Really?  That's the amount of info you give us and expect a valid answer?

Dan - Owner
http://www.Hi-TecDesigns.com

The antenna need to be above PCB on PCB.



 

Quote:

... -3dB beamwidth of around 120 deg... with High Gain as possible.

I agree with Dan that your question is almost perfectly ill-defined.

The way to achieve "high gain" and 120° [azimuth?] beamwidth (implying low gain) is to narrow the elevation beamwidth.
 

I can construct suspended plate configuration with narrow ground plane resulting in 120 deg. beam width and gain ~ 6 dBi.
Gain can be further increased by making an array of plates ...
Such design will be feasible on UHF...
MacGyver is right - you need to explain what you want: your operating frequency, requirements in detail, spatial constrains.

The frequency BW need to 8%

Hi,

I summerize the parameters:

Antenna with a -3dB beamwidth of around 120 Deg.
frequency of operation 4.5GHz
Frequency BW :350MHz
The antenna can be above PCB or on PCB.

Please advice.

8% bandwidth, look up thick patch antenna. That'll work.  

Hi Higgler,

Can i receive beamwidth of 120 deg ? (i know that the beamwidth of patch antenna is around 90 deg maximum)

Patch beamwidths depend on the dielectric you choose and they aren't 90 degree beamwidths as a rule (they can be 60 degree up to 180 degree or more. Patches can be curved also to widen the beamwidths if needed.

If you use high dielectric, the patch is smaller and beamwidth is wider. Er= 8, thickness 0.2 Lambda in the dielectric (about 0.15" thick) will work, that's a 0.18 Lambda wide antenna. That'll give you very wide beamwidth.

Hi Higgler,

IF you increase the Er to 8 the BW is decrease.

I need BW of up to 7%.

Did you see patch antenna with this high BW and high Beamwidth ?

Doron

Best Patch antenna book is Pozar's book, big yellow one. http://www.amazon.com/Microstrip-Antennas-Analysis-Design-Arrays/dp/0780310780/ref=sr_1_2?ie=UTF8&qid=1326039161&sr=8-2 Has all the details in it to make a thick patch antenna on page 77.

Here's notch antenna design info, it's just a slot in your ground plane edge and you cross a feed line on the other side of the circuit card and it's extremebly wide bandwidth. You can control the beamwidth easier based on size.
 http://www.piers.org/piersonline/pdf/Vol3No7Page984to986.pdf

The radiation is different than the patch. Patch antenna radiates perpindicular to the plane of the circuit card, notch radiates in the plane of the circuit card.

 

Hi Hilger,

In the article from piers.org the is no data that they reach to beamwidth of 120 degree.

The book i need to purchase.

There is a lot material how to increase the BW but less material how to increase the beamwidth.

Can you send me a meterial for antenna with beamwidth of 120 degree ?

may be: Printed dipole or pifa or ifa better ?

Doron

 

Wide beamwidth equals small antenna. Narrow beamwidth equals large antenna.
PIFA and Printed Dipole and Notch are nearly related. PIFA is a notch with a very thin layer/line of metal on the top and ground plane on the bottom, similar could be said about a dipole too without a bottom groundplane. Dipole has a 73 degree beamwidth by the way, you can bend it and make it look like an arrow to widen the beamwidth. Notch is easier to design and build though due to it's very wide bandwidth (2:1) or 100% bandwidth (as opposed to 7%).

The notch slot depth sets the beamwidth.
 
You can change the beam from pointing backwards with 1/3 Lambda slot depth to being a narrow beam in the forward direction with the slot width large and tapering widely. Cross the probe 1/4 wavelength from the slot short circuit in the back.

Set the slot length at around 1/2 wavelength or sligthly less and your beamwidth will be very wide. With copper tape extending the slot opening off the end of the ground plane, you can shape the beamwidth to as narrow as you'd like. That's a nice feature in this antenna.

With really wide beam antennas, energy bounced off everything and your patterns get bumpy. 120 degree beamwidth antennas are much more difficult than a 20 degree or 50 degree beamwidth antenna to get a good smooth pretty antenna pattern shape (due simply to antenna energy scattering everywhere, then energy reradiates off scatterers).  

As a general rule, beam width is proportion to lambda  di  vided by aperture

TTFN

FAQ731-376: Eng-Tips.com Forum Policies
Chinese prisoner wins Nobel Peace Prize

Yup agree, should have said electrically small, electrically large, would've been clearer.  

Hi Higler,

I have a question from different subject: Why the Antenna Handset designer not design the Antenna to Dipole which has the ability to be indifferent to human (As a GND) ?

RXTX

I think PIFA's are much smaller. So size only.

Dipoles with thin wires are usually one frequency, narrow band. PIFA's can be many frequencies in a small size.

Hi Higler,

About the antenna with beamwidth of 120 degree, it is look that the slot antenna feed complicated.

Do you have article for slot antenna with BW:7% and beamwidth of 120 degree ?

RXTX

By the way - the antenna need to be with Linear polarization.

Notch/Slot is easier than any other antenna. Bandwidth is so wide, dimensions are not critical. It also makes feeding it easy too and the center conductor is short circuited to ground and you will never have static damage to your transmitter or receiver with a grounded center conductor antenna. Patches are open circuited center conductors and not as safe as a shorted notch due to static.

Dimensions are easy, everything is 1/4 wavelength except the slot opening, make that 0.050" to 0.1" roughly.
1/4 wave dimensions are;
1) distance feedline crossing slot to short circuit in the back
2) distnace feed line crossing slot to open circuit in the front
3) distance from the slot to the top of the antenna
4) distance from the slot to the bottom of the antenna

essentially a square thin antenna. You can make the antenna much taller also, it just needs a half wavelength height for efficiency. If you don't have enough room, VSWR can be ok subsituting the 1/4 wave dimension with closer to 1/3 wavelength and still get 2:1 VSWR and super wide antenna patterns.

example, wavelength = 12 inches. Slot
Antenna becomes 6" tall x just over 6" deep x thickness. Cut slot in center. attach coax across center slot (ground on one side of slot, right in the center of the slot, center conductor of coax. is shorted to opposite side of short). Then for VSWR tuning, and wider bandwidth, on the front opening part of the slot, make it wider and V shaped, so the slot is small at the coax. connection and wide (up to 1/4 wavelength, 3" wide at the open end). If you keep the slot thin at the front opening, it's more challenging to get good VSWR and it's narrower band, but you can make it work with a narrow slot, not sure how well though.
It'll work, I've done it alot with a wide opening in the front of the slot. But the antenna pattern will only be very smooth if you add antenna foam absorber all around the antenna.

 

Hi Higler,

For a mass production issue: do you have idea how i can not solder the coax manually to antenna ?

Yes,
for short circuiting the center conductor; you can use printed circuit techniques, and use a via to ground. Using coax. though, you can use any mechanical screw into the circuit board and hold it down that way. Benefit of wide bandwidth is insensitivity to using imperfect mechanical assembly.
for open circuiting the center conductor; just pass the center conductor 1/4 wavelength across the slot, instead of shorting it on the far side of the slot. Spray coat the metal with an insulator and then glue the center conductor onto the spray coasted metal(proper length 1/4 Lambda) past the slot.

If it connects to a transceiver, marketing it as "static damage insensitive shorted center conductor" might help. It works either way though.  

Hi Higler,

See attached, Please review and let me that i drew it correct.

I am still do not understand how to connect the center conductor from the PCB, Please explain.

What about to use dipole antenna with the arms like a arrow - will it achieve beamwidth 120 degree ?

What about patch antenna with beamwidth expander above the patch ?

RXTX

• http://files.engineering.com/getfile.aspx?folder=1584be4a-ec0f-47e7-bc94-53

Search "vivaldi antenna" on Google and click on the top of the Google search for Images and you'll see a huge collection of photos and I think you'll understand the design alot better.

Vivaldi antenna is a notch antenna, use my "make everything one quarter wavelength" note above.

Get a piece of thin metal, cut with scissors the slot, add a coax. and you have an antenna.

your sketch has a cavity, I don't think I mentioned having a cavity.

good luck.
what application is this for that needs high volume?

Hi Higler,

OK, I understand the feeding issue - see attched fo example.

By the way the vivaldi antenna beamwidth is not smoothing and there are a lot of ripple.

Is the Slot antenna has such of ripple ?

I need that the beamwidth will be 120 degree for Elevation & Azimuth.

RXTX

Hi Higler,

OK, I understand the feeding issue - see attched for example.

By the way the vivaldi antenna beamwidth is not smoothing and there are a lot of ripple.

Is the Slot antenna that you present has such of ripple ?

I need a smoothing beamwidth at 120 degree for Elevation & Azimuth.

RXTX

• http://files.engineering.com/getfile.aspx?folder=d27df32f-8456-4425-9292-d8

The notch antenna = The vivaldi antenna. They are the same, like calling something a car and automobile is the same. Vivaldi is a man's name, notch is a physical description.

The two feeding options are shorting the feed at the opposite side of the slot and leaving the feed open circuited, with a feed length of 1/4 wavelength (electrical length if in a dielectric) past the slot. Sometimes instead of a simple transmission line shape, an oval, round or other shape is used in place of a simple feed line for greater bandwidth.

Recall "The antenna pattern of a very wide beamwidth antenna (120 degrees is very wide) depends on what's around the antenna much more than it depends on the antenna itself".

You can make a vivaldi or notch antenna anywhere from 1/3 Lambda (air) square for a backwards beam (very low gain, very wide beamwidth), up to one wavelength square having about +6 dBi gain. I'd suggest making two antennas or three antennas, 1/2 wavelength and 3/4 and one wavelength square and you'll see the difference in beamwidths when you measure them.  

These antennas can actually have ultra high gain too if you make them long enough.

good luck.   

For the frequency specified (4.5GHz), the effective antenna aperture should be 3 inches or less, using Rayleigh criteria.  However, too small an aperture will result in an even larger beamwidth and degrade its apparent gain.

TTFN
FAQ731-376: Eng-Tips.com Forum Policies