Synchronizing Torque

[raisinbran]

First - I hope that everyone is enjoyng the Holiday Season and spending some relaxing time with family and friends.

We manufacture gearing and couplings to coupler diesel engines to three phase generators. The generator size is 2- 2.5 MW. The generators are 60 Hz, and the output voltage ranges from 480 V to 4.16 kV. These generators are connected through a circuit breaker to the utility and/or other similar generators.

I have never worked with generator protective relays, but I imagine that the circuit breaker control scheme also includes a relay to synchronize the generator to the utility before the circuit breaker is allowed to close. My questions are: What are the proper relay settings for phase angle difference, voltage magnitude difference and frequency difference?

If the circuit breaker is allowed to close outside of these limits, I imagine that there would be a large transient torque that would try to speed up or slow down the generator. How would I calculate the value of that torque.

Here is wishing everyone a safe and prosperous 2009.

Regards,
Raisinbran

 

[7anoter4]

The generator protection which will trip the breaker does not control the synchronizing process.
There are combined generator driving and control systems which may include the protection also. The generator synchronizing process control is acting on governor mounted on the diesel it self in order to fit the frequency and on the generator excitation in order to control the voltage. When the control, checking the frequency and the voltage, states they are in the setting range then the synchroscope checks the angle and taken into consideration the breaker connection time the breaker is given a signal to close.
I don't remember using a gear box between motor engine and generator. Usually there are so constructed that they don't need adaptation. But coupling is always provided. I think the coupling is provided with a kind of mechanical protection which would break before the damage [some thing like a fuse].
The synchronizing torque has to be minimum possible and less than the maximum permitted torque, of course.
But if something could go wrong with the control the torque may be very high and the breaker could not trip in time as the tripping time is about 50-100 milliseconds and the maximum torque would occur in 8-10 milliseconds only.
In this case the generator acts as in shortcircuit case and the maximum torque could be =1.3*Voltage^2/x"d [in p.u.] Voltage [in p.u.] could be 1.1 up to 2, at least.

 

[7anoter4]

I forgot to mention: there are test devices that may test the synchronizing settings and calculate the permissible maximum error of frequency, voltage and angle, taken into consideration the breaker connecting time.
As the breaker function depends upon many factors as coils and breaker mechanism which may corrode or may present a lack of lubrication, the connecting time could be longer, in time. The setting test has to be done then frequently.

 

[waross]

The worst case for maximum torque and mechanical damage in an out of sync closure will be at a phase angle of about 90 deg.
The voltage will be about 70% max.
The worst case for electrical damage will be at 180 deg. The torque will be zero and the voltage will be 200%
Size the coupling for the load, and use protective devices to avoid closing out of sync.
I have seen square drive keys sheared off in generator couplings due to out of sync closures.
I have seen a motor stator spin in the case when the motor was plugged to stop. I imagine that the same thing could happen to a generator.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rmw]

Synchronizing out of phase not only stresses mechanical components (shears coupling bolts and/or keys, breaks shafts, etc-none of it good) it plays hell with the generator stator end windings as well. They 'tweak' like butterfly wings flapping and do great harm to the blocking and bracing and insulation as well as the end winding conductors themselves. (Sorry EE's that's about the best a ME could do to describe what the end windings go through on a huge transient like a serious mis-synchronization).

Don't calculate it; just don't do it. It is one of those "uncool" things that just isn't done in a power plant and when it is done, usually there is great grief to pay including loss of employment. It ranks right up there with serious mis-steps like hitting the ignitors with a furnace full of fuel. Your name will be talked about in the plant for years.

Then there is the poor bozo who from time to time closes the breaker with a generator stationary.........Some service provider company or OEM Service departments is going to get rich off of that one.

rmw

 

[ScottyUK]

Most machines are braced for a three phase fault across the terminals. An out of sync closure under worst case conditions is likely to be about double that in terms of the forces on the machine. Something will likely break in the engine.

For ballpark figures for a synchroniser I'd look at something like 10% voltage, 20[°] phase, 0.1Hz [Δ] freq as the settings for the backup check sync relay and expect the synchroniser to be achieving more like 2%, 5[°], 0.05 Hz. Those figures are from a much bigger machine, and Bill is going to tell me I am far too conservative. 'Conservative' was one of those things which prevented bits of turbogenerator being scattered across Teesside.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[slavag]

Scotty, first of all, I wish to you a good luck in your new job. Im very happy for you. ( btw, thanks to RMW for this information)
Second, xmmm, Im more coservative:
1. for autosynchronazing 1-2deg, 2% delta voltage and 0.02Hz delta.
2. For back up synchrocheck relay 7-8%deg, 5% delta voltage
and 0.05Hz delta.
This is a our setting in last "first synchronazing" for the 150MW block.
Happy New Year.

 

[waross]

Hi Scotty;
Despite our differences of scale, I don't have a problem with your settings. I have seen very small (350 KVA, 600 KVA) machines survive greater phase angle differences BUT I didn't like it and took action to prevent future events. The manual syncs are now generally within less than 15 degrees and the couplings don't break anymore.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ScottyUK]

Hi Slava, Bill,

Looks like the modern synchronisers are better than the ones I've been working with until recently. I've only seen a couple of rough synchronising events when the synchroniser was behaving poorly but still within the limits of the check sync relay. The gas turbine was not happy, presumably because of some effect on the compressor as the machine was thrown into sync with the grid. There was a lot of vibration right down the machine for a few seconds then things settled: the couple of seconds seemed to last about an hour when we were in the local control package!

'Conservative' is definitely the way the go: if the synchroniser and the AVR and governor are all capable of consistently meeting tight tolerances then the settings should be closed in to take advantage of that. My settings for the backup were pulled from the old REG216 relay, and the synchroniser behaviour is from some tests years ago on the Brush PAS-510. Things have moved on since they were designed - thanks for the up-to-date information. It might be of use in my new job.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[GTstartup]

That's interesting Scotty, I didn't know that the ABB REG216 had a synch check feature. Live and Learn.

 

[slavag]

Hi GTstartup.
REG216 have all, isn't standard application, SC in REG216 for the big generator. Usually we use autosynchronaizer with synchcheck functionality. As I remember, Scotty's REG216 is old one, about 20years, autosynchronaizers was old too. Today we use SYN5302 ( of ABB, for example) with 7set points and additional synchrocheck functionality.
Best Regards.
Slava.
Sorry, REG216 is my favorit for the genrator/unit protection

 

[ScottyUK]

Hi Slava & GTstartup,

The REG216 I'm thinking of is early 90's vintage. It's a damned good relay, better in many ways than the newer types. The check sync function had user-definable digital inputs to define the relay behaviour with live / dead line and live / dead bar. The newer ones do it automatically and take away some of the flexibility.

The old PAS-510 synchroniser was a cute piece of analogue design but really has been superceded by later uP based types. It is much easier to repair though, and that really does matter when one fails during holiday season!


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[rmw]

Scotty,

I look at these events purely through mechanical (engineer) eyes and since your (former) 701's were single shaft machines, if you sync'd her with the machine turning slower than the grid, when it was suddenly sped up to match the grid speed, it tried to climb up the bearing walls and shook and shuddered as it fell back and got herself settled back onto the oil wedge. If it slowed down, it tried to dig itself into the wedge and complained until it re-established the proper oil wedge again. Not good things to be doing with a machine that size. No wonder she complained.

rmw

 

[ScottyUK]

You ever been to that site? You know far too much about our machines! The explanation makes sense. It also supports Slava's opinion that the backup check sync relay was probably set with too generous a window and should have blocked the synchroniser from issuing a close command.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Hoxton]

I do not have access to the current IEC / BS standards for ac generators at the moment, but unless they have changed recently, then they call for the design to meet only a three phase short on the machine terminals. Faults further away having a reduced effect due to cable and tranaformer impedance etc.

Forces due to faulty synchronising are not a design criteria. I thing this is because the forces are many times higher and to design for this would result in the generator being large and expensive - the foundations and prime mover shaft and coupling would also have to be stressed accordingly!

I have a foundation loading calculation for a german gas engine - they calculate for a phase - phase fault only (I guess this is the worst "real" event).

So use the best synchroniser and check synchroniser you can afford! Commission and maintain them well. Also, for mains parallel operation, use a mains failure relay, as a short term mains interuption can lead to a faulty synchronising event when the mains returns.

Remember it is the smallest generator which comes off worst.... my 1MVA generators are never going to frighten Scotty's 701's!

 

[ScottyUK]

Hi Hoxton,

I am not convinced that a small generator will fare worse than a big one, quite the opposite. The big machines are designed very close to the limits of the available materials and have smaller design margins than a small set. With turbo machinery it is often the LP turbine of a steam set that is most vulnerable to a massive transient torque because of the highly stressed long slender blading. You don't need much flexure in the blades before moving bits start hitting stationary bits. It's all relative - a big machine is still quite small relative to the combined infeed from the grid and other generating units at a large power station's transmission substation. In a test of strength between the grid and a machine there is only one winner.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Hoxton]

First returning to the post, it seems industry standard to design for short circuit torque. The generator manufacturer will give you this value.

Scotty, my comment about faulty synchronising affecting the smallest unit worst is generally correct, but is a matter of scale. In the UK we have around 40,000 MW or generation capacity connected at any one time. My 1MW generator is going to come off worst. As the ratio of the two systems rating comes closer, then even the larger unit can be damaged. I know that larger units can be more sensitive to sychronising phase angle error than smaller units.

Even so, when I worked for Scotty's favourite generator manufacturer, some years ago, we had a case in the middle east where we had several sites each with a 1MW and 2MW diesel set in parallel, feeding factories, with a grid connection in parallel also. The 1MW generators were ours, the 2MW were from another UK manufacturer. After a while we had core packs spinning in the stator frames, pulling the phase connections out of the endwinding! The radial dowel pins which secured the core pack were shearing.

Invesigation revealed that there was no mains fail relay, overhead line faults caused the supply to be interupted and auto reclosers were restoring the mains, out of synch with the gensets.

There were no reported failures of the 2MW generators. Why?

1) Different construction - 2MW stator core was built into the stator frame, not a separate core pack wound outside the frame and inserted as a pack. A stronger, but more expensive construction.
2) The 1mw took the brunt of the force as it was the smaller unit.

Of course, we expected the 2MW endwindings to fail later, due to fatigue, but by then the protection had been upgraded.