Arc flash on secondary of 480V Transformer 1

[Eleceng01]

I keep coming across this problem and I am hoping to find out what everyone else is doing. I have a pad-mounted 12.47kV/480V, Wye-Wye, 500kVA transformer with a C12 bay-o-net fuse on the primary.

My arc flash analysis shows that the hazard exceeds category 4 between the transromer secondary and the main breaker in the MCC.

How, if at all, is everyone handling this? Thanks.

 

[dpc]

This is very common. Note that this also includes the main breaker section.

Solutions:

Work De-energized
Investigate other/better primary fuses
Install pad-mounted VFI or Vista switch, or recloser and provide secondary relaying to trip the primary device.

The only way to lower the arc-flash level in this situation is to find a way to clear the faults more quickly.

Pad-mounted transformers with 480 V secondaries are a MAJOR arc-flash hazard.

 

[Eleceng01]

dpc,

Thanks for the quick feedback and advise.

If I understand correclty the only way you could de-energize is by opening the feeder breakers on the bus. My understanding is that you can not operate the main breaker with these high arc flash values on the main. Do I understand this correctly?

 

[alehman]

It's common for utilities to select primary fuses only to protect their system from a fault in the transformer. Commonly they will never open for an arcing fault on the secondary. This was recently experienced by a contractor I know when working in the incoming line section of a 480V switchboard. An arc was accidentally started and burned until the utility crew arrived and pulled the primary fuses. The electrician was severely burned, but thankfully he eventually recovered.

 

[Modula2]

Eleceng01, if I read this correctly, your switchgear room must be category 4 due to the characteristics of the primary device. This is a problem.

 

[dpc]

If you have a main breaker, the arc-flash level downstream of the main breaker will probably be less than upstream, since the main breaker can clear the fault.

If no main breaker, then NFPA-70E would say that none of the breakers on the bus can be locally operated while energized. You certainly do not want to open any doors or otherwise provide exposure to live parts.

If you need to work on the main breaker, it would be necessary to de-energize it first by opening the primary circuit (somehow).

One partial solution is to install a main secondary breaker at the transformer. If this breaker has appropriate time-current characteristics, it will lower the arc-flash level at your main switchboard to a much lower level. But it's only moving the problem back to the transformer, not eliminating it.

 

[alehman]

dpc,
I'm confused by your statement that if there is no main breaker, none of the breakers on the bus can be operated while energized. Why would operating any of the branch breakers be more hazardous than operating the main (assuming all breaker are applied within their fault interrupting rating)?

I've had clients ask me to install main breakers at transformers as suggested, but as you say it just moves the problem.

 

[davidbeach]

If there is no main, and something goes wrong during the operation of one of the branch breakers, there is nothing to interrupt the fault until the transformer high-side fuses operate. If there is a main, it would trip in response to a fault in one of the branch breakers.

 

[alehman]

DPC stated, "If you need to work on the main breaker, it would be necessary to de-energize it first by opening the primary circuit (somehow)." To me "work on" means maintenance activity other than normal operation.

So the underlying assumption here is that breakers or switches exposed to that incoming line fault level cannot be operated under any condition? This was discussed in thread238-184025.

This interpretation would seem to present a serious problem for many facilities.

 

[dpc]

The issue of arc-flash hazards while operating breakers with the doors closed is somewhat ambiguous in NFPA 70E. If by "operated" you mean manually opening or closing the breaker while standing directly in front of it, there is certainly an arc-flash hazard associated with this, since the closed door does not really provide complete protection.

The tables in NFPA 70E list arc-flash protection required for operating 480 V breakers with the doors closed, even though technically there are no exposed live parts.

So my recommendation would be to treat manual breaker operation just like any other maintenance activity for 480 V arc-flash determination. And this does result in serious operational issues for unit substations with no main breaker.

 

[Eleceng01]

Thank you all for the discussion. It appears - like I thought, that this is a difficult problem. I agree with dpc that the main breaker can not be operated (i.e. opened/closed) but is useful in lowering the arc flash value on the bus.

 

[stevenal]

Remote opening and closing might be implemented in order stay out of the zone.

But actually working on the main is a problem. DPC says to de-energize. This can be accomlished by pulling the high side elbows. NFPA 70E 120.1 (1-3) are now satisfied. After tagging comes step 5, testing for potential. This puts our worker back in the zone where proper PPE is required. This is risk category 4 per Table 130.7(C)(9)(a) if under 65kA and 1s. The high side fuse will likely take longer than that to clear.

 

[tommom]

Consider using a 750 kva transformer. 500 kva transformer often has a very low impedance (300's do also), allowing a very high arc-flash value. 750 kva transformer standard impedance is 5.75 +- 7.5%, giving a far more reasonable (but still PRETTY high) value.

Wye-wye connection may also be a problem, allowing zero-sequence fault current to flow through the transformer.

 

[dpc]

The primary fuse type can make a big difference. Your options may be limited in this case because it is a pad-mount, but I have found that clearing times can be much faster just be replacing an older EJ-1 9F60 type fuse with a newer 9F62 even with the same current rating.

I'm always a little uncomfortable relying on a primary fuse for a low voltage arc-flash rating since it could be changed out with a different size/type in the future. But if you're stuck with existing equipment, it is a low-cost option that can sometimes get the energy level down below 40 cal/cm2 (still a whale of a lot of energy, but within the realm of being able to operate locally with appropriate PPE).

When you actually do the arc-flash calculations, it is generally the case that you are better off with MORE fault current at 480 V, not less, because of the impact on clearing times. Seems counter-intuitive, but it usually works out that way.

 

[apowerengr]

Agree with dpc, increasing impedance will most often increase arc-flash energy due to inverse characteristic of clearing device. You could check with the serving utility - if the primary to the padmount is radial they might consider downsizing the tap fuse and/or the bayonet. Alternately you could consider installation of a bolt-on current-limiting fuse at the transformer secondary spades however typically such an installation sized for the service size will be way to large to effectively current-limit given the available fault current (might work ok in this case since you've got a 500kVA padmount - higher fault current than a 750) or downsize the fuse for the load/transformer (600 amp Class L). Downside is if the fuse blows you're waiting for the utility to open up the transformer and mostlikely your coordination won't work between the c-l fuse and the main breaker. I've started working with CPS regarding a new bayonet to better mitigate this issue - more to follow.

 

[ruggedscot]

Would the use of a restricted earth fault help to reduce the risk by picking up on a line earth arc and back tripping the hv feeder ?

I know that trying to co-ordinate protection can be a problem in such a situation - we use restricted earth fault protection to protect the transformer secondary and bus bars feeding the main breaker. One issue with this is that the neutral earth link has to be at the panel and not the transformer itself. The protection CTs being before the earth link.

Also this may not protect against a phase to phase fault and such a fault would depend upon the hv protection seeing the fault and tripping.

Rugged

 

[dpc]

Most of the 480V secondaries are solidly-grounded. Ground fault protection can be applied at 480V, but it can be difficult to coordinate with the downstream phase protection. If you have a 480V ground differential zone that included the main breaker, that would allow for fast clearing of the ground faults, and would lower the energy from an arcing fault that involved ground.

But IEEE 1584 arc-flash calculations are based on three-phase faults, so even if you reduce the energy level for a ground fault, the required PPE does not change.

I have a hard time convincing some plant engineers that converting to a high-resistance grounding system at 480 V will drastically lower the risk of having an arc-flash event because the Hazard/Risk Category doesn't change. If it doesn't reduce the required PPE, they often don't see the point.

 

[ruggedscot]

The only trouble with doing a high resistance neutral earth bond on 480v is that you would then have to consider that every neutral is possibly carrying a pd that would be considerably above the earth potential. This may be a concern with some equipment espicially if you are running a lot of switch mode and drive style equipment.

rugged

 

[scrump1]

In the studies we do, we most often come across high incident energy on the main breaker on the 480V bus. Obviously, the main protects the bus, but not itself, thus the bus will typically have much lower incident energy. (Amazingly, some clients insist that the main protects itself while it is being racked in/out. . .)

We use many different approaches to remediate the incident energy depending upon the clients' particular situation and/or budget. I feel that fuses are a bad idea in most situations because it then makes protective device coordination difficult at best.

If you have solid state relays, it is possible to have a 'maintenance' setting with a TCC set for faster trip time. It would be activated by a maintenance switch. This may not be advisable if you are having large motors automatically starting while the relay is in maintenance mode - or if you have forgetful electricians. Another option to explore is a differential relay that includes the downstream main in the zone of protection, since differentials have a faster trip time. Also, you could just replace your equipment. I have one client whose equipment is 30+ years old and was planning to have a major upgrade soon, so it did not make sense to spend money on equipment that was to be replaced. Instead, he has decided to just accelerate his replacement schedule – and that project now has the blessings of his plant management too! So, it can really depend on your situation which path will work best.

These were just a few ideas, I hope they are helpful.

Best regards,
-Shane