Ground to Earth

For general topics and questions that do not fit into any of the other categories or forums.
Post Reply
sonofmitch
1/2 Star Member
1/2 Star Member
Posts: 7
Joined: Thu Dec 01, 2011 11:57 pm

Ground to Earth

Post by sonofmitch »

What's best to ground my table with bare solid copper wire or stranded?
Mitch
User avatar
see&see
2.5 Star Elite Contributing Member
2.5 Star Elite Contributing Member
Posts: 218
Joined: Tue Jun 24, 2014 12:04 pm
Location: Minden, Nevada
Contact:

Re: Ground to Earth

Post by see&see »

The solid wire will usually stay tighter in a connection considering temperature and flexing over using a screw type connection with stranded wire.

A ground connection to earth is just wishful thinking unless there is enough moisture at the base of the rod to conduct a fault discharge. The deeper the rod extends in earth the better.

I've also as a backup connected my table frame to the supply panel ground. The ground/neutral circuit is normally bonded in the main panel to the transformer neutral which is always a good backup ground connection.
You marry into the PlasmaCam family and must accept the fact your software and hardware are proprietary. It's a for better or for worse engagement with overwhelming security. PlasmaCam controls the computer, table, hardware to their advantage IMO. :x
User avatar
SegoMan DeSigns
3.5 Star Elite Contributing Member
3.5 Star Elite Contributing Member
Posts: 822
Joined: Sat Jan 06, 2018 4:45 pm

Re: Ground to Earth

Post by SegoMan DeSigns »

see&see wrote: Thu Mar 14, 2019 4:03 pm A ground connection to earth is just wishful thinking unless there is enough moisture at the base of the rod to conduct a fault discharge. The deeper the rod extends in earth the better.
Stand bare footed on the dry dirt, touch a live wire and get back with us on the results..
User avatar
see&see
2.5 Star Elite Contributing Member
2.5 Star Elite Contributing Member
Posts: 218
Joined: Tue Jun 24, 2014 12:04 pm
Location: Minden, Nevada
Contact:

Re: Ground to Earth

Post by see&see »

SegoMan DeSigns wrote: Thu Mar 14, 2019 4:33 pm
see&see wrote: Thu Mar 14, 2019 4:03 pm A ground connection to earth is just wishful thinking unless there is enough moisture at the base of the rod to conduct a fault discharge. The deeper the rod extends in earth the better.
Stand bare footed on the dry dirt, touch a live wire and get back with us on the results..
What do you mean by live wire? Voltage? Amperage?

A current as little as 100 to 200 mA (0.1 to 0.2 amp) is lethal.

It's totally clear you don't have a clue what you are talking about... :roll:
You marry into the PlasmaCam family and must accept the fact your software and hardware are proprietary. It's a for better or for worse engagement with overwhelming security. PlasmaCam controls the computer, table, hardware to their advantage IMO. :x
User avatar
SegoMan DeSigns
3.5 Star Elite Contributing Member
3.5 Star Elite Contributing Member
Posts: 822
Joined: Sat Jan 06, 2018 4:45 pm

Re: Ground to Earth

Post by SegoMan DeSigns »

Yes and it takes very little amperage to insert noise into the system.
User avatar
see&see
2.5 Star Elite Contributing Member
2.5 Star Elite Contributing Member
Posts: 218
Joined: Tue Jun 24, 2014 12:04 pm
Location: Minden, Nevada
Contact:

Re: Ground to Earth

Post by see&see »

SegoMan DeSigns wrote: Thu Mar 14, 2019 11:12 pm Yes and it takes very little amperage to insert noise into the system.
Yep, that was my point, a ground to earth is useless especially when considering plasma table RMI suppression unless it's deep enough in moisture laden earth. Just like standing on a wet floor using a bad electrical tool without GFCI protection. As little as a 30 volt leak in a tool can kill you..

Many AHJ's in the USA make us put special material around earthing rods when the soil is mostly clay that dries out, shrinks and breaks contact with the rod. There's a lot of good info on the net on this subject..
You marry into the PlasmaCam family and must accept the fact your software and hardware are proprietary. It's a for better or for worse engagement with overwhelming security. PlasmaCam controls the computer, table, hardware to their advantage IMO. :x
tcaudle
4 Star Elite Contributing Member
4 Star Elite Contributing Member
Posts: 1353
Joined: Tue Feb 28, 2012 6:47 pm

Re: Ground to Earth

Post by tcaudle »

Grounding for safety (60 hz) and for noise are two different animals. The workclamp of a plasma is not ground . Its a positive voltage in relation to the torch electrode (like a car battery setting on your work bench) The workcalmp signal is pretty noisy being as how its a lot of PWM noise and harmonics .

While a few mils of current across you heart will kill you it also takes enough voltage to break the skin resistance barrier for the current to flow.

In order for noise to adversely effect a control system it has to get into it in a large enough level to overwhelm the internal signals and that cna take a lot of power since transmission though the air is poor. Most noise gets inot a system though a connection (conducted noise) . Best conductor between two systems is a common ground circuit. Break the common grounds and make the connection a higher impedance (give a longer, less direct path or give it a lower impedance path to ground) and you "decouple" the noise between systems. RFI is another story. The higher the frequency the better it travels in the air (short wavelength but that is for the techies) RFI is easier to block with simple shielding and metal barriers. Low frequency is easy to stop with just air and distance as long as there is no easy (low impedence ) path between the noise emitter and the control.

There are some things that will concentrate the noise and make it more directional... A coil of wire will became a focused antenna (out 90 degs from the plane of the coil) so do not coil up extra wire on higher current conductors and lay them close to your controller or the plasma itself.

So table grounding is for noise While not all ground is a good shunt for normal AC it can be for the lower current mid frequency plasma noise.
Grounding for noise needs to be lots of small conductors onm a heavy wire. Why? Its called "skin effect" and it works from the fact the higher frequency waveforms travel on the outside skin of the conductor so the more surface area there is , the better the conduction .

Since the voltage on a torch wants to flow back to to the source through a connected RETURN (common) very little primary energy will flow to some non realted or connected system like the AC ground. Its always seeking a return path to the source. If you hook a 12v buld up to one leg of the battery on the bench and touch the other side of the lamp to an external object or ground the light will not light up. If you ground either side of the battery external and then touch THAT object it lights. There can be no current flow unless there is a path for the current to return.

EMI and RFI gets complex and there are entire textbooks on how to measure and control them. What works in one place may not work in another.
User avatar
SegoMan DeSigns
3.5 Star Elite Contributing Member
3.5 Star Elite Contributing Member
Posts: 822
Joined: Sat Jan 06, 2018 4:45 pm

Re: Ground to Earth

Post by SegoMan DeSigns »

Thank you for taking the time to explain that to the masses, there has been multiple web postings of machines cured form the dreaded noise with a ground rod and well thought wiring plan. Jim Colt has a similar write-up that he posts.
User avatar
see&see
2.5 Star Elite Contributing Member
2.5 Star Elite Contributing Member
Posts: 218
Joined: Tue Jun 24, 2014 12:04 pm
Location: Minden, Nevada
Contact:

Re: Ground to Earth

Post by see&see »

Another great analogy Tom, thanks, I like reading your stuff.. Yes, EMI is complex and sometimes akin to black magic..

Not related to EMI but I brought up a point over on the PlasmaCam forum regarding a flaw you might already know about.. If you don't want to comment on a competitors product I understand perfectly.

My concern was that the table controller will trip a GFCI when used in the electrical supply to the table. Ok, this indicates a fault in the controller box or a design flaw in the circuits IMO.. There seems to be an engineering principal that hasn't been followed, again, IMO. Additionally, their video manual instructs us specifically to NOT use a GFCI in the supply circuit..

The problem arises, as a licensed contractor I cannot legally eliminate the GFCI circuit on an industrial shop floor.. Technically both of my tables are illegally connected to the power source..
You marry into the PlasmaCam family and must accept the fact your software and hardware are proprietary. It's a for better or for worse engagement with overwhelming security. PlasmaCam controls the computer, table, hardware to their advantage IMO. :x
tcaudle
4 Star Elite Contributing Member
4 Star Elite Contributing Member
Posts: 1353
Joined: Tue Feb 28, 2012 6:47 pm

Re: Ground to Earth

Post by tcaudle »

If you know how GFIC works you might understand why its hard to get industrial equipment to not fault. The GFI works by sensing a difference in the safety ground and the hot and a phantom ground (i.e. a person grounded . An alternate ground path. it trips on just a few mils of phantom current. The probelm with current is it is subject to phase shift into different kinds of loads and also huge "inrush currents " into a typical switching type supply or inverter circuit. Some power supplies with big filter caps have non-linear current waveforms (is called Power Factor) a perfect Power Factor (PF) is 1.0 . A .5 PF is bad . Most inverter type power supplies and systems have bad PF numbers. Some cheaper GFIC sensors will trip if the PF is low or the inrush current is large. To mitigate the high in rush (which is easily 10 X the running current) you can do a "softstart" (electronically or via a bank of resistors ) sometimes the softstart itself can cause a GFIC to trip.

There are safety specs about leakage of a power system. Its measured by putting up to 5,ooo volts on the safety ground of the unit or the chassis ground of the system using a device called a HIPOT . We have three of them. It does not deal with PF just leakage for safety. So a unit can be safe but still trip a GFIC because a lot are made for linear loads .

The electrical codes and the safety rules were written when CNC machines were huge and ran on 480 3P and everything was lethal. Shorts cause explosions and sparks so the rules were written to make those systems safe with wire type, wide component spacing , lockouts , sealed enclosures and coded wiring.
Suddenly you have PC's and wireless devices and microprocessors all crammed into small spaces so they developed new categories but never changed the industrial specs or the CNC machine specs for smaller machines.

A lot of larger inverter type supplies (like plasma cutters ) have built in Power Factor Correction (PFC) and reflect a more linear load back to the power lines .

I do not think a control that trips a GFIC is defective or dangerous its simply not the type load that should be on a sensitive circuit designed for appliances and resistive loads. Sometimes all it takes is a different kind of GFIC but going though a bunch of circtuits can get expensive
User avatar
see&see
2.5 Star Elite Contributing Member
2.5 Star Elite Contributing Member
Posts: 218
Joined: Tue Jun 24, 2014 12:04 pm
Location: Minden, Nevada
Contact:

Re: Ground to Earth

Post by see&see »

Good concise writeup on the setbacks of industrial GFI with high cost and older industrial safety specs being the main setback.. Not too many techies even mention power factor or in rush current and if a facility doesn't have an in-house person like yourself it never gets addressed..

The PlasmaCam controller appears to be especially fragile to any stray outside current judging from the number of warranty claims and replacement discussions over on their forum. 99% of the PC controller issues are probably operator and install related with the lowest impedance path for stray current being through the controller and it's supply panel grounding circuit IMO... That could dovetail into the GFI trip, but again, just my guess as I've had no trouble with either of my two controllers.

Speaking of 480 3ph I purchased an industrial shop a couple years ago, the price was right and built in 2000 with a nice high dollar 480 service and two professionally installed 480 panels inside. From that point the workmanship went downhill when the owner hired some local sparky to finish the electrical for the offices and the shop inside/outside 120 receptacles. The owner got in trouble for not having GFI's in the work area and tried to patch them in using the existing shared neutrals (I discovered later) with not much success at all of course..

They had high watt metal halide high bay lights wired at 120 volts with relays in the ceiling required because of the long run.. The step down transformers were the ancient oil filled type which additionally served as heaters in the winter. I replaced those with one 75KVA 3ph dry type transformer.

What a mess! I had my work cut out for me. I went with 277 volt LED's on the high bay lights with single pole switching. Installed 20' high bay 3ph fan and replaced most of the wiring inside. I stayed with 480 on the compressor, swamp cooler, forklift charger and the Hypertherm units. Actually, it turned out to be a fun project being as it was our own building.
You marry into the PlasmaCam family and must accept the fact your software and hardware are proprietary. It's a for better or for worse engagement with overwhelming security. PlasmaCam controls the computer, table, hardware to their advantage IMO. :x
Post Reply

Return to “CNC Plasma Cutters General Forum”