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>
嗨专家,我们知道测量电缆应该足够好以确保测试结果的紧凑性,但在某些场景中我们必须使用非常长的高损耗电缆(~1dB)以便通过诸如室或连接器来实现连接。
屏蔽盒,所以我有一些问题:1。采用标准校准步骤(SOLT或ECal使用..)后,这根长电缆是否有任何错误? 2.如果是Q1,那个案件遭遇了最大的错误,我认为回报损失会受到卡尔丢失的影响,但我不知道如何从数学角度解释它,有没有这方面的教程? 3.如何减少或消除这种影响? 需要更多校准步骤? 谢谢... 以上来自于谷歌翻译 以下为原文 Hi experts, We know that the measurement cable should be good enough to ensure the caccuracy of test resut, but in some scene we have to use very long cable with high loss (~1dB) in order to achive the connection though such as chamber or shielding box, so i have some questions about that: 1. After standard calibration steps(SOLT or ECal used..) was adopted, is there any error bought by this long cable? 2.If yes to Q1, which case was sufferring the biggest error, I thought return loss would be affacted by the calbe loss but i don't know how to explain it from mathmatics point of view, is there any tutorial for this ? 3.How to reduce or enliminate the effect? More calibration steps needed? Thanks... |
|
相关推荐
4个回答
|
|
你提到:*非常长的电缆具有高损耗(~1dB)*这是非常小的损失....你的意思是10dB?
甚至10dB也不是很大,取决于频率。 更大的问题是:电缆有多长? 超过几米的任何东西可能需要一些照顾。 您可以做的最重要的事情是确保扫描处于步进扫描模式。 如果您保持斜坡扫描模式(通常是默认模式),则可能会遇到问题。 请参阅以下一个或两个链接:http://na.support.keysight.com/vnahelp/tip14.html http://na.support.keysight.com/pna/files/eld.html一旦得到照顾 ,在电缆末端执行你的校准......这将消除电缆的大部分影响。 一定要使用好的电缆,尽量不要移动它。 正如乔尔博士所说:“电缆很邪恶!” 如果您的电缆确实有很高的损耗并且您在电缆末端执行了校准,那么任何回波损耗测量可能会非常不稳定。 我们需要知道您使用的是哪种分析仪,电缆的最大损耗以及您希望实现的规格。 以上来自于谷歌翻译 以下为原文 You mention: *very long cable with high loss (~1dB)* That is very little loss....did you mean 10dB? Even 10dB is not that much depending upon frequency. The bigger question is: How long is the cable? Anything over a couple of meters may require some care. The most important thing you can do is to ensure the sweep is in step-sweep mode. If you remain in ramp sweep mode (which is typically the default), you could run into problems. See one or both of these links: http://na.support.keysight.com/vnahelp/tip14.html http://na.support.keysight.com/pna/files/eld.html Once that is taken care of, perform your cal at the end of the cable...this will eliminate most of the effects of the cable. Be sure to use a good cable and try not to move it. As Dr Joel says: "Cables are evil!" If your cable really does have high loss and you perform your cal at the end of the cable, then any return loss measurements may be very unstable. We would need to know what analyzer you are using, the maximum loss of the cable, and what specs you are hoping to achieve. |
|
|
|
胡bbs2 发表于 2018-11-16 16:19 嗨Jvall,我遇到的问题是腔体滤波器的回波损耗测量。 在我的工作台中将其转好并具有良好的值(通带> 22dB)之后,我将DUT移动到倒角测试平台,但发现回波损耗性能下降到~20dB。 仪器和校准键是相同的(E5071C& N4432A),唯一的区别是工作台使用普通测试电缆而测试台使用1(3米电缆)+ 1(开关盒)+1(3米电缆)测试设置为一个端口 。 我曾经相信经过良好的校准后,电缆的效果会被消除,但是,我听说有些FAE,当存在高损耗时,校准不会很好,所以我想知道它的背景,以便估计多少 我在我的长椅上得到了测量误差。 谢谢... 以上来自于谷歌翻译 以下为原文 Hi Jvall, The porblem I met is the return loss measurement of cavity filter. after it was turned well with good value(>22dB in pas***and) in my workbench, I move the DUT to chamfer testbench but found the return loss performace declined to ~20dB. the instrument and calibration key are the same(E5071C&N4432A), the only difference is that workbench uses normal test cable while test bench uses 1(3m cable)+1(switch box)+1(3m cable) test setup for one port. I used to believe after well calibration the effect of cables would be removed, however, I heard frome some FAE that when high loss exist, calibration would not performes well, so I would like to know the background of it in order to estimate how much measurement error I got in my benches. Thanks... |
|
|
|
hu_wflllllllg11 发表于 2018-11-16 16:27 首先,我对5071C并不熟悉。 第2位。 确保使用步进扫描或等效物......或者至少减慢扫描速度,直到看不到任何变化。 3。 您是否知道在测试的最高频率下电缆的丢失? 作为一个粗略的例子; 如果您的分析仪具有20dB的原始方向性并且您的电缆在* * *方向上具有10dB的损耗,那么任何返回的信号可能处于或接近良好分辨率的极限。 你仍然可以纠正一些效果,但它会非常不稳定。 无论如何,我们确实需要知道电缆的损耗。 更好的解决方案可能是在电缆末端放置一个定向耦合器或桥接器; 但这通常更难以管理。 我不是这方面的专家......也许乔尔或其他人都会参与其中。 以上来自于谷歌翻译 以下为原文 First, I am not that familiar with the 5071C. 2nd. Make sure you are using step sweep or something equivalent...or at least slow the sweep until you see no changes. 3rd. Do you have any idea as to the loss of the cable at the highest frequency you are testing? As a rough example; if your analyzer has raw directivity of 20dB and your cable has 10dB of loss in *one* direction, then any returned signal could be at or near the limits of good resolution. You can still correct for some of the effects, but it will be pretty unstable. In any case, we really need to know the loss of the cable. A better solution might be to place a directional coupler or bridge at the end of the cable; but that is generally more difficult to manage. I am not an expert in this...maybe Joel or someone else will chime in. |
|
|
|
hu_wflllllllg11 发表于 2018-11-16 16:27 如果测试端口电缆在-34 dB的水平上存在不稳定性,则可以考虑您看到的差异(20 dB与22 dB)。 对于测试端口电缆,-34 dB回波损耗稳定性不是很好。 要测试测试端口电缆和开关盒的质量,请在室内的测试端口接口处使用ecal进行校准。 然后换上简短的使用数据 - > mem和data - mem(数据MINUS内存,不是数据/ mem)。 迹线应降至-60 dB左右(仅来自测量人员的噪声)。 现在切换开关盒,看看迹线是否爬升。 然后摆动电缆,看看迹线是否爬升。 如果迹线爬升到-34 dB,则有原因:测试端口电缆不稳定。 这个测量在我的书的第596-599页详细描述:http://www.wiley.com/WileyCDA/WileyTitle/productCd-1119979552.html如果这是一个温度室,你试图在宽范围内进行精密测量 在温度范围内,您可能需要使用Calpods向上移动到PNA。 它们专为此应用而设计。 以上来自于谷歌翻译 以下为原文 The difference you see (20 dB vs 22 dB) can be accounted for if there is an instability of the test port cables at the level of -34 dB. -34 dB return loss stabiltiy is not very good for a test port cable. To test the quality of the test port cable and switch box, take and do a calibration using ecal at test port interface inside the chamber. Then put on a short and use data->mem and data - mem (data MINUS memory, not data/mem). The trace should drop to below -60 dB or so (just noise from the measuremen). Now switch the switch box and see if the trace climbs up. Then wiggle the cables and see if the trace climbs up. If the trace climbs up to -34 dB, you have your cause: test port cable instability. This measurement is describe in detail on page 596-599 of my book: http://www.wiley.com/WileyCDA/WileyTitle/productCd-1119979552.html If this is a temperature chamber and you are trying to do precision measurements over wide temperature ranges, you might need to move up to PNA with Calpods. They were designed exactly for this application. |
|
|
|
只有小组成员才能发言,加入小组>>
1349 浏览 0 评论
2405 浏览 1 评论
2234 浏览 1 评论
2088 浏览 5 评论
2981 浏览 3 评论
1310浏览 1评论
关于Keysight x1149 Boundary Scan Analyzer
793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
1097浏览 0评论
1349浏览 0评论
小黑屋| 手机版| Archiver| 电子发烧友 ( 湘ICP备2023018690号 )
GMT+8, 2025-1-31 14:47 , Processed in 1.136035 second(s), Total 50, Slave 45 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