请问ECAL校准的温度会影响测量结果吗？

> {quote：title = Danhen 2014写道：} {quote}> * 2 - *考虑到问题1，我想知道文献中是否有人提到温度对校准标准的影响？
>问候>>丹尼尔我在另一个帖子中注意到你是硕士学位的学生。
你应该自己做一个文献检索。
有时候，如果其他人可以推荐论文，那将是有帮助的，我知道有时这里的人们已经建议了一些关键词来搜索，这是我从未想过的。
但我原以为有人会发表关于温度依赖于几乎所有水的性质的材料 - 密度，粘度，介电常数等等。你可以找到自己的论文。
我觉得你的校准标准只是水，这是正确的吗？
戴夫



     以下为原文

  > {quote:title=Danhen 2014 wrote:}{quote}
> *2 -* Considering the question 1, I would like know if there is someone reference in literature about the influence of temperature in calibration standards ?
> Regards 
> 
> Daniel

I note in another thread you are a masters degree student. You should be doing a literature search yourself on that. Sometimes it is helpful if others can suggest papers, and I know sometimes people on here have suggested some key words to search for, which I would never have thought of. But I would have thought someone would have published material on the temperature dependance of just about every property of water - density, viscosity, permittivity etc etc. You can probably find your own papers on that. I think you calibration standard is just water, is that correct? 

Dave

嗨drkirkby，“配置校准”-85070E的最常见选项有三种校准标准。
它们是开放式，短型和水。
但是，只能改变水的温度。
我认为其他标准的温度也应该修改，但是这个程序不允许这个选项。
我找到了关于校准标准的反射系数与其各自温度之间关系的文章。
但我还没有找到大量的信息。



     以下为原文

  Hi drkirkby,

There are three calibration standards on most common option of "Configure Cal" -85070E . They are Open, Short and Water. However only temperature of water can be changed. I  think that temperature of the others standards also should be modified, but this program don't permit this option.

I've finding articles about the relationship between the reflection coefficient of the calibration standards and its respective temperatures. But I haven't find considerable quantity of information yet.

> {quote：title = Danhen 2014写道：} {quote}>我对Ecal校准，85070E软件和校准标准有一些疑问;
> * 3 - *在我的测量设置中，我有ECAL，ENA E5070,85070D探头，85070E软件和所有重要配件。
我怎样才能测量校正后的系数反射？
这些论坛上一个非常普遍的问题是人们没有提供足够的信息。
你提供的不仅仅是很多，但我认为还不够。
在这种情况下，*您测量ECAL，但没有提到什么型号或选项。
*你说“ENA E5070”，但据我所知，有一个A和B版本。
它也会有一些软件版本，对支持你的人有用。
*我查阅了* 85070E *探针上的手册，可以看到有各种选项。
我不知道D是否也是如此，如果是，你应该给他们。
*您尚未提供85070E软件的版本。
我经常想到自己做一个类似的探测，所以我真的会对探测结束的一些特写照片以及你对它的结构和短路的信息感兴趣。
我经常想到使用APC7作为探针。
可以用APC7校准套件校准，然后一个校准表面。
我认为人们需要移除夹头，测量内导体的衰退，并观察建模。
无论如何，这只是我的一些想法 - 没有足够的时间来探索我的所有想法。
但回到你在另一个线程中询问的关于校准标准的温度依赖性的问题，我认为短路或开路都不重要，这可能就是为什么85070E软件不允许你设置它们。
这是我的逻辑。
1）短路的目的可能是将平坦表面推向探头。
如果短路的温度很高，它会膨胀，但仍然会与探头接触。
所以我认为它根本没有任何区别。
电导率的变化理论上会产生一些影响，但我认为与其他误差源相比，它的影响可以忽略不计2）开路的目的是获得一些相位参考 - 我希望最佳值是相角180
远离短路的度数。
如果探头周围的空气温度发生变化，我相信对空气介电常数的影响会非常小，除非你做的事情就像使用液化空气一样。
使用普通的VNA和校准套件，手册说它应该在1摄氏度内校准和使用。我相信如果环境温度变化还有很多变化 - 所有电缆，半导体等的长度和直径。正如我所说，我
从来没有使用过这些，尽管我已经在很多期刊上读到了这种技术。
在测量人体乳腺组织的介电特性以进行癌症检测方面似乎有很多工作 - 在我看来，在医学物理学领域工作了数十年，完全浪费时间，尽管没有特别关注。
我认为你真的需要安捷伦员工的技术支持 - 不是像我这样感兴趣，但对此知之甚少的人。
我刚刚尝试运用我的工程背景来对所发生的事情进行有根据的猜测。
显然，加热空气或提高海拔高度会使每单位体积的分子比低空冷空气更少。
我觉得有点难以相信这种影响会非常显着。
Dave编辑：drkirkby于2014年7月23日下午1:09



     以下为原文

  > {quote:title=Danhen 2014 wrote:}{quote}
> I've some questions about Ecal calibration, 85070E software and calibration standards;  

> *3 -* In my measurement setup I have ECAL, ENA E5070, 85070D probe,  85070E software and all important acessories. How could I  measure the corrected coeffecient reflection ?

A really common problem on these forums is people not providing sufficient information. You have provided more than many, but still insufficient in my opinion. In this instance, 

* You measure ECAL, but no mention of what model or options. 

* You say "ENA E5070", but as far as I can tell, there is an A and a B version of this. Also it is going to have some software version, which would be useful to someone supporting you. 

* I looked up the manual on the *85070E* probe, and can see there are various options. I don't know if that is true of the D too, but if it is, you should give them. 

* You have not given the version of the 85070E software. 

I've often thought of making a similar probe myself, so I would really be interested in some close up photos of the end of the probe, and information you have about its structure, and the short. 

I've often thought about the use of an APC7 as the probe. One could calibrate that with an APC7 cal kit, and then one has  a calibrated surface. I think one would need to remove the collet, measure the recession of the inner conductor, and look at modelling that. Anyway, that is just some ideas I have - never enough time to explore all the ideas I have. 

But going back to what you asked in another thread about the temperature dependance of the calibration standards, I don't think that of the short or open will matter at all, which is probably why the 85070E software does not allow you to set them. This is my logic. 

1) The purpose of the short is likely to be to push a flat surface against the probe. If the temperature of the short is high, it will expand, but willl still be in contact with the probe. So I don't think it makes any difference at all. The change in electrical conductivity would theoretically have some effect, but I think its effect would be negligible compared to other source of error

2) The purpose of the open is to obtain some phase reference - I would expect the optimal value is a phase angle 180 degrees away from the short. If the temperature of the air around the probe changes, I believe the effect on the permittivity of the air would be incredibly small, unless you did things like use liquified air. 

Using a normal VNA and calibration kit, the manuals say it should be calibrated and used within 1 deg C. I believe there is far more to change if ambient temperature changes - length and diameter of all cables, semiconductors etc. 

As I say, I have never used ones of these, though I have read about the technique in a number of journals. 

There seems a lot of working going into measuring the dielectric properties of human breast tissue for cancer detection - a complete waste of time in my opinion having spent decades working in the area of medical physics, although not specifically on that. 

I think you really need technical support from Agilent staff here - not someone like myself who is interested, but has little knowledge of it. I've just tried to apply my engineering background to take educated guesses as to what would happen. 

Clearly heating air, or raising altitude would give you less molecules per unit volume that cold air at low altitude. I find it a bit hard to believe the effects of that are going to be very significant. 

Dave

Edited by: drkirkby on Jul 23, 2014 1:09 PM

简而言之，水的电特性随温度而变化。
它们在标准温度如25℃下是众所周知的，因此可以用作校准标准。
如果随后改变温度，则特性将发生变化，测量显示当然也会随着电气特性在高温下的不同而与标准室温相比发生变化。
因此，您必须控制用作校准标准的水的温度。
只要环境温度为20-30℃，Ecal就内置了温度控制。机械标准的性能与温度几乎没有变化，因此开启或短路时需要考虑温度。



     以下为原文

  In short, the electrical characteristics of water changes with temperature. They are well known at standard temperatures such as 25C, and so can be used as a calibration standard.  If you then change the temperature, the characteristics will change, and the measurement display will of course also change as the electrical properties are different at high temperature as contrasted with standard room temperature. 

Thus, you must control the temperature of water used as a cal standard.

Ecal has built in temperature control as long as the environment temperature is 20-30 C

Mechanical standards have almost no variation in performance vs temperature so temperature needed be taken into account for opens or shorts.

嗨乔尔博士，谢谢你的回答。
然而，仍然有一些奇怪的东西;
即使短路的性能不随温度变化......与探头接触时短路，会通过导热改变探头的温度。
如果可以使用温度值设置三个校准标准，那么校准的准确性将更好。
每个校准标准具有一个反射系数，该系数作用于VNA的系统误差。
我确定，不是吗？



     以下为原文

  Hi Dr Joel,

Thank you for you answer.

However there is still something strange; even if the performance of short don't change with temperature... The short when in contact with probe, will change the temperature of probe by heat conducting.

If the three calibration standards could be set with temperature value, then the accuracy of calibration would be bettter. Each calibration standard have a refflection coefficient which acts on systematic erros of VNA. Am I sure, aren't I ?

> {quote：title = Danhen 2014写道：} {quote}>嗨Joel博士，>>如果三个校准标准可以设置温度值，那么校准的准确性会更好。
每个校准标准具有一个反射系数，该系数作用于VNA的系统误差。
我确定，不是吗？
你可能会确定，但我不认为你是对的......虽然理论上可能短时间的温度会影响反射的价值到很小的程度，但我们认为变化不足以使任何变化
实际差异所以不包括它。
但是，水的温度对水的反射系数有显着影响，因此必须包括在内。
但是，如果你不相信我（为什么你应该？）并且作为读者的练习，请执行以下操作：谷歌铜的扩展系数（我们使用BeCu，但它主要是铜）然后计算
电气长度约为17 ps的短路每长度变化。
（提示，首先必须将电气长度转换为物理长度，假设c = 3e8）。
由于长度变化，计算10 GHz的相位变化。
考虑到open没有改变，计算由于长度变化对源匹配的影响。
假设方向性误差等于源匹配误差，最后，计算源匹配误差对mag 1角度90的反射测量的影响。这是一个很好的作业问题....问你关于这个问题的任意数量的问题
，每个问题都需要啤酒。



     以下为原文

  > {quote:title=Danhen 2014 wrote:}{quote}
> Hi Dr Joel,
> 
> If the three calibration standards could be set with temperature value, then the accuracy of calibration would be bettter. Each calibration standard have a refflection coefficient which acts on systematic erros of VNA. Am I sure, aren't I ?
  You might be sure, but I don't think your are right...

While it is theoretically possible that the short temperature affects the value of the reflection to some very small extent, we don't consider the change large enough to make any practical difference so don't include it.  However, the temperature of the water has a dramatic affect on the reflection coefficient of the water, so it must be included.

But, if you don't trust me (and why should you?) and as an excercise to the reader, please do the following:

Google the coefficient of expansion for copper (we use BeCu, but it is mostly copper) and then compute the length change per deg C of a short with an eletrical length of about 17 ps. (hint, you first have to convert electrical length to physical length, assume c=3e8).  The compute the phase change at 10 GHz due to length change. Considering that the open is not changed, compute the effect on source match due to the change in length.  Assume directivity error equals source match error, and finally, compute the effect of the source match error on the measurement of an reflection of mag 1 angle 90.

What a nice homework problem....

Ask as many questons as you like about the problem, each question costs a beer.

关心Dr_joel，非常感谢您的回答，遗憾的是我在最后几天无法访问该论坛。
我还有一些关于校准标准的问题：1 - 在一个端口的误差校正矩阵中，有测量和理想反射系数的值。
我想知道这些理想反射系数是否是所有频率和温度的常数，或者它们是否随频率和温度而变化2 - 关于“开放式同轴探头”和测量的其他问题。
对测量的影响是样品的温度还是同轴探头的温度？
您已经提供了有关校准非常重要的信息。
谢谢你的帮助。



     以下为原文

  Regards Dr_joel,

Thank very much for your answers, unfortunately I couldn't  to access the forum in the last days.

I've still some questions about the calibration standards:

1 - In the matrix for errors correction  of one port, there are values of measured and ideal reflection coefficient. I would like know if these ideals reflection coefficients are constants for the all frequencies and temperatures or if them change with frequency and temperature 


2 - Other question about "open ended coaxial probe" and measurements. What influences on measurements is the  temperature of samples or the temperature of the coaxial probe head ?


You've given  informations very important about the calibration.  Thank you by help.

> {quote：title = Danhen 2014写道：} {quote}> 1 - 在一个端口的错误校正矩阵中，有测量和理想反射系数的值。
我想知道这些理想反射系数是否是所有频率和温度的常数，或者它们是否随频率和温度而变化>标准模型随着开路和短路的频率而变化，假设负载是理想的（反射= 0）
，但可以为任何标准添加基于模型或基于数据的标准。
所有机械标准都假定为温度恒定。
我不是材料测量软件的专家，但我知道使用水的标准取决于水的温度。
>> 2 - 关于“开放式同轴探头”和测量的其他问题。
对测量的影响是样品的温度还是同轴探头的温度？
>我相信探头会稍微改变（例如，由于加热膨胀可能会变长），但效果很小。
如果您有一个热样品，并且您想要加热探头以确保它与样品相同，则可能需要加热开口和短路，当然也需要水。
但你可能喜欢冒险使用



     以下为原文

  > {quote:title=Danhen 2014 wrote:}{quote}
> 1 - In the matrix for errors correction  of one port, there are values of measured and ideal reflection coefficient. I would like know if these ideals reflection coefficients are constants for the all frequencies and temperatures or if them change with frequency and temperature 
> 
The model of the standard varies with freq for open and short, load is presumed ideal (reflection = 0) by default, but it possible to add a model based or data-based standard for any of the standards.  All mechanical standards are presumed constant with temperature.  I'm not an expert on the material measuerments software but i understand that the standard using water DOES depend on temperature of the water.
> 
> 2 - Other question about "open ended coaxial probe" and measurements. What influences on measurements is the  temperature of samples or the temperature of the coaxial probe head ?
> 
I beleive the probe would change slightly (for example, it might get longer due to heating expansion) but he effects are small.  If you have a hot sample, and you want to heat the probe to ensure it is the same as the sample, heating the open and short is probably needed too, and the water of course.  But you could like risk using

我很遗憾地说你的问题形象不完整且缺乏背景，所以我只能用一些假设来回答：1）我假设你想知道在高温下样品的特性，2）我认为
当温度从冷变为热时，样品的特性变化X％3）我假设用作标准之一的水的特性在知道Y％，Y的性质时有一个已知量的变化。
大约3％（只是一个猜测）4）我假设，但相信是真的，探针的特征从热到冷变为Z％，并且Z可能在5％的数量级5）
我假设，但是相信是真的，开放和短暂的特征从热到冷变化很小，m（对于最小）并且大约为1％6）我假设使用的错误
测量热样品到局部冷却的冷探针是T％，也许T是5％？
7）我假设探头的特性变化较小，可能是Z / 10，当测量冷标准时局部冷却（因为只有一部分探头被冷却）8）我认为冷是室温然后
相对精度将是（假设误差不相关，除了开路，短路和探头，其误差可能由于膨胀而随温度线性变化）类型1：sqrt（Y ^ 2 + 2 * m ^ 2）
类型2 Z / 10类型3 X类型4不适用，为什么要这样做？
类型5：使用冷探头并测量样品热：T



     以下为原文

  i'm sorry to say your questions are ill-formed and lack context, so I can only answer them with some assumptions:

1) I assume you want to know the characteristics of a sample at a hot temperature,
2) I assume that the characteristics of the sample change by X% when its temperature changes from cold to hot
3) I assume that the characteristics of water, used as one of the standards, changes by an known amount with an error in knowing the properties of Y%, Y being on the order of 3% (just a guess)
4) I assume, but believe is true, that the characteristics of the probe change by Z% from hot to cold, and that Z is perhaps on the order of 5%
5) I assume, but believe is true, that the characteristics of the open and short change by a very small amount from hot to cold, m (for minimal) and that is on the order of 1%
6)I assume that the error of using a cold probe to measure a hot sample, to localized cooling, is T%, maybe T is 5%?
7) I assume the characteristics of the probe change by a smaller amount, maybe Z/10, when it is locally cooled if measuring a cold standard (since only a portion of the probe is cooled)
8) I assume cold is room temperature

Then the relative precisions would be (presuming the errors are not correlated, except for the open, short and probe which have an error which likley changes linearly with temperature due to expansion)

Type 1: sqrt(Y^2+2*m^2)
Type 2  Z/10
Type 3  X
Type 4 Does not apply, why would you do this?
Type 5: Use a probe cold and measure the sample hot: T

关心Dr_joel，非常感谢您的信息。
这些信息非常有用。
在不久的将来，我将分享有关同轴探头的一些应用的信息。



     以下为原文

  Regards Dr_joel, 

Thank very much by your informations. These informations will be very useful. In the near future I will share information about some applications of coaxial probe.

嗨Dr_joel，请原谅我这个新问题，在类型5中，您是否认为校准“在样品的相同温度下”或“在样品的不同温度下”？
您是否知道一些纸张或书籍显示如何量化这些数量？



     以下为原文

  Hi Dr_joel,

Excuse-me by this new question, 

In type 5,  did you consider the calibration "in same temperature of the sample" or "in different temperature of the sample" ? 

Do you know some paper or book that show how quantify  these quantities in numbers ?

在您的表中，您始终将探头和样品放在相同的温度下。
我建议使用另一种类型：探针在一个温度下校准并测量样品，样品处于不同的温度。
这似乎是一种可能发生的合理情况。
不幸的是，我知道没有参考，但你可以做一些实验，并公布结果！



     以下为原文

  In your table, you always had the probe and sample at the same temperature.  I was suggesting another type: probe calibrated at one temperature and measureing the sample, where the sample is at a different temperature.  It seems a reasonable situation that might occur.

Unfortunately, I know of no references, but you could do a few experiments, and publish the results!

非常感谢Dr_joel每个人都提供的帮助非常重要。
之后，我将在这里发布关于我的硕士学位项目的一些考虑因素和结果。



     以下为原文

  Thank very much Dr_joel

The help that everyone has been providing is very important. After,  I'll post here some considerations and results about my Master degree project.