+ Post New Thread
Results 1 to 19 of 19
  1. #1
    Newbie level 5
    Points: 48, Level: 1

    Join Date
    Aug 2019
    Posts
    9
    Helped
    0 / 0
    Points
    48
    Level
    1

    Patch bowtie matching problem

    hi, im designing bowtie antenna on fr4 substrate in CST. Feeding is the quarterwave trasmission line. But it is poorly matching. I played with dimensions and used optimizer but still cant get the anything valuable. Its been almost 2months and its so frustrating for not having a good result. If u give me any suggestion, it will really mean a lot...

  2. #2
    Full Member level 3
    Points: 1,113, Level: 7

    Join Date
    Oct 2017
    Posts
    151
    Helped
    23 / 23
    Points
    1,113
    Level
    7

    Re: Patch bowtie matching problem

    Could you post an image? There are many possible configurations. Possible reasons: wrong feeding, wrong ground plane, but need more information on your design.
    p.s. I think coplanar feeding bowtie slots are simple to optimize.


    2 members found this post helpful.

  3. #3
    Newbie level 5
    Points: 48, Level: 1

    Join Date
    Aug 2019
    Posts
    9
    Helped
    0 / 0
    Points
    48
    Level
    1

    Re: Patch bowtie matching problem

    It is not slot antenna. By the way I didnt use ground plane at the back.
    I ve used now after your post but there is no significant change in s-parameter.

    Click image for larger version. 

Name:	bowtie.png 
Views:	6 
Size:	16.4 KB 
ID:	155250

    Click image for larger version. 

Name:	s-parameter.png 
Views:	9 
Size:	12.9 KB 
ID:	155251



  4. #4
    Super Moderator
    Points: 260,253, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    45,450
    Helped
    13829 / 13829
    Points
    260,253
    Level
    100

    Re: Patch bowtie matching problem

    A bow-tie antenna is a dipole antenna, but the picture seems to show a monopole antenna. It may work somehow as an antenna, but surely not with the useful wide band characteristic expected from a bow-tie.

    Anyway, before designing a matching circuit, you should look at the input impedance curve.


    3 members found this post helpful.

    •   AltAdvertisement

        
       

  5. #5
    Newbie level 5
    Points: 48, Level: 1

    Join Date
    Aug 2019
    Posts
    9
    Helped
    0 / 0
    Points
    48
    Level
    1

    Re: Patch bowtie matching problem

    it is acting like a monopole cuz the transmission line, right? but otherwise how can i feed the antenna? i dont want to use center-fed. there are some articles used transmission line in bowtie. My design procedure is like that, first design a bowtie antenna by using discrete port in the center (so there is no transmission line at this point) and then by considering input impedance, i design a transmission line.

    By the way the wideband characteristic could be because of not using ground plane? cuz when i simulate the same antenna with ground plane still get s-parameter like -4 or -5dB but the curve shape is more sharp at the end and has kinda narrowband characteristic.

    Quote Originally Posted by volker@muehlhaus View Post
    But your microstrip lines "signal" conductor is connected to BOTH legs of the bowtie. You didnt show line ground in your model.

    This is very different from connecting signal conductor to one side, and ground conductor to the other side.

    As FvM stated, you created a monopole antenna.
    So what can i do to fix that? 😕 i will use ground plane 🖒 if you give any other tips that would be great. Also i didnt get the problem of the transmission line being connected to bowtie? Isn't it how it suppose to be feed?



  6. #6
    Advanced Member level 5
    Points: 14,159, Level: 28

    Join Date
    Apr 2014
    Posts
    2,218
    Helped
    883 / 883
    Points
    14,159
    Level
    28

    Re: Patch bowtie matching problem

    Quote Originally Posted by turuk View Post
    and then by considering input impedance, i design a transmission line.
    But your microstrip lines "signal" conductor is connected to BOTH legs of the bowtie. This does not create a differential signal between the two arms.
    As FvM stated, you created a monopole antenna.

    For Bowtie, you want to connecting signal conductor to one side, and ground conductor to the other side.


    1 members found this post helpful.

  7. #7
    Advanced Member level 5
    Points: 33,952, Level: 45

    Join Date
    Oct 2001
    Posts
    4,848
    Helped
    1440 / 1440
    Points
    33,952
    Level
    45

    Re: Patch bowtie matching problem

    Here is an example including the microstrip balun:



  8. #8
    Newbie level 5
    Points: 48, Level: 1

    Join Date
    Aug 2019
    Posts
    9
    Helped
    0 / 0
    Points
    48
    Level
    1

    Re: Patch bowtie matching problem

    i saw couple of designs like that but can i design it without balun structure? cuz probably it will be a bit challenging for me to design it.

    - - - Updated - - -

    you mean for each arm there should be 2 transmission line? but i saw a paper that used single feed :/ here is the picture
    Click image for larger version. 

Name:	bowtie-patch.png 
Views:	3 
Size:	59.6 KB 
ID:	155253



  9. #9
    Advanced Member level 5
    Points: 14,159, Level: 28

    Join Date
    Apr 2014
    Posts
    2,218
    Helped
    883 / 883
    Points
    14,159
    Level
    28

    Re: Patch bowtie matching problem

    Quote Originally Posted by turuk View Post
    but i saw a paper that used single feed :/ here is the picture
    Click image for larger version. 

Name:	bowtie-patch.png 
Views:	3 
Size:	59.6 KB 
ID:	155253
    There are many papers out there that make no sense. If one line feeds both arms with the same phase, it is not a bowtie antenna.


    1 members found this post helpful.

    •   AltAdvertisement

        
       

  10. #10
    Newbie level 5
    Points: 48, Level: 1

    Join Date
    Aug 2019
    Posts
    9
    Helped
    0 / 0
    Points
    48
    Level
    1

    Re: Patch bowtie matching problem

    i've tried 2 transmission line for both arm (remade from an article). Is it okay? i get better s-parameter result around -30dB and there is no ground plane cuz i dont know where to place the via like which arm or anywhere else... anyway this design works but dimensions are very big for my work
    Click image for larger version. 

Name:	bt.png 
Views:	5 
Size:	11.4 KB 
ID:	155256



  11. #11
    Advanced Member level 5
    Points: 14,159, Level: 28

    Join Date
    Apr 2014
    Posts
    2,218
    Helped
    883 / 883
    Points
    14,159
    Level
    28

    Re: Patch bowtie matching problem

    Add a 1:1 balun where you go from unbalanced coax to balanced symmetric line.
    Also have a look at the alternative design in post #7



  12. #12
    Super Moderator
    Points: 260,253, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    45,450
    Helped
    13829 / 13829
    Points
    260,253
    Level
    100

    Re: Patch bowtie matching problem

    Presume you want the bow-tie as wide band (e.g. 1:2) antenna, you also need a wideband balun, e.g. a transformer. λ/2 transmission line (as in post #7) is only good for small band.


    1 members found this post helpful.

  13. #13
    Newbie level 5
    Points: 48, Level: 1

    Join Date
    Aug 2019
    Posts
    9
    Helped
    0 / 0
    Points
    48
    Level
    1

    Re: Patch bowtie matching problem

    balun is necessary? cuz i never design a balun structure and kinda scares me Im getting good results by using 2 transmission line design, it has vertical polarization just like a bowtie and s-parameter is like -30dB. but gain is still not that good (around 2). If u have any suggestion to increase gain that would be really great..



  14. #14
    Advanced Member level 5
    Points: 14,159, Level: 28

    Join Date
    Apr 2014
    Posts
    2,218
    Helped
    883 / 883
    Points
    14,159
    Level
    28

    Re: Patch bowtie matching problem

    Quote Originally Posted by turuk View Post
    balun is necessary?
    In real world: yes. Otherwise your antenna will be sensitive to touching the coax feed.


    1 members found this post helpful.

  15. #15
    Newbie level 5
    Points: 48, Level: 1

    Join Date
    Aug 2019
    Posts
    9
    Helped
    0 / 0
    Points
    48
    Level
    1

    Re: Patch bowtie matching problem

    thanks, i didnt know that. Is there any source you can suggest me to look at for balun design?



    •   AltAdvertisement

        
       

  16. #16
    Full Member level 3
    Points: 1,113, Level: 7

    Join Date
    Oct 2017
    Posts
    151
    Helped
    23 / 23
    Points
    1,113
    Level
    7

    Re: Patch bowtie matching problem

    maybe this be useful: https://www.hindawi.com/journals/ijap/2013/651040/
    I suggest to simulate separate blocks first. Feeding line with two ports, ensure S21 is good over required bandwidth, and no radiation losses. With some simple ports, then add two SMA ports. Finally use one SMA port and bowtie. Are you sure feeding line impedance is matched to SMA connector? Try to increase feeding line length by λ/4, if S11 become much worse, then something is not matched (bowtie to feeding line, feeding line to SMA connector).


    1 members found this post helpful.

  17. #17
    Newbie level 5
    Points: 48, Level: 1

    Join Date
    Aug 2019
    Posts
    9
    Helped
    0 / 0
    Points
    48
    Level
    1

    Re: Patch bowtie matching problem

    i've tried to increase the length by λ/4 as u said, there were no significant change in s11. and im doing the simulations of the single transmission line rgt now. Thank u for the suggestions one problem is that antenna is fed through the left transmission line and the other transmission line + bowtie arm is fed by that line. this shifts the radiation pattern to the left which something i dont want to. Is it okay to feed the antenna like this? here is the image Click image for larger version. 

Name:	bwt.png 
Views:	3 
Size:	10.9 KB 
ID:	155322



  18. #18
    Advanced Member level 5
    Points: 14,159, Level: 28

    Join Date
    Apr 2014
    Posts
    2,218
    Helped
    883 / 883
    Points
    14,159
    Level
    28

    Re: Patch bowtie matching problem

    By adding the balun, you get a symmetric feed for symmetric radiation pattern.


    1 members found this post helpful.

  19. #19
    Super Moderator
    Points: 260,253, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    45,450
    Helped
    13829 / 13829
    Points
    260,253
    Level
    100

    Re: Patch bowtie matching problem

    You are talking about "λ/4" but show a 1:3 frequency range in post #3. What's the antenna specification you want to achieve? If it's wideband, λ/4 is meaningless.

    Without balun, the feed line will act as part of the antenna and affect the radiation characteristic, as apparently observed in your latest design.



--[[ ]]--