SMD校准结果准确度如何

> {quote：title = PP写道：} {quote}> 1）在动量模拟中，SMD校准用于DIode和负载焊盘。
但我在论坛上看到，如果存在接地的寄生分流路径，则SMD校准是不准确的。
我想你指的是关于这个主题的帖子。
是的，SMD校准存在基本限制：连接到SMD端口的任何分流元件都是浮动的，并且没有像我们预期的那样连接到全局接地。
但是，我们需要了解这个问题的维度：它只会影响那里连接的S2P的分流路径。
系列路径没问题。
如果S2P数据中的分流路径相关，我们将看到此端口配置出错。
一个例子是S2P数据，其已经包括焊盘或其他大的接地电容。
在您的情况下，垫片包含在Momentum布局中，而S2P组件数据完全（或大部分）是系列元素。
对于这些情况，您可以使用SMD端口校准。
>如果是这样，那么TML（零长度）会给出正确的结果吗？
如果使用TML（ZL）校准，结果的准确度如何。
我会留下这个没有答案的。
也许安捷伦人想要加入？
> 2）由于所使用的负载是SMD元件，并且没有分流寄生路径为其进行探测，因此SMD校准应该起作用。
对，就是这样。
>但是我们如何计算负载一端的准确电压。
负载的另一端通过电路接地，但是当使用差分端口时，接地端的电压不为零。
我已经附加了布局，“终端2”接地，并想要计算“终端1”的准确电压。
对于使用SMD校准的EM仿真，端子1和2分组为一个差分端口。
这为我们提供了端口两端的电压（端子​​1和端子2之间的电压差），但是我们没有从端子1到地或从端子2到地的电压。
在您的情况下，端子2连接到全局接地端，端口上的电压和端子1上的电压几乎相同。
（我写的“几乎相同”，因为来自通孔有一些串联电感/电阻）> 3）需要计算SMD负载电阻上的功率（使用SMD校准）。我已经使用负载上的差分电压来计算
权力，这种方法是否正确？
是



     以下为原文

  > {quote:title=PP wrote:}{quote}
> 1) In momentum simulation the SMD calibration is used for the DIode and load pads . But i read on the forums that the SMD calibration is inaccurate if there is a parasitic shunt path to ground.

I think you are referring to my posts on this topic. Yes, there is a fundamentation limitation of SMD calibration: any shunt elements connected to the SMD port are floating, and not connected to global ground as we would expect.

However, we need to understand the dimension of this problem: it only affects the shunt path of S2P connected there. The series path is ok. We will see an error from this port configuration if shunt path in the S2P data is relevant. One example would be S2P data that already includes pads, or other large capacitance to ground.

In your case, the pads are included in the Momentum layout, and the S2P component data is completely (or mostly) series element. For these cases, you can use SMD port calibration.

> If so then will the TML(zero length) gives correct result ? How accurate are the results if TML(ZL) calibration is used.

I'll leave this unanswered. Maybe the Agilent guys want to jump in?

> 2) As the load used is SMD component , and there is no shunt parasitic path to groound for it , SMD calibration should work for it. 

Yes, exactly.

> But how do we calculate the exact voltage at the one end of the load. The other end of the load is grounded through the circuit, but as differential ports are used the voltage at the ground end is not zero. i have attached the layout, "Terminal 2" is grounded , and want to calculate exact voltage at the "Terminal 1".

For EM simulation with SMD cal, Terminal 1 and 2 are grouped into one differential port. This gives us the voltage across the port (voltage difference between Terminal 1 and 2), but we don't have the voltage from Terminal 1 to ground, or from Terminal 2 to ground.  In your case, where Terminal 2 is connected to the global ground, voltage across the port and voltage at Terminal 1 are almost the same. (I wrote "almost the same" because there is some series inductance/resistance from the via)
 
> 3)Need to  calculate the power across the SMD load Resistor(SMD calibration used).I have used the differential voltage across the load to calculate the power, is this method correct?

Yes

谢谢回复。
我有另一个问题。
----对于SMD校准的EM仿真，1号和2号终端分为一个差分端口。
这为我们提供了端口两端的电压（端子​​1和端子2之间的电压差），但是我们没有从端子1到地或从端子2到地的电压。
在您的情况下，端子2连接到全局接地端，端口上的电压和端子1上的电压几乎相同。
（我写的“几乎相同”，因为通孔中存在一些串联电感/电阻）您是在谈论这里的原理图（使用微带线模型）模拟还是EM（布局模型）模拟？
在我们获得的EM模型仿真结果中，端子2（比如说：端口2）电压与端子1（比如说：端口1）电压相同，但符号相反（即使端子2连接到全球接地端）
通孔）。
这有效吗？
因为正如你所说的“几乎相同”，它在我认为的两个Teminals都没有考虑相同的价值。



     以下为原文

  Thanks for the reply. I have another question.

----  For EM simulation with SMD cal, Terminal 1 and 2 are grouped into one differential port. This gives us the voltage across the port (voltage difference between Terminal 1 and 2), but we don't have the voltage from Terminal 1 to ground, or from Terminal 2 to ground. In your case, where Terminal 2 is connected to the global ground, voltage across the port and voltage at Terminal 1 are almost the same. (I wrote "almost the same" because there is some series inductance/resistance from the via)

Are you talking about the schematic(with Microstrip Line models) simulation here or the EM( layout  model) simulation? In the EM model simulation results we obtained, the Terminal 2(lets say : port 2 ) voltage is same as the Terminal 1(lets say : port 1) voltage but with opposite sign(even though Terminal 2 is connected to the global ground though vias).  Is this valid? Because as you said  "almost same", it doesnt account for same value at both the Teminals I think.

> {quote：title = PP写道：} {quote}>您是在谈论这里的原理图（使用微带线模型）模拟还是EM（布局模型）模拟？
>在我们获得的EM模型仿真结果中，端子2（比如说：端口2）电压与端子1（比如说：端口1）电压相同，但符号相反（即使端子2连接到全局接地）
通过）。
不要混淆。
它仅取决于您声明的节点作为测量电压的参考。
如果我们将您的PCB接地视为电压测量的参考，我们不能说V（T1）= - V（T2）因为V（T2）然后接地（根据定义，这就是我们声称的地面
参考。



     以下为原文

  > {quote:title=PP wrote:}{quote}
> Are you talking about the schematic(with Microstrip Line models) simulation here or the EM( layout  model) simulation? 
> In the EM model simulation results we obtained, the Terminal 2(lets say : port 2 ) voltage is same as the Terminal 1(lets say : port 1) voltage but with opposite sign(even though Terminal 2 is connected to the global ground though vias).  

Don't be confused. It only depends what node you declare as your reference for measuring the voltage. If we think of your PCB ground as the reference for voltage measurement, we can't say that V(T1)=-V(T2) because V(T2) is then grounded (by definition, that's what we have declared as our ground reference.