是德科技
直播中

徐铭潞

7年用户 276经验值
私信 关注
[问答]

如何校准ENA以最小化任何不确定性误差

大家好,我是论坛的新手,我刚刚开始使用网络分析仪,所以请耐心等待:)我正在尝试找到一些有关如何使用ENA通过无源元件准确测量电气延迟的信息和指导
E5071C。
我的ENA E5071C 9kHz-8GHz带有E-Cal套件(85093-60010 300kHz-9GHz,2个端口)。
我试图准确地测量从输入1到输出2的2.0GHz-2.9GHz的循环器的电延迟(第3端口终止50欧姆)。
但我不知道我做得对吗?
我担心的是,我不知道我是否只使用E-Cal正确校准ENA。
我还需要做任何其他事情来获得准确的读数吗?
任何人都可以指导我如何校准ENA以最小化任何不确定性误差,以获得循环器或隔离器的最准确的电延迟测量?
当我按下Mark-> Delay时的测量时,每次在电气延迟和相位角上都会得到不同的读数。
哪一个应该具有修复价值?
我是否必须将相位参考角度设置为固定点?
在我的阅读中,当我按下Mark-> Delay the Mark-> Ref时,它似乎指向155deg左右。
我做得对吗?
仅供参考,我附上了阅读的屏幕截图。
如果有人会就此提出建议我会非常感激。
谢谢!

以上来自于谷歌翻译


     以下为原文

  Hi All,
I am new to the Forum and I'm just starting to work with Network Analyzer so please bear with me :)

I am trying to find some information and guidance on how to accurately measure Electrical Delay through a passive component using the ENA E5071C.

I have an ENA E5071C 9kHz-8GHz with an E-Cal Kit (85093-60010 300kHz-9GHz, 2 ports). I am trying to accurately  measure the electrical delay of a circulator that is 2.0GHz-2.9GHz from input1 to out2 (The 3rd port is terminated 50ohm). But I don't know if I am doing it right?

My concern is that I don’t know if I am calibrating the ENA correctly with just the E-Cal. Do I need to do anything else to get the accurate reading?  Could anyone instruct me how to calibrate the ENA to minimize any uncertainty errors to get the most accurate electrical delay measurement of a circulator or isolator?

Also on the measurement when I press Mark->Delay I get a different reading each time on the Electrical Delay and Phase Angle. Which one should have the fix value? Do I have to set the Phase reference angle to a fix point? On my reading it seem like it's pointing at around 155deg when I press Mark->Delay the Mark->Ref. Am I doing it correctly?

FYI, I have attached a screen capture of the reading.

If anyone would advise me on this I would really appreciate it.
THANKS!   
附件

回帖(2)

孔德羲

2019-2-15 15:52:38
最好是将S21轨迹的格式设置为延迟:然后它将显示作为频率函数的群延迟。
不要忘记打开平滑,并使用至少2%,但我会建议5%。
(您可以将格式显示为Logmag,或Phase,或Delay,或VSWR或许多其他内容)。
延迟标记将标记值的组延迟轨迹的值取为标记值,​​并计算线性相位斜率以从相位轨迹中减去,以使相位在标记值处平坦。
我相信它使用20%的推断平滑,所以如果你想看到值来自哪里,用延迟格式将轨迹上的平滑设置为20%,如果标记读数跳跃(由于噪音)那么你将
看到相位跟踪跳跃。
我建议使用1kHz IF BW进行最佳测量。
当然,所有这些都在第5章中完全解释... http://www.wiley.com/WileyCDA/WileyTitle/productCd-1119979552.html如果Ecal连接器与DUT(被测设备)上的连接器相匹配,那么只需
连接它,按ecal,你就完成了。
我们会照顾其他一切。

以上来自于谷歌翻译


     以下为原文

  Best is to set the format of an S21 trace to Delay: it will then show you the group delay as a function of frequency. Don't forget to turn smoothing on, and use at least 2 percent, but I would recommend 5 percent.  (you can show the format as Logmag, or Phase, or Delay, or VSWR or a lot of other things).

Marker to delay takes the value of the group delay trace at the marker value, and computes the a linear phase slope to subtract from the phase trace to make the phase flat at the marker value. I believe it uses an inferred smoothing of 20% so if you want to see where the value comes from, set the smoothing to 20% on the trace with the delay format, if the marker reading jumps around (due to noise) then you will see the phase trace jump around.

I'd recommend a 1kHz IF BW for best measurements.

Of course, all this is completely explained in chapter 5...http://www.wiley.com/WileyCDA/WileyTitle/productCd-1119979552.html

If the Ecal connectors match the connectors on you DUT (device under test) then simply connect it, press ecal, and you are done.  We take care of everything else.
举报

孔德羲

2019-2-15 16:16:37
所以,在你的第一封电子邮件中,你说你想测量循环器的延迟,但你推断它只有一个值。
从第二个图中你可以看到延迟对于不同的频率有不同的值,所以现在你做什么?
您是否想要说明整个频段的平均延迟或(并且这更常见)整个频段的延迟偏差,有时是整个频段的相位偏差。
对于相位偏差,使用Mrk延迟相位跟踪,然后使用标记参考ref并使用其他几个标记来读取峰峰值变化。
如果您只需要延迟,只需阅读延迟。
但是,你知道你真正想要确定的品质因数吗?

以上来自于谷歌翻译


     以下为原文

  So, in your first email you said you wanted to measure the delay of the circulator, but you inferred it would have only one value.

From the second plot you see that the delay has different values for different frequencies, so now what do you do?  Do you want to say the average delay across the band or (and this is more common) the delay deviation across the band, and sometimes the phase deviation across the band. 

For phase deviation, use Mrk to delay on the phase trace, and then marker to ref and use a couple of other markers to read the peak-to-peak variation.

If you only need delay, just read the delay. 

But, do you know what figure-of-merit you are really trying to determine?
举报

更多回帖

发帖
×
20
完善资料,
赚取积分