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.

Behavior of capacitor at microwave frequencies

Status
Not open for further replies.

sini

Newbie level 5
Newbie level 5
Joined
Jun 15, 2004
Messages
10
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
82
modelitics ads

Hi all,
can you explain me behavior of vertically mounted parallel plate capacitor at 10GHz. I want to use this capacitor for DC decoupling between two microstrips. How can I simulate this capacitor (I need S-parameters this two port subcircuit) ? I tried to simulate it at MWO - behavior was strange.
 

The suppliers of uwave capacitors produces very low ESR and very low incuctance Capacitors. But often they tend to be optimistic when they measure their capacitors; the Series Inductance is only the "phisycally internal" inductance.
If the cap is vertically mounted, The current flow 1st along one arm, then into the cap, the along the other arm. The path along the two arm is an additional inductance not quoted into the data sheet.
As rule of thumb the additional inductance is 1nH/mm (Grove Formula).

Also consider the "gap" effect on the microstrip!
Also consider the ESR due to the solder alloy!
 

sini said:
Hi all,
can you explain me behavior of vertically mounted parallel plate capacitor at 10GHz. I want to use this capacitor for DC decoupling between two microstrips. How can I simulate this capacitor (I need S-parameters this two port subcircuit) ? I tried to simulate it at MWO - behavior was strange.

Hi

Since you are working at 10 GHz, it's required that the capacitor you intend to use has a higher self resonance frequency ; you can check the SRF of the capacitor from the data sheet or enquire from the manufacturer's website; for decoupling at 10 GHz and above it's better to avoid using multilayer capacitors and use only single layer parallel plate capacitors as these have higher SRFs and are much smaller in size and you may not even require simulation!; still if you intend to simulate the capacitor you can contact the manufacturer for the equivalent circuit parameters or the S-par file

8)
 

Hi,

well, even if you have measured S parameters of the capacitor that you are going to use, the acctual performance will be affected how and on which substrate you are going to mount it, the pad sizes etc. Company called Modelitics made libraries of lumped elements mounted on the pads that account for the influence of the substrate etc. Thay are availabel for MWO, so you should get them. Restricted set of models is available for downloading from AWR site (although I have to say I never managed to download them becouse connection was timed out)

flyhigh
 

Get ATC capacitor, the S-parameter is provided. You can simulate and check out the Series and parallel resonant frequency...
 

Good idea,
where can I download s-parameters of single layer ATC capacitors ? What is the best for DC blocking of 2 microstrips ( 2 stages...) - parallel or series resonance ? What is alternative model of single layer vertically mounted parralel plate capacitor?
 

Hi all,
At this frequency, SMt capacitor xseem vey critical, due
parasitc package effect, siz of component, pcb effect...
U can see on Murata web if they sale this type of capacitor (you can dowload S parameter file),
i know that for 402 size frequency are limited to 6Ghz,
maybe u need to use a lower size 201.

Bye
 

Hi,
You need a CD decoupling for 10GHz for this application you can find at ATC as well as Dielectric Lab , Murata etc capacitors for low ESR and high SFR. ATC has the 500 series. The problem for your application is the parallel resonance which come after the series one. If you work around the series resonance you have no problem because you need DC block.
If you look for very low resistance you might pay for that (ATC). If you can use dies the best is MIS capacitors which gives very low loss.
Remember that always there is a parallel capacitance to ground threw the substrate, so for very acurate capacitance you have to have S parameters (or measurements) using the real substrate and line dim. you use in the circuit, or you can EM simulate (I use Momentum of ADS). But again as I understand it your application is not complicated one.

D.J
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top