Please help with 4000a tie breaker sequence of operation

[fandi]

Hello All,
We have a N.O. 4000A tie breaker of a twin 4000A switchboards (they're identical switchboard with same ground, installed at the same time). Each swbd is fed from each utility transformer. One of the transformer is being demo'd due to light loads at the facility and a new cogen to feed both swbds.
To combine the two swbds into one, the N.O. tie breaker needs to be converted to N.C. breaker.
At this time, the cogen is not being connected to the swbds yet. We simply turn off the main breaker of the swbd which its upstream transformer will be demo'd and close the tie breaker. Per the peak demand in the latest year electric bill, the remaining transformer should handle two swbds combined (we don't plan to add significant loads in the future) but we still want to make sure to come up with a sequence of operation of permanently closing the tie breaker.
Any input idea would be highly appreciated!
Thanks a lot.

 

[don_resqcapt19]

As far as I know, breakers are not NC or NO like contacts...you just open or close them for manual control, and they open automatically for fault conditions.

 

[fandi]

As far as I know, breakers are not NC or NO like contacts...you just open or close them for manual control, and they open automatically for fault conditions.

Thanks, Don. Regarding breaker tripping in fault condition, should we match the tie breaker settings with the settings of the main 4000A breaker that stays (the other main breaker will become spare)?

 

[GoldDigger]

As far as I know, breakers are not NC or NO like contacts...you just open or close them for manual control, and they open automatically for fault conditions.

A breaker which closed on overcurrent would be a strange beast indeed!
Now a contactor could be NO or NC with the coil deenergized.

 

[jim dungar]

Thanks, Don. Regarding breaker tripping in fault condition, should we match the tie breaker settings with the settings of the main 4000A breaker that stays (the other main breaker will become spare)?

Questions I ask concerning Ties:
Are the loads divided in such a way there would a benefit from having only half of system with power?
Do the existing branch breaker settings have coordination issues with the main breaker settings?
Are the multiple sources of power intended to be backup systems or to meet load demands?

 

[fandi]

Questions I ask concerning Ties:
Are the loads divided in such a way there would a benefit from having only half of system with power?
Do the existing branch breaker settings have coordination issues with the main breaker settings?
Are the multiple sources of power intended to be backup systems or to meet load demands?

Thanks, Jim. I'll try to answer if I understand your questions correctly

Questions I ask concerning Ties:
Are the loads divided in such a way there would a benefit from having only half of system with power?
The combined load of two swbds is about 2000A which is about half of one 4000A swbd alone. The owner wants to maximize the cogen utilization by connecting the cogen to the both swbds' loads. Hence the idea of permanently close the tie breaker. On the other hand, the utility company doesn't want their transformer to be in idle mode so they want to remove one upstream transformer.

Do the existing branch breaker settings have coordination issues with the main breaker settings?
I will have to model all the breaker settings (one main and all feeder breaker of both sides) into SKM to find out.

Are the multiple sources of power intended to be backup systems or to meet load demands?
There are two sources of power in normal condition: utility and cogen (to offset the load demands) so it's not a backup system.

 

[big john]

There's nothing in the breakers themselves preventing them from all being closed at the same time.

They are most likely mechanically interlocked with a key switch system on the doors, or less likely, electrically interlocked through a system of auxiliary switches and shunt-trips.

You need to determine what style of interlock exists in the switchgear itself, that would go a long way to figuring out how to defeat it.

 

[ATSman]

Double Ended Switchboard

Double Ended Switchboard

You have what we call a double ended switchboard or substation: T1/ M1 - Tie - M2/ T2.

Two possible configurations:
1. M1 & M2 closed, Tie open.
Automatic Transfer Scheme where when 1 side looses power the control scheme
trips the main on that side and closes the tie. All load is fed from other
main. System transfers back to normal config once power is restored.
2. M1 & M2 closed, Tie open.
Fixed scheme, no automatic operation where as big john has mentioned, each
main is Kirk-key interlocked with the Tie, i.e. one main has to be open to
close the Tie. This prevents paralleling of T1 & T2.

Except in special cases, power transformers are not normally connected in parallel due to circulating currents (mis-matched impedance) between them. Also the increased available fault current affecting feeder breaker ratings.
So, you should be ok by either removing the auto transfer scheme or Kirk-Key interlocks.

 

[Bugman1400]

The relay settings for the tie are typically coordinated with the mains. You do not want the settings to the same. This is because, if the tie is closed and Bus#1 & Bus#2 are being fed from the one source, in the event that a fault occurs on a feeder from Bus#2 and its breaker fails, the tie breaker will trip before the mains which will keep Bus#1 energized.

 

[jim dungar]

... the tie breaker will trip before the mains which will keep Bus#1 energized.

And why is that the preferred condition?


Why would you assume that the loads directly connected to the main are more desirable than those on the other side of the Tie?
At least one data center, that I know of, has its most critical loads on the "Bus 1" of the switchgear, given the choice they would prefer that the Tie had been supplied as an isolation switch only.
There is another customer that keeps one of their transformers unloaded, as a backup only. They should have ordered their line ups as Main-Main, but the design engineer's boilerplate design did not consider that.

My point is that there is no one always correct answer.

 

[kingpb]

The tie being in the configuration provides some increased reliability to keep at least half the system energized, maybe, if that even matters to the operation.

Is the cogen intended to supply power to the utility or is it to help offset existing load? If the cogen cannot handle all the load, how will you load shed if the utility is lost? In this scenario the tie could be very helpful in dropping half the load, while at least running the cogen, then do a manual load shed of the other half to provide as much power as possible, and manually close the tie. Add an additional DEG to pick up the non-covered cogen load and you can operated in island mode.

As you can see there are multiple scenarios, none of them necessarily better than the other, it just comes down to how do you want to operate; how critical is the operation. Without detailed loads and what the facilities operation is, it is not really possible for us outsiders to recommend a preferred option.

 

[Bugman1400]

And why is that the preferred condition?


Why would you assume that the loads directly connected to the main are more desirable than those on the other side of the Tie?
At least one data center, that I know of, has its most critical loads on the "Bus 1" of the switchgear, given the choice they would prefer that the Tie had been supplied as an isolation switch only.
There is another customer that keeps one of their transformers unloaded, as a backup only. They should have ordered their line ups as Main-Main, but the design engineer's boilerplate design did not consider that.

My point is that there is no one always correct answer.

My assumption is not that Bus#1 has the critical loads. In the configuration of only being fed by 1 xfmr via M1/Bus#1, it makes more sense to have critical loads on Bus#1 if they can be easily swapped. If they can't then, oh well. However, in the event of a Bus#2 fault or a failed breaker fed from Bus#2, isn't it better to trip the tie bkr instead of dropping both Bus#1 and Bus#2. I assume you understand that a fault on Bus#1 or a failed bkr fed from Bus#1 will result in the loss of all loads.

Perhaps I'm missing your logic but, I do agree with your point. Sometimes its just company preference of operation that works best for them.

 

[jim dungar]

,, isn't it better to trip the tie bkr instead of dropping both Bus#1 and Bus#2...

I don't know. It depends on which loads are on which bus.
A paper mill in this area is configured as North Bus - Tie - South Bus - Generator. All of the facility utilities are fed from the North bus. What advantage exists for tripping the Tie?
Properly coordinated branch breakers means the only time the Tie should trip is when there is a fault directly on one the buses. An arcing fault internal to a closely coupled LV switchboard likely requires the entire board be de-energized.

 

[fandi]

There's nothing in the breakers themselves preventing them from all being closed at the same time.

They are most likely mechanically interlocked with a key switch system on the doors, or less likely, electrically interlocked through a system of auxiliary switches and shunt-trips.

You need to determine what style of interlock exists in the switchgear itself, that would go a long way to figuring out how to defeat it.

Hi John,
It was confirmed from the facility's electrician that there's no electrical interlock. There are kirk keys but they are not functional.
Thanks.

 

[fandi]

You have what we call a double ended switchboard or substation: T1/ M1 - Tie - M2/ T2.

Two possible configurations:
1. M1 & M2 closed, Tie open.
Automatic Transfer Scheme where when 1 side looses power the control scheme
trips the main on that side and closes the tie. All load is fed from other
main. System transfers back to normal config once power is restored.
2. M1 & M2 closed, Tie open.
Fixed scheme, no automatic operation where as big john has mentioned, each
main is Kirk-key interlocked with the Tie, i.e. one main has to be open to
close the Tie. This prevents paralleling of T1 & T2.

Except in special cases, power transformers are not normally connected in parallel due to circulating currents (mis-matched impedance) between them. Also the increased available fault current affecting feeder breaker ratings.
So, you should be ok by either removing the auto transfer scheme or Kirk-Key interlocks.

Dear ATSman,
MSA main breaker's kirk key has been removed and there's only a hole now. MSB main breaker (which will become spare after the upstream transformer has been removed) and the tie breaker still have Kirk key. And the electrician confirm they're not functional.

 

[fandi]

The tie being in the configuration provides some increased reliability to keep at least half the system energized, maybe, if that even matters to the operation.

Is the cogen intended to supply power to the utility or is it to help offset existing load? If the cogen cannot handle all the load, how will you load shed if the utility is lost? In this scenario the tie could be very helpful in dropping half the load, while at least running the cogen, then do a manual load shed of the other half to provide as much power as possible, and manually close the tie. Add an additional DEG to pick up the non-covered cogen load and you can operated in island mode.

As you can see there are multiple scenarios, none of them necessarily better than the other, it just comes down to how do you want to operate; how critical is the operation. Without detailed loads and what the facilities operation is, it is not really possible for us outsiders to recommend a preferred option.

Hello kingpb,
Thanks for your input. The cogen only helps offset the existing loads. The cogen doesn't connect directly to the double ended swbds. All the loads fed from the two swbds are non-critical. Here's how the system is set up:
- The cogen connects to a 2000A swbd, lets call it 'cogen swbd'.
- At the cogen swbd, there's a 2000A motor operated breaker (MOB) connect to the double ended swbds. In the event of power outage from the utility company, this MOB will trip and the cogen will switch from 'Utility mode' to 'Island mode'.
- All critical loads will be fed from this 'cogen swbd'. There's no load shedding required because the cogen can handle the critical loads.

So if the kirk-key system has been removed, do we need a sequence of operation for the tie breaker? As others suggested, we could set the tie breaker so that it will trip before the main (at MSA) in case there's fault downstream of MSB so we don't bring the entire facility down.

 

[fandi]

One thing I forgot to add, from the pictures of the three breakers (2 mains and one tie), they have the same breaker settings, which is not necessarily a good thing. With MSB's main 4000A CB being converted to 2000A feeder breaker (by replacing the plug) for the cogen, the tie should have tripped before the MSA's main.

 

[Bugman1400]

I don't know. It depends on which loads are on which bus.
A paper mill in this area is configured as North Bus - Tie - South Bus - Generator. All of the facility utilities are fed from the North bus. What advantage exists for tripping the Tie?
Properly coordinated branch breakers means the only time the Tie should trip is when there is a fault directly on one the buses. An arcing fault internal to a closely coupled LV switchboard likely requires the entire board be de-energized.

I'm not sure if we are referring to the same voltage level. Are you referring to <600V? I was referring to MV. Even for properly coordinated feeder breakers, the failure of the feeder breaker to clear a feeder fault should result in the tripping of the tie breaker and not take down the other mains. So, in your example, if a fault occurs in a feeder off the South Bus and the feeder breaker fails to clear the fault, wouldn't you want to trip the tie breaker and keep the North Bus energized?

 

[jim dungar]

I'm not sure if we are referring to the same voltage level.

Part of the problem of saying "always do things this way"
That is why I recommend determining the purpose of the Tie and the result of it opening, before deciding what settings it should have.