Every electronic product has to answer two questions before it ships, flies, sails or drives: does it pollute the radio spectrum, and can it survive the pollution of others? Those are emissions and immunity, and the answers are only worth anything if they were measured in a room built to make the measurement true. Neometrix’s EMI\/EMC test laboratories are shielded anechoic chambers engineered to do exactly that \u2014 \u2265100 dB of shielding so the outside world disappears, absorber linings until the room measures like open country (normalised site attenuation within \u00b14 dB), and test suites running from CISPR through MIL-STD-461, sized from bench equipment up to complete vehicles on an in-floor dynamometer.<\/p>\n
It has to be a Faraday cage \u2014 \u2265100 dB of shielding, a factor of ten billion in power \u2014 so a city’s radio chatter can’t reach the receiver inside, and the lab’s own kilowatt-class test fields can’t reach the city outside. And it simultaneously has to measure like open country: a bare metal room is a hall of mirrors for radio waves, where every measurement would arrive twice, reflected. That tension never fully resolves; it’s engineered around, continuously.<\/p>\n
The shield is lined with ferrite tile for the low bands and hybrid pyramidal absorbers above them, until the room’s normalised site attenuation tracks an ideal open-area test site within \u00b14 dB, point by point, frequency by frequency. Getting a shielded metal box to behave, electromagnetically, like an empty field is the core achievement of the absorber design.<\/p>\n
Air enters through honeycomb waveguide vents that pass air through but cut off RF below their design frequency. Power enters through filtered feedthroughs. Doors seal on knife-edge contacts through beryllium-copper finger stock. The shield is only ever as good as its worst penetration \u2014 which is why every single crossing of the boundary is deliberately engineered, not treated as a minor mechanical detail.<\/p>\n
Before any product is ever measured inside it, the chamber itself is validated: shielding effectiveness to EN 50147-1\/IEEE 299, normalised site attenuation to ANSI C63.4, site VSWR to CISPR 16-1-4, and field uniformity to IEC 61000-4-3, with 17025-traceable calibration carried in the scope of supply. A lab that hasn’t proven itself first has no business proving anything about a customer’s product.<\/p>\n
The same underlying physics scales across five configurations: a 3-metre chamber for bench-scale products, a 10-metre chamber for machines, a MIL-STD-461-compliant room for defence electronics, an automotive hall with a turntable and an in-floor chassis dynamometer, and a fully anechoic RF chamber quiet to 40 GHz. What changes across all five is scale and configuration, not the underlying engineering discipline.<\/p>\n
Rarely in the steel itself. They fail at the door whose finger stock has fatigued after ten thousand closures, at a penetration someone added for one more cable and never filtered, and at ageing absorbers whose loss has quietly drifted until the normalised site attenuation walks out of tolerance \u2014 while the lab keeps issuing reports it can no longer actually stand behind. That’s why the room is treated as the first test article, validated at handover and re-validatable after every change.<\/p>\n