Why Measuring Your Coax Shield with a VNA Still Doesn’t Prove Your Choke Works
Many hams think they’ve cracked the code:
“Just connect the VNA to the braid of your coax, clip on some ferrites, and measure insertion loss. If S21 drops, the choke is working!”
Unfortunately… this is still not a valid way to measure common-mode suppression.
What They're Doing
Here’s the typical process:
- Take a piece of coax (maybe RG58, or even RG402).
- Strip both ends so the shield is accessible.
- Connect Port 1 and Port 2 of the VNA directly to the shield — center conductors are left unconnected or floating.
- Add a ferrite choke somewhere along the shield.
- Run an S21 sweep and interpret any insertion loss as “choke performance.”
It looks clever. After all, common-mode current does flow on the outside of the shield. So measuring “through the braid” seems logical, right?
Why That’s Still Wrong
Here’s why this approach is flawed:
1. You’re Not Exciting Pure Common-Mode Current
By connecting the VNA across the shield, you're exciting a signal on the shield relative to ground, but:
- There's no return path reference (ground plane or counterpoise), which means current paths are undefined.
- You’re not simulating a real-world common-mode condition — like the shield acting as an unwanted radiator due to imbalance or ground current coupling.
So what you’re really doing is driving the braid as a conductor, not exciting or evaluating common-mode behavior under realistic operating conditions.
2. The Result Depends Heavily on Setup Artifacts
The shield becomes a poor-quality transmission line in this setup, and any S21 attenuation may come from:
- Skin effect at high frequencies,
- Reflection loss at the open ends,
- Or capacitive/inductive leakage rather than ferrite suppression.
The VNA doesn’t know it’s supposed to be “common mode.” It’s just measuring loss between two nodes with undefined mode structure.
3. Ferrite Behavior Isn’t Linear in This Mode
Ferrites are most effective on common-mode current flowing relative to an external reference (i.e., an actual ground or antenna system imbalance). When you apply differential current between two ends of the shield — especially if they’re not grounded — you may push the ferrite into a nonlinear or misleading regime, again skewing results.
How to Do It Right
If you want to truly measure choke effectiveness on common-mode currents, you need one of the following:
- A current probe around the coax, measuring real current flow with/without the ferrite.
- A balanced injection method, using a balun or hybrid coupler to send identical voltage on center and shield — simulating real-world common-mode excitation.
- Or use K9YC’s test jig method, with careful referencing to ground and measurement of current suppression across the ferrite.
The Takeaway
Even if you're connecting your VNA to the braid — not the inner conductor — you're still not measuring true common-mode suppression unless your test setup faithfully recreates real-world current paths and mode structure.
Don’t confuse “loss on the braid” with “choke effectiveness.”
Your ferrite might look like it’s doing something — but in reality, you're just measuring an artifact of a test setup that doesn't reflect actual RF behavior.
Interested in more technical content like this? Subscribe to our notification list — we only send updates when new articles or blogs are published: https://listmonk.rf.guru/subscription/form
Questions or experiences to share? Feel free to contact RF.Guru or join our feedback group!
Written by Joeri Van Dooren, ON6URE – RF, electronics and software engineer, complex platform and antenna designer. Founder of RF.Guru. An expert in active and passive antennas, high-power RF transformers, and custom RF solutions, he has also engineered telecom and broadcast hardware, including set-top boxes, transcoders, and E1/T1 switchboards. His expertise spans high-power RF, embedded systems, digital signal processing, and complex software platforms, driving innovation in both amateur and professional communications industries.