Magnetic field around coaxial cable

[Tomsik]

Hello guys, I need advice on the magnetic field around the coaxial cable 50ohm.
I've always thought there is zero magnetic field around the coaxial cable.
But when I try to connect BNC cable with termination to AWG, for example, 50MHz sine wave, and I measure level with a spectrum analyzer I can see frequency. It's small but isn't 0, why?
Have I got bad cable? Or theory calculates with ideal cable?

[radiolistener]

analyzer I can see frequency. It's small but isn't 0, why?

Because coax cable braid is not ideal and too thin.

Also there is possible common mode current on the outer side of braid due to RF leakage from equipment shell or connector. In order to eliminate it, use RF choke with ferrite or air core:

Have I got bad cable? Or theory calculates with ideal cable?

Any coax cable have some RF leakage. Better cable with a good thick double braid has less RF leakage and Chinese cheap cable with weak and leaky braid has more RF leakage. The difference between good and bad cable is leakage level.

Here is example of dual shielded coax cable and quad-shielded coax cable:



There is also good cable for symmetric lines RD-200 (РД-200), it consists of two coax cables within shared braid:



And note that RF leakage may also happens on your equipment, in such case RF current flows on the outer side of coax cable braid and it leads to RF emission. Basically your coax cable works like antenna in such case. This is why it's important to prevent common mode currents with help of RF choke.

[jmelson]

When you send signals through coax, most of the energy is contained between the inner and outer (braid) conductor.
BUT, the current in the inner conductor induces a return current in the braid.  The braid has finite resistance, PLUS some inductance.  That return current, flowing in the shield, induces a small voltage drop across the shield.  That is your leakage.  Rigid coax is better, but only reduces the effect, it can never be totally removed.

Jon

[T3sl4co1l]

Try it like this:

1. Probe near the end of the cable, to avoid transmission line effects (the frequency is still relatively low so this should be okay)
2. Join grounds, from the cable connector to the probe connector.  Minimal lead length preferred, maybe can just hold the two together with the connectors touching.

This should minimize common mode interference, without needing a ground plane and filtering setup.  Note that ferrite beads on the probe, or generator output, won't have much effect as the impedance is quite high (100s to k ohms).  With a short between them, the impedance is low, and at least a significant effect should be seen if it is due to CM, and further improvement may be seen with adding ferrites.

Common mode coupling would most likely originate from the generator, from having a poorly wired/grounded connector or internal paths that end up as conducted emissions.

Also, a good probe should be fairly immune to common mode (electric fields) in the orientation shown, but who knows.

Tim