Reducing the amp rating of E-rated fuses below 100% FLA of xfmr to mitigate the arc flash?

[bdn2004]

Below is a TCC for a 2500kVA, 12.47kV-480V Delta-Wye grounded transformer and it's 150E primary fuse. This is the existing installation. The full load amps on that transformer is about 116A. The recommended size per the literature I've read says to size the fuse between 100-150% of the FLA = 116A*1.5 = 174A....So the 150E looks ok.

It can be shown that the arc flash can be reduced on the secondary main if that fuse is reduced to a 100E fuse and the transformer will still start. But this is less than 100% of the full load amp rating. On the other hand, looking at the fuse curve for that - which is on the second post - it doesn't look like the 100E would ever trip even at 116A.

Would this be a good idea ?

 

[bdn2004]

 

[dpc]

There are a lot of considerations. The primary one - what is the actual peak demand on the transformer? If you don't know that, I don't think I would be comfortable going below transformer FLA. You really need to have a good handle on what the maximum possible load is on the transformer, even if it rarely occurs. Also, what is the magnitude of the reduction in arc-flash and is it worth it?

What multiplier are you using for the transformer inrush? The "inrush point" shown on the TCC is a very crude rule-of-thumb for avoiding nuisance tripping. The default in EasyPower is 8X FLA. I use 10X. You need to stay conservative with this constraint when fuses have to be replaced.

The 150E looks pretty good. The 100E looks to be cutting it a little too close. Maybe the arc-flash issue can be lived with, or it can be solved another way.

Cheers,

Dave

 

[davidbeach]

That number in front of the E is intended to be higher than the highest continuous load. Don’t be lured in by the defined minimum melt point.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[bdn2004]

Thanks. I don’t see any real advantage in reducing this fuse, just an idea.

But I’ve seen this done successfully at a Plant I used to work at. As in they changed out the fuses 10 years ago to 80E on similar transformers and there’s not been any issues of those fuses tripping. And that facility switches feeders and transformer loading configs around more than most places.

I’ve another large company’s engineering standard that states that the inrush current on the transformer can only be determined by testing ... and if that hasn’t been done - use 12x as the multiplier. And that eliminates this idea too.

 

[Mbrooke]

Reducing the fuse size only increases the risk expensive outages and total facility black outs.

While some S&C E rated fuses can be loaded above their rating indefinitely or for short periods of time, doing so shifts the fuse curve during said loading such that you may lose selective coordination.

Further inrush tolerance goes down. A quick trip and reclose just at the right point in the sine wave or voltage dip that causes motors to stall may cause nuisance blowing of the primary fuse.

All of this assumes overloading as indicated by the manufacturer.

Loading beyond recommendation can cause annealing of the fuse link. Annealing can greatly lower the current carrying capacity of the fuse such that a down stream fault or even normal load can trigger nuisance blowing.

You're asking the right questions and I applaud you for that, but also keep in mind there is a reason why those fuse curves start at 200-240% of their listed rating.

I'd recommend other alternatives to mitigate arc flash.

 

[eeprom]

It might be easier to deal with this on the secondary. You are going to have one very high point for IE, and that's your first point of disconnect. If you use a fused disconnect on the secondary, then use that fused disconnect to feed your MCC/Switchgear, then you have limited the high IE to one specific point, and a spot that is very unlikely to need servicing. I know this is not a cheap solution, but it will solve your problem.

 

[LionelHutz]

I believe these are both 5kV class E fuses, but you run the same risk of this happening if you under size a 15kV class E fuse. E rated fuses generally aren't full range fuses and don't react well to overloads in the 100% to 175% current range.

or

 

[Mbrooke]

Any idea why the react so poorly?

 

[bdn2004]

Geez. I thought these type of fuses are supposed to be self-protecting for any current up to the interrupting rating.

That doesn't look very self-protecting to me.

 

[bdn2004]

Mbrooke,
Interesting....

"Loading beyond recommendation can cause annealing of the fuse link. Annealing can greatly lower the current carrying capacity of the fuse such that a down stream fault or even normal load can trigger nuisance blowing."

A definition "Annealing": "is a heat treatment that alters the physical and sometimes chemical properties of a material to increase its ductility and reduce its hardness, making it more workable. It involves heating a material above its recrystallization temperature, maintaining a suitable temperature for an appropriate amount of time and then cooling."

For example on this .... say the continuous current is 116 Amps, the 100A fuse curve shows it's not gonna blow. But you're saying that continuous current makes that fuse run hotter than it's designed for and is actually changing its physical properties and changing the time-current characteristic. Yikes.

I'd like to have a link to that science.

 

[LionelHutz]

E and R rated fuses aren't full range fuses. They won't clear until the overload is above the minimum melting current. Between rated current and the minimum melting current they will overheat and burn.

 

[Mbrooke]

What about N fuses? I've looked at time current curves which seem to start at 125%...

Can you tell me more about full range fuses? Do you have examples or links of such?

 

[davidbeach]

Per the NEC, at voltages where the E and R rated fuses are applied, there are no requirements to protect against overload. Fuses may be set at up to 3 times ampacity and relays at up to 6 times ampacity; very different than in the low voltage realm. Higher voltage systems are more generally protected against overload by design than are lower voltage systems.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[Mbrooke]

@Deleted Reply: I agree, but to be fair breakers often start at or above 125%.

No shame in what you wrote, I like that you point this out. I love conversations like this.

In fact in parts of Europe fused installations must go one wire size larger.

 

[bdn2004]

I'll put it back in then, thanks...(just thought it wasn't really pertaining to an E-rated fuse, the subject of the thread)

My point is that that this very common, 600V, fuse rated at 200 Amps ... the top of the curve stops at 350A at 5 minutes. In other words it has the same issue as the E-rated fuse, it's not much of an overcurrent device.

I did take a look at the cable it protects: 3/0...and that lies well to the right of the fuse. So it protects that wire so it's doing it's job. Maybe this is just the "conservative" approach the Code and Standard writers came up with in almost all installations, to make sure we don't screw it up ? just a theory.

 

[bdn2004]

And if you look at this 200A, low voltage molded case circuit breaker's TCC curve with it's wide band at the same long times....it's practically the same as the fuse.

 

[Mbrooke]

Table 310.15 B 16 and Table 310.16 are very conservative taking into account fuse/breaker pickups and then some. A 135% of more overload will not damage the conductor.


By starting at 125% instead of 101% assures that a breaker will not trip below 100% from ambient variants in temp, heat build up within the panel board, manufacturing tolerances, ect.

 

[LionelHutz]

It's not simply where the curves start. Some fuses can handle any current for any period of time without bad things happening to them and others can not. E and R fuses fall in the second category.

 

[Mbrooke]

@LionelHutz: why is that? My knowledge in regards to fuses is actually very limited.