I am looking to set up a test jig to test radiation susceptibility. This is to do a small in house test to check if my board is susceptible to radiation. The tests frequencies are the standard test frequencies covered by FCC guidelines.

Now, to do this I want to build a circuit to do that. How do I proceed for such a thing ? Any tutorials that I can look into ?

Are there any test circuits ?

  • \$\begingroup\$ Near field Susceptibility levels are far higher than unintended radiation levels for far-field interference e.g. FCC mask. I have written corporate standards for these tests, which consists of V/m profiles in each band and current arcs from open coils creating impulse EH fields. You have to choose what level of immunity is right for your product and environment which is custom. but since then EU has standards you may follow. \$\endgroup\$ Jan 10, 2017 at 16:04
  • \$\begingroup\$ EMI susceptibility includes pulsed radar, impulse radiated swept f CW, conducted impulse and conducted CW and ESD 10x all user access sides . I recall we used 10kV no effect, 15kV user intervention allowed, 25kV no damage \$\endgroup\$ Jan 10, 2017 at 16:09
  • \$\begingroup\$ @TonyStewart.EEsince'75 , thank you for your tips. Radar and such sound very expensive. Anything I can achieve in a lab environment? \$\endgroup\$
    – Board-Man
    Jan 10, 2017 at 16:11
  • \$\begingroup\$ Radiated easiest method is get a power relay and drive coil thru NC contacts which turns into a pretty good EMI generator then use the wire loop and an antenna and bring along side any IO cables to find immunity threshold. Same for conducted noise AC coupled to line with a X cap 1nF It can be battery powered. The abrupt contacts and arc gap creates impulse EH field all the wave up to the contact gap half wavelength and then harmonics.. but then filtered by wire inductance length, so short is stronger. \$\endgroup\$ Jan 10, 2017 at 16:14
  • \$\begingroup\$ You can then roughly measure E field with a 10M scope probe and 1m wire at various distances parallel to field disturbance with a ground plane if possible. Then record the level with antenna approximating distance to target and interferer to target being the same , then record the level of tolerance at 3 levels, No effect, User intervention required, ( eg. CPU reset), Fault (i.e. EOS damage if any or Latchup( which causes hot chips reset by power cycle.) Then decide what level of exposure fits your product for acceptance criteria. \$\endgroup\$ Jan 10, 2017 at 16:24

1 Answer 1


People have come to me, requesting assistance in modifying their PCBs, often with failure in the 400-600MHz region. My response was "See that long PCB trace running from the MCU to the sensorIC? Its a quarter-wave at 400MHz, over Er=4 PCB material, unloaded at one end, loaded at the other (quarter-wave transformer---bad news, high voltages generated). Install a 10Kohm near the sensorIC."

The sensor IC was getting upset, via its SPI output pins.

With 10Kohms installed, to dampen RF energy and to mostly prevent RFI entrance to the sensor IC, the project moved ahead.

I suggest you specifically test at frequencies where the top/bottom traces will store-energy, as quarter-wave or halfwave striplines.

I also like the nasty-sparky NormallyClosed power-relay test idea.

  • \$\begingroup\$ I see your relay and raise you the infamous 'Calrec CB', used by a UK audio mixer manufacturer for immunity testing back in the day. The thing consisted of a kettering ignition system taken from a junked car with the points driven by a motor and two telescoptic radio aerials brazed onto the sparkplug (Forming an adjustable dipole), it reputedly used to 'test' the RFI immunity of every telly within a few miles of their site. No screened rooms back then. \$\endgroup\$
    – Dan Mills
    Feb 1, 2017 at 20:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.