RF Coupling from Antenna

[GBaron]

We are working on a product that incorporates an RF module and some digital/analog circuitry.

I would like to know what we need to do to prove that the design is solid (from a signal Integrity point of view on the Digital side).

The design uses a 922MHz transceiver with a 3W output and a a 10cm antenna.

I would appreciate any thoughts or useful contacts (Silicon Valley, CA).

TIA,

Gareth Baron

 

[VEBill]

"...signal integrity...digital..."

There's at least two aspects of the RF/digital interaction.

1) Self-EMI margin

Does your system have an embedded antenna ? If so, then you have to make sure that the self-EMI margin is acceptable. Self-EM interference margin is the safety margin (in dB) between the 3 watts of output RF and the amount of power that would interfere with its own circuits. If the margin were very low, then you might have nightmares when mass production starts.

One option for testing would be to temporarily increase your output power by way of an external amplifier (patched back into the same embedded antenna). Another approach would be to measure the amplitude of the coupled 922MHz RF (noise) on your digital signals. This can be very tricky to make sure that your 'scope probe doesn't change the coupling.

It might be worth checking this parameter over temperature to ensure that your design is solid under all conditions.

If your product has just an antenna jack (not an embedded antenna), then be sure to test for the minimum distance for the end users to keep the antenna away from the box. Try this with bad antennas as well as good antennas.

2) What is your signal margin over the maximum path?

This is easy. Simply dial-in some attenuation to see what signal margin exists over your design goal's maximum path. Or just measure the actual maximum 'Line Of Sight' path and make sure it meets or exceeds your design goal.


Obviously, there's also all the normal design issues of making sure that the design works with part variations, temperature range, etc., etc., etc. ...

 

[GBaron]

OK, maybe I should explain a bit more about the design.

The design incorporates a pager module which is encased in a faraday cage. The transmitting frequency is 922MHz and it has digital signals coming from the module.

Our design has this module attached to it (on the PCB) plus a load of other circuitry.

My concern is tha tthe antenna (externally attached helical whip) can generate enough power (3W max) to interfere with the digital traces on the PCB. The board does not currently have a faraday cage around it.

What I would like to do is measure the antenna power picked up in the digital traces to see if they are a cause for concern.

TIA

Gareth Baron

 

[photistor]

Review EMI standards and test proceedures. While you may not need to meet MIL-STD's, they may be a good reference.
MIL-STD-461 and MIL-STD-462

(There are commerical equivalent standards, I can't remember them right now)

Research the tests labled CE (conducted emmisions)& RE (radiated emmissions). If you perform these tests (or simplified versions)you can deterimine if the faraday cage is solid enough for your needs. Basically, you need a spectrum analyzer and a "sniffer" antenna to test for "radiation from the product. Then you'll need an antenna & transmitter to see how the product reacts to an RF environment. Remember, you're not just checking you transmitter's carrier, but the harmonics as well. Also, the clock frequencies can be a real gotcha.

To get an idea of test setup eqipment visit:

http://www.djmelectronics.com/emc-antennas.html

http://www.aratech-inc.com/

 

[VEBill]

Gareth's concern is about self-induced (internal) EMI. This is a (slightly) different problem than external EMI.

I can't add any suggestions beyond my point "1)" above.

 

[GBaron]

OK, so I got the main question answered in a roundabout way.

So, the Digital signals pick-up RF from the 3W antenna. We may have possible noise induced into the digital traces. How would one go about measuring the pickup in the traces ? I'm assuming tha the scope may also parasitically couple to the antenna and pickup also.

Please let me know how this is normally done because I'm currently thinking Copper tape, solder and cardboard to make a temporary faraday cage.

Thanks,

Gareth.

 

[VEBill]

Is there a self-EMI problem now? Any signs? If not, then the remaining question is 'How much EMI margin (in dB) have you got?' & '...over temperature?'

A 'scope will be untrustworthy for the obvious reasons: coupling and/or loading.

If it was me, I'd simply stick additional power (maybe try 5, 7, or 10 watts) into the same antenna under (as close as possible to) the same conditions. If it works under those conditions, then you'd know that you've got a reasonable margin of safety. If it goes insane at 3.1 watts, then you've got some work to do.

Good luck.