完善资料让更多小伙伴认识你,还能领取20积分哦, 立即完善>
大家好!
感谢您的光临。 我已经模拟耦合传输线好几天了,我不知道直接模拟它的方法。我首先模拟四端口单端S参数然后用一些方程将这些数据转换成混合模式,如Sdd11 .i有几个问题。 问题1:有没有直接的方法通过耦合传输线模拟diff_pair信号的S参数? 问题2:当我使用理想的tclin时,结果比pclin2好得多,我想知道阻抗不匹配,你有没有用线速来计算PCB耦合线的阻抗?你能告诉我怎么样 计算阻抗? 我是一个使用广告的新人,我认为问题有点低级。非常感谢你的回答。我附上了原理图和数据显示窗口。 data.jpg231.9 KB 以上来自于谷歌翻译 以下为原文 hello everyone! thank you for your visit. I have been simulating coupled transmission lines for several days,i don't know the way to simulate it directly.i simulate the four ports single-ended S-parameters first and then use some equations to turn these data into mixed mode such as Sdd11.i have several problems. problem 1:is there a direct way to simulate S-parameter of diff_pair signals through coupled transmission lines? problem 2:when i use the ideal tclin,the results is much better than pclin2, i am wondering that the impedance doesn't match,have you ever use the linecal to calculate the impedance of PCB coupled lines?could you tell me how to calculate the impedance? i'm a new one to use ads,and i think the question is a little bit low-level.thank you very much for your answer.and i attach the schematic and data display window. 附件
|
|
相关推荐
5个回答
|
|
谢谢你的精彩回复。
对于第一个问题,我已经尝试了一种方法并且它有效,我将在稍后尝试第二种方法。谢谢你! 关于第二个问题,你的回答让我清楚地了解了Zodd和Zdiff之间的关系。 但是对于一条耦合线,固有阻抗是sqrt(Zodd * Zeven),我计算这个阻抗使用lincal并匹配耦合传输线的四个端口,结果表明它们不像我预期的那样匹配 .i我现在正在尝试使用si9000来计算PCB耦合线的阻抗。 非常感谢你! 以上来自于谷歌翻译 以下为原文 thank you for your wonderful reply. for the first question,i have tried the approach one and it works,and i will try the second approach later.thank you! as for the second question,your answer makes me clearly about the relationship between Zodd and Zdiff. but for a line of coupled lines,the intrinsic impedance is sqrt(Zodd*Zeven),i calculate this impedance use the lincal and matches the four ports of coupled transmission lines,the result indicates that they don't matches as well as i expected.i am now trying to use the si9000 to calculate the impedance of PCB coupled line. thank you very much! |
|
|
|
testd00111 发表于 2019-1-22 15:25 > {quote:title =摆渡今生写道:} {quote}>关于第二个问题,你的回答让我清楚地了解了Zodd和Zdiff之间的关系。 是。 这是您在实际应用中所需要的。 >但是对于一对耦合线,固有阻抗是sqrt(Zodd * Zeven),我计算这个阻抗使用lincal并匹配耦合传输线的四个端口,结果表明它们不匹配我 预期。 你误解了固有阻抗。 固有阻抗是与材料相关的数学属性,但对您的应用没有实际意义。 对于差分线,您需要将线匹配到Zdiff。 以上来自于谷歌翻译 以下为原文 > {quote:title=摆渡今生 wrote:}{quote} > as for the second question,your answer makes me clearly about the relationship between Zodd and Zdiff. Yes. This is what you need in practical applications. > but for a line of coupled lines,the intrinsic impedance is sqrt(Zodd*Zeven), i calculate this impedance use the lincal and matches the four ports of coupled transmission lines,the result indicates that they don't matches as well as i expected. You misunderstood intrinsic impedance. Intrinsic impedance is a mathematical property related to the material, but has no practical meaning for your application. For your differential line, you need to match the line to Zdiff. |
|
|
|
lmksa 发表于 2019-1-22 15:34 是的!我知道,我将耦合线作为两条独立的线并匹配每条线的两个端口(这是我原来的方法,四端口结构)。 我已经尝试过你提到的方法,那个拥有Balun的方法,它适用于mclin,但是当我使用PCB耦合线时,会出现匹配问题。我认为我应该专注于计算PCB线路的阻抗。 谢谢你的善意答复,我应该多练习。 以上来自于谷歌翻译 以下为原文 yeah!i know,i take the coupled line as two separate line and match two ports of each line(it's my original approach,the four port structure). i have tried the approach that you mentioned,the one who has the Balun ,it works well for mclin, but when i use PCB coupled line,the matching problem comes.i think i should concentrate on the calculation of PCB line's impedance. thank you for your kindness answer,i should practice more. |
|
|
|
testd00111 发表于 2019-1-22 15:54 > {quote:title =摆渡今生写道:} {quote}>是的!我知道,我把耦合线作为两条独立的线并匹配每条线的两个端口(这是我原来的方法,四端口结构)。 这不是正确的方法。 如果使用4个独立端口进行模拟,并单独匹配每个端口,则会激发不同的模式(不严格区分)。 这就是为什么你的方法给差分操作带来不好的结果。 如果要为差分模式设计线路,请将其与差分模式匹配。 就那么简单。 >我已经尝试过你提到的方法,那个拥有Balun的方法,它适用于mclin,但是当我使用PCB耦合线时,会出现匹配问题。我认为我应该专注于计算PCB线路的阻抗。 您的线路可能存在差分阻抗错误。 另一个可能的问题是从差分模式到普通模式的模式转换,例 如果两条迹线的长度不同。 以上来自于谷歌翻译 以下为原文 > {quote:title=摆渡今生 wrote:}{quote} > yeah!i know,i take the coupled line as two separate line and match two ports of each line(it's my original approach,the four port structure). This is not correct way to do it. If you simulate with 4 independent ports, and match each port individually, you excite a different mode (not strictly differential). That is why your method gives bad results for differential operation. If you want to design the line for differential mode, match it to differential mode. Simple as that. > i have tried the approach that you mentioned,the one who has the Balun ,it works well for mclin, but when i use PCB coupled line,the matching problem comes.i think i should concentrate on the calculation of PCB line's impedance. It is possible that your lines has wrong differential impedance. Another possible problem is mode conversion from differential to common mode, e.g. if the length of the two traces are different. |
|
|
|
lmksa 发表于 2019-1-22 16:05 感谢您的建议。 我试图调整匹配阻抗并找到最佳匹配阻抗。 结果非常好。 以上来自于谷歌翻译 以下为原文 thank you for your advice. i tried to tune the matching impedance and find the best matching impedance. the result is very well. |
|
|
|
只有小组成员才能发言,加入小组>>
2091 浏览 1 评论
1850 浏览 1 评论
1707 浏览 5 评论
2577 浏览 3 评论
将设备连接到PC并通过asic格式读取数据然后我的网络分析仪挂起
1558 浏览 4 评论
497浏览 1评论
关于Keysight x1149 Boundary Scan Analyzer
355浏览 0评论
1405浏览 0评论
N5230C用“CALC:MARK:BWID?”获取Bwid,Cent,Q,Loss失败,请问大佬们怎么解决呀
268浏览 0评论
2254浏览 0评论
小黑屋| 手机版| Archiver| 电子发烧友 ( 湘ICP备2023018690号 )
GMT+8, 2024-4-20 12:10 , Processed in 1.192901 second(s), Total 69, Slave 64 queries .
Powered by 电子发烧友网
© 2015 bbs.elecfans.com
关注我们的微信
下载发烧友APP
电子发烧友观察
版权所有 © 湖南华秋数字科技有限公司
电子发烧友 (电路图) 湘公网安备 43011202000918 号 电信与信息服务业务经营许可证:合字B2-20210191 工商网监 湘ICP备2023018690号