Continue to Site

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.

Questions on modeling SMD capacitors

Status
Not open for further replies.

wangyufn

Junior Member level 3
Joined
Jul 29, 2005
Messages
29
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
Shanghai, China
Activity points
1,507
Hi, Dear all
I have a question on modeling SMD capacitor. I used a self-developed test fixture on the laminate and measured SMD capacitors with calibrated micro-probes, in both series and shunt mode. After de-embedding, I used series mode data to derive the SMD parameters and it can agree with vender’s data. So, I think my de-embedding calculation is OK. However, my questions is why the resonate frequency in series and shunt mode is different? For example, S11 in series mode has a resonate trap at 2.7GHz, while S21 in shunt mode has a resonate trap at 4.3GHz. The equivalent model can not explain it.
I guess that in shunt mode, because GND terminal of SMD capacitor is connected to a large ground plane, the parasitic inductance and resistance of this terminal is missing. Is this reasonable? Any better explanation of the resonate frequency shifting?

Thank you.
 

For any resonance it should contain both L and C and this fact we knew.
I guess in shunt mode the inductance is getting minimal and C might marginally increase when compared to the model of series mode model.

The result is pushing down the resonance frequency.
 

Thanks for your reply, Kspalla.

-----Quoted from your reply-----------------------------------------------------
I guess in shunt mode the inductance is getting minimal and C might marginally increase when compared to the model of series mode model.
-----Would you please give me a more detailed explanation of this change?

The result is pushing down the resonance frequency.
-----Do you mean that resonance frequency in shunt mode should be lower? Actually in my measurement, it is higher than in series mode.
 

I mean increased resonance frequency.

Added after 3 minutes:

When we place in shunt mode,
The transmission line is 50 ohm and is simpler so the capacitor height will not be effecting for current flow.
When we mount in series mode the current should reach till the height of cap and hence result in additional lead inducatance which will not come in shunt mode.
 

Kspalla, your comment is quite valuable. Thank you a lot.

There are some more questions:
(1) If the parasitic inductance is closely related to signal path, it mean that it can be very sensitive to soldering tin. And is it possible that it has a wide range in different runs of measurement? Up to now, I haven't done any experiments to see this variation yet.
(2) Is it possible to split parasitic inductance into two parts ? One is intrinsic, the other is related to height of capacitor, soldering tin and so on.
(3) Do you have any suggestions to bring these measured data into EM co-simulation? For example, what is the best way to add ports for SMD in ADS Momentum?
 

Some one also suggested to check the grounding of via holes, which also affects the resonance frequency. I think this is reasonable too.
 

This seems pretty logic to me. In case you measure S11, in fact you measure the impedance of the capacitor + the grounding vias which return the current to the ground plane. This makes the resonance lower.

I would rely on S21 for modeling the inductor. You could also consider replacing the inductor with a transmission line of certain impedance. Works well also.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top