是德科技
直播中

李剑

7年用户 150经验值
私信 关注
[问答]

低频去嵌入的最佳方法是什么

大家好,我正在寻找创建评估板的意见。
我的目标如下。
我有一个装在QFN封装中的器件,它安装在评估板上。
该器件具有200欧姆差分输入和输出端口。
频率范围为10 MHz至500 MHz。
我只能访问ENA(事实上是E5071B)。
我需要在评估板的SMA连接器上进行测量时,为器件生成S参数(单端和平衡)。
我计划使用接地共面波导,路径中可能有50到100次转换。
以下是我的一些想法:TRL不实用,因为Line必须是40英寸长。
SOLT因为它是低频,只需创建SHORT(接地通孔或0欧姆SMD接地),OPEN,LOAD(50欧姆SMD接地)和一条线。
适配器表征除非有我遗漏的东西,否则我需要探测一个连接是不切实际的。
建模使用ADS去嵌入设备。
两个想法:想法1.创建一个直线。
我必须找到它,但ADS确实有办法从1个文件生成2个S参数文件。
限制是:M用于S参数测量数据。
L表示左S参数数据。
R代表正确的S参数数据。
S22L = S11R S21L = S21R S12L = S12R这是一组递归方程。
想法2.为SMA连接器查找或创建S参数文件。
创建一个模拟连接器,50欧姆传输线,转换和100欧姆传输线的电路
将此与去嵌入元素(NEG2)一起使用。
我确实浏览了论坛,但每个人似乎都在谈论高频率。
我正在寻找有关上述方法的利弊的意见。
如果我错过了解决方案,请告诉我。
如果我错过了关于这个主题的上一篇文章,我道歉了。
在此先感谢您的帮助。
此致,Mike Virostko集成设备技术

以上来自于谷歌翻译


     以下为原文

  Hello All,

I am looking for opinions for creating an evaluation board.  My goal is the following.  I have a device, in a QFN package that is mounted to a evaluation board. This device has 200 ohm differential input and output ports.  The frequency range is 10 MHz to 500 MHz.  I only have access to an ENA (E5071B in fact).  I need to generate the S-parameters (both single ended and balanced) to the device while making the measurement at the SMA connectors of the evaluation board.  I plan to use grounded coplanar waveguide and there could be a 50 to 100 transition in the path.  Here are few ideas I have:

TRL
    Not practical since the Line must be 40 inch long.

SOLT
     Because it is low frequency, just create the SHORT (grounded via or 0 ohm SMD to ground), OPEN, LOAD (50 ohms SMD to ground) and a line.

Adapter Characterization
    Unless there is something I am missing I would need a probe for one connection which is impractical.

Modeling
     Use ADS to de-embed the device. Two ideas:

Idea 1.   Create a through line.  I would have to find it, but ADS did have a way to generate 2 S-parameters files from 1 file.  The restrictions are:

     M is used for the S-parameter measured data.
     L is for the left S-parameter data.
     R is for the right S-parameter data.

     S22L = S11R
     S21L = S21R
     S12L = S12R

This is a recursive set of equations.

Idea 2.  Find or create a S-parameter file for the SMA connector.  Create a circuit that models the connector, 50 ohms transmission line, transition, and 100 ohm transmission line.  Use this with the de-embedding element (NEG2).


I did browse the forum but everyone seems to be talking high frequency.  I am looking for opinions on the pros and cons of the above approaches.  If I missed a solution, please let me know. If I missed a previous post on this subject, I apologized.

Thanks in advance for you help.

Sincerely,

Mike Virostko
Integrated Device Technology  

回帖(2)

孔德羲

2019-2-15 16:15:51
在您的情况下,最好的方法可能是进行良好的SMA校准,然后使用自动端口扩展,为S参数提供匹配和损耗校正。
只需要一个开放的夹具,并使用ENA上的APE功能(我不确定ENA的代码是什么样的APE,你可以gggggooogle它)。
在高达几GHz时工作得非常好,并且错误可能比任何其他方法都要小。

以上来自于谷歌翻译


     以下为原文

  probably the best way in your situation is to do a good SMA calibration, and then use Automatic Port Extension which gives match and loss correction for the S-parameters.  Just have an open fixture, and use the APE funciton on the ENA (I'm not sure what rev of code of ENA had APE, you could gggggooogle it).  Works very well up to a few GHz, and errors are probably smaller than any of the other methods.
举报

李剑

2019-2-15 16:28:19
你好Joel,APE于2005年7月加入ENA。
我没有意识到这一点(尽管那是在我离开安捷伦之前的4个月,而我在过去的8年里只使用过PNA)。
这是我没有写的想法之一,因为我认为这只是在PNA上。
感谢更新。
另外,您的书“微波元件测量”非常棒且写得很好。
我没看过整本书,但它有所帮助。
Mike Virostko集成设备技术

以上来自于谷歌翻译


     以下为原文

  Hello Joel,

APE was added July of 2005 to the ENA.  I was unaware of that (even though that was 4 months before I left Agilent and I only used PNAs for the last 8 years).  That was one of the ideas I did not write because I thought it was only on the PNA.  Thanks for the update.

On a side note, your book "Microwave Component Measurements" is great and well written.  I have not read the whole book but it has helped.


Mike Virostko
Integrated Device Technology
举报

更多回帖

发帖
×
20
完善资料,
赚取积分