neutral ground(ing) resistor

[SphincterBoy]

Does anyone out there use neutral grounding resistors on the neutral(s) of the distribution substation transformer WYE secondary windings?

I have a situation involving a distribution substation with high ground fault current, in which the electronic feeder protection module (I'll use the acronym POS to describe it) is tripping and going to lockout after one pickup (shot) of a high-magnitude ground fault (undocumented anomoly). I don't want to change the protection module, but I'm exploring using a neutral resistor to lower the fault current to the point where the module will not immediately go to lockout, but reclose (as intended). We have other situations similar to above, but the protection modules will trip/reclose several times trying to free the dead squirrel off the insulator before going to lockout.

I have read that utilities will install neutral resistors to limit ground fault current, and lower nuisance outages.


What is been your experience?

 

[davidbeach]

I would think that adjusting the settings of your relay would be a whole lot easier and less expensive than adding a grounding resistor. Proper relays settings need to be determined for each feeder in each substation, you can't expect to use the same setting everywhere without expecting trouble.

 

[dpc]

If you are talking about an overhead utility distribution circuit, I don't recommend resistance grounding. Generally the problem is not having enough ground fault current to relay when an OH line falls on the ground. A resistor just makes that problem worse.

Also, you won't be able to serve line-to-neutral loads if you add much resistance.

Neutral ground resistance is used for industrial facilities to limit damage from ground faults.

I've never seen it used on a distribution feeder.

 

[jghrist]

Look through the relay manual to see if there is something like a High Current Lockout. On Cooper recloser Form 6 controls, this is a separate setting to prevent reclosing for close-in faults.

For other relays, this may be programmed through logic with a separate instantaneous element.

 

[SphincterBoy]

Many thanks for the great comments everyone!

To DavidBeach: We did an extensive bus/fault study years ago, and have gone over the settings of these devices again and again, with no setting to describe the high GF trip/lockout problem. The manufacturer (Cooper) have said this is an internal function which cannot be shutoff, and was designed to go to lockout after one(1) exceptionally high GF. The module is the IM30 series which Cooper introduced some years ago. Very European, but w/many problems. JL Blackburn advised against neutral resistances as well, citing difficulty in detecting GF far away from the substation. I am going to advise installing shield wires over the distribution lines to stop the lighting flashovers.

To DPC: You are correct. As I said above, JL Blackburn also said "NO" to neutral resistances. And under normal loading conditions, such resistances will be "cooking" 24/7 due to single phase loads.

To JGChrist: We have scoured the manual of the IM30 modules, and nothing was mentioned about the undocumented function. We only found out through a support person at Cooper, who admitted this was a problem, and offered no workaround. FYI: We also use the Form 4C controls, which we like, but are wondering as to their accuracy, which Cooper said is lacking in the Form 4C. We like the Form 6, but there again, what about the accuracy (?), which is extremely important.

 

[dpc]

As far as the Form 6 vs Form 4C - about the only similarity is the word "Form" at the beginning. The Form 6 is a completely new product - it's the same thing as the Cooper Idea relay - just repackaged a little. We have used them on a number of distribution feeder applications - no problems yet.

I'm not familiar with the IM30, but I've had problems with some other "Cooper" relays that were made for them in Italy.

As for accuracy - for very high current faults, the CTs may be more of the limitation rather than the relays.

 

[jghrist]

It's always good to eliminate the faults where you can, but wouldn't a more cost-effective solution be to replace the relay?

Shield wires on distribution lines are not all that effective. A lot of distribution flashovers are from induced voltage from nearby strokes which wouldn't be affected by the shield wire. With typical distribution line BIL and grounding, backflashes to the conductor from strokes to the shield wire will also cause a lot of trips. Generally, arresters will provide better protection, but then there are problems with arrester failures.

You could also install phase current-limiting reactors. This may cause voltage regulation problems and would surely be more expensive than changing the relays.

 

[SphincterBoy]

To jghrist: We will ultimately replace all these modules, which are currently mounted in a metalclad switchgear lineup, which is only 6 years old.

Needless to say, we got burned, and we will do almost anything else to avoid changing these modules. Changing them involves cutting a bigger hole in the plate steel face to accomodate the replacement..........not an easy task considering the number of cubicles.

The IM30 relays were made in Italy, which at the time were the only small, low-cost, integrated relay on the market, and they were very appealing. Oddly enough, Cooper still sells them.

We think we know where the lightning flashovers are occuring, and installing shield protection on an otherwise unshielded line may be worth a try. We have installed shield wires on other feeders, with great success.

 

[stevenal]

More documentation is needed on this undocumented feature. Does it operate on a set level of secondary current, or a set multiple of a pickup value? Maybe CT ratios or relay settings could be adjusted to affect the level of primary current needed to go to lockout. Or maybe Cooper can suggest a surgical method of disabling it, or upgrade the firmware. You have reclosing relays that by design don't reclose? Sounds to me like the problem is Cooper's. I'd be leaning harder on them for a fix.

 

[dpc]

Cooper might give you a really good price on some new Form 6 controls if you make enough noise...

 

[SphincterBoy]

Good point, stevanal and dpc. I haven't checked if the Italian firm which makes the modules for Cooper have made improved versions.

I'll see if newer, revamped versions are available. The question is: Will Cooper charge a fee for replacements?

I have d/l'd the latest documentation for the modules, and nothing has changed in six years.

When we first installed these six years ago, we received a defective unit, and Cooper sent out a replacement unit right away. We noticed the output relays had been changed in the replacement unit, which brings up another bad feature of these particular units: Output contact too small (ampere rating).

We had to install interposing relays between the output contacts and the circuit breaker trip mechanisms, since the output relay contact ratings were below the circuit breaker tripping requirement.........big mess. Now you see why I call these P.O.S.

 

[cflatters]

In the UK, DNO's do commonly use NER's to limit earth fault current on their distribution network to values which do not give rise to excessive fault current.

The NER's also assist in the limitation of substation EPR.

We also commonly encounter NER's on clients 11,000V systems limiting fault current to 800A or 1000A

 

[SphincterBoy]

cflatters: Thanks for the comment. However, I'm afraid I'm at a loss as to what DNO and EPR are acronyms for. Can you please elaborate? Thanks!

 

[stevenal]

Earth is UK speak for ground. Neutral earthing resistor and earth potential rise. Already shot down above for American style systems serving single phase loads line to neutral. Don't know about DNO. Distribution network operator?

 

[cflatters]

DNO = Distribution Network Operator
EPR = Earth Potential Rise

 

[stevenal]

Woo hoo, nailed it.

 

[dewone]

Not familar with the type relay do you know if it is reading actual neutal current returning to the tranformer Y or is it a zero sequence. Where an unbalanced load could be a part of the problem.

 

[SphincterBoy]

dewone: the Cooper IM30DRE overcurrent protection module does not measure neutral current via a neutral CT, but calculates it based on phase imbalance (zero sequence).

The IM30DRE, being a digital computer, has all the trimmings; CPU, A/D, D/A, relay contact outputs, display, etc. I don't think it's internal architecture is all that whippy, but at the time (2000) it was, to my knowledge, one of a handful of small and integrated feeder protection modules on the market.
Now, Siemens and Areva make some really small, nice, and sophisticated modules.

ABB used to (I haven't checked recently) sell a fully-integrated preotection module which allowed for the connection of a neutral CT. I wonder what affect on performance a separate neutral CT has on GF detection vs. virtual values as in the case of the IM30DRE.

 

[dewone]

The 4 wire neutral is perfered for accuracy. Zero Sequence is based on Pick-up or timing the imbalance of the load at a given set point and then timing that out till trip. It is based on the assumption that the imbalance is going to ground. That said it is not well equiped for loads that aren't balanced. I think they derived from motor protection and migrated into power as a simple GF adder. 4 wire systems can take the imbalance because they measure the actual current returning to the transformer wye. Much easier to manage the known than the assumed unknown. They really should call the Zero sequence units (Phase unbalance).
Now to get your problem defeated. Short the GF CT out at the CT terminals. That will elliminate the circuit all togather.
Unless your required to have the GF at that location. In which case perhaps get a more flexible GF unit external to the Cooper unit and fire it's trip circuit via shunt trip.