SOLT与不同的s11和s22为2端口cal

您好，基于您为屏幕截图命名的内容（“一对适配器的s11和s22”），我们可以假设您将两个适配器连接在一起，而适配器＃1的另一端连接到PNA的端口1，
和PNA的端口2的适配器＃2的另一端是这样的吗？
但是，即使您只是在端口1和端口2之间一次连接每个适配器，通过连接类型的S11测量也会测量PNA端口2的负载匹配（从端口2反射回来的信号）
并且在端口1）处接收，并且S22正在测量端口1的负载匹配。当您的校准仅为1端口校准时，在校准期间没有测量到第二端口，因此1端口校准没有
纠正第二个港口的比赛。
但是，当您执行2端口校准时，在该校准期间进行测量，以便2端口校准对两个端口的负载匹配进行校正。
1端口的cals必须假设负载匹配是理想的，这就是为什么当你应用两个1端口的cals时，在测试端口终止的适配器的S11和S22测量看起来更相似。



     以下为原文

  Hello, based on what you named your screenshot ("s11 and s22 of a pair of adapters"), we might assume that you have the two adapters connected together, and the other end of adapter #1 connected to port 1 of the PNA, and the other end of adapter #2 to port 2 of the PNA, is that the case?  However, even if you were just connecting each adapter one-at-a-time between ports 1 and 2, an S11 measurement of a thru type of connection is measuring the Load Match of the PNA port 2 (signal that reflects back from port 2 and received at port 1), and S22 is measuring load match of port 1.  When you have a calibration that is only a 1-port cal, nothing about the 2nd port was measured during the cal, so the 1-port cal does no correction for the match of that 2nd port.  But when you do the 2-port cal, thru measurements are made during that cal so that the 2-port cal makes corrections for the load matches of both ports.  The 1-port cals have to assume that load match is ideal so that would be why the S11 and S22 measurements of your adapter(s) terminated in the testports look more similar when you apply the two 1-port cals.

谢谢。
如果我正确读到这个，你的意思是这是由2端口校准中使用的公负载之间的差异造成的吗？
这导致端口匹配差异？



     以下为原文

  Thanks. 

If i read this correctly, do you mean this is caused by the difference between the male and female loads used in 2 port calibration? which as a result caused port match difference?

我不清楚你在测量什么以及如何在1端口校准情况下终止设备。
对于1端口校准，输入反射是输入端口重新感知和输出端口终端反射的总和。
S11（测量）〜= S11 + S21 * S12 * S11（终止）通常，任何实际装置的S11和S22都不相同。
连接器的尺寸和物理尺寸不同（男性与女性等）。
即使连接器属于同一类型且校准标准相同，S11和S22仍可能不同。
许多物理因素导致了差异。
如果用于两个端口的校准标准相同，则S11和S22对于非常低损耗的设备更相似。
校准标准的差异只会增加差异。



     以下为原文

  It is unclear to me what you were measureing and how the device was terminated in a 1-port calibration situation.
For a 1-port calibration, the input reflection is the sum of the input port relfection and the reflection of the ouput port termination.  
      S11(measured)  ~=  S11 + S21*S12*S11(termination)
 
In general, S11 and S22 of any real device are not the same.  The connectors are not the same dimensionally and physically ( male vs. female etc.).  
Even if the connectors are of the same type and the calibration standards are the same, S11 and S22 can still be different.  Many physical factors contribute to the differences.  S11 and S22 are more similar for very low loss devices if the calibration standards used for both ports are the same. Differences in calibration standard just add more to the difference.

实际上，我采访了Lucux，在这种情况下，我希望S11和S22非常接近，当然不会有10 dB的差异。
DUT是一个非常短（20毫米长）的f-f适配器。
当然，高达6 GHz，这不长。
进行了新的校准，其中使用的设备插入到未知的通道中，因此PNA两侧的标准非常相同。
相同的开放，短期和负载。
使用相同负载的S11和S22的单独测量实际上是相同的。
但是，当插入2端口校准时，S11和S22之间存在很大差异。



     以下为原文

  Actually, I spoke to Lucux and in this case, I would expect S11 and S22 to be very close and certainly not 10 dB difference.  The DUT is a very short (20 mm long) f-f adapter.  Certainly, up to 6 GHz, this is not long.  A new calibration was done where the device used was inserted into an unkown thru cal, so the standards on both sides of the PNA are exacly the same. Same Open, Short and Load. Individual measurements of S11 and S22 using the same load are virtually the same.  However, when inserted in the 2 port cal, there is a large difference between S11 and S22.

您是否从1端口校准中绘制了S11和S22数据？
您认为您的开放，短期和负载标准有多好？
适配器DUT是否与在unknonw直通校准期间使用的相同？
你能显示S21和S12数据吗？
您的测试端口连接器，测试端口电缆的重复性如何？
连接器和电缆重复性误差很容易导致S11和S22在-40 dB左右的差异。
肯



     以下为原文

  Do you plots of your S11 and S22 data from the 1-port calibrations?  
How good do you think your open, short and load standard are? 
Is the adapter DUT the same one you used during the unknonw thru calibration?  Can you show the S21 and S12 data also?
How repeatable are your test port connectors, your test port cables?  Connector and cable repeatability errors can easily caused the differences in S11 and S22 at around -40 dB level. 

Ken