Imagine you are tasked to quote an arc-flash study for an industrial facility with a potential to generate over 5000 arc-flash labels. You have a large mix of MV and LV equipment in this client's facility - there are some tie breakers between the MV and LV equipment as well. Let's throw some redundant utility feeders and gensets into our hypothetical scenario for fun. At a high level based off the info I provided, it's going to be an expensive study. Would you rather:

A) Quote the facility as one big study and deliver the work to the client in one scoop - expensive option but least technically challenging
B) Quote the facility in multiple smaller studies and deliver the work as you work upstream or downstream or however you want to split the system up - least expensive option but more technically challenging
C) Do something else

I can see the cost advantages to option B, but I feel that poses a technical challenge for the study engineer whereas option A poses a monetary challenge for the client. Splitting scope up on a very large facility seems like a nightmare to me for all aspects of the process (short-circuit, coordination, and arc-flash). For short-circuit, if you start at the service entrance and only complete to the last MV level before you step down the voltage, you will want to ensure all downstream inductive loads are captured and modelled as they will affect your calculated fault current values upstream. For coordination, it would make sense to start on the LV side and work your way upstream (we want the device closest to the fault to trip first). However, as I'm sure many of you already know, you need to have your fault current magnitudes at that level in your system to configure the device settings. And finally for arc-flash, well, you can't do it unless you've done the other two studies first...

Open to hear the forum's thoughts on this.

Maybe split the scope up like this:

1. Data collection on a T&M basis with the Owner providing at least one man very familiar with the distribution system to assist and provide all available drawings and shop drawings. Owner's man needs to have keys and authority to access all areas. Include all distribution and equipment down to say, 25 or 50 hp. Nothing smaller at this point to help reduce cost. Small motors will not likely have a great affect on ratings and can be accounted for with a safety margin when looking at ratings vs available fault current in the Short Circuit study.

2. Fixed fee or T&M not-to-exceed for creating the model and running a SC study. Fee provided after data collection is done. Owner corrects any deficiencies found (code violations and over-dutied equipment and protective devices). Include a safety margin for ratings as small motors have not been included in the study. Include time for researching existing series combination ratings on failed (inadequately rated) equipment and devices.

3. Fixed fee to perform coordination study. Perform study after any deficiencies in 2. have been corrected. Include inputting new breaker and relay settings or require the Owner to do that. Owner to replace fuses, if any.

4. Fixed fee to perform arc flash study and print and affix labels. Perform study after all items found in 3. have been completed. The possibility exists here that the Owner may or may not want to sacrifice coordination on a case by case basis to lower IE and therefore want to go back and tweak a few settings.

5. Provide quote to go back and update studies for under 25 or 50 hp equipment or provide generic labels and company Electrical Safety Program policy says no label = no live work permitted in the meantime.

I would say this is very client dependent. Modeling and the calculations part are the easy bits. Data gathering is where the studies live or die.

Questions to ask:
What is their staff turnover? Let's face it, some places rotate employees like crazy. If they are unlikely to have managers/engineers/electrical workers who can see the project through, I would definitely recommend breaking it up.

What is the status of their configuration management? If they know what equipment they have, what their settings are, what the maintenance history is, and have drawings reasonably up to date. this could probably be done in one go.

What is their tolerance for pain? No, wait, that is not the right question. How well can they deal with the inevitable cost increases that will come with the study when they find out that they thought they had good configuration management but really didn't? Are they going to be able to push through and finish or will they leave the study incomplete because they ran out of money?

Several years ago I completed a similar study for a very large automotive manufacture. The job was bid as a single project which was scheduled to span two years. This was the directive of the customer so we worked together to define a clear scope. Given the nature of their business the power distribution system changed daily. At the end of the two year period we had over 20,000 buses which had been analyzed and labeled. EasyPower was the only software capable of modeling a system of this size. A great relationship was built and significant work continues to this day.

I encourage you to define a scope, and depending on this both sides should be prepared for adjustments along the way. For example, depending on how long this project takes you to complete is there much chance that additions to the power systems will occur? Any chance facility personnel or other contractors may adjust device settings? How will these be worked into your scope? What is the nature of their process / how long will this take you? If they have washdown areas or much equipment outside will any means of marking the equipment that you make be washed away?

Feel free to reach out if you'd like to discuss further. Good luck!

1. The philosophy you outlined is very ideal. Especially, for projects where you don't have existing drawings to bid off. I've always found data collection to be the most difficult and time-consuming aspect of arc-flash studies. I try to take the burden off myself by modeling the client's one-line drawing (if available) ahead of the visit. That way I'm merely doing field verification when I'm at their site. If the visit is fairly relaxed, I'll even pull out my laptop to start selecting make and model types for protective devices. We all know that's not how it goes sometimes. T&M is the way to go for projects with little to no information handy before bidding. I've not met many clients who are fans of T&M, though.

2. Agree with everything here. I recently had a project where the client had several overdutied devices. I didn't provide them with labels for the overdutied device or anything connected downstream. We agreed that they would get labels for the affected devices once they correct the issue. The device replacement typically occurs when they have a planned outage, which can sometimes be months from when the study was completed. Also a good point about including time for researching series combinations. The process can be time-consuming depending on the quantity of overdutied devices and the number of unique combinations you have to vet.

3. The last portion of what you wrote is tough to convey to clients who aren't necessarily knowledgeable about what we do - the fact that you need to update device settings if they are changed in the study to have proper IE values on the label. To me, it's easier to recommend settings and then hope they get implemented in the field before the labels are sent. I know most of the time this never happens. It's possible to document existing settings and create a comparison table of some sort to depict existing vs. new settings, but this is very time-consuming.

4. No disagreements here. I've seen lots of clients want tweaks in this stage too. In addition to what you said, I'll have a client or two redline the shit out of the inputs I enter in the "comment" section of the arc-flash label because they want something more descriptive about where the label is supposed to go and that's fine.

5. I typically go by IEEE and exclude 3-phase and 1-phase motors less than 50 hp, but you bring up a fair point. Even though I don't think it's necessary to include them, the customer might want to include them if they want labels for the motor disconnect switches.

The first part says it all. That is why I like to do my own data collection. I know folks who will send others to do it for them and then get upset when they see what data ended up being collected.

I haven't come across too many places where I've seen the same staff as when I did something for them a few years ago. There are a few outlier cases where there isn't consistent turnover and when it comes time to do a study, the process is seamless and I love it. Breaking up the study seems sensible when the staff faces throughout the years aren't consistent especially if it is a monstrosity of work as I described in our fairytale scenario.

Only once before have I had all the info fed to me in one go, but it was a new construction job and the study was only $2,000 - I cherished it. I've never had this happen with existing facilities.

Putting this question in my back pocket for a rainy day.

I grew up on SKM, but am now leaning toward EasyPower. How did you work with the client to establish the procedure to tackle changing elements of the system? I imagine the easiest path would be to provide them with a study and one-line drawing of what equipment/devices were present at the time of bidding. Once they receive the full study and drawings, they can then markup what has changed and send that back to you for an updated study, drawings, and labels.

From reading folks' replies on the thread, I've started to notice that it is essential to be as specific as possible when you outline the scope of work to the client for an extremely large job. Important to do it for all jobs, but special attention should be paid to the jobs with high dollar value potential to avoid confusion later down the road.

Exclude under 50hp, yes for arc flash but no for short circuit, I think.

Don't have time to look it up now but I don't think IEEE says you can exclude under 50 hp motors for short circuit calcs but I think it does say that for arc flash study calcs.

I think the check box to exclude them in SKM is under the Arc Flash evaluation tools, not Dapper/short circuit tools or setup. Going from memory on it though.