MCC Bus Bar Bracing

[wroggent]

I was recently at one of my company's plants looking at their MCCs because they wanted me to determine if they have capacity for an expansion. While I was looking at the MCCs I noticed the nameplate say the MCCs have a 10kAscr. The refinery that the plant is attached to includes the plant in their arc flash program, and thus there are arc flash labels produced by the refinery on the MCCs. For two of the three MCCs the arc flash label indicates and available fault current of 93kA, which I'm assuming is symmetrical but on the label it is simply stated as "Bus Bolted Fault". This prompted me to check the size of the transformer feeding the MCCs. It is a 5MVA 13200d/480y transformer with 5.35% impedance. Assuming the line side of the transformer is an infinite bus (I don't know the available short circuit MVA or current), the available fault current is given by 5000/(3^0.5*0.48*0.0535)=112kA. The 112kA seems to agree with the arc flash labels since I am assuming an infinite bus. I am concerned about the available fault current being so much higher than the bus bracing rating. I forgot to write down information about the main circuit breaker for the MCC, but I'm pretty sure it's an MCCB with a 1600A trip unit. Are there any MCCBs that are current limiting? I'm thinking that because it's an MCCB the MCC will experience the full fault current for a fault at or close to the bus bars (although I plan on looking up the TCC for the main breaker when I get information about it). In other words, I don't think the breaker will interrupt the fault before the current reaches the symmetrical (or otherwise considered as transient, between sub-transient at synchronous) value. If the bus bars were to short, what sort of physical damage could I expect to occur? Lastly, could the MCCs be modified to increase the short circuit withstand rating or would they have to be replaced entirely. It's a little surprising to me that they've been operating those MCCs for 15 years straight without an issue.

Thanks for any insight.

P.S. The short circuit withstand rating is given as a symmetrical current. What does this say about the asymmetrical (symmetrical + exponential decay) fault current it can withstand? Anything for 3 cycles??

 

[Kiribanda]

wroggent,
Just for my own curiosity. .
What is the HRC rating and the PPE requirement indicated in the AF labels at 93kA?
Is it possible to upload a sample of the AF label for the members?

 

[dpc]

On new MCCs, the SC rating on the nameplate is based on the lowest rated interrupting device in the MCC when it is shipped. It doesn't really tell you anything about the bus bracing. You'll need to check with the manufacturer. But obviously, if there is available SC of 93 kA then a 10 kA MCC is a big problem.

I'd assume the bolted current on the arc-flash label is symmetrical. The IEEE 1584-2002 arc-flash calcs are based on symmetrical three-phase bolted fault currents as a starting point.

 

[jraef]

dpc is right, the label is likely reflecting that someone specified some sort of untested custom unit in the MCC lineup somewhere that could not be labeled at more than 10kAIC, which has the undesirable effect of lowering the rating of the entire MCC. A lot of people don't realize that when insisting on custom units. I don't know of any MCC mfrs that offer anything below 35kA bracing, most are 42kA minimum. But that is still an issue if you have 93kA at the terminals.

I think that technically, current limitation at the Main CB to allow for lower ratings below it is only valid if the entire structure has been tested and listed like that. I don't know of anyone that would do it though, because they all offer 100kA bracing and unit listing options, and that would cost considerably less.


"Will work for salami"

 

[wroggent]

Kiribanda,

The HRC is 1 and FR Shirt and Pants are the required PPE. The HRC is 1 for all of the columns that have buckets, i.e., columns downstream of the main breaker. The column with the main breaker is labeled something like "No PPE Available." I only have a picture of one the the labels indicating an HRC of 1.

I am not really familiar with arc flash calculations, but the label really doesn't make sense to me. My understanding of arc flashes is that as available fault current increases and clearing time increases, the incident energy and corresponding HRC increases. I went through the IEEE Std. 1584 method that is shown in NFPA 70E 2009 ed. If I calculated everything correctly, the clearing time would need to be about 0.035 seconds in order to have an arc with an energy of 3.6 cal/cm^2 (15 J/cm^2) at 18 inches (which is what seems to be shown on the label). A clearing time of 0.035 seconds seems unrealistic for an MCCB. Do you think they simply took the HRC from table 130.7(C)(9) of NFPA 70E 2009 ed. for 600 V Class Motor Control Centers?

dpc and jraef,

I was luckily able to find an old spec for the MCCs. The spec indicates the bus bracing was available in 2 flavors; 65kA and 100kA. The MCCs they purchased have the 65kA bracing option. That's not the 10kA I was fearing however I think this is still an issue (due to 93kA available). I don't see anything on the spec that would lead me to believe that there is some type of custom device installed in these MCCs. Thus, I'm not sure what is the cause of the 10kA withstand rating. Which items should I be looking at for determining the withstand rating? Bus bar bracing, MCP short circuit rating, ...? The MCCs are old Advantage series Culter-Hammer. I'll have to contact Eaton about this, but do you suspect I'll be able to get new 100kA braced bus bars or need to get a new MCC altogether?

(And yes, there is another MCC behind this one which is also right up against the wall offering only one path of entrance and egress when in between them. *facepalm*)


Thanks

 

[LionelHutz]

It might be possible to upgrade the MCC with the 100kA bus components, but you won't get new UL 100kA ratings plates. UL does not allow a company to field installing rating plates.

The arc flash energy does sound low but you're talking about a 2 cycle clearing time and a high fault level so it might be possible.

You can't install some type of current limiting (fuses) at the main MCCB. The reason is that the 65kA fault rating is achieved by taking the incoming test cables and shorting them then adjusting the power source until 65kA flows. Then, the cables are connected to the MCCB and the MCC is tested with a short. Well, 65kA won't flow any longer since the impedance of the MCC components will lower the fault current. This means you can't match fuses to the MCC without knowing what the actual test curren was.

The custom device doesn't have to be anything special. Any breaker or contactor or overload or other power component which doesn't have the proper testing to be applied at 65kA into that MCC would limit the rating. Look at the 50hp-200hp motor starter components since the components in that range have to test to 10kA minimum.

 

[dpc]

The first thing I would do is to verify the accuracy of the 93 kA on the label. Just because someone prints a label that says 93 kA doesn't make it accurate. There is a lot of nonsense being printed on arc-flash labels these days. If it is accurate, then the 65 kA bracing and 10 kA short circuit rating are definitely problems. As LionelHutz says, you can only get new rating nameplates from the manufacturer. I suspect the 10kA rating is there for a reason - probably the one already mentioned by JRaef.

 

[jraef]

It's not the bus bars, it's the support structure for the busbars that is different. Changing it in the field would require near complete disassembly of the structures and installation of additional or different bus support structures. I seriously doubt it would be less expensive than buying new, especially considering down time costs, it would take a long time.

Individual buckets would need to be listed for 100kAIC as well. I know that for A-B, most starter buckets are already rated for 100kAIC, but it is optional on larger ones, so C-H may be the same way. But each unit is supposed to have a label inside that specifically tells you what it is rated for. If you have even one unit that is not labeled for 100kAIC, the entire MCC is not. If none of them have labels indicating 100kAIC, you are much better off buying new.

As to the 'custom" unit, I didn't mean to imply that it was something odd, it could have been something very subtle. For example if a customer insisted on having a specific circuit breaker on a starter as opposed to the standard one or one of the tested and listed options, it could have been that the starter had not been SPECIFICALLY tested and listed with THAT EXACT circuit breaker, so substitution will cause that unit to be "untested", and thus the entire MCC rating gets downgraded.


"Will work for salami"

 

[waross]

Is it feasible and economic to consider series reactors to limit the available fault current?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jraef]

I'm working on one of those kinds of projects right now for an aggregate plant. Brand new MCCs were ordered with 42kA bracing because the Available Fault Current was 38kA, turns out in the hubbub of buying the machinery and changing HPs, they upgraded the size of the transformer and nobody thought to re-evaluate the AFC, now it's 51-53kA depending on which MCC is involved... oops. So we are trying to engineer in some CL reactors to drop the AFC to 42k, but it's getting ugly. I happen to think that they should dig a trench and run the cables around the building a few times, it would cost less.


"Will work for salami"

 

[waross]

Good idea Jeff.
I've seen something similar done. We had numerous electrical rooms, each with a pair of transformers and associated Power Distribution Centers. The designers called for a minimum of 100 feet of cable to feed each transformer, to limit the available fault current to the rating of the equipment. The extra cable was looped in the cable tray for the close in units. I've gotten the fault current calc's down to the equipment rating a couple of times by adding in the cable impedance.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Gragas]

If you are going to try that Waross, depending on cable size wouldnt it be an option just to have a 480-480 transformer in that feed?

Random bonus of possibly helping loadflow? ha

 

[jraef]

Reactor / 1:1 transformer, little difference in either case. The issues are size, added switchgear, heat loading (if indoor) and added losses, plust the costs associated with of all of the above. Hence, my project "getting ugly".


"Will work for salami"

 

[jghrist]

If you add impedance with either looped cable or a reactor, you will need to revisit your motor starting calculations.