Question, In the commercial power producer plant, is it mandatory to have a arc flash study/analysis on generator exciter field system? An electrician is task to do a field maintenance on 90 MW plant while the generator is online. Voltage and amperes varies base on load demand. Our generator excitation system is rated at 250 vdc at 750 amperes. On the past, we have our low voltage gloves on, safety eye glass, face shield, hearing protection and dust mask. Job scope is to check field brushes by pulling on brush pig tail one at a time (22 brushes total) to see if it move freely, wipe down carbon dust accordingly.

In my opinion yes. We have been seeing more DC system arc flash but not on the exciter.

If you're doing energized work, it's required.

However every time I've run the DC calculations on excitation, there never seems to be enough available fault current to amount to anything over 1.2 cal/cm2 even using the modified Lee DC fault calculation that you can find in Annex D of 70E.

Just a couple of questions back. Is it essential that you carry out the brushgear maintenance with the generator online? Have you considered NFPA 70E 110.1 (H) hierarchy of risk control measures, elimination of the hazard?

You may be interested in a perspective for energised brushgear maintenance from here in the UK which may give some food for thought. The hierarchy of risk controls have been enshrined into legislation here and across Europe for many years now and there is a duty to examine work practices to see if hazards can be eliminated.

Some years ago, I was asked to carry out a risk assessment review on this activity in one of the largest coal fired power plants in Europe and also a new gas fired station for the same client. Several of the larger power producers in the UK share best practice which gave me a platform to consult with other competitor companies about how they carry out brush gear maintenance. In those consultations I was quite surprised at the range of reactions to my questions about the need to carry out the activity online. Most comments were against energised work considering that the hazards were not just electrical, many tons of rotating plant spinning at 3000 revs per minute (50Hz) being another. There was an argument for observing brush sparking under load but this was in the minority.

I'm asking the above questions as risk assessment reviews often reveal a certain amount of custom and practice built over many years and do not stand up to scrutiny bearing in mind the expectations of current legislation and safety standards. The work that I did uncovered many other improvements which could be adopted should online work be seen as necessary. These improvements fit into the range of controls including Substitution, Engineering Controls, Administrative Controls and PPE.

For certain tasks for proper maintenance of a synchronous or DC machine, YES you need to do it online.

Going back to the OP, according to 70E Annex D.8 using 250 VDC and 750 A with a conservative 2 seconds (no intentional tripping time) of 2 seconds, I get 0.91 cal/cm2 at 18" which is the typical working distance for general cases using 70E. Given that with most brush rings you can't get that close a more realistic distance is probably appropriate. This is well under the 1910.269 cutoff of 2 cal/cm2 and just above the NFPA 70E cutoff of 1.2 cal/cm2. HOWEVER it is also a ridiculous case because it uses maximum power transfer which is the same as the Lee equation for AC...it never happens in reality and the actual incident energy will be much less than this.

In general this will be the kind of results that we will get. Generator excitation whether or not it has any protective relaying is firing from an SCR bridge with most utility generators. The SCR itself severely limits the output voltage and current. In the case of a GE MD824 DC motor and genset that I looked at with a mining excavator, the stall current was 6000 A at 12 VDC. At the peak power point things were much more reasonable but didn't change the fact that incident energy was still much less than 1.2 cal/cm2. It certainly sheds light to me on considering the damage from a flashover vs. the injury potential and the fact that I was never able to find and document any injuries from flashovers.

If you aren't "utility interactive", either not grid connected at all or else set up in some way as a backup system, then yes NERC does NOT apply. NERC/FERC doesn't have any mandatory arc flash rules though so whether or not the equipment falls under these regulations is immaterial. Yes NERC worms their way into the maintenance side of things but it has little to do with the safety aspect of the system except indirectly. NERC regulations are pretty "open" too compared to OSHA. They give very broad systemic-style requirements as opposed to specific requirements and let the utilities develop their own rules.

I'm hesitant to suggest that OSHA 1910.269 does not apply. If it's the primary power source then 1910.269 does apply since this applies to "generation, transmission, and distribution". The definition is a little vague by at least one letter of interpretation makes it very clear that at least cogens would fall under 1910.269. If it doesn't, then OSHA's general duty clause still applies and thus NFPA 70E applies rather than 1910.269 rules. Both contain arc flash requirements, so there is no "escape" either way.

In EU, Canada, and Australia, I've been able to confirm that similar logic applies in that some kind of SAFETY regulation applies, irrespective of whether or not any regulations regarding power production apply.