First, assume everything in my following response is inaccurate, not applicable or a lie and that I don't know anything and that you should not follow any methods or procedures I describe in my response as they might get you or someone else killed.
That said, if this were my task here is how I MIGHT go about it. Or I might not.
Wait for schedules shutdowns or ask the hospital to schedule shutdowns as needed to gather the data.
Another option might be to have the electrician collect the data for me, especially if they were hired by the hospital directly and not by me or my company. Whatever their policy is, whether NFPA 70E compliant or not, is up to them, and reflect that my report (data collected by others, report based on the data supplied by others, client would not shut equipment down to collect the data etc.)
The other approach could be that I would proceed with the job and wear daily wear PPE (20+ calorie jeans, leather boots, 8-12 calorie shirt, 12 calorie arc rated gloves, and safety glasses) and stay well away from equipment, never entering the limited approach boundary if at all possible and absolutely do not enter the restricted approach boundary. Incident energy drops quickly with the square of the distance away from the equipment. For example, an IE of 8 calories at 18 inches is only 2.6 calories at 36". I would use that logic for most of the distribution, but not for large, 480V equipment fed from transformers over, say, 500 or 750 kva without some type of protection at the transformer secondary, especially in a smaller room. More on this later.
Attachment:
working distance.jpg [ 30.92 KiB | Viewed 187414 times ]
I wouldn't wear the balaclava and face shield unless I suspected I was entering the arc flash boundary. This would just be based on experience. Under 400A branch 208V volt panels in larger rooms or corridors are probably covered with daily wear and keeping my distance. Put the balaclava and face shield on if I had to be within say, 6 feet or so. Again, this is just my opinion on one way, right or wrong, to handle this.
For service equipment or separately derived systems with larger transformers, especially with 480V secondary, I would have to handle on a case-by case basis and determine whether or not a shutdown is required. If I can get breaker data without removing covers and feeder conductor data by opening the downstream equipment great. That would only leave the service entrance conductors, which I might be able to get from the utility. Opening up the service equipment is a different matter and can be problematic or risky. If the electrician can get the data for you under whatever his company's safety program is, without you being anywhere close, then go with that, or gather the data during a shutdown.
RE: CT ratios, I don't know what kind of relays or metering (if any) you have. You might be able to get that from the relay itself along with the settings. If not, and the gear has metering, you could possibly look at what the meter says the gear is drawing and then clamp on an Amprobe to the CT lead for the relay and do the math to determine the ratio without having to open up anything that exposes the MV bus.
Another option would be to first use the NFPA 70E Category or Table method where you are able to get conservative PPE requirements.
Another approach would be to spend some time modeling everything you can without opening equipment up, with breakers set to their max settings and assumed feeder conductor sizes based on the OCP and be conservative on top of that.
I could write a book on this with a lot of "ifs" and "buts" but these are my basic thoughts on if I had to do this project without the benefit of any drawings or product data.
