Bus Ducts have been used in facility designs for years. Some were installed before much was understood about arc flash hazards. This refresher will focus on the modeling technique for bus ducts in 480 vac and 208 vac systems. The discussion will also expand on what variables to consider when the bus duct installation presents an arc flash hazard.

*See the full transcript of the webinar below*.

## EasyPower OnSite Intro |
## Utility Short Circuit & Arc Flash |
## Arc Flash Hazard Reduction |
## Changes in NFPA 70E 2015 |

See all of our training videos

**Jim Chastain**: Good morning everyone, welcome to the EasyPower Tuesday refresher, this is Jim Chastain and I appreciate everyone joining us this morning for discussion about arc flash involving bus ducts. As we are want to do, we like to start off with a poll and get a little bit of feedback from the audience at the same time and so today our plan questions are going to include how frequently you use bus ducts in your studies and then request for feedback on future webinar topics. So let's go ahead and launch the first poll question. I appreciate your participation and some feedback and as I mentioned before the feedback has more to do with making sure that we're covering the topics requested, and in [00:01:00] some cases steering the topics of discussion as we're proceeding. I appreciate your feedback and your attendance today. Here's how the audience shapes up today. That makes a fair amount of sense. So let's move on to the second topic, and it's more some feedback on how we can provide information that's useful to the audience going forward and so if you would let us know where you would like topics covered.

By all means participate in the discussion today through the questions list and certainly after the fact if you have the individual questions you're welcome to contact techsupport@easypower.com or send us an email provide and collect feedback that way. and [00:02:00] here's how request shapes up there. Protective coordination is usually a fairly highly requested subject. So again, appreciate everyone for participating and let's move on to the discussion this morning. So, apparently a lot of exposure to bus duct modeling and systems that require bus duct protection. So what we expected to do today is to cover some of the considerations that you should be taking into account as you're both clicking data and conducted the study. They are instruments for a system distribution of electrical power that are fairly pervasive you can see in both commercial and industrial installations in and general the bus ducts are assembled or manufactured in [00:03:00] predetermined sizes that are assembled on site. The links are fixed and the tools for the bus duct permit a variable equipment drops to be able to be moved around which gives a lot more flexibility to a plants configuration rather than having to re-cable either temporarily or, or almost impossible with permanent cabling when loads need to be moved around. The problem is seldom protective devices distributed along the bus ducts and it's this length that affects the incident energy, and that ends up being generally the issue that we need to address when we are doing the modeling. So the technique is relatively easy when you think about it.

Basically, you determine the overall length of the bus [00:04:00] duct that you're trying to model and then break it up into average lengths, and so this decision and the granularity involved depends a lot on you as the system modeler or analyzer. What makes sense in terms of the energies involved and the loads involved and some of the distances from the equipment that people are operating to the bus duct itself. So the total bus load represents local drops at each length. So we're grouping effectively lumping the loads averaged at these, located at these average lengths and so I'll talk more about that. It's easier, I think when you see the example. The incident energy at the bus drop connection will increase and that's where the real hazard is and it's nice to see with the EasyPower tools effectively what that means. [00:05:00] so let's take a typical installation. So we're about a 40 volt system and for hypothetical purposes this particular building as a single bus duct 500 feet long and 1600 amp capacity and so just for this illustration were using 100 feet average bus lengths as my bus parameter. and what we're going to do is at each, at each connecting bus we're going to lump the loads and anything that's within 50 feet either upstream or downstream of that bus will be included in the total loads for the drop. now as we, as we talk about that you could see that if you have very large loads it may make more sense to have smaller average bus lengths and that you need to take into account which are considering at model system. [00:06:00]

So, each drop in this case will have an individual fuse and the main protective device on the incoming to the bus itself is going to be 1600 amps to cover the total capacity. So let's see what the EasyPower model looks like. Alright, so what I've done here is we have a low voltage power circuit breaker as my incoming switch gear and then I have an individual fuse going to the bus duct itself and so in this case we selected a generic 1600 amp copper bus duct and this is modeled from the EasyPower equipment palette by utilizing. In fact, let me just [00:07:00] redraw this from scratch here. So it's this icon on the, in my case the second row is called bus way and I've installed one to the first bus and them I'm going to connect a bus way to the second bus and it's going to be the same characteristics as open up the bus way data dialog box you can see that we've selected a generic manufacturers standard type we're using 100 length amp capacity of 1600 amps and its copper and from that we calculate the average impedances. I'm going to use this as my average, if you will bus duct length so I'm going to copy and paste this into the second bus that I've installed. [00:08:00]

Since I have five connections I'm going to copy the second part hold down control and select both of these items. Now do a control C and then a control V allows me to paste this over here and then reuse this length connection as we go along. So I'm going to do it again until I get the 500 feet that I was looking for. there's 1, 2, 3, 4, 5. now as I indicated earlier the problem is that for this total length the only protective device we have is this, in this case the main fuse that we selected so if I have a load connecting at this end, this bus, at the end bus then it's [00:09:00] gonna be protected by this of fuse and as we seen even in the most simple examples the incident energy will increase as we go through this leg as we go through this bus tour so let's see if we can since I've already indicated the capacity we've copied from first bus I can go ahead and fault the bus a look at the current and what we see is that the current, if we look at this a little closer, the current will stay essentially, because of the impedance will drop at each, at each bus. So here it's 42,000 amps 37,000 amps 33,000 amps etc. Now again, the counterintuitive nature of arc flash is at work here.

So our first thought is "well if this lower current then we may have [00:10:00] lower energy" but as we go and calculate the incident energy what we see is that that energy actually increases along the way. So where we start off with our first bus at 5.6 calories the last bus down here ends up at 21 calories, and this is real and the problem again is, if we look at our range for arc flash, the problem is the trip time because we're depending on that primarily fuse will increase to the point that we have a very elevated incident energy. So, that the problem turns out less with the load, in fact let's start putting some loads in here and kind of show where that leaves us. The problem is actually with the bus duct distribution itself. Alright, so I'm just going to activate this load, a typical load, and just throw in some [00:11:00] large numbers on there but because our drop then will be fused or protected for that particular section now I can go and in spite of having a high incident energy up here, my protective device will, will give me a respectable incident energy calculation when I look down at the, at the load that we're working on. So if to look at what's going on there we're seeing we have a one calorie incident energy because it's being protected by that upstream fuse. So, if we go back and look at the, and look at how this results in the overall scheme of things let's go ahead and put a load down here at the end.

So the question [00:12:00] becomes how, if I'm lumping the loads, and let's kind of, I've created my, a subsystem here that gives me a bus duct load and so we're just going to drop it on this last bus. Alright, so this is a fairly hefty load and it represents the lumped sum of all the loads within the last 50 feet where as this, this particular lump total will be the total 100 feet here and 50 feet downstream. So consequently, at each bus were lumping the loads within 50 feet of that bus with the exception of very first link here. So now that I've done that, what, and because we're, if we're considering this is as a lump load, what do we put in as the protective device? Well, if we're adding up all the current then potentially there's some argument to be made that we [00:13:00] should add up the total current that could flow through the protective devices. But what that, what that gives us is an artificial, an artificial number when it comes to calculate incident energy. So the fact is even though we're averaging the lumps out here the individual, the possibility of a fault at any individual load is much greater than it would be on the whole, the whole total and so, again as the person doing the analysis the size of this particular protective device, and when I'm thinking about worst case, isn't going to be the sum of the total, it'll be the largest one that's coming off at this particular drop and again what we effectively are doing, when you think about it the incident energy in terms [00:14:00] of an individual load is dependent upon, let's go back to short circuit is dependent upon the fault that's going to happen at that particular load. And so what we're doing here is faulting, it's faulting a specific bus and the current that's involved has less to do with the location on the bus duct and more to do with the total that we've included as far as the load.

So it actually increases the total current that this bus is gonna see but it really doesn't do anything in terms of the incident energy at this particular point if we have our protective device sized for an individual load. So the point I'm trying to make, and this is a decision you need to make as you're doing data collection as you're looking at the analysis in [00:15:00] terms of the overall contribution to the place where the most hazard exists which will be at the two drop for the bus type lumping the loads is a pretty good approximation but in terms of the downstream protection this can, like I said, an argument could be made for either summing all in parallel or just sizing this for the worse case which would be the largest device which again would give us the largest amount let through energy. So, we come back and fault the bus, let's just fault the single bus. Look at our arc flash and the incident energy there it ends up still being a fairly reasonable number even though when we look at the whole system were getting some very elevated energies at the bus duck itself.

So the problem ends up being in the way I've [00:16:00] modeled this is I drop a cable from my bus drop and then if I want to I could put individual disconnects at the loads depending upon how the physical setup is for my system and in either case as long as I'm let's see what I want to put here. I want to put a breaker. In this case I'm using a 600 frame and a 300 amp molding case circuit breaker. I still have very good protection as far as operator of safety at the loads but I need to really caution folks about how a exercising any switch gear up here that's next to the hazard, next to the a bus duct itself. Okay, I feel like I kind of stumbled through that, so [00:17:00] the technique is this, I come up with a scenario where I can break my total bus duct up into average segments. What I chose in this case was 100 foot segments and then I lump the loads that are within 50 feet and I use that as my load for each of these virtual drops. And then that gives me a reasonable assumption in terms of the fault contribution and each of these bus duct drops on the bus itself and as far as a downstream I'm using the largest load which would have the largest protective device at each of the drops my virtual bus to give me the loading that's going to be downstream.

So that's what I would label the equipment and I have, again, much higher energies [00:18:00] calculated and basically don't work on the bus duct while it's hot. Okay, I'm thinking I stomped all over that one. Let's see where else we are. Now, the same thing can be seen when we look at even a 240 volt bus and the significance here as it is in a real plant, as we move or change loading at each drops we remove the drops around as long as it stays within the same 50 foot vicinity, radius from the virtual bus we don't have to change the main bus all we are going to do is change the loading at each of drops that we've setup, so it's a relatively minor change in terms of the calculations in EasyPower and frankly will change the system loading because it's all, that's all dependent as far as labeling that's dependent upon [00:19:00] the protective device that's localized either for that load or coming off the drop. To show that in lower voltage system here's a situation where that's exactly the case and here of you can see we've come up with a 42 foot average length that we are using for each the segments and, and then when we've lumped the loads, this case these are relatively small loads and we can model then either as the individual, in this case a motor, aindividual motor or as a group of motors like we would in an MCC.

Now, so the same discussion applies, so if we'd done a group modeling what are we going to for the protective device and in that case at least I would argue or suggest I [00:20:00] want to use the largest single breaker because that would have the greatest amount of let through to give me the calculations for downstream energy and if we go to look at our current we can see the same thing, we just want to look at this series of energies along the bus duct So the bus originates here from these two center points and you can see we have a decrease in current as we go down the bus duct, and this side it's 69 and 67 over here it's 72 and 7000 and as we calculate incident energies we can see that it actually goes up, although this is much more civil in terms of what the initial calculation, the initial energy is because of the shorter length of the bus But you can see [00:21:00] it starts at 1.2 calories and goes to 1.5. and over here it starts a 0.7 and ends up at 7.8. and this is all based upon the fact that we have the initial current is protected from the same upstream fuse and that the load energies will be a factor of the localize protective device. And so it'll be again, a fraction of whatever is going on at the bus way. update column said pretty much just a-Ok, so that is pretty much the technique and the different things you need to take into consideration.

I think most of this discussion would be included in the assumptions that you make for the study so yes, that's what this example is, is specifically a [00:22:00] 240. Are you asking me where it came from? This actually just came from a VAR customer's file that was sent in for assistance in modeling. And so it can be done pretty much straight forward in the EasyPower modeling. Alright, so that pretty much covers what I wanted to stay wanted to check to see if there's any questions. I appreciate everyone attending today and the feedback on the polls and survey, and I invite you all to check the website for updated schedules and new announcements on Webinars and we look forward to talking to you in the future. Thank you one and all. [00:22:47]