As a power systems analysis consultant we regularly get asked to carry out arc flash studies. We are always happy to do them, as it is a useful exercise in most cases. However, what we often see are clients asking for arc flash studies that are are not relevant or required. We have seen no end of arc flash reports that have pages and pages of meaningless tables and results.
We have put together a list of some common misconceptions and misunderstandings that we hope you find useful!
- An an arc flash study is not a replacement for a short circuit study.
- An arc flash study is not a replacement for a protection study.
- A good protection study is more important than an arc flash study!
- If you want to be thorough and safe you need (in order of priority):
- A short circuit study,
- A protection study,
- An arc flash study.
- The IEEE 1584 method is only valid up to 15kV – above this the results are questionable at best – misleading at worst. They can be calculated using the ‘Lee Method’ but it breaks down the higher the voltage.
- Arc flash studies for open terminal 33kV, 66kV, 132kV etc. switchgear are not relevant / valid. The risks with these are controlled / managed in a different way.
- If the switchgear exceeds its fault rating, then an arc flash study is useless and no amount of PPE will save you.
- Safe approach boundaries are associated with work being carried out on switchgear / equipment. Not so much for just walking past the switchgear. There are some videos on YouTube of such events – but they are pretty rare. See the NFPA 70E guide below.
- There are old and nasty switchgear panels out there that are dangerous and unsafe. These are the ones that really need an arc flash study, and wont contain an arcing fault.
- Do not confuse an arc flash with electrocution or a fault. An arc flash is a specific kind of fault.
- You cannot have an arc flash if the insulators are under SF6 – there is no mechanism for it to occur, instead we talk about burn through times – this is the time take for an arc to penetrate through the insulating medium.
- This means that fixed pattern 33kV GIS does not need an arc flash study. If you have vacuum insulated switchgear with SF6 breakers you could potentially have a risk but it is low (see below)
- Arc flash is really associated with switching and maintenance – so if it is fixed pattern gear, the risk is already very low.
- A lot of modern switchgear is arc resistant (AFLR) – that means the risk is generally contained if the doors / panels are closed and it is not being worked on (see above).
- Transformer terminal boxes are not relevant and do not need assessing – as they are not open and not maintained / switched live.
- If an arc flash result is high (>40cal/cm2) it usually (not always) means you are using the wrong method, have made some bad assumptions or the protection settings are wrong!
- If an arc flash result is really high (>100cal/cm2) then there is something seriously wrong in the study. Go back to the beginning and rework from first principles.
- North American switchgear and practice is very different to UK / EU approach. Do not follow the IEEE 1584 approach blindly.
- LV switchgear design can vary significantly – it is important to understand the different types of enclosures that are present and how they do / do not help.
- There is not (yet) a good way of carrying out an arc flash study on a DC system. They can be done, but the results are questionable. Some methods are given in NFPA 70E, but they are simple.
- NFPA 70E Table 130.5(C) has a useful summary of when an arc flash risk can / cannot realistically occur. Have a read of it – it can help significantly when carrying out risk assessments.
