Synchronization and Breaker Angles 1

[drchaos]

I have been reading some of the old threads about synchronizing generators, and the necessity to have the angle as small as possible (<20 degrees at least).
Do any similar standards apply to closing breakers on the rest of the transmission system, not necessarily near generators?
Kundur has a rule of thumb that as long as the delta P of any generation nearby is less than 0.5pu, then the angle is ok.
Does anyone have any similar rules of thumb about angle differences / voltage difference for reclosing generally?
Is 45 degrees ok? Is 60 ok?

 

[slavag]

Hi.
More or less same rules as for generator synchronization:
1. dU about 5-10%
2. dphi about 10-15deg
3. dU about 0.05-0.1Hz

Best Regards.
Slava

 

[drchaos]

Thanks Slava. I read some IEEE survey from 1984 on auto-recloser settings, and a lot of the utilities surveyed would allow angles of 45 degrees or higher, especially at say 400kV or higher.
Do you know is it less onerous on the system to allow larger angle differences at higher nominal voltages?

 

[slavag]

Hi.
Yep. it's another story.
Im have not expirience with 3-ph AR for such voltages.
In my area used only 1-ph AR.
But, I can say, what I know.

Isn't depend directly on voltage level.
It's stability system issue and, very important point, on the dead AR time. I see once Areva settings for the 330kV lines- it was 40deg. and slip is about 100mHz.
This is a point, SLIP.
You setting will depend on:
1. if it connected parts of synchrinaized stable system, in this case window of SC is small.
2. Long dead time ( slow AR), or maybe short lines.
3. Short dead time AR but with high slip, maybe possible say, connectionof asynchronized systems.
Next point, it's supervision time,it is include CB closing time.

Those parameters are calculated together

slip=(2xdPhi)/(ts x 360deg)
ts is SC supervision time. Now, please take in account, for case 1 slip about 50mHz, for case 2 is about 100mHz, for case 3 is about 200mHz and supervision for case 1 about 1sec, for case 2 about 400ms, and for case 3 about 100-200ms.

Now you can calculate what is a dPfi setting.

Sorry, it's only theory, but, hope is help. I havenht any documentation.

Best regards.
Slava

 

[drchaos]

Thanks Slava, That's very useful information.

We are looking at the impact of different angle settings on generators - how much does their power output wobble when breakers are closed. In particular, management are looking at the synchro-check settings when energizing blackstart paths, and synchronizing islanded subsystems. They want to know if there's any issues using the same synchro-check settings in normal operation and in an emergency black-start situation.

 

[slavag]

Hmm.
Emergency black start is another issue, actually, you will close CB on the dead bus/line, or Im not understand this situation. Yep!, problem is a islanded subsystems!!!
From my point of view, need check stability generator and CB data for such asynchronies closing.
From my point of view, possible use other settings for synchro-check.
Try found something in manuals of Areva-P143, ABB-RE_670, Siemens relays-recommendation of setting for asynchronies sytems.
Maybe, check moment on the generator, it be less ( twice) than 3-ph bolted fault.
next criterion, maybe, level of generator current, for example: not more than twice of nominal generator current.

If you found something on the issue, please attac here link, it's intresting for me too.

Best Regards.
Slava

 

[m3ntosan]

drchaos,

The allowable power angle is highly system specific. An allowable value is impossible to provide as the tolerances of different systems and different areas of the same system vary widely. Some systems may accept no more than a 20° angle while others may close CBs with 70° angles during restoration conditions.

I don't think you can use the same settings for normal operation and restoration condition. In restoraion case you will end up with high angle diff and you'll need to switch off the synch-check and do the job manually with eyes on the synchronoscope. In restoration you may allow greater differences than in normal operation.

Closing with high power angle diff, you'll get a large power flow once the CB is closed.

If the voltage magnitudes are not closely matched, a large Mvar flow appears across the CB as it is closed. The large Mvar flow can cause sudden changes in voltage and protective relay operation. The allowable voltage magnitude
differences across the open CB are system specific.

If the frequencies on either side of an open CB are not matched prior to closing, a large MW flow appears across the CB as it is closed. The large MW flow is a response to the initial frequency difference as the system rapidly seeks to establish a common frequency once the CB is closed.

 

[ScottyUK]

A possibly bigger problem under black start conditions, certainly for gas turbine sets, is the size of block load you apply. Any step load increase greater than 15-20% will likely result in a compressor stall and a unit trip, which is scarcely what you need in a black start. You will need a good model of the machine(s) in question because the behaviour is a complex interaction between the network and the rotating masses of the turbine and generator, the AVR system, the governor system and the PSS.


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

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

 

[m3ntosan]

I guess there are many problems regarding black start condition, but that's not the topic
Anyway a gas turbine still has better frequency response than a steam turbine so theoretically should be capable of picking-up relatively large blocks of load with an acceptable frequency dip.
However in a restoration condition the system operator should apply small load blocks. A practical example is: for 500 MW unit with 300 MW spinning reserve available, you should restore load blocks with a maximum around 25 MW.

 

[ScottyUK]

At risk of dragging this thread even further off topic, "...a gas turbine still has better frequency response than a steam turbine so theoretically should be capable of picking-up relatively large blocks of load with an acceptable frequency dip." may be true, but the frequency drop caused by a large block load onto a machine at sync idle will cause the compressor discharge pressure to drop faster than the governor loop can increase the power turbine output to maintain frequency, and the unit will trip. We invested a large amount of money modelling this scenario with one of the major GT manufacturers, and the outcome was that even with perfect governor servo response - something which we came close to achieving - the unit would still trip with a step load greater than about 20% of rating.

The idea of a 500MW unit with 300MW of reserve in a black start situation is a flight of fancy: no one builds 500MW gas turbines, and conventional thermal plant isn't a realistic option for black start because of the internal auxiliary loads. The ST on a single-shaft combination won't be contributing much, if anything, to the unit output until the GT has some load on it.


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

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

 

[m3ntosan]

I was thinking to an early stage of restoration process (islanded operation), not to a start from scratch situation (when you'll probably use HPP or help from neighbours), and not looking particulary to gas turbines, I guess it's a rule of thumb for all types of generation to go with small load blocks in such situations (except maybe low-head hydro).

What do you mean by "conventional thermal plant isn't a realistic option for black start because of the internal auxiliary loads" ? To try and start from cold with own diesel generation? Conventional thermal plant and nuclear power plants usually have good capacity to isolate on internal auxillary load, and operate like this for hours, so maybe you don't need to start from scratch.

May you grow up to be righteous, may you grow up to be true...

 

[ScottyUK]

A true black start is to bring up a grid from dead, not from a state where plants have islanded off and are sustaining themselves. Typically a big diesel set will establish auxiliaries for a light aero-derivative turbine like an LM2500 or LM6000, which will in turn provide enough internal power to the station for a heavy frame gas turbine to achieve cranking speed and accelerate up to sync speed.

It would be 'possible' for something like an LM6000 (or group of them) to establish auxiliaries for a thermal plant, but the start-up times for a big thermal station are too long to be of practical use in re-establishing a grid within a reasonable timeframe. Gas turbine stations are well suited to the rapid response needed, but have perhaps the worst ride-through capability of any prime mover as frequency falls. In an islanded situation they are vulnerable to large step loads because there is very little stored energy in the form of the rotating masses of other machines on the system to dump energy into the system as they slow down. The energy stored in the rotating masses of the machines provides the primary buffer against system collapse under a step load until the governors have a chance to open up to pick up the load in the longer term.

I think we should start a new thread if we continue this diversion!


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

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

 

[m3ntosan]

I guess the inertial stage is the 2nd buffer, followed by the governors response (primary response/reserve in ancillary services). The first buffer is the electromagnetic energy - The MW the generator supplies (sustained for less than a second) does not come from the turbine or inertia but rather from energy stored within the magnetic field of the generator.

Are you familiar with the ancillary service in the UK? It looks a bit different than the UCTE/ENTSO one (or maybe just the terms) - maybe that's the chance to move this in to a different thread!

- Primary response/reserve
- Secondary response/reserve
- Fast tertiary reserve
- Slow tertiary reserve

There's a wikipedia article (find below), but it looks a bit messy to me. And no straight info on National grid site either.


May you grow up to be righteous, may you grow up to be true...

 

[m3ntosan]

Almost forgot the purpose of this thread

"Does anyone have any similar rules of thumb about angle differences / voltage difference for reclosing generally?"

Here's a rule of thumb:

1. 10% difference in voltage magnitude for 400 kV and 20% for <220 kV
2. voltage angle difference less than 45 deg
3. loading on the equipments (OHL, underground cables, autotransformers, transformers) part of the open loop/mesh/ring to be closed, less than 70% of the rated capacity (except those equipments you know the load will decrease after CB closing)
4. the lowest voltage in the open loop/mesh/ring to be more than 95% of the rated voltage for 400 kV and 90% for <220 kV.

Some settings from my utility:

- 35 deg angle difference for automatic reclosing synch check (old non-digital relays)

- BCU: 25 deg angle, 10% voltage magnitude and 0 frequency difference.

- digital protetion relays: 15 deg angle, 10% voltage magnitude and 0.05 Hz frequency difference. More restrictive settings used for synch of separate/different areas.

That's how you find out if you have an open loop/ring/mesh or separate/different systems/areas.


May you grow up to be righteous, may you grow up to be true...

 

[esee135]

Regarding MW swings during closure, I've seen 0.35pu (with fast decaying swing) of generator MVA swing on a 'good syncronization' or within syncrocheck spec, and issued sycnro-close pulse. I've seen 0.5pu of generator MVA on a manual sync as well. It's quite interesting to watch with a strobe on the rotor.

 

[eddiet1]

The attached paper on "Automatic Synchronization for Generators and Tie Lines" is the best reference document I've seen on this subject. For many engineers, it is the design basis for synchronizing system. Synchronizing systems are ususally designed and installed for decades, so it is prudent to document your assumptions for future engineers to troubleshoot our designs (unless you plan on living forever). Hope this helps.

 

[waross]

Welcome to Eng-Tips eddiet1.

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

 

[drchaos]

Thanks for the paper eddiet1, and to the rest of you for responding with your experiences and wisdom! It's not an area that is covered well in any text books so far as I can see [ though the book "Power System Dynamics" by Machowski, Bialek and Bumby does have some of the maths / physics behind the synchronization process.]