Out of phase synchronization of AC generator 2

[Robert7112]

If you have a 60-hz, 2-pole AC generator running at 3600-rpm, what is the worst case scenario for synchronizing out of phase? I've heard both 120-degrees and 180-degrees. I can't find any technical explanation for the 120-degrees. I know the windings are 120-degrees apart, but that shouldn't have anything to do with it. Comments?

 

[DRWeig]

Hi Robert,

At 180 degrees out, the peak-to-peak potential difference between your generator and whatever you're synchronizing with is maximum, so the most current would flow if you closed at that time, which would generate the most instantaneous torque and throw the pieces of your coupling the farthest.

Did you ever see the way we synchronized in the old days, with a set of three light bulbs? We'd adjust governors until the beat frequency between the two sources was close enough that the lights slowly cycled from bright to dark and back. Once this little light show was sufficiently stable, we'd time the dark period once and the next time the lamps went dark we'd wait half the period we timed and slam the tie breaker shut. It was quite a thrill, especially when the machine was big, the other source was the utility, and you missed by a few seconds....

Happy generating, hope all turns out well for you!

Old Dav

 

[aesigler]

The worst case is 180 degrees out of synch. If this should happen expect major damage possible to the rotating equipment, drive coupling, flex plates, crank shafts, rotors, base mounts, etc.

 

[dpc]

Just to clarify - the responses above are referring to 180 **electrical** degrees of phase angle difference.

 

[busbar]

...or the “six o’clock” position on a correctly operating synchroscope.

 

[jbartos]

Suggestion: 180deg is the worst case scenario for a voltage displacement angle between the same phases and 120deg is the worst case scenario between a and c phases in abc rotation. Try to draw sinusoids and measure the worst case of a voltage difference on the ordinate axis.

 

[don01]

a star to Jbartos,here I do recall standing in great fear during my apprenticeship when then alternator was put on line watching the phase lights slowly dance. It was explained to us that it could be said the 120° out of sync. was possibly worse and that any thing was to be avioded like the plague.
I suppose we could say the result of either is going to be very very nasty and the degree of damage somewhat academic.

regards
Don

 

[TheBlacksmith]

Yes, I have done it with lights, the scope and automatic system. The automatic was the worst the first couple of times, as I am a control freak and felt helpless just watching, but it was the smoothest. I saw the results of someone else paralleling two Navy machines (read as way overbuilt for their rating) about 180 out. The mechnical parts held, but the breaker being paralleled across blew its front cover off so violently that it stamped "extra" washers around all the cover bolts. Luckily, no one was standing there.

Blacksmith

 

[tomatge]

Maybe to summarize, it doesn't matter whether the angle is 120degrees or 180degrees - in either case you're well and truly in in trouble !!!

 

[jlazucena]

About the 120
In the old days when we had to manually synchronize, synchronizers had two voltmeters to see the difference in the magnitude of the voltages and a scope to see the differences in speed and angle. There were lights as a back up to the meters and scope.

The procedure more or less was:
Use the voltmeters to obtain the same voltage.
Use the scope to obtain the same speed (The needle should be spinning very slow, some times the needle would get standstill - motionless)
Check the lights to make sure that the synchronous scope was working properly
Close the breaker when the electrical angle (The same that the angle in the scope) was close to zero.

Problem: The scope could get standstill for a different phase, and since the system phases are at 120 electrical degrees, the scope would mark 120 degrees… but a careless operator could try to close the breaker (Since the lights were dim and the scope still) and would produce a short circuit at 120 degrees

About the 180
The worst case is 180 degrees out of synch.

 

[busbar]

To add insult to injury, there are both "bright-lamp" and "dark-lamp" synchronization methods.

 

[Robert7112]

These responses are great. At the suggestion of jbartos, I played with sinusoids and phasor diagrams for hours; a tough exercise for a mechanical engineer. I did confirm that 180 was the worst case, but depending on the exact moment of breaker closure, 120 degrees could be just as bad. After all the effort, I finally reached the conclusion offered by tamatge. If you are looking at the pieces afterword, its really academic. I saw the aftermath of a out-of-phase breaker closure on a 100-MW generator running at 3600-rpm. Nothing came apart, but the stator twisted about 1/4 inch and the step up transformer split the casing on two sides. Since no one was looking when it happened, we can only guess how far out of phase it was.

 

[jbartos]

Suggestion: The automatic synchronization prevents the breaker closure, when the voltage difference between the breaker open contact terminals is excessive, e.g. more than 10V. However, the automatic synchronizer has to have specified:
1. Upper voltage limit for both potential inputs, which is variable from a minimum, e.g. 110V to 140V AC accurate to within e.g. ±2% of full scale.
2. Lower voltage limit for both potential inputs, which is variable from a minimum, e.g. 90V to 120V AC accurate to within e.g. ±2% of full scale.
3. Differential frequency limit, which is variable from a minimum of e.g. 59.5Hz to 60.5Hz and accurate to within e.g. ±5% of full scale.
These are necessary requirements to establish the safe synchronization.
E.g., if there is a "180deg out of phase" one phase across one open circuit breaker contact, the differential voltage may be satisfied, e.g. equal to zero, however, if the upper voltage limit is not 110V and lower voltage limit is not 90V, then the circuit breaker will not close.

 

[peterb]

A couple of observations on jbartos' last post -
- The differential frequency accuracy needs to be a LOT better than 5%, which is equivalent to 3 Hz. A number of 0.5% is closer to the mark.
- When there is a 180 degree out of phase condition, the differential voltage measured across an open breaker pole is definitely NOT zero. It will in fact be 200%.
- In addition to the factors mentioned, the automatic synchronizer needs to be set to anticipate the breaker closing time. This ensures that the closing signal is given ahead of time and that the breaker contact closes at the instant of syncronism.

 

[ccfowler]

I've seen the effects of an automatic system closing the breakers at exactly 180 degrees due to a control wiring error. The unscheduled outage extension was significantly longer than the originally scheduled outage. Fortunately, it was all due to wrecked equipment--no personal injuries. (Injured careers would, of course, be another matter.)

It is best that out of phase synchronization be considered only in the realm of theory and conjecture. In the physical realm, any and every effort should be employed to assure that synchronizations always be as near perfect as possible. Whether it is 180 or 120 degrees out of phase, you REALLY do not want to be around when it happens.

 

[Shortstub]

Responding to Robert7112's query:

Maximum current flow for an out-of-phase closure occurs at 180 deg. Following is the mathematical proof. Neglecting resistance components, then:


Ic = (Es1+Es2)/(Xs1+Xbt+Xs2), where, in per unit,

Ic = current flow at closure.
Es1 = source 1 Volts.
Es2 = source 2 Volts.
Xs1 = source 1 reactance.
Xs2 = source 2 reactance.
Xbt = reactance between them (eg, transformer).
Xt = sum of above, reactances.

Let theta equal the angle between the sources, so that Delta(E) = the vector sum of Es1+Es2. Then holding Es1 at 0deg, and Es2 is varied from theta equals 0 to 180deg, yields the following relationship:

Delta(E) = sqrt[(Es1^2+Es2^-2xEs1xEs2xcos(theta)]/Xt

Simplyfing for the case where Es1 and Es2 magnitudes are equal, and using the trignometric identity, the equation for Ic becomes:

Ic = 2xsin(theta/2)/Xt, and

Imax occurs when theta = 180deg.

 

[Metalguy]

1. 7000+ HP diesel-generator thru one set of master & slave con rods thru side of block during full-load test.

2. Generator ripped out-of-synch rather violently.

3. Inspection of gen. pole pieces revealed one which had bend all the anchor bolts, yet all pole pieces had seen same decel. rate.

4. Out-of-synch REALLY affected one pole piece!

 

[Shortstub]

Responding to "metalguy's" observation. Not all phases have equal electrical current flows.