Author Topic: Grounding  (Read 1216 times)

0 Members and 1 Guest are viewing this topic.

Offline dfroula

  • Forecaster
  • *****
  • Posts: 551
Grounding
« on: July 01, 2014, 11:04:30 AM »
Thinking about grounding again, what would be the best solution? A single-point ground consisting of a ground rod or ground wire bonded to, say, rebar on a deep concrete foundation footing? This, connected to the grounding terminal of the controller, with all other grounds isolated (power supply, shielded cables to router, etc.)?

This seems to violate the idea that an antenna should be grounded before it enters a building.

How about grounding the mast and e-field coax shield outdoors to the same ground as from the controller. Or, does the outdoor grounding through the coaxial cable from the e-field antenna alone provide sufficient grounding to both the controller an h-field shields, so that a separate ground to the controller terminal and/or h-field shields not be required?

Don
WD9DMP

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: Grounding
« Reply #1 on: July 01, 2014, 11:22:09 AM »
Thinking about grounding again, what would be the best solution? A single-point ground consisting of a ground rod or ground wire bonded to, say, rebar on a deep concrete foundation footing? This, connected to the grounding terminal of the controller, with all other grounds isolated (power supply, shielded cables to router, etc.)?

This seems to violate the idea that an antenna should be grounded before it enters a building.

How about grounding the mast and e-field coax shield outdoors to the same ground as from the controller. Or, does the outdoor grounding through the coaxial cable from the e-field antenna alone provide sufficient grounding to both the controller an h-field shields, so that a separate ground to the controller terminal and/or h-field shields not be required?

Don
WD9DMP

Thanks for starting this, Don.  The past few days have been a good review for me on this topic, needless to say...

The design of the station makes grounding an interesting exercise.  In theory, grounding the coax (or even putting lightning arrestors in the line) between the E-field amp and the preamp shouldn't cause problems, as long as four conditions are met:

1) The connection remains DC-coupled (the preamp is powered via the connection
2) The characteristics of whatever is inserted in the line are flat in terms of gain vs. frequency at VLF and LF frequencies (this is a wideband receiver)
3) Tapping the line or inserting the lightning arrestors doesn't cause a change in impedance of the line (the characteristics matter for system tuning)
4) No ground loops are created

In practice, these things may be a tall order: most coax lightning arrestors, for example, tend to be AC coupled, and tend to be designed for the MF range up.

Grounding the coax at your "single point ground" premises entrance is a good idea from a lightning protection and grounding "best practices" standpoint, but in theory may cause a ground loop if the station is also grounded at the controller (also a good idea, from a digital noise standpoint)--UNLESS the controller ground is also grounded at the same point, tied together with low-inductance, low-resistance grounding strap.  If the station is located near the single point ground, it shouldn't be an issue, but who knows?  This is a very high sensitivity system.

Polyphaser grounding literature

Personally, I intend to do all of my grounding through the station controller, tied directly to the premises single point ground.  If I end up putting the E-field antenna outside on my roof, I don't have an easy way to tie that entrance point to the single point ground through a separate path, anyway--it would like be fed through a hole in the attic wall.  I will count on the fact that my chimney and the surrounding trees will likely make more attractive targets.

I'm happy to read others' thoughts on this.
« Last Edit: July 01, 2014, 11:36:33 AM by n0ym »

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: Grounding
« Reply #2 on: July 01, 2014, 11:38:06 AM »
To clarify, I do think a direct ground tie at the controller is a good idea, due to both analog and digital sections being on the same board (the lower the impedance connection from the controller ground plane to your ground reference, the better).

Offline 92merc

  • BismarckWeather.net
  • Forecaster
  • *****
  • Posts: 1129
  • BismarckWeather.net
    • BismarckWeather.net
Re: Grounding
« Reply #3 on: July 01, 2014, 12:59:10 PM »
As my high school teacher used to say, "Ground is ground, the world around".

My controller and H field amps are "near" my main house panel located in the attached garage.  I attached a thick solid wire to that same grounding rod and run it up the wall and to a terminal block.  I then ran one wire to my H field antenna shields, and another to the controller.  But I didn't ground the amps.

I think the amps don't need a ground, or more precisely are using a floating ground.  In other words, grounding it again may make things worse.  I think that is the case with the E field coax.  When I had asked earlier, they recommended not grounding that coax for lightning reasons.   If lightning is going to hit the E field, a little grounding unit won't help.
https://www.BismarckWeather.net
Davis VP2, Cumulus, WeatherDisplay, Blitzortung, Saratoga Scripts, NOAA Stream via PI

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: Grounding
« Reply #4 on: July 01, 2014, 01:04:23 PM »
As my high school teacher used to say, "Ground is ground, the world around".

My controller and H field amps are "near" my main house panel located in the attached garage.  I attached a thick solid wire to that same grounding rod and run it up the wall and to a terminal block.  I then ran one wire to my H field antenna shields, and another to the controller.  But I didn't ground the amps.

I think the amps don't need a ground, or more precisely are using a floating ground.  In other words, grounding it again may make things worse.  I think that is the case with the E field coax.  When I had asked earlier, they recommended not grounding that coax for lightning reasons.   If lightning is going to hit the E field, a little grounding unit won't help.

The amps establish a split rail for signal ground.  However, they do this midway between 5V and 0V rails.  The 0V rail is tied directly to the controller ground lug through the cabling between controller and amps (it's also tied directly to the power supply 0V/ground return), and is tied electrically to the coax shield to the preamp--minus ohmic losses in the conductors.

It's those "ohmic losses" that could cause problems if you use separate grounds for the different components/coax shield.  Separate grounds, and different resistance to ground, can cause different ground potentials for circuit points that ostensibly should be at the same voltage.

Just to be clear, I think it probably wouldn't cause problems if the coax braid and the controller ground lug were both tied to the same ground, as long as the tie were very low ohms/impedance and didn't constitute long, separate runs to ground.  But if I had to choose just one ground point, it would be on the controller board.  It's a better way to go.
« Last Edit: July 01, 2014, 03:17:27 PM by n0ym »

Offline Cutty Sark Sailor

  • WxElement panel
  • Forecaster
  • *****
  • Posts: 3303
    • Frankfort Weather - TwinHollies WeatherCenter
Re: Grounding
« Reply #5 on: July 01, 2014, 01:17:08 PM »
Don't ground the coax. Ground only at the controller. If anything, view it all as a split rail or virtual ground, since the Efield preamp and Amp assume that stature. The coax is integral to the filtering.
When messing with megavolts, nothing's sure when it comes visiting, but the probe end would technically be at least 10k ohms (R1) above ground, and the monster at the door might seek an easier path. If he wants in that badly, he'll get in anyway. More likely to go around to the back door and bust in through your wall wart.
 

Offline Cutty Sark Sailor

  • WxElement panel
  • Forecaster
  • *****
  • Posts: 3303
    • Frankfort Weather - TwinHollies WeatherCenter
Re: Grounding
« Reply #6 on: July 01, 2014, 01:21:44 PM »
Only external ground on the system should be at the controller. It's been said over and over by both the developers and others. One Ground. Controller. Ground. Block.
 

 

anything