是否导出了具有截断内部的同轴电缆的S参数？

大约25年前，我为75欧姆4长槽型N连接器创建了校准系数（4个插槽可以让50个可以配合而不会断开手指，6个短插槽可以提供更好的性能但是总是坏掉）。
在研究计算开放的方法时，我偶然发现了一篇论文，解决了类似于同轴电缆发射到圆形波导中的问题，我记得解决方案基于cyldrical属性，是以bessel函数的形式。
如果你想进一步看，你的googleing和我的一样好。
我最终使用HFSS来计算基于两个极端的S参数：一个是最大的电容（最长的中心引脚，最大的中心导体尺寸，最小的外导体尺寸，一个在最小的电容上，并选择平均值作为valeus
确定校准常数和偏差，以根据开路中的误差确定剩余源匹配。



     以下为原文

  about 25 years ago, I created the cal coefficients for the 75 ohm 4-long slot type Nconnector (the 4 slot would allow a 50 to maybe mate without breaking the fingers, the six-short slots is better performance but always broke).  In doing research for a method to compute the open, I stumbled across a paper solving something like a coax launching into a circular waveguide, and I remember the solution being basing on cyldrical properties, was in the form of a bessel function.  Your googleing is as good as mine if you want to look further.

I ended up using HFSS to compute the S-parameters based on two extremes: one is the largest capacitance (longest center pin, biggest center conductor size, smallest outer conductor size, and one on the smallest capacitance, and chose the mean as the valeus to determine the cal-constants and the deviation to determine the residual source match based on errors in the open.

> {quote：title = Dr_joel写道：} {quote}>大约25年前，我创建了75欧姆4长槽型N连接器的校准系数（4个插槽可以让50个人交配而不会断开手指，
六个短插槽是更好的性能，但总是打破）。
在研究计算开放的方法时，我偶然发现了一篇论文，解决了类似于同轴电缆发射到圆形波导中的问题，我记得解决方案基于cyldrical属性，是以bessel函数的形式。
如果你想进一步看，你的googleing和我的一样好。
谢谢乔尔 - 我会更进一步。
我以前在HP工作时发现了一篇论文“使用高频结构模拟器建模精密连接器”。
这篇论文非常有趣 - 它看起来像是在会议上发表的那种东西。
我个人从未见过75欧姆N，但我可以想象你会遇到与50和75欧姆BNC更常见混合相同的问题。
我从没想过用谷歌搜索同轴电缆 - >圆形波导。
我会尝试的。
一个快速的谷歌找到了很多关于同轴电缆的论文 - >圆形波导过渡，但当然*通常*完全不同，因为在波导的侧面通常有*探头。
但它给了我一个新的搜索词来看。
我想如果连接到同轴电缆并且直径足够大，圆形波导会传播RF信号。
当波导在截止频率以下使用时，我当然感兴趣。
>我最终使用HFSS来计算基于两个极端的S参数：我可以看到HFSS是一种可行的方法 - 可能是最诚实的方式 - 虽然我希望安捷伦的EMPro也能做到这一点。
我自己在HFSS上建模，但我感兴趣的是它是否可以通过分析来完成 - 更多的是出于学术兴趣而不是其他任何事情。
也许Ken Wong对此有所了解 - 他似乎在校准标准方面做了大量工作。
戴夫



     以下为原文

  > {quote:title=Dr_joel wrote:}{quote}
> about 25 years ago, I created the cal coefficients for the 75 ohm 4-long slot type Nconnector (the 4 slot would allow a 50 to maybe mate without breaking the fingers, the six-short slots is better performance but always broke).  In doing research for a method to compute the open, I stumbled across a paper solving something like a coax launching into a circular waveguide, and I remember the solution being basing on cyldrical properties, was in the form of a bessel function.  Your googleing is as good as mine if you want to look further.


Thank you Joel -  I will look further. I had previously found a paper "Modelling precision connectors with a high frequency structure simulator", which you wrote when working at HP. It is quite interesting paper - it looks like the sort of thing delivered at a conference. I've personally never seen a 75 Ohm N, but I can imagine you get the same problems as the more common mixing of 50 and 75 Ohm BNCs. 

I never thought of googling for coax -> circular waveguide. I will try that. A quick Google found many papers for coax -> circular waveguide transitions, but of course that is *normally* quite different, as there is *usually* a probe in the side of the waveguide. But it gives me a new search term to look at.

I would imagine a circular waveguide would propogate an RF signal if attached to coax and the diameter is large enough. I was of course interested when the waveguide would have been used below the cutoff frequency. 

> I ended up using HFSS to compute the S-parameters based on two extremes: 

I can see HFSS is one way to go - probably the best way to be honest - although I would expect Agilent's EMPro could do it just as well. I have modeled it iin HFSS myself, but I was interested if it could be done analytically - more out of academic interest than anything else. 

Perhaps Ken Wong knows of some work on this - he seems to have done a lot of work on calibration standards. 

Dave

我刚刚发现了一篇论文。
乔尔暗示“圆形波导”让我走上正轨。
有趣的是，作者都来自伦敦大学学院，在那里我完成了M.Sc（微波和光电子学）和博士（医学物理学）。
其中一位作者（布莱恩戴维斯）是我在M.Sc.的讲师。
他教了一个名为“字段的计算机建模”的模块。
无论如何，对于其他感兴趣的人来说，这是关于这个主题的一篇论文，但还有更多。
Razaz，M。;
Davies，BJ，“从同轴到圆形波导突然转变的电容”，微波理论与技术，IEEE Transactions on，vol.27，no.6，pp.564,569，1979年6月doi：10.1109 / TMTT.1979.1129672 keywords
：{电容测量;同轴元件;方程;误差分析;区域1;区域2;标准化;波导不连续;波导结;波导过渡}，URL：http：//ieeexplore.ieee.org/stamp/stamp.jsp？tp
Razaz＆amp; amp;＆amp; isnumber = 25065屏蔽的7mm线的边缘电容由Razaz＆amp;
戴维斯在1 kHz时为78.88 fF，在1 GHz时为79.917 fF。
这些值与我在HP 11866A APC-7（7 mm）校准套件上测得的值并没有太大的不同，后者由屏蔽内部组成，但没有像85050B校准套件中的开放标准那样将内部固定到位。
我对HP 11866A open参数的估计值为C0 = 81.5，C1 = 1216.8，C2 = 131.18，C3 = 8.6032。
这些是基于使用85050B校准套件的测量。
https://community.keysight.com/message/59563#comment-59563
本文http://www.gb.nrao.edu/electronics/edir/edir228.pdf中给出的HP 11866A开放的C0值为79 fF，这与Davies和Razaz的计算机模型一致，比我的好一些
测量值为81.5 fF，但没有那么显着的差异。
戴夫
编辑：drkirkby于2013年7月30日晚上8:29



     以下为原文

  I just found a paper on this. Joels hint of the "circular waveguide" got me in the right track. 

The funny thing is, the authors are both from University College London, where I did both my M.Sc (Microwaves and Optoelectronics) and Ph.D (Medical Physics).  One of the authors (Brian Davies) was a lecturer of mine on the M.Sc. He taught a module called "Computer modelling of fields". 

Anyway, for anyone else interested, this is one paper on the subject, but there are many more. 

Razaz, M.; Davies, B.J., "Capacitance of the Abrupt Transition from Coaxial-to-Circular Waveguide," Microwave Theory and Techniques, IEEE Transactions on , vol.27, no.6, pp.564,569, Jun 1979
doi: 10.1109/TMTT.1979.1129672
keywords: {Capacitance measurement;Coaxial components;Equations;Error analysis;Region 1;Region 2;Standardization;Waveguide discontinuities;Waveguide junctions;Waveguide transitions},
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1129672&isnumber=25065

The fringing capacitance of a shielded 7 mm line was computed by Razaz & Davies to be 78.88 fF at 1 kHz and 79.917 fF at 1 GHz. Those value are not wildly different to what I measured on a HP 11866A APC-7 (7 mm) calibration kit which consists of a shielded inner, but nothing to hold the inner in place like on the open standard in the 85050B cal kit. My estimates of the parameters for the HP 11866A open were C0=81.5, C1=1216.8, C2=131.18, C3=8.6032. Those were based on measurement using an 85050B cal kit. 
https://community.keysight.com/message/59563#comment-59563
 

The value for C0 for the HP 11866A open given in this paper 

http://www.gb.nrao.edu/electronics/edir/edir228.pdf

is 79 fF, which agrees with the computer model of Davies and Razaz somewhat better than my measurement of 81.5 fF, but not so dramatically different. 

Dave
 
Edited by: drkirkby on Jul 30, 2013 8:29 PM

如果您搜索“屏蔽同轴开路”，您将找到其他参考。
Waveguid手册涵盖了这种结构之一。
其他出版物包括：W.C。
Daywitt，“具有空心中心导体的有损同轴开路的复合导纳”，Metrologia 24,13-22（1987）M. Horibe，N。Ridler，“毫米矢量网络的同轴开路初级标准的特征化和验证”
分析仪校准“，第76届ARFTG会议2010年12月开放式结构的边缘区域对截断的同轴线的角半径非常敏感。
没有可以处理它的紧密形式的解决方案。
对于具有电介质支撑的截断线，边缘场的一部分在空气中，而部分边缘在电介质支撑中。
没有良好的3D E＆amp; M模拟器，该结构太复杂而无法分析。
基于TRL或偏移短校准技术进行测量要容易得多。
不得在校准中使用开路。
我做了两个并得到了很好的协议。
肯



     以下为原文

  If you search "shielded coaxial open", you will find additional references. The Waveguid Handbook covers one of such structure.  Other  publications include:

W.C. Daywitt, "Complex admittance of a Lossy Coaxial Open Circuit with a Hollow Center Conductor", Metrologia 24, 13-22 (1987)

M. Horibe, N. Ridler, "Characterization and Verification of Coaxial Open-circuit Primary Standards for Millimeter Vector Network Analyzer Calibration", 76th ARFTG Conference  Dec. 2010

The fringing field of the open structure is very sensitive to the corner radius of the truncated coaxial line.  There is not close form solution that can deal with it.

For truncated line with a dielectric support, part of the the fringing field is in air and part of it is in the dielectric support.  That structure is too complicated to analyze without a good 3D E&M simulator.  It is much easier to measure based on TRL or offset short calibration techniques. An open must not be used in the calibration.  I did both and got good agreements.

Ken

> {quote：title = kenwong写道：} {quote}>如果搜索“屏蔽同轴开路”，你会发现其他参考资料。
谢谢。
> Waveguid手册涵盖了这种结构之一。
其他出版物包括：我没有注意到它。
我从亚马逊那里买了一本相当便宜的书。
我再看一遍。
> W.C.
Daywitt，“具有中空导体的有损同轴开路的复杂导纳”，Metrologia 24,13-22（1987）这听起来无关紧要，但我会明白的。
有时在论文中有非常有用的信息，这些信息似乎与人们所期待的完全无关。
当然，还有一些案例，即论文看起来与其标题非常相关，但实际上没有有用的信息。
> M. Horibe，N。Ridler，“用于毫米波矢量网络分析仪校准的同轴开路初级标准的表征和验证”，第76届ARFTG会议2010年12月我将尝试追踪这一点。
>开放结构的边缘区域对截断的同轴线的角半径非常敏感。
没有可以处理它的紧密形式的解决方案。
我可以想象，如果它被切成一个封闭形式的解决方案可能是可能的。
尽管如此，还是需要有人比我更好。
>对于带有电介质支撑的截断线，边缘区域的一部分在空气中，而部分区域在电介质支撑中。
没有良好的3D E＆amp; M模拟器，该结构太复杂而无法分析。
我会在HFSS中尝试这个。
设置应该是一件相当琐碎的事情。
>基于TRL或偏移短校准技术进行测量要容易得多。
不得在校准中使用开路。
我做了两个并得到了很好的协议。
我的8720D没有选项400，它增加了第四个接收器，所以我只限于TRL *，虽然我不认为如果我用一些好的衰减器改进原始匹配会太糟糕。
至少对于高反射装置，应该有足够的信号。
我不知道8720D是否支持同轴电缆的偏移短校准。
我将不得不检查手册，因为我从未在同轴电缆中看到它的一个例子 - 只有波导。
>肯谢谢你的帮助肯。
戴夫



     以下为原文

  > {quote:title=kenwong wrote:}{quote}
> If you search "shielded coaxial open", you will find additional references. 

Thank you. 

> The Waveguid Handbook covers one of such structure.  Other  publications include:

I did not notice it. I bought a copy of that book fairly cheaply used from Amazon. I'll look again. 

> W.C. Daywitt, "Complex admittance of a Lossy Coaxial Open Circuit with a Hollow Center Conductor", Metrologia 24, 13-22 (1987)

That sounds irrelevant, but I'll get it. Sometimes there is very useful information in papers which seem totally unrelated to what one is looking for. Of couse, there is there are also the cases when papers seem very relevant from their title, but actually have no useful information. 

> M. Horibe, N. Ridler, "Characterization and Verification of Coaxial Open-circuit Primary Standards for Millimeter Vector Network Analyzer Calibration", 76th ARFTG Conference  Dec. 2010

I'll try to track that down. 

> The fringing field of the open structure is very sensitive to the corner radius of the truncated coaxial line.  There is not close form solution that can deal with it.

I could imagine that if it is squarly cut a closed form solution might be possible. It would need someone better than me to come up with one though. 

> For truncated line with a dielectric support, part of the the fringing field is in air and part of it is in the dielectric support.  That structure is too complicated to analyze without a good 3D E&M simulator. 

I'll try this in HFSS. It should be a fairly trivual thing to set up. 

> It is much easier to measure based on TRL or offset short calibration techniques. An open must not be used in the calibration.  I did both and got good agreements.

My 8720D does not have option 400 which adds a 4th receiver, so I am limited to TRL*, though I don't suppose that would be too bad if I improved the raw match with some good attenuators. At least with high reflection devices, there should be plety of signal. 

I don't know if the 8720D supports an offset short calibration in coax. I'll have to check the manual, since I have never seen an example of it in coax - only waveguide. 

> Ken

Thank you for your help Ken. 

Dave