Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

HFSS solution frequency question...

Status
Not open for further replies.

haydar

Advanced Member level 4
Joined
Sep 10, 2004
Messages
114
Helped
16
Reputation
32
Reaction score
10
Trophy points
1,298
Activity points
991
hfss solution frequency

hi
I simulated a structure with a solution frequency 63.85 MHz, which was desired frequency.
I repeated simulations with solution frequencies 100 MHz and 200 MHz.
Their S11 results were different from each other.
Which one was correct? I need 63.85 MHz, what do you suggest?
thanks.

Note:solution type is driven modal.
 

grandizer

Member level 1
Joined
Jun 7, 2005
Messages
37
Helped
6
Reputation
12
Reaction score
0
Trophy points
1,286
Location
Germany
Activity points
1,562
solution frequency

As u increase the solution frequency it would be better , but this would take more time to simulate.
usually the small element take it is lambda/30 try zo chek convergence first , ensure the delta s is correct , chek the pass number, and then what type of simulation type do u use?
after that be sure u using the required port, and then check for linear , fast or interpolating simualtion..
after checking all this and the error still exist , try to chech your structure again..
I hope that will hwelp u
 

    haydar

    Points: 2
    Helpful Answer Positive Rating

haydar

Advanced Member level 4
Joined
Sep 10, 2004
Messages
114
Helped
16
Reputation
32
Reaction score
10
Trophy points
1,298
Activity points
991
solution frequency hfss

grandizer said:
As u increase the solution frequency it would be better , but this would take more time to simulate.
usually the small element take it is lambda/30 try zo chek convergence first , ensure the delta s is correct , chek the pass number, and then what type of simulation type do u use?
after that be sure u using the required port, and then check for linear , fast or interpolating simualtion..
after checking all this and the error still exist , try to chech your structure again..
I hope that will hwelp u

hi
there is no error reported. the only problem different results.
 

maverick09

Member level 4
Joined
Mar 9, 2005
Messages
72
Helped
6
Reputation
12
Reaction score
5
Trophy points
1,288
Activity points
2,120
solution frequency in hfss

well set the solution frequency to the maximum frequency you are trying to simulate in the frequency range.

Then try setting the frequency at the center frequency in the frequency range.

Try checking the convergence in the matrix data....the one which gives you the converge nearest to the set convergaence value, should give you accurate results.
 

    haydar

    Points: 2
    Helpful Answer Positive Rating

asdfaaa

Full Member level 4
Joined
Aug 31, 2004
Messages
227
Helped
56
Reputation
112
Reaction score
14
Trophy points
1,298
Activity points
2,530
frequency solution hfss

check the convergence in matrix data whether it is converged first;
try to reduce the convergence criteria and re-run the simulation again;
the difference of simulated S11, what kind of difference is it? for example, whether all the resonance are still there? whether the resonant frequency shifted? If only the magnitude of resonace different, most possible reason is the sweep type you choosed, you may try discrete type, avoid to use interpolation type.

Regards,
 

    haydar

    Points: 2
    Helpful Answer Positive Rating

haydar

Advanced Member level 4
Joined
Sep 10, 2004
Messages
114
Helped
16
Reputation
32
Reaction score
10
Trophy points
1,298
Activity points
991
impostare hfss la solution frequency

hi asfadaa
all results were converged, I checked them.
difference is frequency shift.
thanks in advance.
 

asdfaaa

Full Member level 4
Joined
Aug 31, 2004
Messages
227
Helped
56
Reputation
112
Reaction score
14
Trophy points
1,298
Activity points
2,530
resonant frequency hfss shift

what is the convergence criterial you set? what kind of sweep type did you select?
and what is the frequency shift between two simulations?
If you can post the porject file, I can spend some time to take look of it.

Regards,
 

    haydar

    Points: 2
    Helpful Answer Positive Rating

itaifrenkel

Full Member level 2
Joined
Oct 9, 2004
Messages
147
Helped
36
Reputation
72
Reaction score
17
Trophy points
1,298
Location
Israel
Activity points
1,859
hfss solution frequency significance

haydar shalom,

How are you doing :)

1)
Copper skin depth in 100 Mhz is 6.7um.
Copper skin depth in 67 Mhz is 8.18um.
If you are using copper thicker than 25um than the losses should be calculated correctly in both frequencies. Otherwise you will need to solve inside the metal, or use the special skin depth meshing.

2) There could be impedance mismatches. When comparing the S-parameters always compare also Z0 (assuming waveport solution). If you are using a lumped port then Z0 will stay the same but S11 will be different because of an impedance mismatch.

3) You can always use 100Mhz for mesh adaptation and then add a single point frequency sweep on 67 Mhz. This is the best way to compare 100 Mhz and 67 Mhz results.

Regards,
Itai
 

    haydar

    Points: 2
    Helpful Answer Positive Rating

haydar

Advanced Member level 4
Joined
Sep 10, 2004
Messages
114
Helped
16
Reputation
32
Reaction score
10
Trophy points
1,298
Activity points
991
hfss solution freq

itaifrenkel said:
haydar shalom,

How are you doing :)

1)
Copper skin depth in 100 Mhz is 6.7um.
Copper skin depth in 67 Mhz is 8.18um.
If you are using copper thicker than 25um than the losses should be calculated correctly in both frequencies. Otherwise you will need to solve inside the metal, or use the special skin depth meshing.

2) There could be impedance mismatches. When comparing the S-parameters always compare also Z0 (assuming waveport solution). If you are using a lumped port then Z0 will stay the same but S11 will be different because of an impedance mismatch.

3) You can always use 100Mhz for mesh adaptation and then add a single point frequency sweep on 67 Mhz. This is the best way to compare 100 Mhz and 67 Mhz results.

Regards,
Itai

hi itai
I was out of town, thank you I am fine. what about you?
1) I use 2D surfaces, not 3D objects as copper.
2)I use waveport. how can I compare Z0?
3)I simulate same sweep range with two different solution freq. Isn't that right?

THANKS A LOT
 

me71

Newbie level 1
Joined
May 8, 2006
Messages
1
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
egypte
Activity points
1,284
hfss single point solution frequency

what is the solution frequency for pifa antenna
 

Iman_b

Newbie level 3
Joined
Aug 25, 2010
Messages
3
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
Melbourne,Au
Activity points
1,303
Re: hfss solution frequency

hi
I simulated a structure with a solution frequency 63.85 MHz, which was desired frequency.
I repeated simulations with solution frequencies 100 MHz and 200 MHz.
Their S11 results were different from each other.
Which one was correct? I need 63.85 MHz, what do you suggest?
thanks.

Note:solution type is driven modal.

Hi heydar
How is it going? Did you finally find the reason for the frequency shift in resonant frequency due to solution frequency? I have exactly same problem and I couldn't find the reason in this conversation here. I appriciate if you email me the reason if you found it
Iman_B_1999@yahoo.com
Regards
 

haydar

Advanced Member level 4
Joined
Sep 10, 2004
Messages
114
Helped
16
Reputation
32
Reaction score
10
Trophy points
1,298
Activity points
991
Hi Guys
Sorry for late reply but I haven't been using HFSS for a long time. First of all it was optimized for high frequencies, so 64 MHz is too low. Recently I heard that they made modifications for lower frequencies.
Anyways, ideally all should give the same results. I would check the setup first, lumped elements are tricky. If everything looks ok I would trust higher frequencies.
In my case, I was simulating a birdcage coil which is a very complex structure. If you are designing simple loop coils, things are easier. At least you can build one and compare the results.

best
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top