Solving "Faraday Cage" signal loss in a steel-framed modular site office?

Hi everyone,

I am deploying a small MikroTik-based network for a remote site office. The infrastructure is a heavy-duty, steel-framed porta cabin with sandwich panel insulation (structural specs and metal-cladding details: (******* link removed by mod).

As expected, the metal shell is acting as a near-perfect Faraday cage. Our internal hAP ax² is getting almost zero signal from the outdoor LTE mast, and internal clients lose WiFi as soon as they step outside.

External Antenna: Given the steel-foam-steel sandwich panels, is it better to use a MikroTik wAP ac as an outdoor bridge, or would a LtAP mini with an external SMA antenna pigtail through the wall be more stable for a mobile unit that is frequently moved?

Signal Penetration: Has anyone had success with a "Passive Repeater" setup for these types of steel units, or is a dedicated PoE-fed external access point the only viable way?

Grounding: Since the office is a giant metal box, should I ground the outdoor unit’s shield directly to the cabin's steel frame, or should it have a separate dedicated site ground to prevent surge damage?

I want to ensure our site engineers have a stable link inside these industrial units. Any advice on the best hardware/mounting combo for this metal environment would be great!

Thanks!

Hi,

I'll admit that I'm a bit foggy about the situation you describe, you seem to lump in "we'd like internal WiFi", "but also external", "and LTE" and "stable signal inside", so it's not clear what the exact objective is.

What I can confirm is that these sorts of container offices do have a pretty good (or nasty, depending on perspective) RF blocking capabilities. However, they're not nearly as good as most people image. The typical situation is that 5 GHz drops out almost instantly, even directly outside the door the moment it is closed. 2.4 GHz, however, has much better penetration, though not in any way excellent. In many situations this can be adequate. This can't exactly be called a solution, but in my experience it's often "good enough."

The proper solution is to locate the antennas outside. As you describe, the two ways to do this are to extend the RF side or the networking side. I would strongly urge you to go with the networking side extension. The wAP series works well for this.

If you also want an LTE connection, the wAP has an LTE version.

You should consult with your engineers regarding grounding. The setup for these containers varies with jurisdiction, how permanent the structure is, and whether (and in what way) it is supplies with electrical power. The external device doesn't need anything special: it should be grounded to the metal structure it's on. As for the penetrating data cable, you can play around with different surge protectors and arrestors, but basically you have to accept that a sufficiently inconvenient lightning strike can destroy these things. It helps is the external device is close to the container, not on any sort of mast and the external part of the cable is short, but these things only help balance probabilities in your favor.

If it is a Faraday cage because you want a Faraday cage, then cross the barrier with fibre optics [not your situation, but more for people who find this on search]

Re grounding, I suggest that any outdoor unit is grounded according to where its power supply is derived.

I am not an RF guy, but from what I was taught years ago, I think that if you poke a small hole through the Faraday cage, such that [nano] amps in is balanced by [nano] amps out, it make little difference whether that hole is poked for ethernet or rf in coax.

My usual practice in this type of situation is pretty straightforward. An indoor AP for indoor wifi and an outdoor ap for outdoor wifi. The client radio device for getting internet from the tower should be outside with a cable of some sort going inside to the internet device there. Usually this is a pretty simple and effective way to handle things, but when in doubt, I always recommend a site survey, especially for any kind of performance critical installations.

A mesh can be a Faraday cage, a plain sheet of metal Is more properly a Faraday shield.

Grounding Is a very delicate theme, in this case, the metal container can be considered a large metal surface/mass that Is insulated from the ground (usually these office boxes are laid on some timber beams to level It and to keep them slightly elevated from the terrain surface) and in most cases (but It depends on local norms and on the actual location) It should be properly grounded with an adequate earthing rod connected through a wire of appropriate section.

For what It costs (some 3 meters of ground wire) I would ground the antenna/external device connecting the wire directly to the earthing rod (It can be the same rod used for grounding the container) and not by bolting It to the metal shell[1].

And yes you will need an internal AP (a wAP Ax in a corner or a cAP Ax on the ceiling would do, but almost any AP would do, usually these containers are 12x2.40 m, so 30 sqm or so or the double if coupled) and an external one, this latter depends on the coverage you want/need outside, the wAP Ax would be Ideal, but It Is not omnidirectional, It covers theoretically 180 degrees, more realistically 90 to 120.
A GrooveA 52 Ac?

[1] Rationale: these sandwich panels are usually made of extremely thin sheets, painted, and though they are usually connected to the (usually L or U shaped) structural profiles (also painted) with rivets there is not any guarantee of a valid electric connection between the various parts composing the container, of course if the container is grounded through a wire bolted to one of the structural profiles, you can connect the grounding wire for the antenna/device to the same profile.