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>
3天内不再提示
德赢Vwin官网
论坛|
我是VNA的新手，我有一个简单的问题。
它是一个8720D VNA，每个测试端口都有一个从3.5mm到N型（母）的适配器。 该机器在一年内由供应商校准; 不确定什么是精确校准的（仅限于3.5mm测试端口，我假设）。 打开机器后，没有任何SOLT校准，S11和S22读数为-6到-7 dB。 两个3.5mm到N的适配器始终打开。 我有一个N型短。 短路时，S11和S22与开路，-6或7 dB大致相同。 频率为2 - 10 GHz。 我的问题是这是否正常？ 我认为打开（不打开校准套件）或短路，S11应该接近0 dB。 虽然它可能不准确，这就是为什么需要进行SOLT校准，但它是否会出现如此大的“损失”？ 谢谢你的帮助！ 以上来自于谷歌翻译 以下为原文 I am new to VNA and I have a quick question. It is a 8720D VNA, each test port has an adapter from 3.5mm to N-type (female). The machine was calibrated by a vendor within a year; not sure what was calibrated exactly (to the 3.5mm test port only, I assume). After turning on the machine, without any SOLT calibration, the S11 and S22 reads -6 to -7 dB. The two 3.5mm to N adapters are always on. I have a N-type short. And with the short on, the S11 and S22 are about the same as open, -6 or 7 dB. Frequency was 2 - 10 GHz. My question is whether this is normal or not? I thought with open (NOT open calibration kit though) or short, the S11 should be close to 0 dB. Although it may not be exact and that is why SOLT calibration is needed, is it expected to see such a big "loss"? Thanks for your help! 
0
|
|
相关推荐
5个回答
|
|
|
通常，迹线应位于中心线上。
但是，不正确的是采样器校正，它校正了迹线的位置。 F / W 6.14及更低版本的D型号没有很好的方法来纠正您遇到的问题。 它有一个服务测试，但如果原始采样器级别超出范围，它在大多数情况下都无法正常工作。 因此，请不要尝试，否则您将真正导致仪器出现问题。 选件000 F / W 7.00及更高的单元确实能够毫无困难地纠正这个问题。 即使迹线未居中，如果正确校准，该装置仍可正确测量。 使用标准执行用户校准时，它将使您现在看到的错误无效。 问候，戴夫 以上来自于谷歌翻译 以下为原文 Typically the trace should be on the center line. However, what is incorrect is the sampler correction which corrects the position of the trace. The D models with F/W 6.14 and lower did not have a very good way to correct the problem you are experiencing. It had one service test but it does not work correctly most of the time if the raw sampler levels are out of range. Therefore don't try it or you will really cause problems with the instrument. A unit with the option 000 F/W 7.00 and higher did have the ability to correct this problem with no difficulty. Even though the trace is not centered, the unit can still measure correctly if properly calibrated. When you perform the user calibration with your standards, it will nullify the error you see now. Regards, Dave |
|
|
|
|
|
> {quote：title = meta5718写道：} {quote}>我是VNA的新手，我有一个简单的问题。
>>它是一个8720D VNA，每个测试端口都有一个从3.5mm到N型（母）的适配器。 该机器在一年内由供应商校准; 不确定什么是精确校准的（仅限于3.5mm测试端口，我假设）。 我原本预计8720D可以用85131F柔性电缆套件进行校准。 也就是说，当我的VNA返回安捷伦进行校准时，没有证据表明这些电缆被使用过。 如果您有兴趣，这里有一份校验证书的副本。 http://boxen.math.washington.edu/home/kirkby/Agilent-standard-calibration-with-uncertainties-for-8720D-vector-network-analyzer.pdf如果你的VNA是由众多笨蛋之一完成的 实验室，比任何人猜测可能已经测试过的和可能没有的东西。 >打开机器后，没有任何SOLT校准，S11和S22读数为-6到-7 dB。 两个3.5mm到N的适配器始终打开。 我有一个N型短。 短路时，S11和S22与开路，-6或7 dB大致相同。 频率为2 - 10 GHz。 >>我的问题是这是否正常？ 我认为打开（不打开校准套件）或短路，S11应该接近0 dB。 虽然它可能不准确，这就是为什么需要进行SOLT校准，但它是否会出现如此大的“损失”？ 我个人不会太在意这件事。 这是我在我看到的，在相同的范围内使用短的低损耗测试端口电缆。 * S12从-0.3到-4.3 dB不等（连接电缆）* S21从+0.14到-4 dB（连接电缆）* S11（电缆末端3.5 mm）+0.7到-5 dB（连接器打开）* S22（电缆末端的母头为3.5 mm）+0.95至-5.6（连接器开路）S12和S21的理想0 dB损耗远远低于开路的理想0 dB回波损耗。 >谢谢你的帮助！ 说实话，如果您真的想知道您的VNA是否正常工作，我会将其发送给安捷伦进行校准 - 而不是某些第三方实验室。 我的一些关于特定实验室的评论对于我认为不应该的问题颁发校验证书。 这个“校准证书”是笑话，无效还是有效？ PS - 我不为安捷伦工作，从未这样做过，但他们是我个人信任的唯一实验室。 戴夫 以上来自于谷歌翻译 以下为原文 > {quote:title=meta5718 wrote:}{quote} > I am new to VNA and I have a quick question. > > It is a 8720D VNA, each test port has an adapter from 3.5mm to N-type (female). The machine was calibrated by a vendor within a year; not sure what was calibrated exactly (to the 3.5mm test port only, I assume). I would have expected an 8720D to be calibrated with the 85131F Flexible Cable Set. That said, when my VNA was returned to Agilent for cal, there is no evidence those cables were used. If you are interested, there is a copy of the cal certificate here. http://boxen.math.washington.edu/home/kirkby/Agilent-standard-calibration-with-uncertainties-for-8720D-vector-network-analyzer.pdf If you got your VNA done by one of the many dodgy cal labs, than its anyone's guess what might have been tested and what might not have been. > After turning on the machine, without any SOLT calibration, the S11 and S22 reads -6 to -7 dB. The two 3.5mm to N adapters are always on. I have a N-type short. And with the short on, the S11 and S22 are about the same as open, -6 or 7 dB. Frequency was 2 - 10 GHz. > > My question is whether this is normal or not? I thought with open (NOT open calibration kit though) or short, the S11 should be close to 0 dB. Although it may not be exact and that is why SOLT calibration is needed, is it expected to see such a big "loss"? I would personally not be too concerned about that. Here is what I see on mine, over the same range with the short low-loss test port cables. * S12 varies from -0.3 to -4.3 dB (cables joined) * S21 varies from +0.14 to -4 dB (cables joined) * S11 (male 3.5 mm on end of cable) +0.7 to -5 dB (connector open) * S22 (female 3.5 mm on end of cable) +0.95 to -5.6 (connector open) So far from the ideal 0 dB loss for S12 and S21, and far from the ideal 0 dB return loss for the opens. > Thanks for your help! To be honest, if you really want to know whether your VNA is working, I would send it to Agilent for calibration - not some third party lab. There are some comments of mine about a specific lab here that issues cal certificates for things I don't think they should be. Is this "certificate of calibration" a joke, incompetant, or valid? PS - I don't work for Agilent, and have never done so, but they are the only lab I'd personally trust to do the job. Dave |
|
|
|
|
|
非常感谢你的答案，Dave :)我看到用户校准是必不可少的，因为VNA是有用的。
我想85054B将是所需的套件，因为我的大多数东西都有N型连接器。 不幸的是，我没有工作。 基本上，我们正在用一对喇叭进行一些双基地RCS（散射）测量。 并且S21参数是唯一需要测量的参数。 此外，首先进行具有已知RCS的参考测试，对实际测量进行归一化。 因此，在我看来，即使没有VNA校准，测量仍然有效。 因为我只是取两个S21参数之间的比例，任何未校准的错误都将被取消。 但是，我有点担心的部分是VNA端口的未校准信号丢失（对于我的情况至少为3dB）可能会损害动态范围？ 我的散射信号相当弱，到达端口2时可能会下降-70dB。所以我试图减少测量系统中所有可能的损耗。 好吧，我猜VNA的测试端口是我无法控制的。 问候。 以上来自于谷歌翻译 以下为原文 Many thanks to your answers, both Dave :) I see that a user calibration is essential, given that the VNA is functional. I guess the 85054B would be the kit needed since most of my stuffs have type-N connecters. Unfortunately, I don't have one at work. Basically, we are doing some bistatic RCS (scattering) measurements with a pair of horns. And the S21 parameter is the only one needed to measure. Also, a reference test with known RCS was done first, to which the actual measurement was normalized. Therefore, it seems to me that even without a VNA calibration, the measurement is still valid. Because I simply took the ratio between two S21 parameters and any uncalibrated error will be cancelled. However, the part I am a bit concerned is that the uncalibrated signal loss at VNA port (at least 3dB for my case) may hurt the dynamic range? My scattering signal is fairly weak and can drop by -70dB by the time it reaches port 2. So I was trying to reduce all possible losses in my measurement system. Well, I guess the test ports of VNA are beyond my control. Regards. |
|
|
|
|
zhaophqin 发表于 2018-12-27 19:02 > {quote：title = meta5718写道：} {quote}>非常感谢你的答案，Dave :) >>我认为用户校准是必不可少的，因为VNA是有用的。 我想85054B将是所需的套件，因为我的大多数东西都有N型连接器。 不幸的是，我没有工作。 我喜欢85054B。 更便宜的选择是85054D。 我会给你一个关于你可能会喜欢的另一个选项的私信。 检查你的收件箱。 >基本上，我们正在用一对喇叭做一些双基地RCS（散射）测量。 您是否使用N->波导过渡进入波导？ 如果是这样，波导校准套件可能更合适。 话虽如此，我制作了自己的X波段（WR90）波导校准套件，它与Agilent X11644A基本相同，但制作成本却低得多。 我没有太多使用它，但已将它用于一些业余无线电目的。 我可以告诉你，它们不是最容易使用的东西。 周围有许多活动件，粘在定位销上，拧上螺丝和螺母。 在与机械物体作斗争时，尽量不要过度弯曲电缆。 使用N cal套件比波导校准套件更容易* - 至少在我有限的经验中。 也许花在波导校准套件上的时间更多，我会更快。 >并且S21参数是唯一需要测量的参数。 此外，首先进行具有已知RCS的参考测试，对实际测量进行归一化。 因此，在我看来，即使没有VNA校准，测量仍然有效。 因为我只是取两个S21参数之间的比例，任何未校准的错误都将被取消。 对此不太确定。 >但是，我有点担心的部分是VNA端口的未校准信号丢失（对于我的情况至少为3dB）可能会损害动态范围？ 我的散射信号相当弱，到达端口2时可能会下降-70dB。所以我试图减少测量系统中所有可能的损耗。 好吧，我猜VNA的测试端口是我无法控制的。 VNA的测试端口可能不受您的控制。 仪器的左下角和右下角是否有SMA-SMA链接？ （忽略R通道上的那个）。 如果是这样，VNA有一个选项，如选项012，允许人们绕过某些应用程序的耦合器丢失，你的应用程序听起来可能就像其中之一。 我想过尝试将选项012添加到我的VNA中。 除一个部件外的所有部件均可从安捷伦获得，但安捷伦的所需电缆之一为零。 （几个月前这是真的 - 现在可能有另一个是无法获得的）>问候。 戴夫 以上来自于谷歌翻译 以下为原文 > {quote:title=meta5718 wrote:}{quote} > Many thanks to your answers, both Dave :) > > I see that a user calibration is essential, given that the VNA is functional. I guess the 85054B would be the kit needed since most of my stuffs have type-N connecters. Unfortunately, I don't have one at work. I'd love an 85054B. The cheaper option is the 85054D. I'll drop you a private message about another option you might like. Check your inbox. > Basically, we are doing some bistatic RCS (scattering) measurements with a pair of horns. Do you go to waveguide with an N->waveguide transition? If so, a waveguide cal kit might be more suitable. Having said that, I made my own X-band (WR90) waveguide cal kit, which is basically the same as the Agilent X11644A, although cost me a lot less to make. I have not used it much, but have used it for some amateur radio purposes. I can tell you they are not the easiest things to use. A lot of moving pieces around, sticking in alignment pins, putting screws and nuts on. Trying not to flex the cables too much as you fight with the mechanical things. Using an N cal kit is a *lot* easier than a waveguide cal kit - at least in my limited experience. Perhaps with more time spent on a waveguide cal kit, I'd be faster. > And the S21 parameter is the only one needed to measure. Also, a reference test with known RCS was done first, to which the actual measurement was normalized. Therefore, it seems to me that even without a VNA calibration, the measurement is still valid. Because I simply took the ratio between two S21 parameters and any uncalibrated error will be cancelled. Not so sure about that. > However, the part I am a bit concerned is that the uncalibrated signal loss at VNA port (at least 3dB for my case) may hurt the dynamic range? My scattering signal is fairly weak and can drop by -70dB by the time it reaches port 2. So I was trying to reduce all possible losses in my measurement system. Well, I guess the test ports of VNA are beyond my control. The test ports of the VNA might not be beyond your control. Are there any SMA-SMA links in the bottom left and bottom right of the instrument? (Ignore the one on the R-channel). If so, the VNA has an option such as option 012, which allow one to bypass the loss of the coupler for certain applications, and your application sounds like it might be one of them. I thought of trying to add option 012 to my VNA. All the parts except one are available from Agilent, but Agilent have zero stock of one of the cables needed. (That was true a few months back - it is possible another is unobtainable now) > Regards. Dave |
|
|
|
|
60user7 发表于 2018-12-27 19:19 再次感谢，戴夫！ 8720D的选项012直接接收器对我们来说听起来不错，但我们的VNA没有这个选项。 我会更多地了解校准选项（电缆或波导），并且你对它们的要点很好。 干杯 以上来自于谷歌翻译 以下为原文 Thanks again, Dave! The option 012 direct receiver of 8720D sounds pretty good for us, but the VNA we have does NOT have this option. I would look a bit more into the calibration options (cable or waveguide), and your points on them were well taken. cheers |
|
|
|
|
只有小组成员才能发言，加入小组>>
1283 浏览 0 评论
2371 浏览 1 评论
2188 浏览 1 评论
2062 浏览 5 评论
2946 浏览 3 评论
1100浏览 1评论
关于Keysight x1149 Boundary Scan Analyzer
749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
918浏览 0评论
1283浏览 0评论
/6 
关注我们的微信
下载发烧友APP
德赢Vwin官网 观察
版权所有 © 湖南华秋数字科技有限公司
德赢Vwin官网 （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
2366
评论
淘帖
2824
