|
|
|
|
s-parameter - RF Cafe Forums
|
guest
Post subject: s-parameter
Unread postPosted: Sun Jun 19,
2005 11:00 am
hi everyone,
my problem is really simple. I
have to measure S11 in a component soldered in a PCB for a frequency
range between 1.1 GHz to 1.3 GHz. In this pcb there are two input tracks
for the inouts of the component and luckily there are also two identic
tracks with open and short termination at theand of them. my question
is what's the procedure to get the right S11? Or otherwise what's the
procedure to move the reference plane od the vector analyzer?
thanks
Top
Guest
Post subject:
Unread
postPosted: Mon Jun 20, 2005 5:15 am
Does not anyone want to
help me :'( ?
Guest
Top
Guest
Post
subject:
Unread postPosted: Mon Jun 20, 2005 6:47 am
The
accurate way to measure the S11 of your component is as follows: You
have to de-embedd the length of transmission line that leads to the
device input. In order to do that, you have to calibrate your VNA as
close as possible to the device.Solder an SMA connector to your PCB.
Solder a 0-ohm resistor between the input of the device to GND (at the
end of the transmission line that leads to the device's input), and
calibrate your VNA in "Short", do the same with 50-ohm "Load" termination,
and with open-circuit "Open". By that you will eliminate the length
of the transmission line, as you take it into account when calibrating
the VNA including that piece of transmission line. Now you can measure
the real S11 of the device, with a single port. Make sure that the output
of the device is terminated with 50-ohm, and of course measure the S11
when the device is powered (with its supply voltage) or else your result
won't show the real impedance of the device.
Top
Phineas
Post subject:
Unread postPosted: Tue Jun 21, 2005
11:26 am
Since you have to de-embed the component, why not get
a similar comonent that hasn't been placed onto a PCB
and measure
the S11 parameter?
Top
Guest
Post subject:
Unread postPosted: Tue Jun 21, 2005 2:10 pm
Phineas wrote:
Since you have to de-embed the component, why not get
a similar comonent
that hasn't been placed onto a PCB
and measure the S11 parameter?
What do you mean?
My idea is this, I calibrate my VNA and
then I measure the short presents on the boards and I get the s parameter
of the board. I measure the componet with the board and I de- embed
the s parameter?? What do you think about it??
guest
Top
Guest
Post subject: De-embedding
Unread postPosted:
Tue Jun 21, 2005 2:35 pm
I recommend care. Normally, an accurate
de-embedding requires at least open, short, and terminated measurements.
See the Agilent application notes on the subject.
Good Luck!
Top
Guest
Post subject: Re: De-embedding
Unread postPosted: Tue Jun 21, 2005 2:55 pm
Guest wrote:
I recommend care. Normally, an accurate de-embedding requires at least
open, short, and terminated measurements.
See the Agilent application
notes on the subject.
Good Luck!
I didn't find anything
on the agilet's website... anyway I'll check out once again...and thanks
everyone
Guest
Top
Guest
Post subject:
Agilent app not on vector analyzer de-embedding
Unread postPosted:
Wed Jun 22, 2005 1:23 pm
Hmm - it's gotten harder to find stuff
on Agilent's website.
I think this is what you need:
http://cp.literature.agilent.com/litweb
... -5329E.pdf
Good Luck!
Top
Guest
Post subject: Re: Agilent app not on vector analyzer de-embedding
Unread postPosted: Thu Jun 23, 2005 1:48 am
Guest wrote:
Hmm - it's gotten harder to find stuff on Agilent's website.
I think this is what you need:
http://cp.literature.agilent.com/litweb
... -5329E.pdf
Good Luck!
thanks a lot...you are
extremely kind :-)
Guest
Posted 11/12/2012
|
|
|
|