Issues with adding high-resistance system ground 2

[peebee]

We have been asked to review the possibility of adding high-resistance grounding to several 480-volt systems which were purchased, but not yet installed. Several potential issues come to mind, including:
- breaker ratings
- breaker ground current sensing
- TVSS ratings
- UPS ratings
- method of ground fault relay/alarming (I vs V sensing)
- VFD ratings
- soft starter ratings
- magnetic starter ratings
- transformer ratings

Any comments on this list? Any critical items I've overlooked?

Anything on my list above that should be taken off (transformers seem like they should *probably* be OK, I'm not quite ready to guarantee it, though -- and I have no idea if most of the solid state stuff like VFD's and UPS's would be OK or not....).

Any thoughts at all would be appreciated.

 

[dpc]

Many of our clients have converted a lot of 480V systems to High-Res grounding with no significant problems that I am aware of.

The only possible breaker issue is the single-pole short circuit test for MCCBs. But the only time this comes into play is if there are two separate line-to-ground faults at the same time, on different phases with one of the faults being on the supply side of the circuit breaker. Under these unlikely conditions, the breaker's single-pole interrupting rating must be used and it is usually lower than the three-phase rating. Most people consider this scenario so unlikely that they don't worry about it, not to mention the fact that a 480V ground fault is likely to be an arcing fault with a fair amount of arc-resistance.

There are no issues with transformer ratings, starter ratings, or any other ratings that I am aware of. The grounding system should provide an alarm when a ground fault occurs. You then must trace the fault using injection of a non-60 Hz waveform that you can track back until it goes away. This is similar to tracing ground faults on an ungrounded system. It is imperative that ground faults be tracked down quickly.

Any existing ground fault relaying based on current will mostly not be capable of detecting faults. However, it can be left in place - there is no need to remove existing relays or replace existing trip units.

Of course, you can not serve any 277 V loads with High-res grounded system, but I'm sure you already know that.

High-resistance grounded system are safer than solidly-grounded (or ungrounded) provided you have trained electricians to respond to, track down, and eliminate ground faults when they occur. If that isn't going to happen, you should stick with the solidly grounded systems - the brute force approach still works well.

 

[DickDV]

dpc's last paragraph is the clincher. Simply put, given the maintenance behavior of most companies, simply DO NOT DO HIGH RESISTANCE GROUNDING. It is exceedingly hazardous because it allows operation with one three phase leg grounded.

This forces the other two legs to full voltage above ground, not what your system or much of your equipment is designed to see. VFD's in particular create their own virtual ground internally in order to do ground sensing properly on their motor leads. This conflicts with a corner-grounded supply even for a few minutes and can result in damage to the VFD.

If you have DC drives, you will have to put isolation transformers ahead of each one to avoid having the whole armature circuit float 460VAC high off ground in the event of a phase ground.

A high resistance or floating power supply is an invitation to some of the ugliest common mode noise issues I have seen in my career. Noise coupled into the ground system depends on the supply grounding system for a return path. Lacking that low resistance circuit, the noise will look for another convenient path---maybe the shield on that sensitive instrumentation, maybe the ground on that critical computer power supply, maybe somewhere else equally undesireable.

Why would anyone want to run with one phase grounded anyway? Do the responsible thing and clear the grounds as soon as they occur.

There is a good case to be made for low resistance grounding. That is, enough resistance in the system grounding bond to limit ground fault current to limit collateral damage when a ground occurs but still clearing the current limiting device, either fuse or CB. This is, in my view, good practice and will not lead to the abuses and equipment problems listed above.

But high resistance grounding or no grounding at all? No! I almost lost my life to a floating supply with one lead in a mud puddle (it had been there for weeks, we later discovered) and I wouldn't wish the shock I got on my worst enemy.

 

[peebee]

How about the solid-state stuff? Any issue there?

 

[dpc]

peebee,

I have heard some concerns regarding some VFDs and high-resistance grounding, but I have worked with clients who have hundreds in service and have never personally heard of a problem with VFDs that could be (logically) blamed on the high-res grounding. But there may be issues that I'm not aware of - maybe others can enlighten us both. Here's a link to a paper by Gary Skibinski that seems to imply that there are actually benefits to high-res grounding for VFDs:

http://literature.rockwellautomation.com/idc/groups/literature/documents/wp/drives-wp019_-en-p.pdf

There have been a lot of IEEE papers on high-res grounding of 480V systems and you should be able obtain more data than you can possibly read. If there is an inherent problem, it should be well-documented because there are a lot of high-res systems out there.

I respectfully disagree with DickDVs assessment on the high danger of high-resistance grounding at 480V. Use of high-res grounding greatly reduces the risk of arc flash injuries because a ground fault will produce no arc. Since over 70% of fault start as ground faults, this is a substantial benefit.

Solidly-grounded 480 V systems have their own set of hazards, including arc-flash risk, fire risk, and equipment damage risks.

If the high-res system is properly maintained, it is much safer than a solidly-grounded system, in my opinion. Proper maintenance is not difficult, it just requires an understanding of the need to keep on top of any ground faults.

Also - Dick mentioned low-resistance grounding - this is strictly limited to medium-voltage systems and should never be done on a 480V system - it would be hazardous due to the high risk of sustained arcing faults.

 

[DickDV]

Low resistance grounding is, in fact, done on 480V systems and is very desireable from a flash and damage reduction standpoint.

There is no risk of sustained arcing since the ground fault current is still high enough to release the main breakers.

Post-Glover makes a whole series of resistors specifically designed for low resistance grounding at 480V. There must be a market for it somewhere!

Any VFD with a CE label on it is vulnerable to corner-grounding damage. The noise filter is designed with line to ground balance and will not sustain corner-grounding for more than a few minutes without leaking out a little bit of that all-important smoke.

Finally, any electrical safety system that depends upon prompt maintenance and trouble-shooting for its integrity is fatally flawed. And that is not intended to be a play on words!

 

[davidbeach]

DickDV, maybe we'll never agree, but there are many high resistance grounded 480V systems that operate safely because proper maintenance is available. In many cases they can operate high a higher level of plant safety than a low or solidly grounded system could because the process is not subjected to the chaos that follows a breaker tripping off one portion of the system in response to a ground fault. You are correct that lack of proper maintenance can be a killer, but proper maintenance does exist.

Where that level of maintenance doesn't exist, one can still get many of the benefits by running a solidly grounded 480V system as a 3 wire (plus ground) system. In large systems the neutral conductor (cables and bus bars) adds a tremendous cost. Three phase loads can be served from the 3 wire distribution without issue, and any 277V lighting loads would be served from the secondary of 480 delta - 480Y/277V transformers. The lack a neutral throughout the system, the lack of a ground fault path from the lighting circuits to the service, and the greatly reduced fault current at the lighting panels can make this type of system a much less expensive (even after buying the necessary 480Y/277V lighting transformers) than the conventional system, and the circuits most likely to have ground faults don't have a zero sequence path back to the service. Other portions of the system could be similarly subdivided.

 

[DickDV]

Being a motor control specialist, I am admittedly a bit out of my element discussing power distribution systems. My comments are simply based on twenty years of observation in the field and many difficult and dangerous experiences dealing with those systems.

My observations lead me to the following conclusions:

As mentioned above, no-neutral, floating or high-resistance systems are definitely less expensive to install. Why this is a good argument to compromise safety is a puzzle to me.

There are many of these floating systems around and some are maintained properly, without doubt. Discerning which are and which are not is terribly important from a life-safety standpoint and is most difficult to do when working in a facility you are not intimately familiar with. The familiar "Oh, yeah, we really watch closely for grounds" statement from the maintenance manager carries no weight with me at all having found thru experience that talk is cheap.

In my territory are a wide variety of chemical, fine chemical, and pharmaceutical plants. Without exception, the plant engineering people reject floating systems and insist on grounded wye sources with neutral. I attribute this to a heightened awareness of safety and the willingness to spend the necessary money to assure it.

Finally, as mentioned before, I am regularly replacing DC drives, DC motors, and AC drives that are damaged by corner-grounding and that experience runs across three different brand names. I even ran into one instance where there was so much leakage current from floating phases to ground that when small (<3hp 460V) AC drives were installed, the drive supply fuses would blow with the drives only idling,--not running! The leakage currents exceeded the fuse ratings and were finding their way back to the source thru the drive virtual ground network. We had to install a drive isolation transformer with primary fusing at 30amps. Now, the drive fusing holds but, occasionally, on high humidity days, etc. etc., the 30 amp transformer fuses blow from the leakage currents.

I hope you can see why I am not impressed by the claims that floating networks are "cheaper"!

 

[dpc]

Dick,

I'm not sure high-resistance grounding is much cheaper than solid-grounding - equipment grounding conductors are still required. And it's not really "floating".

Also, I'd be interested in where you've encountered low-resistance grounding in a 480V system and what level of resistance was used. AFAIK, low-resistance grounding at 480V is not really permitted by the NEC. I'm not a grounding expert, but everything I've read indicates low-resistance grounding at 480V is a not a good idea.

I do agree that maintenance folks pay more lip service to clearing grounds than actually doing the work at times. But I've seen the havoc than can occur from a 480V ground fault on a solidly-grounded system. But at least they don't have much trouble locating the fault - and they don't have much choice either

 

[marbr2005]

This discussion about low voltage systems be solid grounded or not comes from the 60's when I was young and my prefered book was "Electric Power Systems" by Donald Beeman.
I have experienced 460V system (Brazil's no-load system voltage for industrial aplications)resistance grounded (max.ground-fault current around 3,0 Amps) during many years without problems.
1.Current detection via CT and relay at neutral circuit.Alarm is given.
2.Fault finding mode selection starts a pulsing output that shorts part of grounding resistor in order to create a variable fault current that can be be detected using a portable clamp ammeter that identifies the faulty feeder/faulty circuit.
3.To prevent the risks of second ground fault we use ground protection relay (tripping corresponding CB) at each feeder set around 6,0 to 8,0 Amps.
(There is a remark: Second ground fault on other phase at DIFFERENT circuit.See IEEE paper about)
4.We do not mix at the same MCC, inverters/ converters and AC motors starters/others feeders. Basically the resistance grounded system is applied for ordinary loads.
5.Maintenance crew works so fast as they can to clear fault as soon it appears.
Sorry my bad english.

 

[waross]

I think that the issue and the responses depend heavily on the issue of "Will the customer recognize the importance of finding and removing grounds immediately!"
This is the heart of the issue.
In my experience, only about 10% or less of the plants that I have dealt with could be trusted to find and repair ground faults immediately.
I would not use high resistance grounding for any plant that lacked a well trained electrical crew.
I think that the candidates for a high resistance grounding scheme will be large installations where there is a significant saving to be made in cost of a neutral conductor as suggested by davidbeach.
I would be wary of installing high resistance in plants that are heavily influenced by MBA type "Bottom line today" thinking. Even good electrical crews may be frustrated by management interference, as for instance when a production line must be shut down to effect repairs and management will not allow the loss of production.
respectfully

 

[peebee]

Actually, the issue (as I intended it when I started this thread) has nothing to do with the merits, safety, or maintenance of resistance grounded systems. They are all valid concerns, but have nothing to do with the question posted at the top of this thread.

The question is: if the system was originally designed for solid grounding, and we convert it to resistance-grounded, what equipment might not be rated for such operation?

So far above, the only equipment that has been identified as possibly being a problem is VFD's. In addition to that, I'm positive that 480Y/277-volt breakers are not suitable for operation on a resistance grounded system (and would be a code violation). In addition, any Y-connected TVSS equipment would not be rated for a continuous line-ground voltage of 480-volts, and would also require replacement.

To get back to the original question, what other equipment would be affected by this conversion to resistance grounding?

 

[marbr2005]

Maybe these points can be useful:
Breaker(load center feeders) ground current sensing:To apply a core balance CT and a ground relay (50GS) set a little bit higher than the current limited by the grounding resistor. This prevents the risks of a second ground fault on other phase at different circuit.
I x V sensing: If you have inverters, a phase to ground fault at DC bus will cause a DC component circulation.
At AC output the effect will depend of switching frequency. In such a case, conventional magnetic CT is not the best choice.
Transformers ratings, magnetic starters ratings and breakers ratings seem to be OK.
Arresters, if any, should be 100% type.

 

[RRaghunath]

High resistance grounding with pulsing apparatus to detect the faulty feeder is an ideal solution when we are looking for reliable LV system but no problems with detection / planned isolation of the faulty feeder. I understand this is widely practiced in US.

 

[davidbeach]

If I may be allowed to toot my own horn just a bit, there is a paper I wrote that looks at using negative sequence relaying to detect phase-to-phase faults via ground in ungrounded/high impedance grounded systems. The paper is available at

http://www.basler.com/php/download.php?file=NegSeq.pdf&id=262&type=res.

 

[dpc]

peebee,

If by "480/277-volt breaker" you mean single-pole 277V breakers, that concern is a moot point because you cannot have any 277 V line-to-neutral loads on a high-res grounded system.

If you are referring to the single-pole interrupting rating, this is generally not a big concern for the reasons I tried to explain in my previous post.

The single-pole rating issue only applies to molded case breakers. UL specs for molded case breakers allow a lower single-pole interrupting rating. ANSI requirements for Power Circuit Breakers are more stringent.

I would encourage you to talk with a breaker manufacturer about breaker application with high-resistance grounding. But I can tell you that as a practical matter, no one worries about it.

I'm not aware of any other signficant issues.

 

[ihipress]

I am trying to decide if a high resistance ground is the appropriate solution to a problem I am having with an installation.

The SCR's in 4 quadrant DC drives are being damaged during power outages. Semiconductor fuses on the incoming line side also clear.

I am evaluating the semiconductor fuses to be sure they are the correct fuse.

Inspection of the DC motors revealed arcing to ground around the commutator, and it has been suggested that a high resistance ground be installed on the neutral of secondary of the isolation transformers.

Maintenance in the plant is an issue and I don't want to install a high resistance ground unless it is absolutely necessary.

Since the SCR damage only occurs during power outages, I speculate the problem is related to loosing the line when the drives are regenerating.

If the SCR damage is being caused by ground faults, then I think the damage would be random instead of only occurring during power outages.

Until I can make a connection between power outages and the arcing to ground I saw in the motors, I am not totally convinced that adding a high resistance ground will resolve the problem of SCR damage.

 

[moira]

All: Having worked many years involved with both systems, it is a tought call! What are your customer's biggest electrical safety concerns? I'm inclined to give the edge to high resistance grounding. You want to have, let's say , the greatest amount amount of arc-flash protection in organic solvent areas for one. A leg in or to a pump motor can ground out. You can temporally keep running and not suffer from an immediate blast! Having said this; you must have a well trained and fast responding maintance team. In a 40 year work career, I've never seen a ground from two different phases at the same time! Mostly all devices and loads operate the same way. You can have the advantages of, say, 277 VAC lighting systems thru the use of transformers. Keep in mind the added cost of alarm and thumper systems. The age and structual integraty of your systems is of great concern and understanding. You don't want to have grounds in underground feeders every time it rains!

 

[powerjunx]

moira,
High resistance grounding is simply used to limit fault currents. It's usually connected to midpoint of "Y". While, 277V (L-G) is also connected to midpoint of "Y" to cater lighting power supply.