Did we build the right thing right? Testing our new Near-Field Probe kit

Ey builder!

As you might know it, I am building an Open Source Near Field Probe kit​

Since I started, the project has gained some attention. I am excited to see the applications!

The PCBs have arrived! It is exciting to see the designs coming to life.

Once I soldered the connectors I said to myself: let's test them!

(Silence...)

How am I going to test these?

I have been victim of the same trap that my customers fell into.

It looks like I made the same mistake I continuously prevent them to make

1. Plan your tests
2. Prepare the material

Bring spare parts just in case

In this case, I did not planned any specific tests further than having a generic idea.

I spent the last week gathering EMC documentation to keep training your brain. While useful, it does not help us to move us any further in our quest of mastering Near-Field Probes.

Anyways, we keep the same goal, we adapt the plan.

​

Planning the tests of Near-Field Probes

Goals

What we want to achieve with the tests is:

• Did we build the right thing? Ensure that the assembled probes can detect alternating electric signals (basic functionality
• Did we build the thing right? Characterize the frequency response of each probe (specific functionality?

If the answer to both questions is positive, we will have green light for the next stage!

Did we build the right thing?

We built probes to measure, so we need to be able to measure some moving fields. Then, if we want to detect alternating electric signals, first we will need to generate them. I am doing the measurements in an open space, so there will be ambient noise. Having a controlled environment will help to avoid misreadings.

What we need:

• Generate a known alternating signal (Signal generator or Vector Network Analyzer)
• A way to remove the ambient noise from the measurements

Did we build the thing right?

While having a device that measures something is great, it is not enough. We should ensure that they measure what they are intended to measure with a minimum of quality.

• E-field probes should be mostly sensitive to electric fields
• H-field probes should be mostly sensitive to magnetic fields
• Usable in a real environment like a PCB
• Repeatability: the result should be the same (or similar), no matter the moment

Material

• A Spectrum Analyzer: mine is the Siglent SVA 1015X. It has also a Vector Network Analzyer (VNA).
• Interfaces for coaxial interfaces like Tekbox TBCAS1​
• Coaxial cables
• Basic multimeter

Test plan

• Measure the DC resistance at the coaxial connector to check for short-circuits
• Connect the probes to the Port 1 of a VNA and measure the S11
• Connect the probes to a Spectrum Analyzer and capture some electromagnetic signals

If everything goes well so far, the second part of tests will be something like the following:

• Inject a signal of a known frequency on a terminated cable
• Measure the signal with both types of probes
• Change the frequency and type of the signal
• Characterize the response of the probes

Test results

As you can see, there are some issues with the H-field probe. I currently working on finding it the root-cause to fix it and design a new version. In the meantime, we can focus on the one that works, the E-field probe.

I will detail the test results on the next chapter. Stay tuned!

Tell me! For what would you use the Near-Field probes kit? Hit reply and let me know!

Happy building! ⚒️

Ignacio