Closing a Tie Breaker when the Services are out of Phase

[ed downey]

Can someone help me understand the Math and the physical properties that happen when the following occurs:

Two separate 4000A 480V 3 Phase 4 Wire services with an automatic tie breaker between the two services.

If the tie breaker malfunctions and tries to close causing the two services to be tied together and they are out of phase.

I have always understood that the current will continue to rise while the two services are tied together until something breaks.

How would I calculate the change in current associated with this circumstance or if there is something else that would happen that I am not thinking of that would help also.

Thanks for the help in understanding this.
Ed

 

[shepelec]

I can tell you there would be one heck of a bang. You would be closing in to a phase to phase short.

 

[mcclary's electrical]

It would not be a phase to phase short unless you swapped phases. The frequency would probably not match though.

 

[iwire]

It would not be a phase to phase short unless you swapped phases.

Read the title and the first post again.

 

[ohmhead]

Well when you transfer to other source if you have lots of motors online they are going to backfeed into system and if its out of phase the voltage difference is like the 4th of july !

Yes its a kinda short circuit bucking phases or connecting a phase to b phase .

But iam just a simple conduit electrician and dont ever listen to me .

 

[charlie b]

Usually, when you place two sources in parallel, the control system will ensure that their voltage levels, frequencies, and phase angles are close to one other, or it will not allow the tie breaker to close. What this means is that (looking only at Phase A) as the voltage rises and falls 60 times a second, when the voltage on Source 1 is approaching its positive peak, the voltage on Source 2 is also approaching its positive peak. That way, when the two sources are connected to each other, the voltage between one and the other is very small, with the result that only a small amount of current will flow from one to the other at the moment of transition. Thereafter, meaning after the momentary transient and after things settle down to a steady state operation, the two sources will act as one, supplying the same voltage, the same frequency, the same phase angles.

But if the two sources are out of phase, it could mean that the voltage on Source 1 might be approaching its positive peak at the same moment that the voltage on Source 2 is approaching its negative peak. If the control system fails, and if it allows the tie breaker to close at that moment, the voltage across the tie breaker's contacts at the very moment of closure will be twice the peak value of the system voltage.

How would I calculate the change in current associated with this circumstance. . . ?

It is a simple Ohm's Law situation. Divide twice the system voltage by the resistance of the contacts in the tie breaker. Since the tie breaker is likely to be sort of big, and its contacts sort of big also, their resistance will be sort of small. So dividing a very high voltage by a very small resistance will get you a current value in the tens of thousands, if not hundreds of thousands, of amps. Anyone standing close to the tie breaker at that moment is likely to have a very bad day.

 

[winnie]

Just some musings here:

At the most basic level, the same thing will happen that occurs when you close any circuit: Current will flow so that the total voltage around the circuit is zero.

The problem is that we normally think of a service as a 'stiff' voltage source; one where the output voltage doesn't change with current. Normally we connect loads that adjust current flow in response to voltage; the net result is that we can approximate the system as consisting of a stiff voltage source that sets the voltage, and various loads that consume current at that voltage.

This approximation falls apart very quickly. One failure is the inability to answer the question: what happens when you connect two stiff voltages sources together.

If you had two ideal stiff voltages sources, and connect them together, then infinite current has to flow. The reality is that there is no such thing as a perfectly stiff voltage source; every source has its voltage versus current characteristic; every source has its impedance.

Consider fault current calculations, and ask what happens if you have a bolted fault right at the transfer switch, shorting only one of the services. You look at the voltage of the supply, its impedance, and the impedance of the conductors up to the point of the fault, and the current contributed by any other sources (such as motors). You basically solve ohm's law for this circuit, and get a very high current since the source impedance is so low.

When the tie fails and tries to connect the two sources, you have to do the same sort of calculation, comparing the voltage across the circuit (created by the phase difference) with the impedance of the circuit (the internal impedances of the supply and the impedance of the conductors connecting everything). But now you have two voltage sources in the circuit, each with their own impedance.

You may also have to go back further, to the mechanical systems supplying the electrical sources; if the phase error is small enough than one service can 'pull' the other into synchronism; you get lots of current flow while everything is settling down, but eventually the two systems are operating in parallel.

-Jon

 

[zog]

Divide twice the system voltage by the resistance of the contacts in the tie breaker. Since the tie breaker is likely to be sort of big, and its contacts sort of big also, their resistance will be sort of small. .[/SIZE][/FONT]

A 4000A 480V breaker contacts are typically around 15-20 microhms with the proper pressure. Thats 0.00002 Ohms.

 

[zog]

You may also have to go back further, to the mechanical systems supplying the electrical sources; if the phase error is small enough than one service can 'pull' the other into synchronism; you get lots of current flow while everything is settling down, but eventually the two systems are operating in parallel.

I have had the chance to do this, destructive testing to determine the max phase angle differences that some switchgear systems could "pull in". The highest we were able to do before the massive fireball occured was a 12 degree difference. At 10 degrees the switchgear and generators were breaking mounting bolts as they tried to jump.

 

[winnie]

The highest we were able to do before the massive fireball occured was a 12 degree difference.

Gosh that sounds like fun (with appropriate protective equipment!)

-Jon

 

[ed downey]

Thank you all for helping me understand what is happening in the circuit. I believe this may have occurred at one of the projects I was on when the electrician drilled into the control wiring at the Tie breaker. Luckily no one was injured.

 

[ATSman]

Current Limiting Issue

Current Limiting Issue

Usually, when you place two sources in parallel, the control system will ensure that their voltage levels, frequencies, and phase angles are close to one other, or it will not allow the tie breaker to close. What this means is that (looking only at Phase A) as the voltage rises and falls 60 times a second, when the voltage on Source 1 is approaching its positive peak, the voltage on Source 2 is also approaching its positive peak. That way, when the two sources are connected to each other, the voltage between one and the other is very small, with the result that only a small amount of current will flow from one to the other at the moment of transition. Thereafter, meaning after the momentary transient and after things settle down to a steady state operation, the two sources will act as one, supplying the same voltage, the same frequency, the same phase angles.

But if the two sources are out of phase, it could mean that the voltage on Source 1 might be approaching its positive peak at the same moment that the voltage on Source 2 is approaching its negative peak. If the control system fails, and if it allows the tie breaker to close at that moment, the voltage across the tie breaker's contacts at the very moment of closure will be twice the peak value of the system voltage.It is a simple Ohm's Law situation. Divide twice the system voltage by the resistance of the contacts in the tie breaker. Since the tie breaker is likely to be sort of big, and its contacts sort of big also, their resistance will be sort of small. So dividing a very high voltage by a very small resistance will get you a current value in the tens of thousands, if not hundreds of thousands, of amps. Anyone standing close to the tie breaker at that moment is likely to have a very bad day.

Would just like to add that although this is the calculated Ohm's Law value,
the actual value of fault current that would flow when the tie closes would be limited by 1) available fault current at the line side of the two main breakers and 2) the short time delay pickup/ instantaneous pickup settings of the three breaker trip units or protective relays (OCPDs.)

 

[mivey]

I have had the chance to do this, destructive testing to determine the max phase angle differences that some switchgear systems could "pull in". The highest we were able to do before the massive fireball occured was a 12 degree difference. At 10 degrees the switchgear and generators were breaking mounting bolts as they tried to jump.

Any video you could share?

 

[mivey]

...the actual value of fault current that would flow when the tie closes would be limited by 1) available fault current at the line side of the two main breakers and 2) the short time delay pickup/ instantaneous pickup settings of the three breaker trip units or protective relays (OCPDs.)

You meant to say the value & duration.

 

[zog]

Any video you could share?

It was military and at a classified location so no. It was pretty cool, until I had to unload 5 fire extinguishers in it.

 

[mivey]

It was military and at a classified location so no. It was pretty cool, until I had to unload 5 fire extinguishers in it.

I can hear it now: "You know, I don't think that one will do what we want, but thanks for the demo." :grin:

 

[dbuckley]

There should be mechanical interlocks on this kind of system toprevent something really ugly from happening. PLCs are fine todo the sequencing, but mechanical is the backstop.

I can well believe that generators were braking free with this sort of mistreatment; if you have a stiff source and parallel an out of phase generator onto that stiff source then the generator will revolve to its correct phase position as quicky as it can, and this will mechanically destroy gensets.

 

[ed downey]

This happened to be two separate utility sources coming from different transformer banks. I think the tie breaker took the brunt on the damage.

 

[zog]

I can hear it now: "You know, I don't think that one will do what we want, but thanks for the demo." :grin:

The engineers from the company that built the gear briefed us before the test and said "You can't break it", we saw that as a challange

 

[wireguru]

The engineers from the company that built the gear briefed us before the test and said "You can't break it", we saw that as a challange

Wow. Saying zog cant break something is akin to saying homer simpson wouldnt eat an unattended donut.