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>
仪器:N5242A固件9.33适配器特性宏:A.02.10问题描述:我们很难在通过时获得一个低方差的校准(在~10GHz之后)我们可以在手动执行校准后得到一个干净的校准
。 (我没有图像,这是在一个安全的区域。)我们在校准的自动化版本中执行的(广泛)步骤是:1。预设2.执行完整的2端口校准,作为参考 路径3.为所有输入和输出执行1个端口cals(每个大约30个)4。为每个1端口和参考运行适配器特征描述宏5.将适配器特性的结果添加为每个端口组合的夹具6.将夹具压平到校准中 7.测试(对于没有代码的我表示道歉,这完全来自内存。)如果我们执行这些步骤,在大多数情况下,我们会得到一个很好的校准。 有时,路径可能具有高差异(在20GHz附近>约0.1 dB),这在同一路径的直接手动校准中不存在。 (我们并不期望两种校准类型完全匹配,只是为了在同一个球场。)我们有什么遗漏,或者我们可以做得更好吗? 如果我能提供更多信息,请告诉我。 谢谢 以上来自于谷歌翻译 以下为原文 Instrument: N5242A Firmware 9.33 Adapter Characterization Macro: A.02.10 Problem Description: We're having difficulty trying to get a cal with a low amount of variance on a thru (after ~10GHz) We can get a clean cal after performing the cal manually. (I don't have images, this is in a secure area.) The (broad) steps we're performing in the automated version of the calibration are: 1. Preset 2. Perform a full 2-port cal, as the reference path 3. Perform 1 port cals for all inputs and outputs (approx 30 each) 4. Run Adapter Characterization macro for each 1-port and reference 5. Add results of Adapter characterization as fixture for each combination of ports 6. Flatten fixture into calibrations 7. Test (My apologies for not having the code with me, this is all from memory.) If we perform these steps, for the most part, we get a good calibration. Occasionally, a path might have that high variance (> approx 0.1 dB around 20GHz), which is not present on a direct manual calibration of the same path. (We're not expecting the two cal types to line up exactly, just to be in the same ballpark.) Is there anything we're missing, or that we could do better? Let me know if there is any more information that I could provide. Thank you |
|
相关推荐
3个回答
|
|
如果无法访问用于创建S2P文件的校准集,我无法具体说明可能发生的情况。
然而,通常,单端口夹具表征方法依赖于反射测量,因此其精度随着夹具的损耗增加而下降。 在更高的频率下,您的夹具会有更多的损耗,这可以解释为什么您看到在最终参考平面上执行的2端口校准与具有嵌入式解嵌件的2端口校准之间存在较大差异。 如果可能的话,尝试以更高的功率,更低的IFBW和更多的平均值来执行你的第二层1端口cals(在你的灯具末端)。 这些动作的任何组合都将改善最终夹具S2P文件的质量。 以上来自于谷歌翻译 以下为原文 without having access to the calsets that you used to create the S2P files, I can't say specifically what may be going on. However, in general, the one port fixture characterization method relies on reflection measurements and therefore its accuracy drops as the loss of the fixture increases. At higher frequencies your fixture has more loss and that may explain why you are seeing bigger differences between a 2-port cal performed at the final reference plane versus one that has the fixtures de-embedded. if possible, try to perform your 2nd tier 1-port cals (at the end of your fixtures) at a higher power, lower IFBW and more averaging. Any combination of these actions will improve the quality of the final fixture S2P file. |
|
|
|
Topcbpcba 发表于 2018-12-27 07:31 谢谢你的建议。 我们采取了我们可以做的部分。 我们无法真正提高功率水平。 我们正在应用电力斜坡。 我们尝试将平均值从5提高到10,几乎没有变化我们尝试将IFBW从1kHz降低到500Hz,几乎没有变化。 有人指出,当我们尝试使用适配器表征宏的手动版本时,它警告我们组件cals有> 12hr delta。 (我们通常不会看到这一点,因为我们使用命令行版本)> 12小时差距对此有多大影响? 谢谢 以上来自于谷歌翻译 以下为原文 Thank you for the advice. We took what parts of it we could. We can't really raise the power level. We are applying a power slope though. We tried raising the averaging to 10 from 5, with little to no change We tried lowering the IFBW from 1kHz to 500Hz, with little to no change. It was noted that when we tried using the manual version of the adaptor characterization macro, it warned us that the component cals had a >12hr delta. (We don't typically see this, as we use the command line version) How much would a >12hr gap affect this? Thank you |
|
|
|
> {quote:title = BrianAllred写道:} {quote}> ...组件cals有> 12hr delta。 (我们通常不会看到这一点,因为我们使用命令行版本)> 12小时差距对此有多大影响? >>谢谢你很难说。 这取决于12小时内发生的事情。 如果环境变化大于校准规范(在帮助中公布)或第一层校准参考平面上的任何类型的电缆运动和/或漂移所允许的环境变化,那么是的,可能会产生重大影响。 尝试在相同条件下同时处理第1层和第2层cals,但尽可能接近时间并查看生成的S2P文件是否有任何差异。 以上来自于谷歌翻译 以下为原文 > {quote:title=BrianAllred wrote:}{quote} > ...that the component cals had a >12hr delta. (We don't typically see this, as we use the command line version) How much would a >12hr gap affect this? > > Thank you it is hard to say. it depends on what is happening during those 12 hours. If there are environmental changes greater than what is tolerated in the calibration specs (published in the help) or any kind of cable movment and/or drift at the 1st tier cal's reference plane, then yes, there could be a significant impact. try doing both the 1st and 2nd tier cals under the same conditions, but as close in time as possible and see if there is any difference in the resulting S2P file. |
|
|
|
只有小组成员才能发言,加入小组>>
1238 浏览 0 评论
2352 浏览 1 评论
2163 浏览 1 评论
2029 浏览 5 评论
2913 浏览 3 评论
983浏览 1评论
关于Keysight x1149 Boundary Scan Analyzer
713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
813浏览 0评论
1240浏览 0评论
小黑屋| 手机版| Archiver| 电子发烧友 ( 湘ICP备2023018690号 )
GMT+8, 2024-11-29 10:02 , Processed in 1.458702 second(s), Total 80, 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