
Antonio

Top

nubbage
Post subject: Posted: Wed Jun 20, 2007 4:07 am

General

Joined: Fri Feb 17, 2006 12:07 pm
Posts: 218
Location: London UK
Hi IR
The SUM(g) term is the sum of all the element values of the normalised low pass filter prototype. An N element LPF prototype will have N values for the reactances of each L and C used.
I got the formula from H.L.Schumacher's paper "Understanding Filters and Multiplexers" Microwave System News Vol 13 No 12, but almost certainly it is stated in matthei Young and Jones as well (which is more readily available).

Top

nubbage
Post subject: Posted: Wed Jun 20, 2007 4:29 am

General

Joined: Fri Feb 17, 2006 12:07 pm
Posts: 218
Location: London UK
Hi Antonio
First off, my last posting should have said an N element LPF has N+! values of g.
Secondly, we would need to know how "narrow" is your "narrow" when you say narrow band?
A good source of filter design information is the Radio Society of Great Britain (RSGB) publication "Microwave Handbook" and in particular Volume 2 chapter 12. It is a largely non mathematical treatment and explains the physics going on. An example is given of a coupled element coaxial multi-cavity filter, resembling a comb-line type, at 1GHz with a bandwidth of 10MHz (a narrow band type in my thinking) that has a band-centre loss of 1.0dB. However, the resonator Qu is estimated to be 1400 to achieve that loss, and you stand no chance of achieving that with strip-line substrate, no matter whether it is edge, face or end-coupled.
The chapter also covers stripline edge coupled designs and presents a BASIC routine for their design.
I am not familiar with hairpin types.
The classical text book on microwave filter design is Matthei Young and Jones without a doubt. A copy of that might help.

Posted 11/12/2012