ST Power Generator

We have 2 PGs 40MW each one,connected to the grid. If island mode is selected, 52G breaker open (or trip), PG1 goes to speed control, and PG2 goes to droop control (KW).

What happened is that we want to disconnect from grid "due to some planned job of public grid" (normally to open 52G). But by mistake or without knowledge, someone has opened 52L breaker (the 2 PGs didn't see that they are in island mode, still running in parallel mode).

After few seconds (8 sec) of the opening of 52L, under voltages of many levels of voltage have been observed (alarms in DCS).Then 52G tripped out by under voltage protection.

The question is: why this under voltage occurred?

Please let me know if you need other information.

Thanks in advance.

Without being able to see the schematic drawings for the generator protection and DCS inputs there's not a lot we can say for sure.

We also don't know what the island load was at the time of the event, or what the power factor of the island load was at the time of the event. It's possible that if the operator(s) had manually switched one unit to Isochronous Speed Control mode (or selected island mode) that the opening of 52G might not have occurred.

It could also be that when the plant detects it is in island mode (separated from the grid) and switches one governor to Isochronous Speed Control Mode (leaving the other in Droop Speed Control mode) that it also sends some kind of signal(s) to the generator excitation control systems ("AVR") to boost their output, and this didn't happen.

Again, without being able to see the schematic drawings and understand the configuration of your plant (all plants are NOT the same), it's difficult to same much more with any degree of certainty.
Dear Sir,

Thank you for the reply. I will try to give you more details:

We were in parallel mode PG1 = 6MW (second warm up timer on), PG2=28MW and the grid=1MW. We want to disconnect from grid (manually). The normal procedure is:

1. Open 52G then automatically PG1 will control speed (frequency) and PG2 will control power. (The logic sees only 52G).

2. Open 52L: to let public grid's technicians do their works (outside our plant).

3. After completion of works: Close first 52L then 52G.

4.PG1&2 return back to parallel mode.

Unfortunately these steps haven't been respected. The operator, by mistake, opens 52L first, so the PGs didn't see that they are in island mode, because 52G is still close. After 52L opened (around 8 seconds) the following events occurred by chronology:

1. Under voltage alarms appeared in the whole plant, in different levels of voltage. (Before 52G trip).

2. 52G tripped out by under voltage protection.

3. PG1 breaker tripped out (if the 52G open and the PG is in 2nd warm up the PG breaker should open).

4. Many motors tripped out by under voltage protection.

5. PG2 tripped out by over speed protection.

6. Of course black out.

My concern is how (why) the under voltage occurred after the opening 52L and before the 52G trip.

Thanks in advance.

PC Load Letter


I am going to make the following assumptions based on experience. Correct me if I am wrong in any of them

A. Your machine has a breathless exciter system with a independent AVR

B. The AVR has like typical AVR's has Voltage control / PF control / MVAR control mode

C. At the time of the trip the plant was importing VAR's from the grid.

D. At the time of trip the governors were in load control or MW control mode and you were exporting power to the grid.

This is a possible explanation of what could have happened.

1. At the time the grid breaker was opened your plant was exposed to a huge VAR defecit. This large VAR deficit is the reason for the voltage drop. At this time your AVR's should have kicked in and regulated the voltage. Either they were not properly tuned or they were not in voltage control mode. (If its connected to the grid you can operate your AVR in MVAR or PF Control - this needs to be switched back to voltage control if the grid opens). This may not be done correctly either.

2. Your governor is either properly not tuned for load rejection or stayed in MW control mode. what is the step load throwoff that the machine experienced. E.g lets say your machine was in MW control of 35MW and the plant load is 30MW. Now you have 5MW export to the grid. When the grid opens there will be a raise in speed due to the 5MW load throwoff in the machine. if the machine was just in droop mode then speed will stabilize a bit above based on the droop percentage. However if it stayed in MW control, the machine will see the 5MW throwoff and try to keep on increasing the fuel to get it to 35MW. That's not possible as the load is only 30MW. thus the extra fuel added to the machine keeps on increasing the speed (not the load ) till the machine trips on overspeed.

I will check these things
1. What mode did the governor operate on when it was paralled to the grid ?
2. Did the mode change once the grid opened ( it seems like it did not )
3. What mode did the AVR operate on when it was paralled to the grid
4. Did the mode change once the grid opened
5. What was the MW and MVAR export/import from the grid at the time of the trip.

I am sure this event can be avoided in the future by proper SOP's. Anyway that's my take on the issue. I have to go and re-insert a rotor, bye bye for now.
Dear Sir,

Thank you for your time.
Hope that the rotor was inserted without problem.

please be informed that we do not export power to grid, we only import to our plant (we are not allowed to export).

Yes we have AVRs. and we produce the MVAR, otherwise we will get penalties from the public grid. We should open the plant breaker to let our PGs go to island mode control (PG1 to frequency, PG2 to power control). As per the logic when the PGs see that the plant breaker opens, they automatically go to island. But the operator, opened the grid breaker (52L) first. The power taken from grid at that time was almost zero (0.5 MW); PG1=6MW and PG2=28MW. The PGs will never go to island control mode since the plant beaker (52G) is still close. So, the PGs were actually disconnected from Grid (but if I can say: they don't know that they are disconnected). So, after grid breaker opened (52L); and plant breaker is still closed; under voltage alarms appeared.

What are the reasons of these under voltages (After 52L open and before 52G trip)?

PC Load Letter


Sorry it's outage season here and things are a bit hectic. Without looking at the trends and knowing the answer to the questions I posted above, it's hard to ascertain what's the exact reason. But one thing is for sure: Voltage dips without faults in the system are always caused by VAR deficit, and it's usually caused by maloperation of the AVR.

If you are comfortable please share your e-mail; and I can send you a few old trip analysis that I have done in the past.

Bye for now.
Dear Absado,

I am Formely AVR GENERATOR PROTECTION PANELS commissioning engineer.

I would like to know how about the Configuration of the plant (electrical single line diagram).

The possible cause of this undervolatge trips can be various.

Looks like one of my experienced issue on Frame 9e Unit.

The cause of the problem was that electrical Breaker (located in the MCC) for sources change over (normal to stand/by) was not operating properly (time delay to adjust) thus causing no operating of bossting flashing for the excitation as CSA mentionned in one of his post.

How could somebody can open 52L while units running?
It can be very dangerous!

I don't have all the diagrams of your plant, but lets begin by searching on the PROTECTION RELAY device the events alarms trips logs to have a better picture of what is going on at your site.

Dear James;

thank you for the reply.

When the operator opened the 52L (grid breaker), the power taken from the grid was almost 0 MW (0.5 MW more or less).

Could you please let me know how can I share the single line diagram?

<b>Moderator&#8217;s Note:</b> At present, you cannot post diagrams directly to You can upload your illustration to a download site and post the url here and share it.
Dear Absado,

You can drop me an email with the document. I will study it and try to give you my opinion on such issue that you got at site.

Also, Did you get some trends (during this event) or views from other devices such Protection relay (s) or AVR ...,? By this way you are be able to know if you selectivity regulation & protection has been respected.

The problem I have been faced was because of very unstable grid causing MVAR fluctuation and Excitation boosting flashing was not operating according properly due to thesettings Undervolatge relay (of Undervolatge and time delay had to be adjsuted during tests) for the Air isolated Switchgear breaker located in the MCC thus making a big peak of MvAR and KV for a short duration time and giving trip to the unit.

Hope this information may help.

<b>Moderator&#8217;s request:</b> Can you please post here what you find and the resolution of this issue. Many people search our archives looking for information and solutions to problems. Thank you in advance.

I look forward to hear from you soon,
My email is: [email protected]