How a $90 VNA Reveals the Hidden Current Paths Behind Radiated Emissions

By Steve Berry, Principal EMC Engineer

We recently had an interesting EMC problem where a customer’s cabinet was radiating because there was a significant voltage difference between the test ground plane and the cabinet frame.

Figure 1 Large Industrial Unit in Anechoic Chamber

Since the client is based in the US and we have been helping them remotely, I wanted to demonstrate the principle in a simple way, so I decided to reproduce the effect at home using my dishwasher—and a $90 VNA!

I have done a quick test on our dishwasher to demonstrate some of the measurement techniques that I think the client could easily carry out in their own lab, rather than having to go to an EMC test house, which can be costly in both time and money.

The idea was to measure how well a dishwasher would radiate when common-mode current was conducted down the power cord.

The general setup is shown below. Note that the earth pin of the power cord was connected to the foil ground plane.

Figure 2 I happened to be renovating my kitchen, so the washing machine became my DUT

I used my NanoVNA H4 for these measurements for convenience and also to show

that you don’t need expensive kit or complex layouts for this sort of investigation.

Firstly I calibrated the VNA with open, short and 50Ω.

Figure 3 Calibration of my VNA

Next I measured the impedance of a 90mm long length of 8mm wide copper braid. The return path was looped back tightly to give the minimum inductance.

Figure 3 Measuring the braid

This braid measured about 40nH, as shown in Figure 4.

Figure 4 Result - Braid

Next I connected the braid to the nearest metal part of the dishwasher, shown in Figure 5 & 6.

Figure 5 Connecting the braid to the nearest metal part of the dishwater

Figure 6 Measuring the Unit using my VNA

Summary of the Dishwasher Measurements

Figure 7 Impedance measurement of the set-up

Figure 8 VSWR of the set-up

• The dishwasher measured 60 cm wide, 58 cm deep and 85 cm high.

• The aluminium foil ground plane measured 112 × 93 cm, with the dishwasher placed centrally on the ground plane.

• The power cord was 2 m long.

• The power cord has an inductance of approximately 2 µH, which follows my rule of thumb of 1 µH/m.

The first resonance was at 7.8 MHz. Above this frequency, the system behaves more like a transmission line.

At 16 MHz, the system behaves more like a 210 pF capacitor. Note that this unit has a plastic lower section, unlike the client's cabinet, hence the relatively low capacitance.

The VSWR plot has a minimum at 21 MHz, so the system should work most effectively as an antenna at this frequency. The VSWR was 1.7, which indicates that it can be a very effective radiator.

Adding more straps to connect the ground plane at the corners should reduce the radiating effectiveness of the metalwork.

What We Learnt

The measurement showed a strong feature at around 22 MHz. More interestingly, it showed that although the power cord has around 2 µH of inductance, the overall assembly actually looks capacitive above about 10 MHz.

So where is the RF current going?

The answer is somewhat counterintuitive.

The RF current is finding a low-impedance path through the capacitance between the base of the cabinet and the ground plane. In this case, the dishwasher has a plastic base, so there is no obvious conductive connection underneath it.

Yet at these frequencies, the capacitance between the cabinet and the ground plane provides an effective RF return path.

This is a useful reminder that at RF, current doesn't necessarily take the path we expect. We tend to think of current flowing through wires, but the easiest path can actually be through the space underneath the cabinet.

The simple VNA experiment was a great way of making this visible—and it shows that even an inexpensive VNA can be a surprisingly useful tool when debugging cabinet radiated emissions.

Sometimes the best way to understand an EMC problem is to reproduce it with whatever happens to be in the kitchen!