1
\$\begingroup\$

I've been working on a Shielding Effectiveness system of 120dB according to IEEE299-2006 standards. The frequency range covered is 9kHz to 40GHz.

I've designed it using an analog signal generator, power amplifier, antennas (varying antennas for varying frequencies), pre amplifier and spectrum analyzer. I would like to know how to choose a cable for the system. I figured RG213 would be a good choice, however if it necessary to use it through out the system all the way from signal generator to spectrum analyzer.

Any ideas or suggestions would be good. In fact, guiding the way on how to select an appropriate cable that would meet the aim of 120dB for the frequency range of 9kHz to 40GHz would be much appreciated.

\$\endgroup\$
  • \$\begingroup\$ If you didn't use a cable you wouldn't have to worry about its shielding capabilities so, why do you need a cable? \$\endgroup\$ – Andy aka Jan 27 '16 at 9:33
1
\$\begingroup\$

RG213 is a badly shielded low cost cable for a few hundred MHz at most.

For your application you'll need 086 size rigid coax, just over 1 mm thick. Rigid or semi-rigid coax has a solid braid, so up to 100 dB screening. (but probably not at 9 kHz). At 40 GHz you'll need very thin coax, which is unfortunately very lossy.

Brace yourself for the price of 40 GHz connectors, hundreds of dollars each. You can't use N-type above 10 GHz or so.

You might consider separating your whole system into different frequency bands. The antennas and amplifiers will already be quite band-specific. Then you could use 141 or 250 size rigid coax for the lower frequencies.

\$\endgroup\$
  • \$\begingroup\$ +1. Excellent point on the cost of connectors. Perhaps the OP should consider physically different ports for the different frequency ranges (which you may be implying). \$\endgroup\$ – Peter Smith Jan 27 '16 at 18:28
  • \$\begingroup\$ Type N connectors mode at about 20 GHz, but for spectrum analysis applications, Type N can be made to work well to 26.5 GHz. For 40 GHz, 3.5mm is sufficient, and they are not too lossy at low frequencies. \$\endgroup\$ – Tom Anderson Dec 7 '16 at 4:22
1
\$\begingroup\$

RG213 is not recommended for GHz application. Attenuation will be too high. You will need to look for low loss microwave cables.

Be careful of the length as well as it will introduce a lot of attenuation.

Using RG214 for <1GHz is good, above that, please go for something like sucoflex or similar.

Do remember that 120dB system dynamic range may not be a true representation in the field due to environment noise.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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