请问怎么测量DUT的S参数？

我找到了David M. Pozar撰写的一本书，微波工程，第2版。
在第217页，有一节关于TRL校准。
这正是我要找的信息。
这是如何进行校准计算的。
但是有一个新问题。
本书中提到的方法似乎不起作用。
让我总结一下Pozar先生的步骤。
该错误可以通过将DUT夹在中间的一对错误框来表示。
首先测量T，R和L标准的S参数。
根据T，R和L的S参数，可以使用公式来计算误差框S参数。
然后将错误框S参数转换为ABCD参数并以矩阵形式相乘以找到DUT ABCD参数。
最后将DUT的ABCD参数转换回S参数。
遇到的问题是，在第一步，第220页的等式4.78。这个等式将有2个答案，为了找到正确答案，伽玛（传播常数）的实数和虚数都必须为正。
这是问题，方程4.78的答案都没有给出实数和虚数的正数。
在我拥有的201个数据点中，有107个不满足这个条件。
请看看Pozar方程式的错误。
是否有其他书籍或方法显示如何使用TRL测量的S参数执行校准计算？
如果你们都知道，请向我推荐。
我将深表感谢。
谢谢你编辑：odanzy于2012年4月30日上午9:26 - 删除了受版权保护的材料的图片



     以下为原文

  I've found a book written by David M. Pozar, Microwave engineering, 2nd edition.
in page 217, There is a section about TRL calibration.

This is exactly the information i am looking for. Which is how to do the calibration calculation.

But there is a new problem. The method mention in this book does not seem to work.

Lets me summarize how Mr Pozar steps.
The error can be represented by a pair of error box sandwiching the DUT.
First measure the S parameters for the T,R,and L standard.
From the S parameters of T,R,&L, a formula can be used to calculate the error box S parameters.
Then the error box S parameters is converted to ABCD parameters and multiply in matrix form to find the DUT ABCD parameters. finally convert back the ABCD parameters of the DUT back to S parameters.

The problem encountered is, at the first step, equation 4.78 at page 220.
There will be 2 answer for this equation, in order to find the correct answer, both the real and imaginary of gamma (propagation constant) must be positive. This is the problem, neither answer of equation 4.78 gives both real and imaginary positive. out of 201 data points that i have, 107 of them do not satisfy this condition.

Please have a look and see whats wrong with Pozar's equation.
Are there any other books or methods that shows how to perform calibration calculation using TRL measured S parameters? 

if you guys know any, please recommend them to me. I will be deeply grateful.
Thank you

Edited by: odanzy on Apr 30, 2012 9:26 AM - Removed pictures of copyrighted material

没有人对手动TRL校准感兴趣吗？
这些天人们如何完成它？
VNA中的一切都是自动化的吗？
但我想知道基本面及其运作方式。
Pozar的“微波工程”一书是描述TRL如何完成并提供计算的唯一来源。
但我发现它为什么不起作用。
我正在参考第2版“微波工程”看第220页，等式4.77a是错误的。
这里有一个错误。
让我们来看看如何获​​得4.77a。
方程式4.74b进入4.76b得到4.77a看看，方程式4.74b：T12 = S12 ^ 2 /（1-S22 ^ 2）式4.76b L12 = [S12 ^ 2 * e ^（ - gamma * L
）] / [1 - S22 ^ 2 * e ^（ - 2 * gamma * L）]等式4.77a（如书中所述，错误的一个）L12 * e ^（ - 2 * gamma * L）-L12 *
S22 ^ 2 = T12 * e ^（ - gamma * L） - T12 * S22 ^ 2 * e ^（ - gamma * L）等式4.77a（正确的）L12 * -L12 * S22 ^ 2 * e ^（ -
2 * gamma * L）= T12 * e ^（ - gamma * L） - T12 * S22 ^ 2 * e ^（ - gamma * L）看看上面的方程式，它在左手边一个粗心的错误
等式4.77a。
术语e ^（ - 2 * gamma * L）位于错误的位置。
我想也许第三版和第四版pazar的微波工程会解决这个错误，但令我惊讶的是，第3版和第4版有完全相同的错误。
使用正确的公式，并使用matlab求解方程，仍有40％的数据不满足伽马的条件，传播常数同时具有实数和虚数值。
现在我想知道，当前VNA中编程的TRL计算是什么？
我们是想用它而忘记TRL是如何工作的？
有人可以告诉我如何，或者我在哪里可以找到TRL的工作公式或算法？
编辑：mtxx于2012年5月14日下午11:10编辑：mtxx于2012年5月14日下午11:11



     以下为原文

  No one interested in doing manual TRL calibration?

how do people get it done these days? everything is automated in the VNA?
but i want to know the fundamentals and how it works.

Pozar's "microwave engineering" book is the only source that describe how TRL is done and provide the calculations.

but i've found out why it didnt work.
i am refering to the 2nd edition of "microwave engineering"

look at page 220,
equation 4.77a is WRONG. there is a mistake here.

lets have a look how 4.77a is obtained. he substitude equation 4.74b into 4.76b to get 4.77a
have a look,

equation 4.74b:
T12 = S12^2 / (1-S22^2)

equation 4.76b
L12 = [ S12^2 * e^(-gamma*L) ] / [ 1 - S22^2 * e^(-2*gamma*L)  ]

equation 4.77a (as stated in the book, the WRONG one)
L12*e^(-2*gamma*L)-L12*S22^2 = T12*e^(-gamma*L) - T12* S22^2 * e^(-gamma*L)

equation 4.77a (The correct one)
L12*-L12*S22^2 * e^(-2*gamma*L)= T12*e^(-gamma*L) - T12* S22^2 * e^(-gamma*L)

have a look at the above equations, its a careless mistake in the left hand side of the equation 4.77a.
the term e^(-2*gamma*L) is placed at the wrong location.


i thought maybe the 3rd and 4th edition of pazar's microwave engineering will fix this error,
but to my astonishment, the 3rd and 4th edition has the exact same mistake. 

using the correct formular, and using matlab to solve the equation,
there are still 40% of the data does not satisfied the condition that gamma, the propagation constant to have both real and imaginary value to be positive.

now i wonder, what is the TRL calculations being programmed into current VNA?
are we suppose to just use it and forget about how TRL works?

can someone enlighten me how, or where can i find a working formular or algorithm for TRL to be performed?

Edited by: mtxx on May 14, 2012 11:10 PM

Edited by: mtxx on May 14, 2012 11:11 PM

至少对于安捷伦来说，我们用于TRL校准（和SOLR）的确切解决方案被认为是专有信息。



     以下为原文

  At least for Agilent, the exact method for solution we use for TRL calibration (and SOLR) is considered proprietary information.

所有权？
好吧，我理解，公司政策。
你知道任何可以计算它的软件吗？
或安捷伦有任何软件可以做到这一点？
我的意思是一些有助于计算和执行校准的功能？
我在Matlab中查看了RF工具箱，不幸的是他们没有这样的功能。



     以下为原文

  proprietary? all right, i understand, company policy.

do you know any software that can calculate it? or Agilent have any software that can do this?
i mean some function that will help calculate and perform the calibration?
i've look through the RF toolbox in  Matlab and unfortunately they dont have such function.

您可以在他们的TRL多线路校准上查看NIST。



     以下为原文

  You might check with NIST on their TRL multi-line cal.

如果您在以下网址获得有关校准的Rytting说明：http：//www.hparchive.com/seminar_notes/a-217.pdf它会告诉您关于校准的所有信息......为什么以及如何使用剩下的专有资料。
你读完之后。



     以下为原文

  If you get Rytting notes about calibration at:

http://www.hparchive.com/seminar_notes/a-217.pdf

It will tell you all you want to know about the calibration... why and how's

Not much proprietary stuff left.. after you read all that.

Doug Rytting论文是一个很好的起点。
有很多关于TRL校准数学的论文发表。
TRL有不断的研究和改进。
快速搜索，例如谷歌学者，关于TRL校准将产生大量论文。
是的，安捷伦的实施是专有的。
肯



     以下为原文

  The Doug Rytting paper is a good starting point.  
There are so many published papers on the mathematics of TRL calibration.  There is continuous research and refinements on TRL.  A quick seach, such as Google Scholar, on TRL calibration will yield tons of papers.

Yes, the Agilent's implementation is propriatary.

Ken