1/2 Cycle fuses and Arc Flash Energy Levels 3

[rockman7892]

When looking at Arc Flash incident energy levels I have seen that any equipment immediately protected by a fuse clearing in 1/2 cycle has a low IE level and in some cases can be as low as a category 0.

What about equipment that is much further downstream of the fuse such as equipment that has 2 or 3 busses and circuit breakers between it an the upstream fuse? Does this downstream equipment also have a very low IE level because of the upstream fuse? In other words does every device downstream of the fuse have a very fast clearing time and therfore a low IE level or only devices that are immediately protected by a fuse?

Any good references on the impact of fuses on Arc Flash results?

 

[rbulsara]

Any device in series of the fault location will have an impact and can be considered. Most software will allow you to select how many level of devices in series can be evaluated.

The important part is to include that particular overcurrent device ID and location on the arc flash label so it can be verified before commencing the live work.

As usual practical and professional judgment need be applied, there are no codes for it.

Fast fuess while provide good protection, do not help in coordination.

Rafiq Bulsara

http://www.srengineersct.com

 

[dpc]

The incident energy is a function of the available fault current, the working distance and the clearing time. It doesn't matter if the fuse is 5 feet or 5000 feet away. Fault currents will decrease as the distance from the source increases because of the increased impedance of the conductors. So there will be a tendency for clearing times to increase for more distant faults.

The arcing fault current at the location of interest must be determined and using that, a maximum clearing time is determined. It has nothing (directly) to do with physical location of the fuse in relation to the fault.

Any good references on the impact of fuses on Arc Flash results

Just contact your nearest Bussmann salesperson. They will be happy to fill your office with data on fuses and arc-flash. There is nothing magical about fuses and arc-flash. It is all a function of the clearing time. And that depends on the fault current available.

David Castor

http://www.cvoes.com

 

[PHovnanian]

The arcing fault current at the location of interest must be determined and using that, a maximum clearing time is determined.

Minimum clearing time?

Maximum clearing time can be quite long for higher arc impedances and are usually set by engineering judgment. And while the instantaneous arc power will be lower, the arc energy (power integrated over time) can be substantial.

 

[rockman7892]

rbulsara

You mention devices being in series. Lets assume however that we are dealing with a perfectly coordinated system. In this case wont the interruption time stricly be a function of the immediate upstream device and no other devices upstream should have an impact on this time?

From the arc flash study I have reviewed it appears that equipment immediately after a fuse has a low IE level and in most cases is a category zero. I think I am attributing this to the fast clearing time of the fuse at the given fault or arcing current.

So lets say that for a given fault current or arcing current on a bus we replace the 1/2 cycle fuse with a cirucit breaker who's relay pickup curve matches the fuse curve exactly. So then for the same level of fault current will the 5 cycle operating time of the breaker compared to the 1/2 cycle operating time of the fuse (same fault current same curve) cause the IE to increase drastically?

 

[Zogzog]

You need to look at the worst case incident energy, often that is at lower arcing current levels where clearing times are longer. If the arcing current is below the current limiting range the fuse will not operate in 1/2 cycle.

 

[dpc]

Minimum clearing time?

No. IEEE 1584 provides equations for approximating arcing fault current based on bolted fault current. Using this arcing current and fuse maximum clearing time for that current, the incident energy is calculated. For systems < 1000 V, a second calculation is done at 85% of the estimated arcing current.

The goal is to estimate the worst-case incident energy.

David Castor

http://www.cvoes.com

 

[rbulsara]

rockman:
Perhaps I don't understand you question, if the device "sees" the current for a longer time for whatever reason, it will have higher arc flash energy.

I do not understand 1/2 cycle fuse thing, a fast acting fuse rated to open within 1/2 cycle of a 50,000A fault current peak, does not open in 1/2 cycle at currents less than that. You still need to look at fuse total clearing time,which are comparable to a breaker relay combinations.

See what zogzog said too.




Rafiq Bulsara

http://www.srengineersct.com

 

[rockman7892]

I guess my confusion stems from the fact that I have always been under the impression that a fuse was much faster acting than a breaker (assuming we are comparing same current magnitudes)with a fuse clearing in 1/2 cycle after pickup while a breaker usually takes 3-5 cycles to clear after pickup.

Maybe I dont understand the whold 1/2 cycle current limiting capabilities of fueses and therefore am confusing the clearing time. Any ideas where I can get a better idea of fuses and their current limiting capabilities?

 

[dpc]

Molded case circuit breakers are much faster than 3-5 cycles at high levels of fault current. Newer breakers will clear faults within 1-2 cycles - some even faster.

The only reason a device is considered "current-limiting" is because it can open and clear a fault before the first peak of fault current - less than 1/2 cycle. This applies to fuses and current-limiting circuit breakers. There is nothing in the fuse that makes it inherently current-limiting. If the fault current level is such that it takes the fuse longer than 1/2 cycle to clear, it is not doing any current-limiting in that application.

David Castor

http://www.cvoes.com

 

[LionelHutz]

In other words does every device downstream of the fuse have a very fast clearing time and therfore a low IE level or only devices that are immediately protected by a fuse?

The simple answer to this would be "no".

If we really simplify things consider the following. The energy in an arc flash is roughly proportional to I^2t. I think you will find that as you move further downstream from the fuse that the current drops quicker than the clearing time increases. This means the fault energy will drop as you move away from the fuse.

OK, so the point of the above? The energy is lower at most points after the fuse not because the fuse has a "magical" very fast 1/2 cycle clearing time, but rather because the fault current and available fault energy is just, well, lower.

So lets say that for a given fault current or arcing current on a bus we replace the 1/2 cycle fuse with a cirucit breaker who's relay pickup curve matches the fuse curve exactly. So then for the same level of fault current will the 5 cycle operating time of the breaker compared to the 1/2 cycle operating time of the fuse (same fault current same curve) cause the IE to increase drastically?

You must use the complete clearing time. The operation of the protection relay does not directly stop the fault current so using the relay operating time is very wrong. In this example, the fault current with the breaker is flowing 10 times longer than the fault current with the fuse. Your arc flash energy will be approximately 10x as high.

 

[Zogzog]

"I guess my confusion stems from the fact that I have always been under the impression that a fuse was much faster acting than a breaker (assuming we are comparing same current magnitudes)with a fuse clearing in 1/2 cycle after pickup while a breaker usually takes 3-5 cycles to clear after pickup. "

Been listening to fuse salemen? There are some slick marketing campagns and "free" training out there from the fuse OEM's that have a lot of people confused. Don't get me wrong, they have value in some cases for arc flash mitigation but they are by no means a magic solution.

 

[Maak]

IEC 62271-200:
"Application of suitable current-limiting fuses in combination with switching devices can limit the short-circuit current and minimize the fault duration. It is well documented that the arc energy transferred during such tests is not predictable by I2t. In the case of current-limiting fuses, the maximum arc energy may occur at current levels below the maximum interrupting rating."

So, current-limiting fuses are effective only if the current is in the current-limiting range. In LV systems arcing fault currents can be less than 50% of the bolted-fault current. (Even less than 30% according to Dr. Sweeting.)

Please see figures 16-41 in IEEE-1584.

 

[rockman7892]

Thanks for all the responses!

To provide an example and a better understanding for myself I am looking at the attached motor bus with 30cycle fault current and corrosponding TCC with fuse.

You'll notice that the one-line to the left of the TCC shows the 30cycle fault current avaliable at the 4.16kV switchgear bus (ER-2 MV Bus) as well as at the motor terminal box. The fault currents are somewhat close however the switchgear bus has an Incident Energy level corrosponding to a category 3 while the motor terminal box has and IE corrosponding to a category 0.

When looking at why the IE levels differ so greatly I look to the TCC and I think it is due to the fact that that at the 30cycle fault current for the ER-2 MV Bus the there is a delay on the instantaneous portion of the upstream breaker curve with the delay being about .15sec. It is possibly this delay that is causing this bus to be a category 3.

The 30sec fault current at the E18M motor terminal box however is shown to the right of the fuse curve at point #1 which I have labeled. At this point it looks like the fault would corrospond to a fuse time of .01 sec or less. Would that fact that this fault current corrosponds with .01sec or less put this fault magnitude in the 1/2cycle clearing or fault limiting range of the fuse?

What if for the sake of discussion the fault current at the motor was 2.0kA as I indicated at point #2 on the curve. In this case the fuse clearing time would be much longer probably about .2 sec or so and thus would put the fuse outside of its current limiting range and lead to a higer IE level. Is this correct?

Although this TCC sheet shows the 30cycle fault current, the Arc Flash report shows the bolted fault current that was used in the IE calculation to be 32.122kA. The momentary fault current shown in the short circuit study is 33.325A. So for the Arc Flash calculation how does the software ususally what bolted fault current to use for the IE calculation when using the integrated method? Why would it just not use the maximum momentary fault current? I have seen some cases where breakers have used the 30cycle currents for the IE calc.

 

[LionelHutz]

I think you've got it.

What if for the sake of discussion the fault current at the motor was 2.0kA as I indicated at point #2 on the curve. In this case the fuse clearing time would be much longer probably about .2 sec or so and thus would put the fuse outside of its current limiting range and lead to a higer IE level. Is this correct?

The IE could be higher than the fuses at #1 or 22.2kA. I'd expect that much extra time would allow a fair bit more energy. It would have to be calculated to really know.

However, the fuse clearing IE at point #2 won't be as high as the breaker clearing IE at point #1. The clearing time of the fuses may be the same but the fault current is 2.2kA lower. Just saying, not sure if this is what you were thinking or not.

 

[rbulsara]

So for the Arc Flash calculation how does the software ususally what bolted fault current to use for the IE calculation when using the integrated method? Why would it just not use the maximum momentary fault current?

I believe most software would check the IE at 38% of the calculated fault current and at 85%, and use the higher result. The whole idea is to calculate most "probable" maximum IE, which does not necessarily occur at the maximum fault current. Not all devices open at the momentary peak current.

Also IEEE formulas are based primarily on testing and not just theories. The formulas were essentially developed to fit the curves, produced by plotting the results of the testing.


Rafiq Bulsara

http://www.srengineersct.com

 

[dpc]

I'd suggest getting a copy of IEEE 1584 and reading that. The rationale for the calculation method is explained in the standard. All the software programs are based on the equations in IEEE 1584, where applicable.

David Castor

http://www.cvoes.com

 

[rockman7892]

O.K. so its safe to sumarize that any fault magnitude to the right of a fuse curve will operate in the current limiting range and operate in 1/2 cycle whereas any fault magnitude to the left of the fuse curve will take longer for the fuse to clear the fault and will operate longer than 1/2 a cycle and outside the current limiging range.

LionelHutz I understand your point about the reduced fault current at #2 possibly not having an effect on IE since although the fuse clearing time is longer the fault magnitude is dramatically reduced which will possibly lead to less IE.

Looking at the TCC its obvious to see that the the MV bus has a higher IE because of the intentional delay on the breaker curve of about .09s. But what if this delay was removed and the instantaneous pickup was carried straight down? Would the bus then have the same IE level as the motor since the fault on the bus in now cleared with no delay. Or will the IE level on the MV bus still be a good bit higher on the bus since the breaker may take 3-5cycle to open upon fault pickup whereas the fuse only takes 1/2 cycle to clear at this point?

My question about which fault current the software used in the IE calculation stemmed from the fact that I have seen all different fault currents (momentary, interrupting, 30cycle etc...) used for different points in the system when using the integrated method for determining the IE levels. Maybe some of the are the 38% and 85% currents as rbulsara mentioned and the software just displays these reduced currents as the maximum bolted fault currents in the report. Report is done in Easy Power.

Or is it possible that if the software looks at a particular fault and realized that the device's time curve is not set to clear the fault in the momentary range that it wont interrupt it until it is in the interrupting range so therefore it would use the interrupting rating. This wouldn't make sense though because I belive the fault current prior to interruption would still be contributing to the IE regardless of the device opening.

 

[rockman7892]

Theoretically would would happen in this application if the motor fuse shown was replaced with a larger motor fuse. I know the goal with the fuse is to keep it as low as possible still allowing for motor operation but what consequences are there with someone replacing it with a larger size.

Obviously the first thing is that at the motor the IE level may increase by moving to a higher fuse size due to the fact that the fault current magnitude might fall to the left of the fuse curve.

But what about other consequences? Since the fuse is only providing short circuit protection for the motor is you risking more motor damage during a fault by installing a higher fuse size?

Just trying to think of a scenario where a plant was in a pinch and only had larger fuses avaliable.

 

[rbulsara]

rockman:
Take a deep breath and start with what dpc said in his last post. If the arc flash calcs were so simple, it would not have taken so long for them to come out as a 'standard', which by all counts is in its infancy, even today.

Also do not confuse arc flash IE calcs for personal protection (that is what the arc flash calcs is about) with equipment protection.

You deal with equipment damage with is own withstand rating and thermal capacity evaluation, independent of the arc flash calcs.

Yes, anytime you change a fuse or device setting, your previous calcs become invalid.

Rafiq Bulsara

http://www.srengineersct.com