Motor contactor drop out

[paulengr]

Paul, nice post. All stuff we used to do that we don't do anymore (and I can't understand why).

Can you point me towards some literature stating what is legally required?

NFPA 70E references 70B and NETA MTS. 70E is not legally required but OSHA requires some standard under the general duty clause. So at two references in legal requirements are weak. NERC mandates PMs for utilities but is a little vague on specifics.

In your experience, what percentage of customers require and/or are willing to pay for such testing on new installations?

Almost none. It’s more of something that kicks in either due to insurance typically FM or Hartford when age is an issue.



Sent from my iPhone using Tapatalk

 

[paulengr]

Paul, nice post. All stuff we used to do that we don't do anymore (and I can't understand why).

Can you point me towards some literature stating what is legally required?

NFPA 70E references 70B and NETA MTS. 70E is not legally required but OSHA requires some standard under the general duty clause. So at two references in legal requirements are weak. NERC mandates PMs for utilities but is a little vague on specifics.

In your experience, what percentage of customers require and/or are willing to pay for such testing on new installations?

Almost none. Most manufacturers do a factory test after assembly so most on site testing is really all about verifying that it survived shipping and installation, mostly looking for wiring issues. It’s more of something that kicks in either due to insurance typically FM or Hartford when age is an issue. The more proactive companies are doing more testing because of arc flash concerns than anything.



Sent from my iPhone using Tapatalk

 

[paulengr]

Of course, because the definition of arcing has changed to 'emission of light with volatizing between electrodes' (paraphrasing) aka garden variety short circuit.

On the other hand 60 years of utility practice has shown otherwise just about weekly.

That is utterly false. The tests in question were three parallel bus bars vertically oriented terminating into a block of insulating material. I don’t recall what they used as insulating material but I believe it’s FRP (glastic). The gap was varied from around 1/4 to 1 inch and arcing tine was monitored along with incident energy which is thermal radiation onto a copper calorimeter. Light and some other variables are monitored but mostly not reported. For some reason these barrier tests greatly enhance the output energy. Larger gaps increase the energy but also if the gap is too wide the arc is unsustainable. Enclosure size also makes a difference. These tests are intended to simulate industrial, not utility conditions. Most of the time in overhead line configurations utilities don’t have a problem with arc flash because it just shoots dow.n the line away from the worker. In generating plants it’s a bigger problem but equipment design and clearance rules tend to dictate towards working de-energized for the most part. Plus high fault current 208/120 systems in generating plants would be pretty rare. In fact one of the curious things about arc flash is it becomes less of a problem as voltage increases. It is mostly current dependent. It is mostly independent of voltage. Except that where tens of kA of fault current is not unusual at 208, 480, and 600, it is pretty rare at 13.5 kV and above because high AIC equipment gets crazy expensive.

 

[mbrooke]

That is utterly false. The tests in question were three parallel bus bars vertically oriented terminating into a block of insulating material. I don’t recall what they used as insulating material but I believe it’s FRP (glastic). The gap was varied from around 1/4 to 1 inch and arcing tine was monitored along with incident energy which is thermal radiation onto a copper calorimeter. Light and some other variables are monitored but mostly not reported. For some reason these barrier tests greatly enhance the output energy. Larger gaps increase the energy but also if the gap is too wide the arc is unsustainable. Enclosure size also makes a difference. These tests are intended to simulate industrial, not utility conditions. Most of the time in overhead line configurations utilities don’t have a problem with arc flash because it just shoots dow.n the line away from the worker. In generating plants it’s a bigger problem but equipment design and clearance rules tend to dictate towards working de-energized for the most part. Plus high fault current 208/120 systems in generating plants would be pretty rare. In fact one of the curious things about arc flash is it becomes less of a problem as voltage increases. It is mostly current dependent. It is mostly independent of voltage. Except that where tens of kA of fault current is not unusual at 208, 480, and 600, it is pretty rare at 13.5 kV and above because high AIC equipment gets crazy expensive.

Utilities also have underground systems, many of which rely on burning clear, and are above 100,000ka:

 

[mbrooke]

Same thing, at 208 volts arcing has been shown to not be an issue in the real world "normally self-extinguishing"

 

[paulengr]

Same thing, at 208 volts arcing has been shown to not be an issue in the real world "normally self-extinguishing"

I would almost agree. But we have a documented fatality. IEEE 1584-2002 state’s that at 208 V or less and under 125 kVA “need not be considered”. But when they started testing under real world conditions it is clearly an issue and definitely self sustaining in some cases. The problem so far is that self sustaining is not really predictable. At 130 VDC (not even AC) it’s not. At 208 and above it can be. Over about 300 V it most certainly is. IEEE 1584 was revised down in the 2018 edition when they found real works conditions where self sustaining arcs occur and because OSHA investigated a case of a fatality at those voltages that was an arc at the top of a breaker in an MDP on a temporary construction panel right off the utility transformer. It’s a bit of an extreme example but temporary construction panels are pretty real world.

https://drive.google.com/file/d/0B6mGRCG7wns_bHZIRDl5VzlDSWM/view?usp=drivesdk

 

[mbrooke]

I would almost agree. But we have a documented fatality. IEEE 1584-2002 state’s that at 208 V or less and under 125 kVA “need not be considered”. But when they started testing under real world conditions it is clearly an issue and definitely self sustaining in some cases. The problem so far is that self sustaining is not really predictable. At 130 VDC (not even AC) it’s not. At 208 and above it can be. Over about 300 V it most certainly is. IEEE 1584 was revised down in the 2018 edition when they found real works conditions where self sustaining arcs occur and because OSHA investigated a case of a fatality at those voltages that was an arc at the top of a breaker in an MDP on a temporary construction panel right off the utility transformer. It’s a bit of an extreme example but temporary construction panels are pretty real world.

https://drive.google.com/file/d/0B6m...w?usp=drivesdk

I think what they are trying to say is the incident energy persisted for some time probably because it occurred before the main.

The issue is that arcing is used as a catch all blanket term for at least a dozen phenomenon...Some not likely some not even physically possible for a given variable... It much more profitable to mandate gimmicks then saying setting the instantaneous setting of a breaker to reach into a fault or current liming fuses do the same thing.

 

[paulengr]

I think what they are trying to say is the incident energy persisted for some time probably because it occurred before the main.

The issue is that arcing is used as a catch all blanket term for at least a dozen phenomenon...Some not likely some not even physically possible for a given variable... It much more profitable to mandate gimmicks then saying setting the instantaneous setting of a breaker to reach into a fault or current liming fuses do the same thing.

I would agree somewhat. The problem is that OSHA does not distinguish arc flash from other injuries and I’m not sure most inspectors would even recognize it. So it gets lumped into either burns or “electrical”. I undertook scouring the reports in detail to try to put together data on causes and try to find what the chance of shock or arc flash is for cases like just walking by (zero) or doing normal operations (1 case in 5 years). There are kind of three injury categories in “electrical”. The vast majority are shocks with or without burns. There are a few startled cases where something happened which may be a shock or just a loud bang and someone trips or falls out of a man lift or off a ladder. Obviously neither shock nor arc flash. The third are burns without direct contact which are obviously arc flash which is a small minority. Of those around 1 in 15 is a fatality. Caulsey who publishes a lot of the statistics that get reported counts all electrical burns as arc flash so we see “50% of electrical injuries are arc flash” which is not true. It’s more like 5-10% at most of bonafide arc flash injuries.

As to eliminating it, I will grant you that years ago we set trip settings and fuses as large as possible to avoid nuisance trips. But often for instance our crude bolted fault estimations were so high that the breakers and fuses could never trip on a fault. It was ok for AIC (sizing) but not coordination. Even then rules like leaving 2 seconds delay between breaker curves was common. Today we go for minimum settings to minimize fault time. Accuracy is a lot more important. So yeah in existing plants you can often just turn things down to minimum. In the worst case as you said turn on instantaneous and forget about coordination during maintenance. I disagree about doing it everywhere. Coordination is important. Ultimately zone selective is the best way but not practical everywhere. Also though attacking the problem head on forces you to get creative. Things like realizing that opening and tripping on the high side of a transformer is generally safer, or tripping a high side breaker with a low side relay. But the big problem today is still that for instance if I order up switchgear or an MCC from somewhere I have no idea what the fuses or trip units are so I can’t plan or fix this stuff until it arrives.