通过PNA 5242A校准涉及50 db衰减器的路径

就快到了。
校准后，您是在校准前进行平均，还是仅在校准后进行平均？
对于50或100个平均值，迹线看起来大致相同，这意味着您看到的噪声不是扫描到扫描噪声，但可能内置于校准。
请确认您在校准期间进行了平均值？
并且在更改平均数时重复校准。
最后，校准后：进行扫描并允许平均计数达到50（或100）;
标准化（数据到内存，数学：内存数据），并重新开始平均并发布结果。
将刻度设置为0.1 dB / div。
除了噪音外，你看起来很近。
根据我在系统中看到的情况，我希望你的例子中的噪声小于0.02 dB p-p。
关键是在校准之前打开平均值以消除校准期间的噪音。



     以下为原文

  Nearly there.  When you did the calibration, did you turn averaging on before calibration, or only on after calibration? The trace looks about the same for 50 or 100 averages, which means that the noise you see isn't sweep-to-sweep noise but maybe built into the cal.   

Please confirm that you had averaging on during calibration?  And that you repeated the calibration when you changed averaging number.

Finally, after calibration: take a sweep and allow the average count to go up to 50 (or 100); do normalize (data into memory, math : data over memory), and restart averaging and post the result.  Set the scale to 0.1 dB/div.

It looks like you are close except for noise.  I would expect the noise in your example to be less than 0.02 dB p-p based on what I see on my system.  The key is turning on averaging before calibration to remove the noise during cal.

你好Dr.Joel和Daras先生，很抱歉延迟回复。
:(我正在休假，因此无法回复你。我已经在您之前的邮件中提到的步骤中附加了有或没有50 dB衰减器的直通响应S21（相位和增益）的屏幕截图。响应看起来更远
最后两个屏幕截图是DUT的响应（相位和增益）。我想要点亮的几点是： - 1）我的DUT必须工作在脉冲模式，但是使用脉冲设置我
给予PNA（10 us Ton，10 msec周期（。1％占空比）IFBW变为50 Hz代替校准时的1 KHz并且读数也受到影响。是否存在IFBW和Pulse的关系
脉冲周期或吨 - 我手动将IFBW改为100 KHz以获得良好的读数.2）使用平均值。
因子= 100，扫描平均值，PNA的响应显着减慢，我的意思是需要2-3分钟才能稳定下来。
出于这个原因，我将平均因子更改为10点，同时进行测量，我的校准发生在100点。
反应变得有点快，但仍然很慢。任何建议我们如何才能提高PNA的响应率？
3）使用Trace（最后两个），我为我的DUT展示了如果我继续我的测试可以吗？
感谢您的耐心和帮助。
此致，罗希特。



     以下为原文

  Hello Dr.Joel and Mr. Daras,

Sorry for the delayed response. :( I was on leave and hence unable to reply you back.
I have attached the Screen shots of the thru response S21 (phase and gain) with and without the 50 dB attenuator in steps you mentioned in your previous mail. The response looks further better. The last two screen shots are the response of the DUT (both Phase and Gain). 

Few points which I would like to hi-light is:- 
1) My DUT has to work in pulsed mode, but with the pulse settings I gave to PNA( 10 us Ton, 10 msec Period (.1% duty cycle) the IFBW changes to 50 Hz in place of 1 KHz which was at the time of calibration and also readings gets affected.Is there any relation of IFBW and Pulse Period or Ton of Pulse?? I manually changes the IFBW to 100 KHz in order to get a good reading.
2) With Avg. Factor = 100, Sweep average, the response of the PNA slows down considerably, I mean it takes some 2-3 minutes to settle down to a value. For that reason I changed the Averaging Factor to 10 points while doing measurement tough my calibration happened at 100 points. The response has become little fast but still its slow.Any suggestion how can we increase the response rate of the PNA??
3) With the Trace (last two) I showed for my DUT is it okay if I proceed with my Test?

Thanks for your patience and help.

Regards,
Rohit.

附件

• Enhanced Response Calibration with 50dB PAD.docx179.3 KB
  

> {quote：title = Rohit写道：} {quote}>你好Dr.Joel和Daras先生，> >>我想点亮的几点是： - > 1）我的DUT必须在脉冲模式下工作
，但是根据我给PNA的脉冲设置（10 us Ton，10 msec周期（。1％占空比），IFBW变为50 Hz而不是校准时的1 KHz，并且读数也会受到影响。
IFBW与脉冲周期或脉冲周期之间有任何关系我手动将IFBW改为100 KHz以获得良好的读数。正确，在脉冲模式下，您需要IF BW> 1 / Ton时间。> 2
）使用Avg。
因子= 100，扫描平均值，PNA的响应显着减慢，我的意思是需要2-3分钟才能稳定下来。
出于这个原因，我将平均因子更改为10点，同时进行测量，我的校准发生在100点。
反应变得有点快，但仍然很慢。任何建议我们如何才能提高PNA的响应率？
在测量过程中，您可能不需要平均这么多，因为您有来自放大器的高功率信号。
您可以将平均值降低到10，甚至只降低1或2，并调整放大器的数据 - > mem和data / mem，看它是否会影响放大器测量中的走线噪声。
> 3）使用Trace（最后两个）我为我的DUT展示了如果我继续我的测试可以吗？
DUT跟踪看起来对我很好，你怎么看？



     以下为原文

  > {quote:title=Rohit wrote:}{quote}
> Hello Dr.Joel and Mr. Daras,
> 
>> Few points which I would like to hi-light is:- 
> 1) My DUT has to work in pulsed mode, but with the pulse settings I gave to PNA( 10 us Ton, 10 msec Period (.1% duty cycle) the IFBW changes to 50 Hz in place of 1 KHz which was at the time of calibration and also readings gets affected.Is there any relation of IFBW and Pulse Period or Ton of Pulse?? I manually changes the IFBW to 100 KHz in order to get a good reading.
Correct, in pulsed mode you need the IF BW to be > 1/Ton time.
> 2) With Avg. Factor = 100, Sweep average, the response of the PNA slows down considerably, I mean it takes some 2-3 minutes to settle down to a value. For that reason I changed the Averaging Factor to 10 points while doing measurement tough my calibration happened at 100 points. The response has become little fast but still its slow.Any suggestion how can we increase the response rate of the PNA??
You probably don't need to average so much during measurment, as you have a high power signal from the amplifier.  You can reduce the averaging to 10, or even just 1 or 2, and evaulate the data->mem and data/mem of your amplifier to see if it affect the trace noise on the amplifier measuerement.
> 3) With the Trace (last two) I showed for my DUT is it okay if I proceed with my Test?

The DUT trace looks very good to me, what do you think?

> {quote：title = Rohit写道：} {quote}> 1）我的DUT必须工作在脉冲模式，但是我给PNA的脉冲设置（10 us Ton，10 msec周期（。1％占空比）
IFBW变为50 Hz代替校准时的1 KHz，并且读数也会受到影响。有IFBW和脉冲周期或脉冲吨的关系吗？我手动将IFBW改为100 KHz以便获得
一个好的读数。确实如此：对于脉冲测量，IF BW> 1 / Ton的脉冲。> 2）平均值。
因子= 100，扫描平均值，PNA的响应显着减慢，我的意思是需要2-3分钟才能稳定下来。
出于这个原因，我将平均因子更改为10点，同时进行测量，我的校准发生在100点。
反应变得有点快，但仍然很慢。任何建议我们如何才能提高PNA的响应率？
您可以在测量中大幅降低平均因子。
您可以尝试10次平均或不平均。
只需看看放大器S21增益的走线噪声。
如果噪声很大（超过0.05 dB），则调高平均值。
但由于增益高，功率很高，所以噪音很低。
> 3）使用Trace（最后两个）我为我的DUT展示了如果我继续我的测试可以吗？
这对我来说很好看。
此外，我认为您可以在校准期间减少平均值，50似乎没问题。



     以下为原文

  > {quote:title=Rohit wrote:}{quote}
> 1) My DUT has to work in pulsed mode, but with the pulse settings I gave to PNA( 10 us Ton, 10 msec Period (.1% duty cycle) the IFBW changes to 50 Hz in place of 1 KHz which was at the time of calibration and also readings gets affected.Is there any relation of IFBW and Pulse Period or Ton of Pulse?? I manually changes the IFBW to 100 KHz in order to get a good reading.

Exactly so: for pulsed measurements, the IF BW > 1/Ton of the pulse.

> 2) With Avg. Factor = 100, Sweep average, the response of the PNA slows down considerably, I mean it takes some 2-3 minutes to settle down to a value. For that reason I changed the Averaging Factor to 10 points while doing measurement tough my calibration happened at 100 points. The response has become little fast but still its slow.Any suggestion how can we increase the response rate of the PNA??

You proably can reduce the average factor a lot in the measurements.  You can try 10 averaging or no averaging.  Just look at the trace noise of the S21 gain of the amplifier.  If it gets much noise (more than 0.05 dB) turn up averaging. But because of the high gain, the power is high so noise will be low.

> 3) With the Trace (last two) I showed for my DUT is it okay if I proceed with my Test?

It looks good to me.

Also, I think you can reduce averaging during calibration, 50 seems ok.

谢谢Joel博士。
谢谢你的耐心。:)问候。
罗希特



     以下为原文

  Thanks Dr. Joel. Thanks for your patience.:) 


Regards.
Rohit