有关N5230C 4端口PNA校准的问题

您可以将任何端口组合用于输入和输出（13-24或12-34）。
只需确保在定义拓扑时将它们连接起来。
如果所有DUT端口都是女性，我建议使用Unknown Thru校准。
唯一需要注意的是，在开始标准4端口校准之前，您需要使用该分析仪执行全局Delta匹配校准。



     以下为原文

  You can use any port combination you like for the input and output (13-24 or 12-34).  Just make sure they are connected as the topology is defined.

I recommend using Unknown Thru calibration if all your DUT ports are female.  The only caveat is that you will need to perform a Global Delta Match calibration with that analyzer before starting the standard 4 port calibration.

只要您只对使用混合模式s参数计算差分/平衡参数感兴趣，“任何端口组合”方案都可以正常工作。
但是，如果您的设备实际上需要使用真差分信号进行激励，那么端口组合限制（差异端口1 = PNA端口1＆amp; 3和差异端口2 = PNA端口2和4）确实适用（假设
你有第二个选择，选项246，在N5230C）编辑：daras在2012年8月20日上午11:21



     以下为原文

  "Any port combination" scheme works fine as long as you are only interested in using mixed-mode s-parameters to compute the differential/balanced parameters.  however, if your device actually needs to be stimulated with a true differential signal, then a port combination restriction (diff. port 1 = PNA ports 1 & 3 and diff. port 2 = PNA ports 2 & 4) does apply (assuming you have the 2nd source option, option 246, in the N5230C)

Edited by: daras on Aug 20, 2012 11:21 AM

亲爱的odanzy和daras，我非常感谢你的帮助。
让我更详细说明一下我的设置：N5230C PNA包括“245：20.0GHz 4端口有效”选项。
我的DUT有4个SMA母连接器（差分输入和差分输出）。
由于我的差分LNA内部有AC接地，我认为它需要用真正的差分信号激励（即两个输入端口信号具有180度的异相）。
另外，我想进一步提出如下问题：（1）对于“直通选择”，以下是选择权吗？
（请假设我设置了差分输入（端口1和3）和差分输出（端口2和4）端口1 - 端口2 - 未知通过端口3 - 端口4 - 未知通过端口1 - 端口3 - 未知直通（2）在“响应>测量平衡>中
拓扑“：我可以选择输入端口（1,3）和输出端口（2,4）。但是，当我执行”向导> SmartCal（引导校准）>选择引导校准端口“时，选择”校准类型选择“为
“4端口校准”，“4端口校准配置”被禁用，如下所示（我无法更改“4端口校准配置”）：选择第一端口：1选择第二端口：2选择第三端口：3选择第四端口
：4这是否意味着我需要选择输入（1,2）和输出（3,4）？我非常感谢你的帮助！-Jay



     以下为原文

  Dear odanzy and daras,

I really appreciate your help. Let me more specify my setup: The N5230C PNA includes the option of “245: 20.0GHz 4-port active”. My DUT has 4 SMA female connectors (differential inputs and differential outputs). Since my differential LNA has AC ground inside it, I think it needs to be stimulated with a true differential signal (i.e. two input port signals have 180-degree out-of-phase).

In addition, I would like to ask further questions as follows:
(1) For the “Thru selection”, is the following selection right? (Please assume I set differential input (port 1&3) and differential output (port 2&4)
port1 - port2 - Unknown Thru
port3 - port4 - Unknown Thru
port1 - port3 - Unknown Thru

(2) In “Response > Measure balanced > Topology”: I can select input ports (1,3) and output ports(2,4). However, when I did “Wizard > SmartCal (Guided Calibration) > select ports for guided calibration” and selected “cal type selection” as “4 port cal”, “4 Port Cal Configuration” is disabled and it shows as follows (I cannot change the “4 Port cal configuration”):
Select 1st Port: 1
Select 2nd Port: 2
Select 3rd Port: 3
Select 4th Port: 4
Does this mean that I need to select input (1,2) and output(3,4)? 

I am very grateful for your help!

-Jay

嗨Jay，我们称之为“True Mode Stimulus Application”或iTMSA的测量类型。
此测量要求您的分析仪在PNA内部和模型上有两个实际源，即选项246.因此，除非您有选项246或4端口PNA-X，否则您无法执行iTMSA测量。
但是，就校准而言，任何4端口校准方法都可以用于iTMSA测量或混合模式s参数方法（一组用于计算差分参数的单端s参数测量）
使用超位置原理）。
至于你的具体问题：1-是任何满足4端口校准的“直通选择”是可以接受的。
2-如果您有一个4端口PNA并且您正在进行4端口校准，那么只有一个有效的端口组合可以校准，即端口1,2,3和4.换句话说，
没有什么可供你选择，所以选择是灰色的。
这样想：如果你有4端口PNA并且想要进行2端口校准，那么你有很多选择 - 校准端口1和2，或1和3，或1和4，或2
3，或2和4，或最后3和4，这是一个组合问题，4个项目一次采取2顺序不重要，给出6选择。
但是，一次只有4个项目只有4个选项，因此灰色选择:-)如果你没有选项246，使用混合模式计算差分参数的方法是完全有效的，如果你的设备
在线性条件下运行。
以下是一些可能有用的信息：本论坛的另一个主题是艾森斯塔特和博克尔曼的论文



     以下为原文

  Hi Jay,
We call the type of measurement you are talking about "True Mode Stimulus Application" or iTMSA.  this measurement requires that your analyzer have two actual sources inside and on your model PNA that is option 246.  So, unless you have option 246 or a 4-port PNA-X, you cannot perform an iTMSA measurement.  However, as far as the calibration is concerned, any 4-port calibration method will work fine for either iTMSA measurements or the mixed mode s-parameter method (a set of single-ended s-parameter measurements that are used to compute the differential parameters using the super-position principle).  
As to your specific questions:
1- yes any of the "thru selections" that satisfies the 4-port cal is acceptable.
2- if you have a 4-port PNA and you are doing a 4-port calibration, then there is only one valid combination of ports that you can calibrate and that is ports 1,2,3, and 4.  in other words, there is nothing for you to select, so the choice is grayed out.  think of it this way: if you have 4-port PNA and you want to do a 2-port calibration, then you have a number of options - calibrate ports 1 and 2, or 1 and 3, or 1 and 4, or 2 and 3, or 2 and 4, or finally 3 and 4, It is a combinatorial problem, 4 items taken 2 at a time where order is not important, gives 6 choice.  However, there is only 1 option for 4 items taken 4 at a time, hence the grayed out selection :-)
if you don't have option 246, using the mixed mode method of computing the differential parameters is perfectly valid, if your device is operating under linear conditions.  Here is some info that might be helpful:
another thread in this forum Paper by Eisenstadt and Bockelman

亲爱的达拉斯，非常感谢您的宝贵意见！
由于我实际上将一个小信号应用到我的DUT（低噪声放大器），我认为它在线性区域运行。
此外，我不需要测量“差异到共同”或“共同差异”。
请让我问你或确认如下：（1）根据你的建议，我认为我仍然可以在DUT的线性区域下测量我的DUT的增益和输入回波损耗。
我对吗？
（2）我可以选择输入端口1和3或1和2吗？
如果我正确设置端口拓扑，两者都可以吗？
（3）当我使用带有电流选项（245）的N5230C时，我可以看到/测量Sdd21，Sdd11（这些是差分测量）和S31，S11（这些是单端测量）。
差分测量和单端测量之间哪些有意义的数据？
（4）我想确定：在端口拓扑中，如果我将输入端口设置为1＆amp; 3并将输出端口设置为2＆amp; 4，那么测量单端增益的S21是否正确？
如果我将输入端口设置为1＆amp; 2并将输出端口设置为3＆amp; 4，那么测量单端增益的S31也是正确的吗？
（我知道，如（3）所示，如果单端测量没用，这个问题没有任何意义。）（5）你能不能再多说一下计算差分参数的混合模式方法
？
对很多问题我很抱歉。
非常感谢你的建议！
-Jay



     以下为原文

  Dear daras,

I greatly appreciate your valuable advice! 
Since I actually apply a small signal to my DUT (low-noise amplifier), I think that it operate in linear region. In addition, I do not need to measure ‘differential-to-common’ or ‘common-to-differential’.

Please let me ask you or make sure with you the follows:
(1) As your suggestion, I think I can still measure the gain and the input return loss of my DUT under the linear region of the DUT. Am I right? 
(2) Can I select the input ports 1&3 or 1&2? Are both okay if I set the port topology correctly?
(3) When I used my N5230C with the current option (245), I can see/do the measurement of Sdd21,Sdd11 (these are differential measure) and S31,S11(these are single-ended measure). Which is the meaningful data between differential measure and single-ended measure?
(4) I would like to make sure: In the port topology, if I set the input ports as 1&3 and the output ports as 2&4, is it correct to measure S21 for the single-ended gain? If I set the input ports as 1&2 and the output ports as 3&4, is it also correct to measure S31 for the single-ended gain? (I know that, as in (3), if the single-ended measure is useless, this question does not have any meaning.)
(5) Could you please a little bit more explain the mixed-mode method of computing the differential parameters?

I am sorry for many questions. Thank you very~ much for your advice!

-Jay

>（1）根据您的建议，我认为我仍然可以测量DUT线性区域下DUT的增益和输入回波损耗。
我对吗？
是>（2）我可以选择输入端口1和3或1和2吗？
如果我正确设置端口拓扑，两者都可以吗？
是>（3）当我使用带有电流选项（245）的N5230C时，我可以看到/测量Sdd21，Sdd11（这些是差分测量）和S31，S11（这些是单端测量）。
差分测量和单端测量之间哪些有意义的数据？
我不确定我是否完全理解这个问题。
Sdd21和Sdd11代表差分增益和差分输入匹配。
如果您的端口拓扑是1-3和2-4，那么S31是差分输入的两个物理端口之间的隔离，而S11只是差分输入的一个端口的单端输入匹配。
>（4）我想确定：在端口拓扑中，如果我将输入端口设置为1＆amp; 3并将输出端口设置为2＆amp; 4，那么测量S21的单端增益是否正确？
是和S43也代表通过DUT的另一条路径中的单端增益。
您还可以将S41和S23视为单端增益的其他组件。
>如果我将输入端口设置为1＆amp; 2并将输出端口设置为3＆amp; 4，那么测量单端增益的S31也是正确的吗？
（我知道，如果在（3）中，如果单端测量没用，这个问题没有任何意义。）是>（5）你能不能再多说一下计算混合模式的方法
差分参数？
我在前一篇文章中引用的论文应该包含这些计算的所有细节 - 比我在这里解释的要好得多，因为撰写论文的人实际上发明了这种方法。
>>对不起，我很抱歉。
非常感谢你的建议！
>> - 杰伊你非常欢迎。



     以下为原文

  > (1) As your suggestion, I think I can still measure the gain and the input return loss of my DUT under the linear region of the DUT. Am I right? 
Yes

> (2) Can I select the input ports 1&3 or 1&2? Are both okay if I set the port topology correctly?
Yes

> (3) When I used my N5230C with the current option (245), I can see/do the measurement of Sdd21,Sdd11 (these are differential measure) and S31,S11(these are single-ended measure). Which is the meaningful data between differential measure and single-ended measure?
I'm not sure if I understand this question fully.  Sdd21 and Sdd11 represent the differential gain and the differential input match.  If your port topology is 1-3 and 2-4, then S31 is the isolation between the two physical ports of the differential input and S11 is simply the single-ended input match of one port of the differential input.

> (4) I would like to make sure: In the port topology, if I set the input ports as 1&3 and the output ports as 2&4, is it correct to measure S21 for the single-ended gain? 
Yes and S43 would also represent the single ended gain in another path through the DUT.  you can also look at S41, and S23 as other components of single ended gain.

>If I set the input ports as 1&2 and the output ports as 3&4, is it also correct to measure S31 for the single-ended gain? (I know that, as in (3), if the single-ended measure is useless, this question does not have any meaning.)
Yes

> (5) Could you please a little bit more explain the mixed-mode method of computing the differential parameters?

The paper that I referenced in the previous post should have all the details of those computations - much better than I can explain here, since the guys who wrote the paper, actually invented that method.

> 
> I am sorry for many questions. Thank you very~ much for your advice!
> 
> -Jay

You are very welcome.

亲爱的达拉斯，非常感谢你的帮助！
我阅读了一些关于混合模式S参数的文档，并使用我的N5230C和当前选项（245）对差分DUT测量进行了以下序列。
你能看一下，让我知道它是否合适？
（1）设置端口拓扑：输入端口（1＆amp; 2）和输出端口（3＆amp; 4）（2）4端口SmartCal（引导）校准（3）测量所有单端S参数，如S11，S21
，S12，S22和S31，S41，S32，S42。
（4）用以下等式计算差分输入回波损耗（Sdd11）和增益（Sdd21）：Sdd11 = 1/2 *（S11-S21-S12 + S22）Sdd21 = 1/2 *（S31-S41-S32 + S42）I
真的很感谢你的帮助！
-Jay



     以下为原文

  Dear daras,

Thank you so much for your help! I read some documents regarding the mixed-mode S-parameters and made the following sequence for the differential DUT measurement with my N5230C with the current option (245). Could you please take a look and let me know whether it is okay or not?

(1) Set the port topology: input ports (1&2) and output ports (3&4)
(2) 4-port SmartCal (Guided) calibration
(3) Measure all single-ended S-parameters such as S11, S21, S12, S22 and S31, S41, S32, S42.
(4) Calculate differential input return loss (Sdd11) and gain (Sdd21) with the following equations:
Sdd11 = 1/2 * (S11- S21- S12+ S22)
Sdd21 = 1/2 * (S31 - S41- S32+ S42)

I really appreciate your help!

-Jay

你好Jay，你实际上不必自己计算Sdd21和Sdd11。
适用于所选拓扑的所有差分和共模参数均可在PNA自己的测量对话框中找到;
你只需要选择它们。
PNA将为您进行计算。
否则你的过程是正确的。



     以下为原文

  Hi Jay,

You don't actually have to calculate Sdd21 and Sdd11 yourself.  All the differential and common mode parameters appropriate for the topology that you have selected are available in the PNA's own measure dialog; you simply have to select them.  the PNA will do the calculations for you.  otherwise your process is correct.

嗨daras，谢谢你的快速回复！
我想问下面的问题。
我的理解是......由于我的N5230C有“245：20.0GHz 4端口有效”选项（非选项246），“真模式激励应用”是不可能的。
因此，我需要使用混合模式S参数。
但是，根据您的回复，可以使用我的N5230C测量差分输入回波损耗和差分增益。
你能告诉我哪里错了吗？
非常感谢，杰伊



     以下为原文

  Hi daras,

Thank you for your quick reply!
I would like to ask the following question. My understanding is that…since my N5230C has an option of “245: 20.0GHz 4-port active” (NOT option 246), “True Mode Stimulus Application” is not possible. Hence, I need to use mixed-mode S-parameters.

However, according to your reply, differential input return loss and differential gain can be measured with my N5230C. Could you please let me know where I am wrong?

Many thanks,
Jay

是 - 选项246仅确定您是否能够将真差分信号应用于DUT，但它与PNA进行平衡/差分测量的能力无关。
在没有真模式激励条件的情况下，差分参数简单地根据在DUT上执行的单端s参数测量来计算。



     以下为原文

  Yes - option 246 only determines if you have the ability to apply a true differential signal to your DUT, but it has nothing to do with the ability of the PNA making balanced/differential measurements.  in the absence of true mode stimulus conditions, the differential parameters are simply computed from single-ended s-parameter measurements performed on the DUT.

嗨daras，这是否意味着我的N5230C可以执行混合模式S参数测量？
谢谢你，杰伊



     以下为原文

  Hi daras,

Does that mean my N5230C can perform the mixed-mode S-parameter measurement as it is?

Thank you so much,
Jay

是



     以下为原文

  YES