Good earth ground

For posts specifically relating to fusor design, construction, and operation.
Post Reply
jlheidecker
Posts: 97
Joined: Thu Jan 23, 2003 3:11 am
Real name:
Contact:

Good earth ground

Post by jlheidecker »

In my old lab I had a large metal spike that went from the floor of my
lab and into the earth 6 feet. Unfortunately I do not have such a good
earth ground in my new lab. Can you guys offer alternatives?

Thanks,
Jason Heidecker
Richard Hester
Posts: 1519
Joined: Thu Jul 05, 2001 12:07 am
Real name:

Re: Good earth ground

Post by Richard Hester »

If you have a water pipe nearby, try that (if it's not PVC).
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Good earth ground

Post by Richard Hull »

Like Richard said, check your pipes. Copper and Iron piping is a relic of early last century in homes. Today, the electrical ground of a house is and must be a driven copper rod, by code and by law. If I demanded a known good, separate ground, in a modern home, I would go to the electrical fuse box and tie on a #8 or #6 insulated copper cable to the box ground and run it as needed to my experimental area.

As an ex-Tesla coiler running up to 15kw systems in my lab, I ran three 10 foot driven ground rods inside a buried 3' length of 28" diameter PVC pipe with a 1" wall thickness.

#00 cable is run along side a #8 bare copper conductor 6 feet into my lab where it connects to a network of under-bench 3/4 inch copper piping. The large sunken PVC pipe forms a pit which I have filled with sand to a depth of 2 feet just outside my lab. I occassionally pour in rock salt and water to saturate the ground below to keep the electrical grip on the earth solid.

The sand and pipe pit keep the salt from diffusing into the upper ground level where it would kill the grass. Seems to do fine.

Attached is a picture of my ground system taken for this post in a driving rain as hurricane Isabel was rising in strength here. Richmond is expected to see the eye of the storm in about 13 hours. The winds are already 35 mph and gusty to 50. Sent home from work at 11AM

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Verp
Posts: 171
Joined: Fri May 02, 2003 3:27 pm
Real name:

Re: Good earth ground

Post by Verp »

I have a strong suspicion calcium chloride would work much better as the salt to make the ground a good conductor. Most soil contains a significant amount of hydrous layered silicates in the form of clay. When the clay component of soil is exposed to a lot of Na+, the Na+ bonds to the negative charges of the silicate layers, neutralizing the net charge. This causes the silicate layers to pack together squeezing out the water and ions between the silicate layers. This makes the clay more compact and less water and nutrient permeable, which is one reason why too much common salt is not good for plants. Ca++ ions also bond to the silicate layers, but the extra + charge repels the silicate layers, expanding the clay and letting the soil hold more water and soluble ions between the silicate layers and in the soil in general. This is also why gypsum and other calcium compounds are used to undo salt damage to soil.

I know this stuff because I took a soils course in my entomological studies. I had never seen such a multidisciplinary field before. Soil determines everything from the local ecology to the construction of building foundations.

Anhydrous calcium chloride is sold as a desiccant and snow melt. Its affinity for moisture is so strong; it gets noticeably hot when dissolved in water. I tried to use that property to unfreeze a sewer line by making a hot slurry with it. It cooled and formed something resembling water soluble concrete. Keep this in mind if you try making a saturated solution of it. I used it as a desiccant to freeze dry a freezer full of wild edible mushrooms by simply letting a large pan of it dry the air inside the freezer. I purchased it as a 50 lb bag of snow melt, so it is not hard to get or expensive.

Rod
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: Good earth ground

Post by John Futter »

I have put in few hilltop RF systems in my time and all have used copper sulphate filled copper pipes buried below the tower base.
the pipes provide a salt path into whatever cracks in the earth so that the RF impedance is minimal for lightning strikes. the conductivity of earth is dictated for lightning strikes over the first 30 feet, so this area is where the most gains are had because of the rise time of the event, fusors with discharges are within the same order of magnitude. The pipes are bonded to the reo steel in the tower base and to the tower itself, and the reo is bonded to the radio equipment.
Result is all equip bounces with the strike equally and nothing gets a differential voltage, these systems have been in place for many years with no failures after multiple lightning hits.
Note copper sulphate is poisonous to the environment so its use must be below the biological soil level and this may not be used if leaching into an active aquifer could cause downstream pollution.
All of this learnt on an international lightning protection course.

The high voltage used by fusioneers is at the small end of problem but the fixes are the same.
dv/dt and di/dt are killers of equipment / experimentors.
tligon
Posts: 587
Joined: Tue Jan 02, 2007 1:58 pm
Real name:
Contact:

Re: Good earth ground

Post by tligon »

The last time I looked at an electric code it was actually calling for two ten foot ground rods, some specified distance apart. There is a minimum spec for the ground conductor to the rods which is worth looking up. Probably 6 AWG at present.

Be aware that the requirements for a building ground are not particularly stringent. I think the spec is 25 ohms or less. 25 ohms is, in my way of thinking, terribly high. Picture what would happen if a 10 kA lightning bolt hit it. But at least if you run two rods, you have some ability to test conductance between them.

Some ground upgrades include ring grounds (a copper conductor in a trench making a ring around the building), and radial ground planes favored by some radio amateurs.

NEC generally requires all building grounds to be tied together. You can enhance your building ground by adding more rods, but having a separate ground not tied to the building ground may violate code.
tligon
Posts: 587
Joined: Tue Jan 02, 2007 1:58 pm
Real name:
Contact:

Re: Good earth ground

Post by tligon »

That "bounce" problem is interesting. We used to have two buildings, a concrete "blockhouse" and a metal building. One day the metal building was hit by lightning. The entire structure evidently "bounced" at least several thousand volts. There was no electrical damage in that building.

The blockhouse was another matter. There was a heavy bundle of ethernet, phone, and security system cable between the two, and equipment in the blockhouse connected to those was largely fried.

Hence the need to tie grounds together.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Good earth ground

Post by Richard Hull »

You should see the ground bouce around an operating 15kw tesla coil. We did a series of tests back in the 90's and upon attachment and break of some 13 foot arcs, the local ground bounced over 4,000 volts.

I killed one washing machine, two TVs one portable phone, three touch tone "twisted pair" phones and my neighbors answering machine. during my tenure with large coils. To this day, we still have only old rotary phones in our home.

So what is a good ground? Depends on who's talkin', what's operatin' and what needs protectin'.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
billwcf
Posts: 37
Joined: Wed Sep 14, 2005 3:52 pm
Real name:

Re: Good earth ground

Post by billwcf »

Hi, I was pretty heavy into Ham radio in the early 80's. I read an article in QST or some Ham publication about a way to achieve a perfect, lasting , ground. The author said to attach an old car radiator to heavy copper wire and bury it under you septic drain field. In that way you would always have a "wet" ground. We built our house in '86 and that's exactly what I did. The septic tank dude thought I was crazy until I explained it to him. It has served us well. I don't guess too many people are building houses these days, but if you do, keep it in mind. Of course the relatively cheap copper radiator and wire back then, now have become a luxury item. Another thought: maybe you can drive the copper rod off to the side of an existing house drain field. Civil engineers, please correct me on that. The drain fields are usually easy to find as the grass seems to be greener and thicker over them. Guess you would have to drive it below ground surface and trench the wire so the mower won't get it. -bill
tligon
Posts: 587
Joined: Tue Jan 02, 2007 1:58 pm
Real name:
Contact:

Re: Good earth ground

Post by tligon »

A good thought, that. My drain field is already in but I know where it is.

The codes where mine is say it is three feet underground. My system uses four parallel 100 ft trenches, 10 ft apart, each 3' wide, and with a perforated pipe down the middle. That should be a pretty easy target to hit with ground rods.

Otherwise, water runs downhill. Look for drainage channels.
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: Good earth ground

Post by DaveC »

For safety, all system grounds should return to the HVPS, using either flat Cu braid or #10 Cu AWG (larger is fine). The HVPS should turn be grounded to the mains ground at the nearest panel.

The panel, depending on local codes, conditions and etc should have the ground rod.

You really don't want ground rods in several places. In almost any location they rarely lower the ground impedance by any important amounts... especially given today's situation with plastic water and gas lines, insulation joints and the like.

You can get an estimate of what a ground rod will do for you, if you think of it as a cylindrical resistance... forget the length, as a first approximation, in a medium of resistivity (whatever your soil is or ground water measures...) . Many places the ground water or soil resistivity is in the 500 - 2000 ohm -cm range.

The resistance between two concentric cylinders of diameters A and B in an infinite medium of resistivity rho... is R = rho/2pi*[Ln (B/A)] . The resistance is scaled by the ground resistivity, and the log of the ratio of rod diameter A and outer cylinder B.

For small short ground rods and long distances... the rod actually looks more like a hemispherical electrode and the relationship there becomes R = rho/4pi * (1/L - 1/D) where L is rod length and D is distance to the other electrode.

The point in all this is that the geometric parts are unity or greater, so the resistance is NOT low, unless the medium is salt water.

I've seen the ground mat of a large switching station, mat was made with with 500 kcmil bare copper on 3 ft centers - 1/4 mile square, swing 1-2 kV during switching operations. And the large Gas Insulated stations can easily swing a number of times that figure.

So the only LOW impedance lines are those that resemble strip lines... with commensurate spacings of conductors... 1mm or less..Other wise you're talking about 10's to 100's of ohms impedance during impulses.. And if you happen to hit a resonance mode, you could have an open circuit....

Dave Cooper


PS.... seems to me this was all discussed a long time back... ??
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Good earth ground

Post by Richard Hull »

Dave is right. This ground business pops up on occasion. Grounding for our purposes is usually not much of an issue unless you plan on pulsed fusor operation or are working with a lot of RF ion gun feeds, etc. We just don't do much impulse work with fusor's.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
Doug Coulter
Posts: 1312
Joined: Sun May 27, 2007 3:18 pm
Real name: Doug Coulter
Location: Floyd, VA, USA
Contact:

Re: Good earth ground

Post by Doug Coulter »

Well, I DO primarily do pulsed work here, and it's a problem. And it's not a DC ground you worry about, it's that bounce which can be either from inductance to the real ground (eg any long run), but also simple induced volts as in a transformer from a big current pulse in some wire that runs near another. No brute force un thought out scheme seems to work all that well. The old star ground system is susceptible to different inductions in the different wires and so on. Usually the best plan is to kill the noise as close to the source as you can, use coax for things that have high current pulses in them, and so on. But also, testing with a scope with reduced current pulses *before* you fry some delicate gear is a good idea.

I have two lighting rods on my lab, primarily to protect the solar panels on the roof. They take direct hits fairly regularly with nothing bad happening indoors (they get a little shorter each time though), except the odd surge suppressor on my internet/phone connection giving up its life for me. Since this is a lab and a workshop, I further have compressed air piped everywhere, a loop of Cu pipe on each floor, tied together at all 4 corners, kind of a sloppy Faraday cage. All this is tied to ground rods, total of 8, driven into the ground all around the building at the corners and wall middles, tied in with short lengths of #2 wire. It's still not perfect, but it's good enough to limit radio reception in here!

What is weird is being right under a lightning bolt. Things actually click and jump around from the enormous one turn transformer's magnetic field. But no boom right away -- the first you hear it is when it bounces off the surrounding hills, it's quite eerie to experience in person. And cool once you know it's not hurting anything.
Why guess when you can know? Measure!
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: Good earth ground

Post by John Futter »

So far all comments have been close to the mark

but beware if you think you are running a steady state system.
If you are using the usual HV supply ie Spellman, Glassman, Bertan, ETC these all have coaxial outputs ie the output cable is usually RG8 or a derivative (RG213, RG214) these cable are capacitors at around 33pf per foot and then there is the capacity of your experiment (feed throughs, grids, whatever) So in effect you are driving a capacitor of around 400 -2000pF.
now you have an arc and its over in a few nS--the dV/dt and dI/dt during this event are in the many many tens, or hundreds of volts per microsecond and tens of amps per microsecond. These levels are way outside of what normal semiconductors can take due to internal inductance and capacitance --especially modern logic circuits with very low circuit to substrate breakdown figures.

As stated inductance in the earth circuit is your enemy and this is directly proportional to width vs length of the earth conductors ie wide and short is good.
Lets look at this from the RF point of view.
A quarter wave transforms zero impedance to infinity ie voltage to current source

so lets put some numbers to this --100Mhz, period 10nS quarter wave length 750mm. So if your breakdown gets itself over in 10nS then any earth wire approaching 750mm or longer is a liability transforming the current of the discharge into a voltage spike. This voltage will rise to infinity unless something breaks down --hopefully not your turbo controller being the weakest link.

This where the Transils (transzorbs), gas discharge tubes, MOV's become your friends and also clamp around ferrite beads over the outer of the coax to your HV supply.

The ferrite cores decouple currents running on the braid of the coax ie stop them and you can redraw the circuit at this point for the rf portion into an isolation transformer inserted in the middle of the coax. This transformer is present until its core saturates where it magically disappears due to core saturation -so these cores (for EMI) tend to have high core loss and relatively low uR.

Lesson
any thin wire over a foot in length is tending to be an inductor raising impedance between connected nodes-- all calculable.
clipon ferrite EMI absorbers work as isolation transformers---- work very well on long coax supply leads from HV supply with one near each end.

Real mother earth has very little to do with local earth due to distance away (inductance) but for safety is a very good reference to steady state ---because of this impedance, only one connection to it should be made to avoid large potential differences between these points under arc conditions.
here endth the lesson
Post Reply

Return to “Fusor Construction & Operation (& FAQs)”