RF Interference Testing

[darrenb]

Hi,

I googled "eevblog rf interference testing" before posting.  This question is for work and not my normal personal stuff - not sure how this goes down here.

Work is looking to do some rf interference testing.  They have been told they need to test at ten times normal output power for interference to equipment from Bluetooth, WiFi (both 2.5 and 5 Ghz) as well as cellular signals.  Basically any consumer type signals that could be in the vicinity.

The problem is that the list of equipment that we've been told we need includes an RF Signal Generator, RF Power Amplifier, Antennas and a Spectrum Analyser.  At the required frequencies these are very expensive to either purchase or rent.

Any ideas for either sourcing this equipment in Sydney Australia or any ideas about how this might be done using alternative equipment would be appreciated.  In short we just have to have a way to show that we are radiating at the required output power.

[tggzzz]

I suggest that the first step is to define the authoratative acceptance criteria.

After that it may well be necessary to hire a specialised testing company.

[darrenb]

Hi,

The testing is being done on an aircraft.  The authoritative acceptance criteria, and expected result at these loads, is that there is no affect on any of the systems when subjected to the RF load.  A full testing plan has been developed which states exactly where and how the tests are to be done.

A specialised testing company is an option.  If possible I'd like to get a clearer understanding of the equipment that would be required no matter who does the testing and options for acquiring it.  For example is a spectrum analyser at these frequencies required if we can show that the output power levels at the required frequency is being emitted through other means?  Is an rf power meter an option?

[darrenb]

Also, for example, is the rf power amplifier likely to be required or could the output be achieved via a modestly high dbm rf generator and a high dbi antenna?  If the antenna has to be directional to achieve the output that would likely work.  We're only looking at 10 x consumer output power so maybe antenna gain would be enough?

[Howardlong]

I'd agree wholeheartedly with tggzzz, it sounds like you really need to get a compliance testing house involved, not just to do the work, but to specify *exactly* what the testing will be bearing in mind the rather casual "ten times consumer output".

RF is a funny thing, and I don't mean funny haha either.  Getting controlled, repeatable measurements is very, very hard. I don't know what category of aircraft you're talking about, but at one extreme you could be talking about needing a very large anechoic chamber.

One thing that concerns me, are there not already standards in Australia for certification? In Europe we have EASA and ETSI approval for each piece of avionics equipment. Admittedly these are unit tests, not a system test made once the devices are fitted. However typically the immunity levels provided by the type approvals process mitigates this to a large degree.

At the other extreme, you could simply use a carrier wave from a signal generator with an HPA and a dipole at various defined frequencies in various locations of the aircraft and see if that affects any of the avionics. The problem with this is that it also means you're going to need to have a means of testing your avionics at the same time, for example your transponder, TCAS, weather radar, GPWS, FMC, ACARS, ILS, GPS, VOR and DME equipment. Furthermore, typically the equipment might not just fail completely, more likely it will become less sensitive due to front end (or other) overload. In addition, it might be the faults occur not with a carrier itself, but it might be the modulation on the carrier. Also remember that a lot of interference is caused by complex mixing products, so testing at one frequency in a band could well be not enough.

There are lots of dimensions to what you've proposed, and it's difficult to see how it could be achieved without creating a set of reasonable technical criteria first.

[tggzzz]

RF is a funny thing, and I don't mean funny haha either.  Getting controlled, repeatable measurements is very, very hard. I don't know what category of aircraft you're talking about, but at one extreme you could be talking about needing a very large anechoic chamber.

Just so. Anything aviation-related will have clearly defined requirements, and those must be satisfied on the exact production equipment. Make one tiny change and repeat the tests.

Hence there is one way in which in-house testing might help them - as an informal mechanism for them to find out relative magnitude of the interference and susceptability of the equipment, before they go to the external testing house. For example, does a ferrite/slot/shielding/screening increase/decrease the effect.

If that's the case then virtually any testing with any equipment would be better than nothing. Google for a reasonable starting point, rent some equpment at the right time, and just have a look.