| Author |
Message |
huiyuzz
Joined: 22 Jul 2004
Posts: 46
|
 10 Aug 2004 5:17 hfss reference conductor |
|
|
|
|
Hi,
I'm using HFSS v9.2. I am wondering how to assign a face of a conductor to ground. Is it just be assigning it to be a perfect E boundary surface??
Pls advise.
Cheers
huiyuzz
|
|
| Back to top |
|
 |
Mr. D.
Joined: 27 Nov 2003
Posts: 33
Helped: 2
Location: Switzerland
|
10 Aug 2004 8:31 hfss wave port reference conductor |
|
|
|
|
Perfect E means simply that the E-Field is normal to the surface -> it simulates a perfect conductor.
To assign a surface to gnd you have to use for example a lumped port and connect the line (you will be asked for) from the conductor to gnd.
If you use wave ports you don't have to do anything.
D.
|
|
| Back to top |
|
|
huiyuzz
Joined: 22 Jul 2004
Posts: 46
|
10 Aug 2004 9:41 hfss ground perfect e |
|
|
|
|
Hi Mr. D.,
Does that mean if i am going to assign the outer surface of the outer conductor of a coax antenna to be the ground. And I am using wave port excitation, I just need to assign the cross-sec of the coax antenna to be my waveport and thats all, I do not touch the outer conductor??
Perhaps another doubt to be cleared: what are terminal lines for in waveports?
Thanx
huiyuzz
|
|
| Back to top |
|
|
wlcsp
Joined: 16 Aug 2004
Posts: 179
Helped: 14
|
17 Aug 2004 11:46 what is meant by hfss? |
|
|
|
|
The perfect-E boundary condition does not necessarily mean ground. Ground in HFSS can be determined by having same reference for more than one port. So if you have the perfect-E face as reference, that means you have a perfect ground (without parasitic resistance and inductance).
To define a port on a coaxial face, just select the face which covers the area of outer conductor. Then, make an integration line pointing from inner conductor to outer conductor. If you do the same for the other port, this means you have the outer conductor grounded. For simplicity (computationally faster) you can define it as perfect-E.
the terminal lines are used if you're interested in giving the port a voltage source (excitation). The terminal lines are also usefull when you have closely coupled lines where you use only wave port for two lines. Here you have to have voltage line for each line.
|
|
| Back to top |
|
|
huiyuzz
Joined: 22 Jul 2004
Posts: 46
|
18 Aug 2004 2:37 hfss return loss draw |
|
|
|
|
*** To define a port on a coaxial face, just select the face which covers the area of outer conductor. Then, make an integration line pointing from inner conductor to outer conductor. If you do the same for the other port, this means you have the outer conductor grounded. For simplicity (computationally faster) you can define it as perfect-E. ***
Hi, wlcsp
I do not quite understand what is meant by the "face which covers the area of the outer conductor" And why do i need another port? Where shld i assign these ports??
Please advise. Thanx
Cheers
huiyuzz
|
|
| Back to top |
|
|
wlcsp
Joined: 16 Aug 2004
Posts: 179
Helped: 14
|
18 Aug 2004 12:49 terminal reference conductor hfss |
|
|
|
|
Let's think about a coax cross-section (2-D). To draw a coax you need two circles. Let's say the first circle you draw is centered at (0,0) with radius 1. The second circle is also centered at (0,0) with radius 2.
There are two ways to assign port on a face. First one is to select the face of the bigger circle. This covers a circular area centered at (0,0) with radius 2. But since the inner circle is normally a conductor (if not PEC), so the wave will not propagate in it.
The second way is the face that covers all area from between radius 1 and 2 as the port.
The calibration line can neglected.
But in 3D case, where you want to have both ends of the coax to be ports, sometimes the calibration line is necessary. You can also make a calibration line from pointing from inner circle to outer circle on both faces. This will define the inner circle (inner conductor) as reference (ground).
Please have a look at the hfss model file.
|
|
| Back to top |
|
|
kenzo4369
Joined: 10 Aug 2004
Posts: 8
|
24 Aug 2004 9:34 hfss select faces |
|
|
|
|
 Hi...wlcsp
If I want to assign a wave port to the dipole antenna in HFSS,
and I wish it has the same effect as the differential port in IE3D,
because I assign normal wave port, the Return Loss in HFSS is very bad(2dB)
my dipole antenna seems like never been excited while simulation...
Thanks for ur help 
|
|
| Back to top |
|
|
huiyuzz
Joined: 22 Jul 2004
Posts: 46
|
24 Aug 2004 11:31 hfss assign voltage |
|
|
|
|
For dipole, u can define a lumped port between the probes, it would be easier to use a lumped port than a wave port in this problem as it is an interior structure.
Then u need to define the integration line from (-) to (+).
Rgds
huiyuzz
|
|
| Back to top |
|
|
Google
AdSense
|
24 Aug 2004 11:31 Ads |
|
|
|
|
|
|
| Back to top |
|
|
kenzo4369
Joined: 10 Aug 2004
Posts: 8
|
26 Aug 2004 5:23 hfss attach surface to object |
|
|
|
|
Thanks huiyuzz.U are realy nice,I'll try lumped port 
Is it correct that the height of the lumped port equal to the thickness of substrate? and the width is equal to the width of dipole's feeding?
Thanks for ur help~~
P.S Sorry My English is so poor 
|
|
| Back to top |
|
|
