EMRFD Message Archive 10481
Message Date From Subject 10481 2014-11-22 19:03:11 Jerry Haigwood Grounding coax cable Hi Folks,
I am building a HF project with lots of modules. Most of these modules
are interconnected using RG174/U. I have a question for all you smart guys
and gals.
There are two schools of thought on grounding the shield of the coax.
Some people ground both ends of the braid. They feel this method will offer
better attenuation of any noise/RF that might want to leak in or out. Other
people are concerned about creating a ground loop and only ground one end of
the braid. Most of my cables will be less than 6 inch long in case that
makes a difference. So, what do you think?
Jerry W5JH
"building something without experimenting is just solder practice"10482 2014-11-22 19:53:02 David Re: Grounding coax cable With HF as opposed to baseband signals, you have less to worry about with ground
loops. If neccessary common mode chokes or baluns may be used to attenuate
ground loop noise at high frequencies.
On Sat, 22 Nov 2014 21:03:08 -0600, you wrote:
>Hi Folks,
> I am building a HF project with lots of modules. Most of these modules
>are interconnected using RG174/U. I have a question for all you smart guys
>and gals.
> There are two schools of thought on grounding the shield of the coax.
>Some people ground both ends of the braid. They feel this method will offer
>better attenuation of any noise/RF that might want to leak in or out. Other
>people are concerned about creating a ground loop and only ground one end of
>the braid. Most of my cables will be less than 6 inch long in case that
>makes a difference. So, what do you think?
>
>Jerry W5JH
>"building something without experimenting is just solder practice"10484 2014-11-23 05:30:22 kb1ckt Re: Grounding coax cable I believe whenever the shield is not grounded that their is two or more wires inside the shield. Think if all the signal inside the cable, and the shield for just shielding. Coax though uses shield as part if the circuit. It shields too once frequency gets high enough that the thickness of the shield is a few skin depths. Then the shield acts as if it was two wires, inside and outside.Shawn kb1ckt
Sent from my iPad10485 2014-11-23 06:11:32 Dan Mills Re: Grounding coax cable If you want the coax to shield RF you must ground both ends (If the
target is just the E field and the line is electrically short you can
float one end, but think about how the loop completes).
In general RF current will flow so as to minimize loop area (And thus
energy stored in the H field), leaving one end ungrounded forces the
return current into the chassis making for a much larger loop are to
both pick up and (potentially) radiate nasties.
Single ended audio suffers because large externally produced currents
can flow between the chassis (With the loop completed via the pains
safety earth connections - which are inviolate), this is less of an
issue at RF.
Now all that said, I know what an approximation to the theory says,
but you really should experiment (And measure).
Dan M0HCN.
10486 2014-11-23 06:22:58 bobtbobbo Re: Grounding coax cable In that situation I have always grounded both ends and never had a problem.
Bob
K1AO10489 2014-11-23 08:53:44 DuWayne KV4QB Re: Grounding coax cable I worked in industrial and medical instrumentation, to ground or not was determined mostly by frequency. In DC to AF signals the currents in the circuit are almost all determined by the resistance in the circuit. Grounding at both ends creates a parallel current path or Ground Loop that may couple unwanted signals from one stage to another. With Video or HF signals the currents are mostly determined by the circuit impedance. If only one end of the coax is grounded, the actual impedance of the other circuit is the circuit impedance plus the impedance of any stray inductance or capacitance in the return path through the chassis. This can cause impedance mismatch issues or possibly instability or parasitic oscillation. In most cases input and output impedance's are matched and usually kept low 50-75 ohm, coax is grounded at both ends and very near the input or output circuit.
Things get much more complicated when you have a mixture of AF, RF, and digital circuitry in the same box. In that case you may have to have separate analog and digital grounds, isolation circuits between different types of circuits.
Do a search on 'analog devices grounding app notes' and take a look at a great selection of application notes on grounding and shielding of different signals.
73 DuWayne KV4QB
10490 2014-11-23 13:04:54 Will Re: Grounding coax cable One way around the problem would be to use a floating link coupling at
one end of the co-ax. Then that end of co-ax would not need to be grounded.
Cheers,
Will,
ZL1TAO
10491 2014-11-24 19:10:43 kb1gmx Re: Grounding coax cable >>>With HF as opposed to baseband signals, you have less to worry about with ground
loops. If neccessary common mode chokes or baluns may be used to attenuate
ground loop noise at high frequencies.<<<
First ground loops occur when there is sufficient current flow regardless of thefrequency to effect a potential difference. That can happen at DC or GHZ.The use of chokes and baluns can work but solving the why of the loop isoften more effective.Allison10492 2014-11-24 20:45:36 Andy Re: Grounding coax cable Follow the currents. Many RF circuits are low impedance (~50 ohms) so the current is as important as the voltage.The coax shield is part of the current path. Break that path and you no longer have just the coax carrying the current. Now the return current must find a different path(s), through whatever chassis and other wires are nearby. Now the path has a very large and open loop (compared to the coax), with much bigger inductance, and the impedance won't be 50 ohms anymore. Plus the return path may be shared by many other signals, and all their magnetic fields are coupled to one another.Andy