0
\$\begingroup\$

For regular DC power adapters I will see a polarity indicator like the following:

polarities

However I have seen some RF circuits that indicate a DC power connection to the circuit over an SMA connector but do not specify a polarity. Is there a standard for the polarity of the center connection?

I was originally going to assume the center connection as positive but then I came across a BNC to post adapter and it had the center connection as negative.

\$\endgroup\$
4
  • \$\begingroup\$ @mkeith it's often ground. In balanced systems, both inner and outer conductor might oscillate relative to ground, and a superimposed DC voltage can be referenced to anything. I'd agree, in such a system, if there's any connection to ground, it's likely to be on the outer conductor, but there's no rule. \$\endgroup\$ Jul 30 '18 at 15:33
  • \$\begingroup\$ @MarcusMüller, upon reflection, I guess that is true. There is no rule that SMA or any coax cannot be used to feed dipole antennas and such which don't really have a "ground." I will delete my comment. \$\endgroup\$
    – mkeith
    Jul 30 '18 at 15:38
  • \$\begingroup\$ I don't know, it's a valuable commentary anyway! \$\endgroup\$ Jul 30 '18 at 15:41
  • \$\begingroup\$ Your answer covers it, I think. \$\endgroup\$
    – mkeith
    Jul 30 '18 at 15:43
3
\$\begingroup\$

I'd expect the outer conductor to be the one that's on ground potential, for practical reasons; I'd expect the bias voltage to be on the center conductor. However, whether that voltage is positive or negative to ground would be defined by the application. So, if you have an LNA that needs positive bias relative to ground, then that center conductor would be positive relative to the outer conductor, and vice versa.

Don't expect a standard; I'd agree, I'd expect a higher probability of the center conductor being positive relative to the outer, but you'll always have to check.

\$\endgroup\$
3
  • 1
    \$\begingroup\$ If the system documentation doesn't explicitly show the polarity for DC bias connection, then it is very likely that the center is positive bias. If I were documenting a system which required negative center bias, I would for sure call attention to it with bold print, etc. \$\endgroup\$
    – mkeith
    Jul 30 '18 at 15:42
  • 1
    \$\begingroup\$ Absolutely! But then again, RF systems with bias voltage tend to be a bit expensive to omit a simple hypothesis test to not e.g. fry an LNA \$\endgroup\$ Jul 30 '18 at 16:04
  • \$\begingroup\$ If I made a system that took a DC bias on an RF connector, then I wouldn't bother to supply any user-facing documentation about polarity, because it has to be used with whatever else was supplied with the system. This is not going to be some mix'n'match 'use another power supply' system. Polarity, voltage, current, their tolerances, impedance, bandwidths etc would be in the technical documentation. \$\endgroup\$
    – Neil_UK
    Jun 10 at 13:12

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.