Isolated Grounds 6

[Chuk1t]

I know that isolated grounds are used to reduce common mode noise;

Who originated the concept. I know that at one time Wang PC's/ terminals were required to be on circuits utilizing isolated grounds, but is the concept still valid for today's computer/data systems and is it worth the extra effort and cost's???

 

[TWW]

I personally do not think that the isolated grounding is that important today (if ever). Since the isolated grounding conductor is ran in the same conduit as the rest of the conductors, common-mode noise will still be induced upon it, Plus the inductance of the single wire (isolated ground) is too high (ie open circuit for the mega-Hz) it is useless in shorting high frequency impulse noise occuring about the ITE. This problem is addressed using Signal reference structures and multiple low length/inductance ribbon bonding to the Srs.

Most imporant is to have all the grounds connected (togeter) per the nec with single bond at the service entrance (or at a padmount transformer) that supplies the building (better, no objectionable current at all.

However, the modular furnature manufactures still incorporate a dedicated circuit w/isolated ground in their harness's, So your customers may want it use it anyway.

 

[busbar]

Two references may be:

IEEE Std 1100-1999 "Emerald Book"

http://shop.ieee.org

“FIPS PUB 94” (withdrawn but interesting and somewhat applicable)

http://www.itl.nist.gov/fipspubs/fip94.pdf

 

[MacLin]

It all depends on the critical nature of the electronic equipment and the location (what other equipment is in the area)that it will be put into service. Not having an isolate ground is OK as long as no other equipment in the area shorts out and causes fault current. The isolated ground is not for high freq noise mitigation, its for ground isolation for critical equipment. Question - Does current always take the path of least resistance?
No - It takes all paths and proportionally goes the path of least resistance. Electronic are typically powered by a 3 prong plug, and the circuitry is not a very high impedance between neutral and ground. When a high fault current occurs, without an isolated ground wire, fault current will enter your electronic thru the ground plug and damage your equipment trying to get to the neutral return. Bottom line - you'll probably be OK as long as no fault occurs in the area. Can you guarentee this? Murphy is alive and well - isn't he?

 

[dpc]

Per NEC, an "isolated" ground is still bonded to the system ground at one point.

In the old days, some control system suppliers used to insist on a "quiet" ground, with a cluster of ground rods not (directly) tied to the main electrical system ground. Since this was clearly an NEC violation, the practice died off, although there are still some die-hards left.

There was never much evidence that this isolated ground did much good. The best approach depends a lot on what you are trying to prevent. Just make sure it complies with NEC to keep your backside covered.

 

[Chuk1t]

Thanks all for your responses to my question on isolated grounding. There seems to be a wide range of opinion on how useful isolated grounding really is.

 

[atozgr]

An isolated ground is an electrical system where there is no intentional connection between the equipment grounding conductor path and the grounding electrode system.
• It fails to meet the basic requirements of the grounding system as it does not provide the effective ground path required by NEC and can result in unsafe and hazardous conditions.
Such systems should be wired as "insulated ground", which is a circuit where the equipment grounding conductor is run as a separate insulated wire back to the grounding conductor bonding point. It is not connected to other grounding points along the way.
From an article by:
Richard P. Bingham is the manager of technology and products for Dranetz-BMI.
© Copyright October 1999 Electrical Contractor Magazine

 

[peebee]

atozgr -- "isolated" and "insulated" mean about the same thing, don't they? And NEC uses them more-or-less interchangeably, see 250.146(D) and the FPN below it.

Chuk1t -- back to your original question -- my understanding is that IG helps about 50% of the time, and hurts about 50% of the time. That's why higher-end audio equipment will often have a ground isolation switch built into it, you set the switch to whatever position gives you less noise.

Also, keep in mind that most such equipment is connected to lots of other grounded equipment via grounded communications cables, which effectively introduce lots of ground loops anyway. Even if you're very careful about isolation transformers, opto-isolators, etc., you're almost guaranteed to inadvertently start introducing some ground loops if you have more than 2 or 3 pieces of equipment.

Give the extra cost, the 50/50 effectiveness rate, and the high chance of screwing up a true IG installation, I generally recommend against IG.

At the same time, if a client wants is, I don't argue with them. I'll give them the above explanation and a chance to back out exactly one time, if they still want the IG after that, well, it's their money! (Many of them like IG simply because it helps ensure that everybody plugs their computer into the right outlet at their desk, the orange one -- that's the single overriding good reason I've heard of for IG outlets!)

By the way, keep in mind that the point of IG is to reduce ground voltages, which it does by breaking all ground loops and stopping all ground current flows (during normal operation). The OTHER way to reduce ground voltages is to reduce ground resistance: V=IR, so reducing either I or R will reduce ground voltage.

The way to reduce ground resistance is to use BIG FAT ground cables, and to keep them SHORT.

The nice thing about the BIG FAT SHORT ground cable concept is that it's really hard to screw up, practically foolproof. Any idiot can understand BIG FAT SHORT. The downside is that BIG FAT copper cables are EXPENSIVE (lucky that short ones are cheaper!). BIG FAT SHORT (my name, not an industry standard) is typically what you will find where ultra-low ground voltages are really really important. IG, maybe. BIG FAT SHORT, always.

 

[ladengr]

Can I terminate isolated ground conductor from the orange recptacle to local panelboard ground bus? This ground bus is bonded to the panel enclosure. The panel is connected to the dry type transformer and its neutral is grounded and bonded as required by NEC.

 

[peebee]

Yup. Your proposed installation would be in complete conformance with 250.146(D) -- and that's a common installation where you only have one or two IG receptacles that you want to power up.

There's no code-required "IG Ground Bus". Code doesn't even refer to such a thing. Panelboard manufacturers offer them, though as they can be very handy for two reasons:

1) They provide a convenient place to land all those extra ground wires, and

2) if you keep all your normal ground wires segregated onto a different ground bus from your IG wires, then it makes it easy to test if your ground system is truely "isolated" -- just remove the IG bus jumper and check the resistance between the IG bus and the normal ground bus. Remember to put that jumper back when you're done, though!

Also, if you've subfed panels a couple times, then the IG bus is a good way to keep the IG feeder cables segregated from the other grounding cables and provides a convenient place to land the IG feeder cable.

But, there is absolutely no requirement to install an IG bus in a panel. It's just a matter of convenience.

 

[PEMBA]

Can I put mutiple IG recptacles on one circuit as long as all the receptacles have seperate IG ground conductors back to the ground bar?

 

[tictok]

I have a lot of systems running our DCS systems and computers. Design standard I use is to install an Isolation transformer with the transformer secondary Ground as my "system" reference ground. Recent harmonic and system analysis have shown this works very well and all the systems are very clean.