Another Grounding Question

[W1RKW]

I trying to determine if I should be grounding the AC ground to my HB rig chassis or not, especially the plate supply.  I'm looking at the ARRL Hand book and some of the PS schematics ground  AC mains power along with the center tap of the secondary to the chassis.  Then there are schematics in the handbook that don't show an AC mains ground to the chassis but only the grounded center tap to the chassis.  I'm thinking that it's just a drawing issue in the various schematics and it is implied that the AC ground should be there even though it's not shown in some instances.  Dealing with HV, to me it would be a good idea to ground the chassis to the utility ground in all circumstances.  Yes/no?

[KL7OF]

There was a thread on this subject last month..Some say that if you have a 4 wire AC service.. eg: 2 hot wires, 1 ground and 1 neutral, you should connect the chassis to the neutral.  That is to say the neutral is floating.  Some say to connect the chassis to service ground.  Some say to connect the chassis to the RF ground.  I remain confused.....What I see is that even with a 4 wire service, there are still only 3 wires coming off the pole  transformer (assuming single phase service)...The neutral wire in a 4 wire system originates from the service panel and is at the same potential as the earth ground wire.....so how is the neutral different from the ground?  At my station I have a 3 wire 220v system with a ground rod driven into the earth directly below the service panel.(Code)  The third wire from the pole is bonded to this earth ground as is the box of the service panel itself.This third wire is also earth grounded at the pole by a pc of #6 solid copper running down the pole. The neutral buss in the service panel is connected to this earth ground as
well....I have a couple ground rods driven into the earth under the drip line of my shack eaves...These two ground rods are bonded together with #6 solid copper and then bonded to the ground rod under the service panel...I have all chassis, all equipment racks and all coax braids bonded to this ground system..I haven't had any ground loop problems and breakers and fuses blow when they are supposed to.  This differs from what others are doing with "floating" neutrals and separate RF grounding..And their systems are working for them as well.  Hence the confusion...   The point that I'm having trouble with is....In a 4 wire system, how is the neutral at a different potential than the ground?...How is a separate RF ground system at a different potential than the service ground?...Earth ground is Earth ground....I am also looking for enlightenment grasshopper...

[WA1GFZ]

my building inspector told me the safety ground and neutral are attached at a single point inside the breaker panel. Then I showed him the 4 #8  solid lines coming in with the 3/4 inch copper water main. He told me to bond the 4 wires to the main. ( this is my RF ground for the shack) then run a lead over to the breaker panel and connect it to the safety ground buss.
I used #4 bare ( NEC 200 amp ground wire size)  Outside the house below the meter socket I have 2- 8 rods 6 feet apart. I ran a #4 from the rods to the safety ground buss in the breaker panel. Neutral only meets safety ground in one spot. This avoids ground loops. I will run a 4 wire 240 line for high power stuff and keep neutral off the chassis to avoid loops and currents on the safety lead. There will be a number og GFI breakers in the house so the last thing I want is currents flowing on the safety ground.
He told me this is the best way to avoid lightning voltage offsets.

[n2bc]

In a standard modern 240/120VAC home setup there are three wires coming from the pole, thru the meter box and into the breaker panel.  There are two "HOT" lines and one Neutral.  In the breaker panel the neutral is connected to a buss that is bonded (connected) to the panel box itself and also connected to the ground rod(s). 

Coming out of the main panel hot goes thru the breaker (single pole for 120V, 2-pole for 240V), the neutral goes to the bonded buss and the safety ground goes to the bonded buss.

While we can all see that at the far end an ohmmeter would show the neutral and ground are connected together, the difference is that the ground should NEVER purposely carry current unless there is a fault.

In a downstream sub-panel, the neutrals and grounds connect to separate buss strips and the neutral is NOT bonded to the box chassis.  The safety ground is always bonded to the box chassis.  The main panel is the ONLY place that neutral and safety ground are connected together.

So... again assuming a modern installation, the neutral should be treated as current carrying and NEVER connected to the chassis of a device. The safety ground if present in the device's cord MUST be connected to the device chassis.  Notice.... the safety ground MAY NOT be in the cord as in a 2 wire cord.  But if it is, it must go directly to the chassis. 

Two other twists regarding neutrals...

You may not need the neutral at all - as in the case of a "Pure" 240V device, it only needs the hots and safety ground.

You can run two 120V circuits with only ONE neutral. This would be the case where a feed from the panel is in say, xx/3 wire (black, red, white, bare).  Black and red go to breakers on different phases and the white neutral serves both of those circuits.  So if the "hots" are 20A (12/3 wire), doesn't the neutral "see" 40A?  No, because the hots are out of phase the neutral will only ever see 20A max.  It's important to look for this if you get inside a panel and need to juggle around breakers.  In this arrangement if you move both hots to the same phase, you WILL overload the neutral.  I think this kind of hookup is still kosher, but it is potentially dangerous if you don't spot it.  To me the risk is not worth the cost saving.

Bottom line after all this windage:  Neutral and safety ground ARE DIFFERENT.  Neutral NEVER goes to the chassis, Safety ground if present ALWAYS goes to the chassis.  LOOK carefully at the panel before you juggle breakers.

Happy 2007 to all

73, Bill N2BC

[WBear2GCR]

In a standard modern 240/120VAC home setup there are three wires coming from the pole, thru the meter box and into the breaker panel.  There are two "HOT" lines and one Neutral.  In the breaker panel the neutral is connected to a buss that is bonded (connected) to the panel box itself and also connected to the ground rod(s). 
<snip>

You can run two 120V circuits with only ONE neutral. This would be the case where a feed from the panel is in say, xx/3 wire (black, red, white, bare).  Black and red go to breakers on different phases and the white neutral serves both of those circuits.  So if the "hots" are 20A (12/3 wire), doesn't the neutral "see" 40A?  No, because the hots are out of phase the neutral will only ever see 20A max.  It's important to look for this if you get inside a panel and need to juggle around breakers.  In this arrangement if you move both hots to the same phase, you WILL overload the neutral.  I think this kind of hookup is still kosher, but it is potentially dangerous if you don't spot it.  To me the risk is not worth the cost saving.

Bottom line after all this windage:  Neutral and safety ground ARE DIFFERENT.  Neutral NEVER goes to the chassis, Safety ground if present ALWAYS goes to the chassis.  LOOK carefully at the panel before you juggle breakers.

Happy 2007 to all

73, Bill N2BC

As I sit here and scratch my head, looking at the NEC book does not help either... so, let's see now, what does one do if one has a three wire run of 220/240vac single phase? Each half of the 220/240 line is 180 degrees out, ok. Fine.

It comes off a panel with 220vac breakers, automatically set so they they must be on opposite sides of the line... no problem.

Now, my gear only wants each leg of the AC to run. The third wire is that safety ground or is that "neutral"?? And what should I do with it??  Does it want to go to the chassis??

OR, should I, or IF I have a local ground rod near the shack/rig, should I "bond" all the chassis to the ground rod (no chance of it going "high" then) and use the "third wire" as an "in the circuit only" ground point and not tie it directly to the chassis?? Is tying the chassis to the local ground rod the eqivalent of the "4th wire" - is that the "ground" nowadays or the "neutral"??

What if the "third wire" is tied to the chassis AND the gear is grounded locally to a ground rod?? No good?

And, what if I have an incoming metal panel through which I run all the RF feeders, including coax, and it is dropped to ground via a ground rod locally? Doesn't that make for a separate ground from the AC system's ground? Is that ok?? How is it different than a local ground rod for the chassis??

And, if the AC system's ground is a goodly distance from the service panel - like 50ft or more, is that really a ground that is worth a damn for anything that is faster than 60cycles - like RF or static, or lightning??

And they call this "wireless"??

           Heeeellllllppppppppppppppppp!!!

                       signed "very confused by it all..."

                       _-_-WBear2GCR

[WA1GFZ]

always put the third wire safety ground to chassis. Now if you have 2 ground systems as I have my inspector suggested I bond them. I ran a # 4 bare lead.  He wanted me to do the water main anyway which I guess is normal but he wanted everything going back to the breaker panel.  I also bought some extra #4 to continue over to the shack The main point with lightning is to have a low resistance between grounds so there won't be a voltage potential between them.

[n2bc]

NEC is about AC safety, not about RF grounding.  NEC does say all grounds must be tied together, again an AC safety thing.

You say: "...my gear only wants each leg of the AC to run."  Really?  Everything is 240VAC?  A pure 240VAC device only needs the two hots, not the neutral. EVERYTHING needs safety ground.

RF grounding is much more complicated.  Protection from lightening is worse yet.  Every wire has inductance. The higher the frequency and the longer the distance and the crappier the ground losses the mode significant all of this becomes. 

It took me 5 years of fiddling to get an RF ground system that worked for me.  3 ground stakes under the tower, 4 or 5 more outside the shack (lost count) each with varied lengths of buried radials. All of this tied to a 6" copper strap that wraps around the foundation of the house and is tied to the AC service entrance safety ground.  Then there's the fiddling with each of the antennas to keep the RF off the feedlines.

Lightening...  tough to test any setup against lightening.  I've had VHF & UHF equipment powered on and connected to antennas for 20 years here.  Never had a lightening hit that damaged the radios.  But I have lost two well pumps 200' below the surface.  Both of those went during very nasty electrical storms.

A friend of mine lives on a hilltop and his tower is pretty much the tallest thing around - full of HF beams and repeater antennas. He's got a fortune invested in lightening devices, so far so good.  Another fellow nearby lives in a valley, feeds all his antennas through a well bonded copper bulkhead and uses shorted PL-259s to 'safe' the feeds.  He's the proud owner of a charcoaled TS-940 that was disconnected from the antennas at the time of the strike. Two of his shorted PL-259s literally exploded. 

In my mind the best lightening protection is a good insurance policy.  The best protection from getting killed by your equipment is to understand and properly use safety grounding.

73, Bill

[W2VW]

Thanks Bill for taking the time out to write this. Time to disconnect the neutral at the sub-panel ground strip here......

[WBear2GCR]

<snip>

You say: "...my gear only wants each leg of the AC to run."  Really?  Everything is 240VAC?  A pure 240VAC device only needs the two hots, not the neutral. EVERYTHING needs safety ground.

RF grounding is much more complicated.  Protection from lightening is worse yet.  Every wire has inductance. The higher the frequency and the longer the distance and the crappier the ground losses the mode significant all of this becomes.
 
<snip>

73, Bill

Well, of course not everything is 240vac!
The stuff that is 240, is of course 240.
I was thinking about the 240 since that was the main topic being discussed...
I actually use a 240-->110-0-110 iso tranformer in most cases where I have a 240vac drop available.
That includes the LR where the audio gear resides.

So, the question remains unresolved in my mind, should the "safety ground" that is local to the shack, and remote from the main panel be sent to a local set of ground rods??

And, if using a large iso tranny off the local drop, with or without a sub panel, would it then be a good idea to take the CT of the secondary and run that to the local ground rod?? That wire then being used to ground all the chassis/the green wire in the line cords (whatever you name it) ?

As far as I can tell any length of ground wire that is not minimalist, when faced with the rise time of a real lightning strike is going by definition to have a huge voltage and current drop across its length... so I am skeptical about the claims and idea of long runs being able to put opposite ends on the same potential.

While not directly related, I have this audio amplifier that has 1/4" thick and 1" wide solid copper buss bars that hold together 6 x 80,000 ufd caps a 75vdc = ~500,000ufd = ~1/2 Farad! Well, there is enough drop in the ground buss bar so that I have to chose the best spot near the center of the bar as the ground for lowest hum spec!! You'd think that such a low Z thing as a copper bar would have negligable current drop across its short length. Not so! Very surprising to me it was/is.  I think this illustrates why I am skeptical about long length "ground" wires being useful in the case of a large pulse.

But then too, two ground rods separated by some distance would probably also have a different potential... ??

                 _-_-WBear2GCR

[The Slab Bacon]

If you are pullung cable for an appliance, or socket for something that uses bothe 110 and 220 at the same time you are supposed to use 4 wire cable with red (220 hot) black (220 hot) white (neutral) (also consider the neutral like a center tap) and bare copper (safety ground)

The reason for this is that the bare ground wire is NOT supposed to be carrying any current unless there is a safety ground problem. It is there to prevent an elctric shock hazard and not to be part of the actual current carrying circuit. You should never use the bare ground wire as part of the current carrying circuit not even for the neutral return. It is there to be a safety ground only.

This may sound a bit redundant especially in a 120v line, to have a seperate neutral and ground that tie together back at the main panel, but that is the way that it is! You always have a grounded neutral, but you should never use the neutral as a ground. The ground is tied to the neutral, but you should never use the ground for your neutral.   that should make it clear as mud!!

                                                                              The Slab Bacon

[w3jn]

So, the question remains unresolved in my mind, should the "safety ground" that is local to the shack, and remote from the main panel be sent to a local set of ground rods??

NO!!

And, if using a large iso tranny off the local drop, with or without a sub panel, would it then be a good idea to take the CT of the secondary and run that to the local ground rod?? That wire then being used to ground all the chassis/the green wire in the line cords (whatever you name it) ?

  You can do that, but run it to the same ground as the service entrance.  You are re-deriving your neutral from the CT of the isolation xformer, and that neutral should be grounded.

As far as I can tell any length of ground wire that is not minimalist, when faced with the rise time of a real lightning strike is going by definition to have a huge voltage and current drop across its length... so I am skeptical about the claims and idea of long runs being able to put opposite ends on the same potential.

As stated before, the NEC is about safety (ie clearing a fault with circuit breakers) not about RF grounds.  You want ALL of your ground rods bonded to the service entrance to be able to do this.  LIghtning protection is another matter entirely, but again in order to *MINIMIZE* potential differences between ground rods, they must be bonded.  Copper wire has much less resistance than the earth.

While not directly related, I have this audio amplifier that has 1/4" thick and 1" wide solid copper buss bars that hold together 6 x 80,000 ufd caps a 75vdc = ~500,000ufd = ~1/2 Farad! Well, there is enough drop in the ground buss bar so that I have to chose the best spot near the center of the bar as the ground for lowest hum spec!! You'd think that such a low Z thing as a copper bar would have negligable current drop across its short length. Not so! Very surprising to me it was/is.  I think this illustrates why I am skeptical about long length "ground" wires being useful in the case of a large pulse.

But it's much better than going thru the earth, as would be the case with 2 ground rods not bonded together.

But then too, two ground rods separated by some distance would probably also have a different potential... ??

  Absolutely.  And most likely enough resistance so that a fault wouldn't pop a breaker - hence the requirement to bond all grounds to the service entrance.