Ever considered applying tolerances to TCC values when determining the fault clearing times? 3

[Wfg42438]

Hello Everyone,

Typically, when running a protection and coordination study or Arc Flash study we use time current curves to determined fault clearing times.

In some older facilities there can be protective devices (Relays, Fuses, etc.) which perhaps have been in service for an extensive amount of time.

I am curious if in such situations if one should consider applying a tolerance to the clearing time obtained from the TCC curve for example a fuse or an overcurrent relay?

Also has anyone come across an overcurrent relay manufacturer that recommends or suggests a tolerance on the predicted TCC times?

I am not dealing with the scenario myself at the moment but just curious to learn how others deal with the items above.

or

if most stick to the TCC times since there are no real guidelines on this that im aware of at least.

 

[dpc]

NFPA 70E and IEEE 1584 says that you need to take into account the "condition of maintenance" of the protective device and its operating time. How you do that is basically up to you. Breaker opening time has to be included with the relay operating time of course. Most software accounts for the breaker opening time automatically.

 

[Wfg42438]

@dpc

I agree with your statement 100%.

In the relay case the FCT= Breaker Opening Time + Relay Tripping time.

My question is based on the equipment's age could one introduce a "%Tolerance" on the relay trip time?

For example something along the lines of :
FCT= Breaker Opening Time + Relay Tripping time + (%Tolerance/100) *RelayTrip Time

from what i recall i dont believe manufacturers give a %tolerance to consider in the relay trip time but i could be wrong.

Again, not proposing everyone does but curious if its a common practice for those who encounter older equipment

 

[thermionic1]

Most electromechanical relays are calibrated to +/- 3%. In my experience, if a CO or IAC was properly calibrated, it will still often fall into this time band 15-20 years later.

I've run into switchgear lineups where every single relay was out of whack. After double and triple checking my own equipment, the relays were recalibrated. I believe maybe the previous tech either had an uncalibrated test kit, or their kit was distorting and giving false values.

There are also cases where originally the coordination study may have called for say 3.2 A and over time that data was lost. The next relay routine and someone thinks the tap should be 3.0 A and adjusts accordingly. I can't tell you how many times in a previous life I was asked to test relays and ask the client for settings and get that deer shined in the headlights look. CO's and IAC's have a little clip on the nameplate designed hold a tag stating the Tap, Time Dial, Inst and CTR, but not everyone uses those.

LV power breakers, especially older units that haven't been retrofitted with a newer trip unit can have a lot of slop. I recall some very old dashpot breakers that were all over the place. Some unit had no oil in the dashpots, some dashpots had small holes. I found some that were consistently different. The root cause in this case was that the dashpot oil had been replaced with a different viscosity oil, causing the times to be substantially quicker than the published curves.

 

[davidbeach]

I have seen older breakers and EM relays exhibit slow behavior the first time they're called upon to do anything after a long period of just sitting there. Subsequent operations have had more normal times. Instantaneous elements have less opportunity to be sluggish after sitting for a long time than time overcurrent elements. But get rid of the EM relays and go with something that will always calculate out the exact same time for the same current. Throw on a couple of cycles for a sluggish breaker and you'll be good. I don't have a good feel for how slow a EM TOC relay can be if it hasn't done anything for a few years.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[Wfg42438]

@thermionic1

Thank you for the feedback from what you have experienced.

From what you have seen where you expect a +/- 3% tolerance it seems considering this in the FCT wouldn't be the worst idea.

However, I do realize an extra 3% tolerance is not going to make a significant impact just adding a tiny error margin.

From your experience have you seen any manufacturer data sheets suggesting the end user to consider a tolerance on the predicted TCC curve times for electromechanical relays?

 

[Wfg42438]

@davidbeach

Thanks for sharing such interesting information on what youve seen out in the field.
Based on what you have shared it seems there is some wiggle room in there for error.

However, there is no set guideline to say what to do in all cases as expected.

 

[stevenal]

When studying coordination, a coordinating margin is typically used that will cover some relay error as well as breaker time. Maximum clearing times for fuses should be just that and well tested by the manufacturers.

For arc flash, NFPA 70E says to consider operating condition and condition of maintenance.