Callide Power Station 6

[hacksaw]

What is the likely cause?

CS Energy, which runs Callide Power Station Queensland AU), said there was a fire in one of the plant's turbines.

"That tripped the other three units, that then was a serious reduction to generating capacity in Queensland," CS Energy chairman Jim Soorley said.

The station was evacuated and a 550-metre exclusion zone put in place as fire crews worked to extinguish the blaze.

 

[img83934]

From the AEMO Preliminary Report all the Feeders into H24 Calvale opened at their remote ends at 1406hrs (ie. at Halys, Wurdong and Stanwell on Zone 2 Protection. This is what disconnected the faulted unit from the system. No CBs in H24 Calvale opened, None.

See this AEMO Preliminary Report:

https://www.aemo.com.au/-/media/fil...nsland-and-associated-underfrequenc.pdf?la=en

This indicates to me that All protection in H24 Calvale switchyard may have been inoperable. The following is a little out of the box thinking, but, according to NOAA models, “a combined CME is (was) forecast to hit Earth’s magnetic field during the late hours of May 25, potentially sparking G2-class geomagnetic storms.” The CME occurred on Sunday 23rd May and it takes a known time for the solar wind to arrive at the ionosphere, but they gave a broad timeframe so their forecast could be out by an hour or so. A CME induced Geomagnetic Storm could render the protection systems at H24 Calvale and maybe many other substations inoperable due to saturated Current Transformers from the DC flowing on the HV Transmission Lines.

Generator Transformers which are configured from 3 x Single Phase Transformers connected externally with a Earthed Star as a 3 Phase Unit are the type most affected by DC Saturation of the Magnetic Circuit. Callide C3 and C4 units are of this formation so they are susceptible to Geomagnetic Storms. Generator Transformer Core saturation could have caused the damage to the unit by indiction of severe 2nd and 3rd harmonics. I do not know what type of Generator Transformers are used for Callide B1 & B2.

The start of the problem as seen by AEMO via SCADA was Loss of Power Output by C4. Harmonics could have caused the shaft to fail between LP Turbine and Generator which would have caused the LP Turbine to be destroyed leaving the generator spinning on line. The piece of shaft on the floor looks like the end pointing upwards has been rubbing against the other part of the shaft spinning at a slower speed and eventually stopping. The interference between the shaft sections at the break would eventually cause damage to the bearing supporting the generator shaft and throw out the section of shaft.

There also seems to have been a failure of electrical protection systems on the generator because it did not trip from Reverse Power. The Generator 20kV Breaker did not open and the Backup Reverse Power inter trip should have opened the 2 x 275kV CBs in H24 Calvale so there seems to have been serious protection problems at H24 Calvale.

 

[img83934]

With reference to the Frequency graph supplied by ByrdJ, the the yellow, purple and red traces are the AEMO Dispatch Targets for C3, C4 and B2 Units respectively. These only change after a trader rebids the unit after it trips or when the spot price changes the target. It does not record the actual time of the unit trip, only the time the rebid entered the dispatch calculation in 5 minute intervals. The frequency trace shows when the trips occurred. Unit C4 reduced generation to reverse power at 1334hrs, Unit C3 tripped when C4 lost excitation at 1344hrs and Unit B2 was disconnected from the system at 1406hrs with the disconnection of the 275kV feeders from Halys, Stanwell and Wurdong at Halys, Stanwell and Wurdong.

I have some more information about the Geomagnetic Storm but I have lost the links to the website so I dare not include them here. Suffice to say that careful examination of the graphs shows that the incident occurred in the middle of the Geomagnetic Storm which lasted for about 8-9hours. Coincidence! maybe. Cause? The storm presence extended from Sydney to Townsville, lying directly along the path of the HV Transmission lines from Bayswater to Townsville.

If I can find the link to the information I will post it on this thread.

Ian.

 

[byrdj]

reports that Unit C3 is back up.

https://www.csenergy.com.au/news/callide-unit-c3-back-online

I would think they have an understanding of the failure to be sure C3 doesn't have the same potentional

 

[RRaghunath]

Any details as to what led to Substation busbar trip??

 

[FacEngrPE]

I notice they hired this guy Dr Sean Brady to investigate. His published investigations make interesting reading. He sounds like a good choice for a lead investigator.

 

[RRaghunath]

If so, soon we will have an informative report from Dr. Sean Brady I suppose!

 

[RRaghunath]

Now the final report is out on the incident. Join the discussion at the LinkedIn below:

https://www.linkedin.com/posts/luke...heddingpdf-activity-6852007371234013184-pbWR/

 

[byrdj]

would it be possible to provide a link to the report?

NEVER MIND. . .i SAW IT

https://aemo.com.au/-/media/files/e...iated-under-frequency-load-shedding.pdf?la=en

 

[MechinNC]

Read the above report (most of it!!)

Sounds like root cause was failure of the turbine control system while work was done on the DC system. Something about changing out a battery charger. Screwed up switching DC power, causing a loss of DC power to several systems. Generator lost excitation, steam admission valves closed, lube oil system shut down... BUT the output breakers for some reason remained closed. Maybe the breaker needed DC to operate. Without excitation, this caused the gen to motor asynchronously, and this continued for like 30-40min. Did not know these gens could motor without excitation, but I worked in other areas of the plant.

Not sure what took out the unit, whether it was the turbine overheating from blades wind-milling with zero steam flow, the gen overheating from motoring (LOTS of MVAR's), or loss of lube flow. Kind of academic sorting that, as motoring for 40min in that condition, any number of scenarios could take it out destructively.

The report also said like 15,000 error messages came in to the ops room during that period. Maybe buried in that were clues to the ops staff what was going on, but buried indeed.

The report did not go into detailed failure analysis of the machinery (unless I missed it), it was more focused on the system wide effects, which is appropriate on that level. The plant itself (and it's insurers) will handle the failure analysis of the destroyed unit. A few electricians and maybe some ops staff are likely out of jobs.

 

[byrdj]

I've only had time to read a few pages.

a motoring generator with exitation does little harm to generator, but bad for turbine (overheating rotor)

a motoring generation with out exitaction is VERY VERY bad for the generator. the motoring is from industion, this results in surface currents on the rotor. the generator slots cut for current carrying bars become gaps for the surface current to ARC accross. even after a couple minutes the wedges restaning the bars are damaged beyound reuse

 

[TugboatEng]

When the main steam valve shuts, the air ejectors for the condenser are still running and should maintain vacuum? Wouldn't that prevent overheating of the turbine rotor?

 

[MechinNC]

Even with air ejectors running, the pressure in the condenser and turbine casing is not zero. Low, but not zero. Flow is zero, so while heating is less than with higher pressure, it is still there. And the blades are big, many of them, and spinning fast. Blades grow from temp and start rubbing the casing. Rub makes more heat and more growth and thus more rub and more heat. Classic smowball effect. Then it seizes and gen rotor inertia breaks coupling. Busted coupling pretty much indicates turbine seized first (probably).

 

[byrdj]

just finished a quick read of the report. greater than 97% deals with the response of all of Australia's grid and other power producers.

even the final summary of the event starts with the event Callide 4 stops producing power.

I use to know Turbine controls and protection. I never concerned my self with the detail of generator and transmittion. that said, the vintage turbines that used actual devices (that is not a computer based control) were design to use control power (125VDC) to trip AND a means to trip on loss of control power. even the interconection from the generator used a power signal to trip and a loss of signal. independant circuits.

so I did not get a positive read on why callide 4 turbine tripped AND generator breaker not open

oh well

 

[MDEAus]

byrdj

I am not surprised this is mostly around the grid response as it is from the grid operator AEMO. The more indepth report to keep an eye out for will be CS energy's independent root cause analysis report if it is ever publicly available.
It does however provide a bit more of a clear picture on what happened just not the why.

 

[TugboatEng]

From my glance through the conclusion, the generator circuit breaker did not open as it required DC power to do so. The downstream breakers did not open because the loss of communication during the loss of DC power freeze-framed the generator output so they never recognized the reverse power.

 

[RRaghunath]

Redundant DC system is the default standard for power stations.
Unavailability of both the DC supplies is extremely unlikely and is disastrous as this incident demonstrates.
Circuit breakers with no-volt coil and circuit breakers with capacitor tripping device are available which ensure the breaker still opens even after loss of both the DC supplies.
It may be appropriate to consider the same to prevent recurrence of such incidents.