尝试用 NanoVNA 测试手上的一个从成都 M 哥那买的车台苗子用的磁吸天线底座的特性阻抗，参考国外的一些教程，最后的做法是： 1. 底座上装了 M 头转 SMA 母，再接 50Ω 负载（NanoVNA 附带的） 2. 底座的线一端接 M 头转 SMA 母，再接 SMA 公-公，再接 NanoVNA CH1 3. NanoVNA 提前校准好 50k ~ 25MHz 段（方法同样是参考网上的，假负载 ~ SMA公-公 ~ NanoVNA CH1，分别用开路、短路、50Ω负载作为假负载点校准里面的按钮，最后保存、应用，测试施密特图里这三个负载正常测量） 4. 开施密特图 R+L/C 模式，从 50k 开始向上找到线第一个与主轴交叉的位置，读当前测量阻抗 5. 将此阻抗与 50 相乘开根号得到传输线特性阻抗这里 25MHz 的选取是参考了某个视频上说用 75/线的长度，我取了 3m，算出来就是 25MHz。我怎么测，那个交叉点处的阻抗都是 1.2Ω 左右（交叉点在 9.5 MHz 左右），最终测量出的值是 7Ω 左右，怎么考虑误差这个值都和 50Ω、75Ω 的典型值相差得太离谱了。各位有何想法吗？我在想是测量原理本身有问题（网上至少看到了另一种做法，目前还没尝试），还是我的操作有问题呢？特别是校准到底应该把待测的同轴线加进去还是像现在一样不加。感谢各位的指正。

借楼问下传输线有阻抗，那SMA/UHF/N头有特性阻抗吗？还是说这些接头太短不会影响S参数例如，有RG58(50Ohm)和RG59(75ohm)，这两者应该都可焊接到UHF头上

说一下我常用的做法，抛砖引玉： #### 准备工作 1. NanoVNA Port 1 连接好辅助电缆（即，随机器附赠的馈线）以及必要转接头。 2. 设置频率 50k - 25M（或者使用 `CW FREQ` 固定在一个频率上） 3. 校准机器（开路、短路、50欧） 4. 底座的线端连接到辅助电缆上。接下来有两种方法： #### 开路/短路阻抗测试法 1. 将任意一条轨迹的格式设置为复阻抗的模。 2. 设置方法为，打开 `DISPLAY` - `TRACE` ，选中一条后返回，`FORMAT` - `|Z|` 3. 移动光标到一个合适频率，通常会高一点（5M 左右），否则示数可能不稳定。 4. 天线底座上保持开路，读取开路阻抗值。 5. 天线底座上连接短路校准件，读取短路阻抗值。 6. 阻抗计算公式： [math:0] \left| Z \right|=\sqrt{\left| Z \right|_{\left( \text{open} \right)}\times \left| Z \right|_{\left( \text{short} \right)}} [/math:0] #### 电容/电感计算法实际上，这是个用 LCR 电桥的测试法，但矢网能够胜任这个电桥的工作。 1. 分别将两条轨迹的格式设置为并联电感与并联电容。 2. 设置方法为，打开 `DISPLAY` - `TRACE` ，选中一条后返回，`FORMAT` - 按两次 `MORE`，分别选择 `PARALLEL C` 和 `PARALLEL L`。 3. 移动光标选择合适的频率，**要求被测线缆的长度需要小于测试频率对应的波长的1%** 4. 天线底座上保持开路，读取电容值，单位为 `pF`。 5. 天线底座上连接短路校准件，读取电感值，单位为 `nH`。 6. 阻抗计算公式： [math:0] \left| Z \right|=31.62\times \sqrt{\frac{L_{\left( \text{nH} \right)}}{C_{\left( \text{pF} \right)}}} [/math:0] #### 实际测试结果晚上拿了一段同轴电缆做测试，两种方法分别测得阻抗为 53.91 和 52.57，比较吻合，考虑是测量误差。

@"[未知用户]"#p38196 有的，任何平行导体都会有特征阻抗。连接器的阻抗需要和线缆匹配，否则接头处也会有信号反射。您提到的头应该都是 50Ohm 的，只能焊接 RG58。焊 RG59 只能说是凑合用，小功率，频率不太高的情况下尚可。

@"BG5VQE"#p38198 感谢，今天上午我去尝试下

@"[未知用户]"#p38199 感谢提醒，您看这样的分析是否正确低频情况下，由于波长很长，而射频连接头的实际长度很短，因而射频头对应为传输线的角度很小，在S11史密斯图上转不了多远，不会大幅影响驻波比楼主给的方法也涉及到同样问题，天线座的物理长度和电气长度都很短，想让史密斯图转半圈对于几个cm的量级来说可能需要GHz的级别，但是这个频率下nanoVNA不准确，传输线本身也有大量损耗所以测不准。

@"[未知用户]"#p38196 有的，比如淘宝上有卖50Ω和75Ω的BNC接头，一种是射频信号使用的，一种是用在视频信号的传输中。这两种接头即使是能连接同规格的同轴电缆，外观看起来也一模一样，但它们的阻抗也是不同的，要求不高的话混用问题不大，但接头处阻抗不连续是肯定会导致驻波高的

@"BH2VZC"#p38196 有的，比如淘宝上有卖50Ω和75Ω的BNC接头，一种是射频信号使用的，一种是用在视频信号的传输中。这两种接头即使是能连接同规格的同轴电缆，外观看起来也一模一样，但它们的阻抗也是不同的，要求不高的话混用问题不大，但接头处阻抗不连续是肯定会导致驻波高的

@"BH2VZC"#p38202 可以这样理解，您说的很棒！

@"BH2VZC"#p38196 @"BG2EWI"#p38207 外观粗看是一样的，但是细节上还是有差别的。硬连在一起搞不好会撑坏/接触不良。

@"BloodBlade"#p38231 是的，严格上来说两种接头的内导体尺寸确实不同

How to use a NanoVNA to measure the characteristic impedance of (coaxial) cables?

chariri

I tried to measure the characteristic impedance of a magnetic mount antenna base I bought from "M" in Chengdu using a NanoVNA, referring to some foreign tutorials. The final method was:

An SMA female connector was mounted on the base, which was then connected to a 50Ω load (the one included with the NanoVNA).
One end of the cable from the base connects to a M connector, which then connects to an SMA female connector, and finally to the NanoVNA CH1.
NanoVNA has been pre-calibrated for the 50k ~ 25MHz range (the method is the same as that described online). Using a dummy load (~ SMA female to male connector) connected to NanoVNA CH1, calibrate using the buttons within NanoVNA to simulate an open circuit, short circuit, and 50Ω load. Finally, save and apply the calibration, then test the three loads in the Smith chart to ensure they are measuring correctly.
Enable the Schmidt plot R+L/C mode, starting from 50k and moving upwards to find the first point where the line crosses the main axis. Read the current measured impedance.
Multiply this impedance by √50 to obtain the characteristic impedance of the transmission line.

The choice of 25 MHz was based on a video I saw that mentioned using a 75-ohm cable with a length of 3 meters, which resulted in a frequency of 25 MHz.

How can I measure it? The impedance at that crossover point is around 1.2Ω (the crossover point is around 9.5 MHz), and the final measured value is around 7Ω. How do I account for this difference, which is so far off from the typical values of 50Ω and 75Ω?

What are your thoughts on this? I'm wondering if the measurement principle itself is flawed (I've seen at least one alternative approach online, but haven't tried it yet), or if there's something wrong with my procedure. Specifically, should the coaxial cable being tested be included in the calibration process, as some suggest, or should it be excluded as I'm currently doing?

Thank you for your corrections.

BH2VZC

I'd like to inquire about renting a room.
If transmission lines have impedance, do SMA/UHF/N connectors also have characteristic impedance? Or, are these connectors too short and therefore won't affect S-parameters?
For example, RG58 (50 Ohm) and RG59 (75 Ohm) cables should both be solderable to UHF connectors.

BG2EWI

BH2VZC Yes, for example, you can find 50Ω and 75Ω BNC connectors on Taobao. One is designed for use with RF signals, while the other is used for transmitting video signals. Even though these connectors look identical and can connect coaxial cables of the same specifications, their impedances are different. If the requirements are not strict, using them together should not be a problem. However, if there is an impedance discontinuity at the connector, it will definitely lead to high standing wave.

BloodBlade

BH2VZC
BG2EWI

At first glance, they look identical, but there are still differences in detail. Connecting them directly might cause damage or lead to poor contact.

BG5VQE

Let me share some of my common practices, to get things started:

Preparation

Connect NanoVNA Port 1 to the auxiliary cable (i.e., the feed line included with the instrument) and any necessary adapters.
Set frequency to 50 kHz - 25 MHz (or use) CW Frequency Fixed to a specific frequency.
Calibrate machine (open circuit, short circuit, 50 ohms)
The cable ends on the base are connected to the auxiliary cable.

There are two possible approaches:

Method for testing short-circuit impedance

Set the format of any trajectory to be the magnitude of the impedance.
The method for setting it up is to open. DISPLAY - TRACE Select one, then return.FORMAT - |Z|
Move the cursor to an appropriate frequency, which is usually a higher one (around 5 MHz), otherwise the reading may be unstable.
Maintain an open circuit on the antenna base and measure the open-circuit impedance value.
A short-circuit calibration component is connected to the base of the antenna, and the short-circuit impedance value is read.
Impedance calculation formula:
|Z| = √( |Z|_{(\text{open})} × |Z|_{(\text{short}))}

Capacitance/Inductance Calculation Method

In reality, this is a test method using an LCR bridge, but Y-Net can handle this bridge effectively.

Set the format for each of the two paths to be a parallel inductor and a parallel capacitor.
The method for setting it up is to open. DISPLAY - TRACE Select one, then return.FORMAT - Press twice MORE, choose respectively Parallel C And Parallel L。
Move the cursor to select the appropriate frequency.The length of the cable to be tested must be less than 1% of the wavelength corresponding to the test frequency.
Maintain an open circuit on the antenna base, read the capacitance value, unit is pF。
The short calibration component is connected to the base of the antenna, and it reads the inductance value, which is measured in units of. nH。
Impedance calculation formula:
|Z| = 31.62 * √((L (nH) / C (pF)))

Actual test results

I tested a section of coaxial cable in the evening, and using both methods, I obtained impedance values of 53.91 and 52.57 respectively. These values are quite close, which suggests that the difference may be due to measurement error.

chariri

BG5VQE Thank you. I will try it out this morning.

chariri

BG5VQE
Just now, while testing, I found that the terminals on the base were already QBG.🤣
I've ordered a new base, so I'll try it out now.

BG5VQE

[未知用户] Yes, any parallel conductor will have its characteristic impedance. The connector's impedance needs to match the cable; otherwise, there will be signal reflection at the connection point. The connectors you mentioned should all be 50 Ohm, and they can only be used with RG58 cables.
Using RG59 for welding can only be considered a makeshift solution, and it may still work in low-power, low-frequency situations.

BH2VZC

[未知用户]
Thank you for pointing this out. Do you think this analysis is correct?
In low-frequency conditions, due to the long wavelength and the relatively short physical length of the RF connector, the angle of the transmission line corresponding to the RF header is small, which limits how far it can be measured on an S11 Smith chart. This has a minimal impact on the VSWR (Voltage Standing Wave Ratio).

The method provided by the original poster also involves similar issues. The physical and electrical lengths of the antenna connectors are both very short, and achieving a full rotation on the Smith chart for such small values may require GHz frequencies. However, at these frequencies, the nanoVNA is inaccurate, and the transmission line itself has significant losses, making it difficult to obtain accurate measurements.

BG5VQE

BH2VZC You're absolutely right!

BG2EWI

BloodBlade Yes, strictly speaking, the internal conductor sizes of the two connectors are indeed different.

BG5VQE

chariri Okay, good luck with the testing!

闽ICP备2021006864号-7

闽公网安备35020602002794号互联网举报中心

用户协议隐私政策社区规范积分规则中继规则转载须知
更新日志联系我们常见问题友情链接运营报告赞助

浩奇科技（厦门）有限公司

是无线电，把我们联系在一起