校准标准的适度相位误差会影响标量测量吗？

对于矢量误差校正测量，校准标准的相位误差将影响幅度测量精度。
在矢量误差校正中，涉及复数减法。
差矢量的大小是相位的函数。
您可以绘制电容模型与高达11GHz的频率，并查看6 GHz以上的电容是否有任何异常变化。
如果它遵循相对平滑的轨迹，则电容模型可能在该频率下可用。
创建了一些开路电容模型，以便在指定的频率范围之上工作。



     以下为原文

  For vector error corrected measurements, phase errors of cal standards will effect magnitude measurement accuracy.  In vector error correction,  complex subtraction is involve.  The magnitude of the difference vector is a function of  phase.

You can plot the capacitance model vs frequency up to 11GHz and see if any unusualy changes in capacitance above 6 GHz.  If it follows a relatively smooth traajectory, the capacitance model is probably usable up to that frequency.  Some open capacitance models were created to work way above the specified frequency range.

85032B Opens用于非常早期的8510校准套件，规格高达18 GHz。
可能使用相同的电容系数。
我认为值得尝试而不是使用适配器交换技术。
相位误差将对反射跟踪产生一阶效应，从而产生反射相位误差。
您可以测量以短路终止的良好适配器，以查看相位响应是否合理。



     以下为原文

  The 85032B Opens were used in a very early 8510 cal kit specified up to 18 GHz.  It is possible that the same capacitance coefficients were used. I think it is worth a try instead of using the adapter swap technique. The phase error will have a first order effect on refletion tracking and hence a reflection phase error. You can measure a good adapter terminated with a short to see if phase response is reasonable.

> {quote：title = kenwong写道：} {quote}> 85032B Opens用于指定高达18 GHz的早期8510校准套件。
可能使用相同的电容系数。
我认为值得尝试而不是使用适配器交换技术。
相位误差将对反射跟踪产生一阶效应，从而产生反射相位误差。
您可以测量以短路终止的良好适配器，以查看相位响应是否合理。
你知道那个cal kit Ken的型号吗？
如果我能找出系数是什么，那将是有用的。
我认识一些拥有8510s的人，所以如果我知道模型是什么的话，也许可以让他们看看VNA中的系数。
我有一家航空公司 - 85055A验证套件的一部分。
我想这比用于查看相位响应的适配器更好。
虽然我现在对我的电缆有所顾虑，但这将成为另一篇文章的主题。
戴夫



     以下为原文

  > {quote:title=kenwong wrote:}{quote}
> The 85032B Opens were used in a very early 8510 cal kit specified up to 18 GHz.  It is possible that the same capacitance coefficients were used. I think it is worth a try instead of using the adapter swap technique. The phase error will have a first order effect on refletion tracking and hence a reflection phase error. You can measure a good adapter terminated with a short to see if phase response is reasonable.

Do you know the model of that cal kit Ken? If I could find out what the coefficients were, it would be useful. I know a few people with 8510s, so it might be possible to get them to look at the coefficients in the VNA if I knew what the model was. 

I do have an airline - part of an 85055A verification kit. I guess that would be a better choice than an adapter for looking at the phase response. Although I have concerns about my cables now, but that will be the subject of another post. 

Dave

我相信这是HP 85054A校准套件。
它很久以前就被淘汰了，我找不到任何对该套件的引用了。
验证工具包的航空公司可以工作。
您必须使用延迟功能来平移相位响应，以查看相位响应错误。
肯



     以下为原文

  I believe it was the HP 85054A cal kit.  It was obsoleted so long ago that I couldn't find any references to that kit anymore.  

The verification kit's airline can work.  You will have to flatten the phase response by using the delay function to see phase response errors.  

Ken

> {quote：title = kenwong写道：} {quote}>我相信这是HP 85054A校准套件。
它很久以前就被淘汰了，我找不到任何对该套件的引用了。
谢谢肯，这很有帮助。
我会尝试找一个有该工具包手册的人。
>验证工具包的航空公司可以工作。
您必须使用延迟功能来平移相位响应，以查看相位响应错误。
>> Ken如何使用航空公司查看阶段错误？
我有一个85055A验证套件，其中包括一个50欧姆的航空公司和一个25/50欧姆的“Beatty标准”，但没有85032F或85054B，我无法使用验证套件。
但我有兴趣知道如何最好地使用该套件的航空公司部分。
关于VNA的不同类型的校准有很多*，其中许多安捷伦应用笔记都专注于它的某些方面。
但我没有看到很多关于如何验证校准是好还是坏的文章。
Rohde和Schwarz在所谓的T-checker http://www.rohde-schwarz.co.uk/file_5737/1ez43_0e.pdf上有一篇论文，但我没有这些设备，尽管我确实想过
从标准同轴T结中制作出一个。
但是如果我知道如何使用它，航空公司就是一个更有用的工具！
Dave编辑：drkirkby于2013年5月15日10:20 PM编辑：drkirkby于2013年5月15日下午11:04



     以下为原文

  > {quote:title=kenwong wrote:}{quote}
> I believe it was the HP 85054A cal kit.  It was obsoleted so long ago that I couldn't find any references to that kit anymore.  

Thank you Ken, that is helpful. I will try to find someone with a manual on that kit. 

> The verification kit's airline can work.  You will have to flatten the phase response by using the delay function to see phase response errors.  
> 
> Ken

How do I use the airline to see phase errors? I do have an 85055A verification kit, which includes a 50 Ohm airline and a stepped 25/50 Ohm "Beatty standard", but not having either an 85032F or 85054B, I can't use the verification kit. But I'm intersted to know how best I can use the airline part of that kit. 

There is a *lot* written about different sorts of calibration for VNAs, with many Agilent application notes devoted to some aspect of it. But I don't see much written about how to verify if a calibration is good or bad. Rohde and Schwarz have an intersting paper on a so-called T-checker

http://www.rohde-schwarz.co.uk/file_5737/1ez43_0e.pdf

but I don't have one of those devices, although I did think of making one out of a standard coaxial T junction. But the airline is probalby a more useful tool, if only I knew how to use it! 


Dave

Edited by: drkirkby on May 15, 2013 10:20 PM

Edited by: drkirkby on May 15, 2013 11:04 PM

> {quote：title = kenwong写道：} {quote}>我们找到了85054A套件的开路电容系数。
>雌性开放[open（m）] 110E-15,25E-27,170E-36,0;
延迟= 0>雄性开放[open（f）] 64E-15,20E-27,36E-36,0;
延迟= 17.5437ps谢谢。
我确实设法找到了手册 - 或者更确切地说是其他人为我找到了它。
http://cp.literature.agilent.com/litweb/pdf/85054-90001.pdf该信息非常有用。
> T-checker概念听起来很有趣。
然而，推导是基于网络无损的假设，这是微波频率无法满足的条件。
是。
我很想尝试从商业T路口开一个，但担心损失会搞砸。
设备R + S的使用设计为尽可能低损耗。
如果你不在德国，我会认为购买其中一个是相当昂贵的。
显然最好是让德国人买它！
>此外，验证参数似乎只是幅度。
我假设如果相位错误，幅度将是错误的，因为您已经说明相位误差会导致幅度误差。
我怀疑错误可能会在特定频率下消除，但不会在一定频率范围内消除。
>航空公司可用于以多种方式验证VNA系统性能。
其中一个受欢迎的是纹波技术。
当您使用短路终止航空公司时，反射波将跟随测试端口的源阻抗。
扫描宽频率范围时，会产生纹波图案。
峰峰值纹波幅度表示系统的近似残余源匹配。
在时域中，相同的测量提供方向性和源匹配误差的相对指示。
见附件。
>谢谢。
关于为此目的使用航空公司，您有什么好的参考吗？
我在某个地方读过，比如*“除非你看到如何使用它们，否则不要使用航空公司”*。
我已经向安捷伦英国询问了他们运行的任何VNA课程，并且被告知没有计划。
我知道过去有过这样的课程，但显然已经没有了。
我只是看了一下，似乎Anritsu过去常常运行一个免费的课程（甚至更好！）但不再是。
我看过utube，看看有没有关于处理航空公司的事情，但没有这样的运气。
>如果您有实际测量的S21，S12 mag和航空公司的相位，您可以使用它来检查测量的磁场和相位精度。
您确实已将短路的mag和阶段响应添加到比较数据中。
或者您可以先测量短路，然后通过短测量标准化您的测量，以获得仅限航空公司的数据。
我真的想找到关于这个方面的更多细节。
似乎有很多关于VNA校准的文章，但更不用说验证校准。
戴夫



     以下为原文

  > {quote:title=kenwong wrote:}{quote}
> We found the open capacitance coefficients of the 85054A Kit.
> The female open [open(m)]  110E-15, 25E-27, 170E-36, 0; delay = 0
> The male open [open(f)]  64E-15, 20E-27, 36E-36, 0;  delay = 17.5437ps

Thank you. I did manage to find the manual - or rather someone else found it for me. 

http://cp.literature.agilent.com/litweb/pdf/85054-90001.pdf

That information is *very* useful. 

> The T-checker concept sounds interesting.  However, the derivation was based on the assumption that the network is lossless, a condition that can not be met at microwave frequencies.  

Yes. I was tempted to try to make one from a commerical T junction, but had the worries that the losses would mess things up. The device R+S use is designed to be as low-loss as possible. I'm led to believe it is quite expensive to buy one of them if you are not in Germany. Apparently it is best to get someone in Germany to buy it! 

> Also, the verification parameter seemed to be magnitude only.  

I assume if the phases are wrong, the magnitudes will be wrong, since you have already stated phase errors leads to magnitude errors. I suspect the errors could cancel at specific frequencies, but not over a range of frequencies. 

> Airlines can be used to verify VNA system performance in many ways.  One of the msot popular is the ripple technique.  When you terminate the airline with a short, the reflected wave will insteract with the source impedance of the test port.  When swept over a wide frequency range, a ripple pattern is generated.  The peak-to-peak ripple magnitude indicates the approxiate residual source match of the system. In time domain, the same mesurement provide a relative indication of the directivity and source match errors.  see attached. 
> 

Thank you. Do you have any good references about using the airlines for this purpose? 

I read somewhere, something like *"do not use airlines unless you are shown how to use them"*. I've asked Agilent UK about any VNA courses they run, and are told there are none planned. I know in the past there have been such courses, but apparently not any more. I just had a look, and it seems Anritsu used to run a complimentary course (even better!) but not any more. I've looked on utube to see if there was anything about handling airlines, but no such luck. 

> If you have the actual measured S21, S12 mag and phase of the airline, you can use that to check the mag and phase accuracy of your measurement.  You do have add the mag and phase response of the short to your comparison data. Or you can measure the short first and then normalize you measurement by the short measurement to get airline only data.

I'd really like to find more details about this aspect. There seems to be tons written about the calibration of VNAs, but much less so about verifying the calibration. 

Dave

对于校准和纠错，相位误差将转移到幅度。
这可能不是验证的情况，尤其是验证处于固定的幅度值。
有很多关于航空公司特征的论文发表。
参考下面的论文。
还要检查这些文件的参考文献。
K.H.Wong，“使用精密同轴空气介质传输线作为校准和验证标准”，Microwave Journal，Dec。1988 pp 88-92 K.H.
Wong，“通过物理测量对校准标准进行表征”，第69期ARFTG会议摘要，1992年6月安捷伦应用说明AN1287-11，“为安捷伦矢量网络分析仪指定校准标准品和试剂盒”事实上，很少有论文发表在验证上。
NIST的Ron Ginley做了几张工作车间报纸。
您可以对它们进行搜索。
Dough Rytting和Roger Pollard也做了一些关于校准和验证的论文。



     以下为原文

  For calibration and error correction, phase error will transfer to magnitude.  That may not be the case for verification, especially the verification is at a fixed magnitude value. 

There are many published papers on airline characteriztion. References to my papers below. Also check references from these papers.
K.H.Wong, “Using Precision Coaxial Air Dielectric Transmission Line as Calibration and Verification Standards,” Microwave Journal, Dec. 1988 pp 88-92
K.H. Wong, “Characterization of Calibration Standards by Physical Measurements,” 39th ARFTG Conference Digest,  June 1992
Agilent Application Note AN1287-11, “Specifying Calibration Standards and Kits for Agilent Vector Network Analyzers”

Indeed very few papers had been published on verification.  Ron Ginley of NIST did a couple work shop papers.  You may be able to do a search on them.  Dough Rytting and Roger Pollard also did some papers on calibration and verification.