Looking for good software to design microwave cavity filters

[Tavarnelle]

site:edaboard.com wasp-net filter

Hi all
I am looking for a good software to design microwave cavity filters
(iris,interdigital etc.).
Does someone can help me,please?
Greeting Tavarnelle.

 

[webdog]

www.wa4dsy.net/cgi-bin/idbpf

Take a look at =Mfilter= by Eagleware. It's a very easy to use and high accuracy program. You find more information at h**p://www.eagleware.com
After designing the filter, simultate and adjust the fitler using physical model and then finally check your design with an EM-simulator and you will have a working microwave filter for sure.


A very simple basic program for DOS was published in Ham Radio years ago, but I did realize a 600 MHz narrow band interdigital filter: I had no problems to acheive my design goals. The author claimed it would work above 1 GHz, and showed a 1300 MHz filter. I migh have a program listing if you are interested.

Good Luck
/WD

 

[Tavarnelle]

Excuse me, webdog,
I know that eagleware permits sintesys and simulation of microstrip (on substrate) and not on AIR... Is it right?

I'd like to design filters on air, for example using a wave guide with iris and / or screws

Btw, I am interested on your software, if it design such filters :wink:

Thank you very much!!

Tavarnelle.

 

[u_f_o]

waveguide filters

hi Tavarnelle!
for waveguide filters you can use Zeland COCAFIL, WASP NET and Mician MicrowaveWizard.
the last is the best (imho).
unfortunately all programs are commercial.
best regards.
u_f_o

 

[webdog]

Excuse me, webdog,
I know that eagleware permits sintesys and simulation of microstrip (on substrate) and not on AIR... Is it right?

I'd like to design filters on air, for example using a wave guide with iris and / or screws

Btw, I am interested on your software, if it design such filters :wink:

Thank you very much!!

Tavarnelle.

Mfilter do design whatever substrate you want! It can be teflon or air!
If you choose the type correctly you can design very easy an interdigital filter made in a metal box with rods and use a setting screw at the hot end to adjust the resonant frequency of each resonator. It does NOT compute for helix type of resonator.

Eagleware do offer a 30 days try-out free, why not have a look at their web-site? They even offer a very good e-book on microwave filter design for free! It is included in their help-files and I think you can find it at the e-book forum here on elektroda "HF Filter Design and Computer Simulation" by Randall W Rhea

/WD

 

[nandopg]

I've been using Parfil and Elliptic from WaveCon (http://www.waveconsoft.com/) very successfully up to 12GHz.

NandoPG

 

[u_f_o]

hi!

I've been using Parfil and Elliptic from WaveCon (http://www.waveconsoft.com/) very successfully up to 12GHz.
NandoPG

it seems these programs have bugs and aren't updated since 2000.
i think you should check its results with some em-simulator. i mean microwave filters.
these programs produce incorrect dxf-files. for example "parallel coupled transformer in" filter dxf-file has incorrect resonator lengthes and resonator number (if your number is greater then 4).
but i agree, these programs are very useful for start point synthesis.
best regards.
u_f_o

 

[nandopg]

Dear friend:
The filter's dimensions outcome by Parfil and Elliptic are very exact!! I already did dozens of microwave filters using these programs. As to updates I don't know, as I got the program by the end of in 1997 and from them on I am very happy with the results I am getting.
I never used dxf files generated by these programs or any other once the dxf standard has been suffering lots of changes along the time.
Again, I reinforce that if somebody needs a good program to design coax microwave filters Parfil and Elliptic are the right choice. Also I would like to ask to those members without the required experience to help others to be quiet.

NandoPG

 

[u_f_o]

hi!

Again, I reinforce that if somebody needs a good program to design COAX microwave filters Parfil and Elliptic are the right choice.

very valuable revision !

I'd like to design filters on air, for example using a wave guide with iris and / or screws.

but our object is not quite coax.

Also I would like to ask to those members without the required experience to help others to be quiet.

very strange declaration from person who uses programs five years and seems did not try to use them for microstrip filters!
ps: i'm very happy to know that Parfil and Elliptic produce very exact results for coax filters.
best regards.
u_f_o

 

[nandopg]

My dear u_f_o,
If you need coax filters build up on cavities, as it is the case in the first post of this thread, you have to use Parfil or Elliptic. No doubts.
The person that post for the first time asked for cavity filters help, right? So, you have to talk about what? Cavity filters, right?
But if you don't have nothing to say about cavity filters, stay quiet!

NandoPG

 

[flatulent]

tangents

There is nothing wrong with going off on tangents as long as it contributes to people's understanding of the subject. I learned about two filter programs that I was not aware of. So far, no one has talked about the original subject of real waveguide cavity filters which are six sided metal boxes which couple energy to each other through holes in their common walls.

 

[nandopg]

Dear flatulent(?)
Please, take a look in the first post of this thread. What do you conclude?

NandoPG

 

[pewang]

eagleware's filter book is very good design reference.

 

[flatulent]

conclusion

Can anyone give advice on six sided metal boxes with coupling holes (iris) between them? There has been discussion on waveguide filter programs that have periodic disturbances in a waveguide and analysis programs, but no discussion of the details of programs for genuine resoanant cavities with light E field coupling between them. I have tried mican and found the filter design was very obtuse and there were no error messages to tell you why it would not go to the next stage.


https://www.engineers.com/ESEPTUM/eseptum.htm is another program but I do not know how well it works.

 

[u_f_o]

my dear, dear, dearest NandoPG!

My dear u_f_o,
If you need coax filters build up on cavities, as it is the case in the first post of this thread, you have to use Parfil or Elliptic. No doubts.

i need not coax filters! look please in the first and third posts of this thread. both post's author is Tavarnelle. do you see word "coax"? i can see words "cavity", "waveguide" and "irises". yes, i can see word "interdigital" too. due to this word you have right tell us your opinions. but you have no right to shut up other people, having greater experience in filter design. look at following inequality:
cavity filters >> coax + waveguide filters >> coax filters
is it right? he he.
stop your advertisment of the bugged expensive commercial programs, stop your spam, PM me, if you want know my opinion about you.
best regards.
u_f_o
ps: when i've desined my first interdigital "coax" filter (in 1974), i've used pen, paper, handbook and electromechanical calculator.

 

[u_f_o]

hello flatulent!
- about eSeptum: seems it use approximate closed form formulas for irises and is not enough accurate. (i tried v3.2, i compared with mician and cst mwo results and with experimental data too).
- about your mician project: what kind of irises had you used? what mician version do you use? please report me your filter parameters, and i'll try it with version which i have. may be your requirements aren't realizable with your irises? if you can pm me your project.
best regards.
u_f_o

 

[nandopg]

Dear UFO,

1- Very interesting the equation you posted above. Unfortunately I could not understand its full meaning. But I will try harder. Sorry.

2- Exactly in 1974 I was using an IBM /360 to write my programs on microwave structures and not only paper, handbook and electromechanical calculator. I am sorry again.

3- It seems you never use Parfil or Elliptic before. So your experience doesn't authorize you to say the program doesn't work.

4- You are not the first user to post in this thread. So what I advised it was not for you.

5- I really don't care what you think about me. I am sorry again.

NandoPG

PS: Please, never ever PM me.

 

[flatulent]

piece of history

For those of you who are not familiar with the past, the IBM 360 was the top of the line computer from IBM at the time. The electronic part was about 2 m high, 1 m deep, and 4 m wide. The input and output devices were about 2 cubic meters in volume each. It required about 4 people to run it. Programs were "written" on punch cards one line per card.

A two transistor amplifier analyzed with SPICE took about 20 seconds to run and the output was printed on paper 0.7 m wide. The "graphics" were done with the location of ____ ------ ........ and ******* symbols in a crude approximation to a graph.



Mechanical calculators were a wonder to behold. They could do + - x / functions to 20 digits. They made wonderful noises as they calculated totally by rotating the display disks which had the 10 digits on their circumfrence. People with musical ability would device numbers to calculate which would make musical rhythm patterns and sounds as the machine changed the digits and shifted to the next one.




Also a reminder that only people who write famous operas and symphonies are allowed to be critical and grumpy towards others on this site.

 

[u_f_o]

hi NandoPG! i'm very glad to see you again!
1 - no comments.
2 - i have soon bored with electromechanical calculator and i have soon written my program for interdigital and combline filters, use it hitherto sometimes and i needn't Patfil for such filters.
3 - yes, i've used Parfil and Elliptic only for microstrip structures (see 2), i never had said they don't work but they have unaccurate results. try please and you make sure that it is right. turn your first attention on upper stopband. try also "parallel coupled transformer in" - second left icon in dialogbox.
there are good ideas in these programs, but i'm irritated by gedanken products.
about dxf - it is very useful tool for microstrips for import-export. they have funny mistake - doubled SEQEND strings. i've patched code and emailed them, but in version 4.2.1 they didn't corrections. they stopped work.
4 - that's good.
5 and ps - our thoughts and feeling are alike.
best regards.
u_f_o

 

[cheng]

Fileters etc.

Hey NandoPG and U F O, cool down please.
You both are some of the most appreciated and valuable members that I know since I am on the board and it is already a long time
I can understand NandoPG for the spam over here is ridiculous and most of the posts are just crap but u_f_o is a very respected member of our small Pro community in here. We have to encourage serious discussions between the vaulable members and not to waste efforts on answering some "generic" and most often stupid posts. I am really pissed by folks claiming things on..."rumours, I heard, I been told, a person i know etc." We are here to share our OWN experience, knowledge, information etc, not others'. It is why I appreciate your posts (not only yours - but to a small pro community too) high. And frankly speaking I've been feeling like you guys many times when I see some question like - "which one is the best...?". That's damn ridiculous and shows 2 things. Either ignorance or just making fun of us.
So, I would invite every one to get to the point and stop spamming our wodnerful board with "generic" verbal construction of the sort shown above.

cheers,

yours truly