Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic 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.

Register Log in

problem desigining a microstrip TL

Status
Not open for further replies.

satishgra

Member level 3
Joined
Mar 29, 2008
Messages
59
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,288
Location
Bangalore
Activity points
1,717
Hi,

I am a pure VLSI guy and have no clue of designing a TL
But, I had to do it for some other reason using HFSS

My design is giving me the same results irrespective of the dimensions used
Tool used : HFSS

quick reply could be helpful

Regards,
satish
 

chemic

Newbie level 5
Joined
Sep 8, 2009
Messages
10
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
France
Activity points
1,350
Hi Satishgra,

what is the length of the microstrip you are simulating, and the frequency range?
 

satishgra

Member level 3
Joined
Mar 29, 2008
Messages
59
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,288
Location
Bangalore
Activity points
1,717
Hi,

Thanks for the reply....
I am more of like interested in getting a hang of HFSS
Hence, i am okay with any results

I am more of a MEMS guy who wanted to understand S-parameter calculation ( I know this is simple)..After looking at the tool, thought it is worth spending some time

I am looking at designing a BPF using micro strip TL


I was able to get rid of the issue...seems to be a problem with the waveport definitions...Can you let me know how to design a BPF using this microstrip...I have done all the Reqd calculations (referred it from pozar)
 

satishgra

Member level 3
Joined
Mar 29, 2008
Messages
59
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,288
Location
Bangalore
Activity points
1,717
My present query is....from my calculations, a microstrip of width 3.2cm for (1 cm substrate) is required to design a microstrip of 50 ohm z0
But, simulations show it is resulting in 4 ohm only

where am i going wrong...and second question

if connect 3 such microstrips to form a T circuit, the resultant impedance is turning out to be 0.002 as aganist expected 6 ohm


I am attaching my BPF (only half done...should replicate it around the axis to complete), microstrip and also a prf report of the design...
 

chemic

Newbie level 5
Joined
Sep 8, 2009
Messages
10
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
France
Activity points
1,350
Well,

first I have to say that I am a newbye on HFSS. I am curious about your waveport definition, what was wrong with it?

On the BPF, HFSS is not the right tool to design it. Or, better, it is not a good idea to begin from it. I would start from a circuit simulation, using ADS or Designer, where you can tune the parameters without re-frawing everything. Then you generate the layout, and you do the fine-tuning using a planar simulator. You can then verify the results using HFSS to take into account the fringing effects and so on. If you start from HFSS you simply will have too many degrees of freedom.
 

satishgra

Member level 3
Joined
Mar 29, 2008
Messages
59
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,288
Location
Bangalore
Activity points
1,717
Yeah...I was defining the waveport in parallel to the substrate....which should mean that the input signal is going to flow through the substrate and will have no impact as to what your microstrip is......

I will look into ADS and see if i can design the same
 

chemic

Newbie level 5
Joined
Sep 8, 2009
Messages
10
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
France
Activity points
1,350
I did the same error in the past ;)

About ADS, it is quite easy to design microstrip subassemblies, as you have parametric models in the schematic window. You connect them, and you can even tell the optimizer to tweak the design dimensions to have a given performance. Then, once you are satisfied with youur simulation, you just have to generate the layout, and simulate it with Momentum (the embedde 2.5D EM simulator). Usually you will need to do small adjustments, and then, if your substrate definition is accurate, you should have a working design since the first time. In this way I designed a Wilkinson Power splitter on a standard FR4 substrate (Not really the most characterized one for RF apps) and it matched almost perfectly the simulations.
 

satishgra

Member level 3
Joined
Mar 29, 2008
Messages
59
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,288
Location
Bangalore
Activity points
1,717
If i do it on ADS, atleast i do see some meaningful result....But with HFSS...i am getting fucked...I dont know where i am going wrong
 

chemic

Newbie level 5
Joined
Sep 8, 2009
Messages
10
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
France
Activity points
1,350
Yes, the problem with HFSS is that, first, you have to really know how you are going to define ports, while in ADS it's a bit simpler (but you can have odd results as well). Second, drawing on HFSS is quite complicated, and I would say impossible above some degree of complexity. It's always better to import designs made with other tools (either PCB design tools or mechanical CAD if you wanna simulate antennas).
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top