Speed Control

N

Thread Starter

nikkumnit

What actually controls the speed when we synchronise a generator with the grid, previously say it was running at 3000rpm at no load, but at synchronisation the speed should go down but how does governing system controls the speed at that time?
 
Oh, nikkumnit,

You've fallen victim to the ivory-tower nonsense printed in a LOT of textbooks and references about Droop Speed Control and full load speed and no load speed.

There has been an awful lot written here on control.com--so much that sometimes it feels like it should be DroopSpeedControl.com. There is a helpful 'Search' function at the top of every control.com webpage that can be used unlock a plethora of information from previous posts.

But, I'm going to take a few minutes to try to directly answer your question.

A synchronous generator (more correctly called an alternator) and its prime mover must be synchronized when it is connected in parallel with any other synchronous generator and prime mover.

That means that the AC sine waves being produced by the "incoming" generator must be in phase with the AC sine waves being produced by all the other generators--which implies that the frequency of all of the other generators must be exactly the same <b>and</b> that the frequency of the incoming generator must be very nearly equal at the time of synchronization. (In fact it can be made to be exactly equal to the frequency of all of the other generators it is being synchronized to--but common practice is for the frequency to be just a little bit higher (a fraction of a Hz). However, once the generator breaker closes the speed is fixed by the frequency of the grid--read on.

Once all the permissives for synchronization have been met and the generator breaker is closed and there is current flowing in the generator stator, the magnetic field produced by the current flowing in the generator stator <b>locks the magnetic field of the generator rotor into synchronism.</b> This means that the generator rotor can't spin any faster or slower than synchronous speed (which for a two-pole generator synchronized to a grid operating at 50.0 Hz is 3000 RPM).

<b>Every synchronous generator synchronized to the same grid operates at its synchronous speed--and no faster nor slower <i>because of the magnetic forces at work inside every generator keeping the generator rotors locked into synchronism with the magnetic fields created by the current flowing in the stators of the generators which appear to rotate around the stator at a speed proportional to the frequency of the grid.</i></b>

That's what synchronism means--every generator rotor is locked into its synchronous speed with every other generator when multiple generators are operating in parallel (synchronized). One generator can't be running at 49.79 Hz, and another at 50.13 Hz, another at 48.6 Hz, another at 49.99 Hz, another at 50.0 Hz, another at 50.92 Hz--they ALL have to be operating at 50 Hz (or whatever the grid frequency is they are all connected to--nominally 50, or 60, Hz).

If any synchronous generator could operate at any speed while connected (synchronized) to a grid with other generators then why would it be necessary to synchronize the incoming generator? Why not just close the generator breaker regardless of the frequency or voltage differential and just start generating? The process of synchronization is rather involved and particular, and there's a reason for that and it doesn't change once the generator breaker closes. Not for any synchronous generator and its prime mover.

It's all about magnetism and the strengths of the magnetic fields inside the generators. There are "two" magnetic fields inside the synchronous generator--the one produced by the DC current flowing in the generator rotor windings, and the one produced by the AC current flowing in the generator stator windings. The one produced by the AC current flowing in the generator stator windings "appears" to rotate around the machine, and it keeps the rotating magnetic field of the generator rotor locked into synchronism with the apparently rotating magnetic fields of the stator preventing any single generator from running faster or slower than the speed corresponding to the frequency--when multiple generators are synchronized together.

So, it's <b>physically impossible</b> for a single synchronous generator and its prime mover to decrease (or increase) speed when it is synchronized with other synchronous generators and their prime movers. It just can't happen. Full stop. Period. (Under normal grid stability conditions.)

If a <b>single</b> generator and its prime mover was connected to a distribution network (a "grid") and no other synchronous generators were connected (synchronized) to that "grid" and the prime mover governor was operating in Droop Speed Control mode and had a droop setting of 4%, when the load was zero the machine (presuming it was a two-pole generator) would be operating at 3000 RPM to produce 50.0 Hz. As load was increased to 25% of the rating of the prime mover, the speed of the prime mover and generator would decrease to 99% of rated, and the frequency would drop to 49.5 Hz.

If the load were further increased to 75% of the rating of the prime mover, the speed of the generator and prime mover would decrease to 97% of rated, and the frequency would decrease to 48.5 Hz. But this <b>ONLY</b> happens when it's a single synchronous generator supplying a load and the generator's prime mover governor is operating in Droop Speed Control Mode.

Droop Speed Control is about how much the load changes--how much the energy flow-rate into the prime mover changes because load and energy flow-rate are directly proportional--when the error between the prime mover speed reference and the prime mover actual speed changes. It's a linear relationship and it can be explained using a simple y=mx+b function, where "x" is the error between the speed reference and the actual speed (Speed Reference minus Actual Speed in most cases).

So, those references and texts which say the speed decreases when the load increases are not being completely honest about the conditions when they say that. As you read on, you will see they are technically correct but they aren't properly stating all of the conditions for their charts and graphs.

In reality when multiple synchronous generators and their prime movers are synchronized together they operate as one "giant" synchronous generator and prime mover--again, because the magnetic forces inside each generator keep every generator spinning at its synchronous speed. And, all of the motors and televisions and lights and computers and computer monitors connected to the grid appears to be one "giant" load to the one generator and its prime mover.

When the amount of torque being provided by "the" prime mover is exactly equal to the amount required by the load and the frequency is at rated (50 Hz in your example) then the power provided by "the" generator is sufficient to supply "the" load at rated frequency.

Let's take a grid with many generators and prime movers supplying a load and the frequency is 50.0 Hz. Now, let's say that someone starts a group of very large AC motors (driving pumps moving water somewhere). And, further let's say that the energy flow-rates into all of the prime movers remained constant when the group of AC pump motors was started. The grid frequency would decrease in this case below 50.0 Hz--how much? By an amount equal to the relation of the increase in load to the total load on the grid.

It might be a lot, it might be very little--but the frequency (speed) <b>will</b> change. And, this DOES match what the texts and references say happens when load is increased on "a" synchronous generator with it's prime mover operating in Droop Speed Control Mode.

So, technically, the texts and references are correct, but the speed of one generator and prime mover doesn't change--the speed of all generators and their prime movers changes. And the speed changes as a function of the relative change in load to the total load.

Now, let's say someone synchronizes another generator to the grid and the load hasn't changed. As the new generator is loaded (the energy flow-rate into the generator's prime mover is increased) if nothing is done to any of the other generator prime movers synchronized to the grid the frequency of the grid will increase--because there's too much energy for the load.

How much the frequency will increase depends on the amount of load the incoming generator is producing in relation to the total load--so it could be hundredths or thousandths or even ten-thousandths of a Hz, which is barely imperceptible. But, the frequency WOULD increase if the load remained constant and the energy flow-rates into all of the other synchronous generator prime movers remained constant. And, this is different than the information in the texts and references, too.

Those texts and references simply DO NOT properly state all of the operating conditions and so they cause a LOT of confusion and are very misleading. They are correct--if they would properly state all of the conditions, but they don't. It's just wrong what many of these textbook- and reference authors do when they don't properly state conditions.

I don't think many of them have ever really synchronized power generating equipment or observed operation of a grid--large or small. They know maths and that's all--and they don't really properly relate the maths to the real world. Worse, they just don't properly state the conditions for their examples--which are usually just a single generator and prime mover driving some load, but they imply that this is true for all generators even multiple generators supplying a very large load.

Again--multiple synchronous generators and their prime movers synchronized together DO act as one--because of synchronism, but the authors don't take the time to fully explain the conditions for their charts and graphs and statements and maths. And, in the process they do a great disservice to many.

Hope this helps.
 
>What actually controls the speed when we synchronise a
>generator with the grid, previously say it was running at
>3000rpm at no load, but at synchronisation the speed should
>go down but how does governing system controls the speed at
>that time?

Technically the grid controls the speed. The frequency of the grid will lock in the speed of your generator (depending on the number of poles in the generator). You can't speed up or slow down any because the generator poles are magnetically locked into the speed of the grid.

So how do we load or unload at the same speed? Speed droop.

We use a proportional controller where the output (gas valve, steam valve) is in proportion to the error (actual speed - setpoint speed (usually referred to as reference speed)). If we want to increase load we raise our speed reference increasing the error, opening the valve. To lower load we decrease the speed reference, decreasing the error, closing the valve. The result is that the generator tries to speed up or slow down, pushing harder or weaker against the grid, but actual speed does not change. If the actual speed does change, say to breaker opening, or load change on a small microgrid, the valve can still respond as the error changes.

In the US we run 4-5% speed droop. This means that the difference between maximum speed reference and no-load speed is say 5%. So at your no load 3000 rpm your fuel or steam valve would be at minimum, and with 5% error (3150 rpm speed reference with 3000 rpm actual) your gas valve would be at full open. This provides for stability with enough responsiveness to hopefully prevent it from overspeeding on a breaker open. It's also used to ensure equal load sharing between units where the size of the units or grid allow.

The alternative to speed droop is "Isoch" control (Isochronous) where you open/close the fuel valve to maintain an exact speed or grid frequency. You can put a generator in Isoch to handle bus load changes while maintaining grid frequency.

Woodward publishes a decent document on the subject that goes into the detail of what I've glossed over here.

-Frank
 
I shouldn't have said the gas valve would be at full open. It would be at its maximum rated load position, which could be for example only 50% of valve open.

>In the US we run 4-5% speed droop. This means that the
>difference between maximum speed reference and no-load speed
>is say 5%. So at your no load 3000 rpm your fuel or steam
>valve would be at minimum, and with 5% error (3150 rpm speed
>reference with 3000 rpm actual) your gas valve would be at
>full open.

-Frank
 
Hi,

>Woodward publishes a decent document on the subject that
>goes into the detail of what I've glossed over here.

Could anyone please inform the name/title of the Woodward document that is referred to in this post?

Many thanks.
 
Probably this one, "Governing Fundamentals and Power Management" Woodward Reference Manual.

Best regards,
 
Oh, nikkumnit,

You've fallen victim to the ivory-tower nonsense printed in a LOT of textbooks and references about Droop Speed Control and full load speed and no load speed.

There has been an awful lot written here on control.com--so much that sometimes it feels like it should be DroopSpeedControl.com. There is a helpful 'Search' function at the top of every control.com webpage that can be used unlock a plethora of information from previous posts.

But, I'm going to take a few minutes to try to directly answer your question.

A synchronous generator (more correctly called an alternator) and its prime mover must be synchronized when it is connected in parallel with any other synchronous generator and prime mover.

That means that the AC sine waves being produced by the "incoming" generator must be in phase with the AC sine waves being produced by all the other generators--which implies that the frequency of all of the other generators must be exactly the same <b>and</b> that the frequency of the incoming generator must be very nearly equal at the time of synchronization. (In fact it can be made to be exactly equal to the frequency of all of the other generators it is being synchronized to--but common practice is for the frequency to be just a little bit higher (a fraction of a Hz). However, once the generator breaker closes the speed is fixed by the frequency of the grid--read on.

Once all the permissives for synchronization have been met and the generator breaker is closed and there is current flowing in the generator stator, the magnetic field produced by the current flowing in the generator stator <b>locks the magnetic field of the generator rotor into synchronism.</b> This means that the generator rotor can't spin any faster or slower than synchronous speed (which for a two-pole generator synchronized to a grid operating at 50.0 Hz is 3000 RPM).

<b>Every synchronous generator synchronized to the same grid operates at its synchronous speed--and no faster nor slower <i>because of the magnetic forces at work inside every generator keeping the generator rotors locked into synchronism with the magnetic fields created by the current flowing in the stators of the generators which appear to rotate around the stator at a speed proportional to the frequency of the grid.</i></b>

That's what synchronism means--every generator rotor is locked into its synchronous speed with every other generator when multiple generators are operating in parallel (synchronized). One generator can't be running at 49.79 Hz, and another at 50.13 Hz, another at 48.6 Hz, another at 49.99 Hz, another at 50.0 Hz, another at 50.92 Hz--they ALL have to be operating at 50 Hz (or whatever the grid frequency is they are all connected to--nominally 50, or 60, Hz).

If any synchronous generator could operate at any speed while connected (synchronized) to a grid with other generators then why would it be necessary to synchronize the incoming generator? Why not just close the generator breaker regardless of the frequency or voltage differential and just start generating? The process of synchronization is rather involved and particular, and there's a reason for that and it doesn't change once the generator breaker closes. Not for any synchronous generator and its prime mover.

It's all about magnetism and the strengths of the magnetic fields inside the generators. There are "two" magnetic fields inside the synchronous generator--the one produced by the DC current flowing in the generator rotor windings, and the one produced by the AC current flowing in the generator stator windings. The one produced by the AC current flowing in the generator stator windings "appears" to rotate around the machine, and it keeps the rotating magnetic field of the generator rotor locked into synchronism with the apparently rotating magnetic fields of the stator preventing any single generator from running faster or slower than the speed corresponding to the frequency--when multiple generators are synchronized together.

So, it's <b>physically impossible</b> for a single synchronous generator and its prime mover to decrease (or increase) speed when it is synchronized with other synchronous generators and their prime movers. It just can't happen. Full stop. Period. (Under normal grid stability conditions.)

If a <b>single</b> generator and its prime mover was connected to a distribution network (a "grid") and no other synchronous generators were connected (synchronized) to that "grid" and the prime mover governor was operating in Droop Speed Control mode and had a droop setting of 4%, when the load was zero the machine (presuming it was a two-pole generator) would be operating at 3000 RPM to produce 50.0 Hz. As load was increased to 25% of the rating of the prime mover, the speed of the prime mover and generator would decrease to 99% of rated, and the frequency would drop to 49.5 Hz.

If the load were further increased to 75% of the rating of the prime mover, the speed of the generator and prime mover would decrease to 97% of rated, and the frequency would decrease to 48.5 Hz. But this <b>ONLY</b> happens when it's a single synchronous generator supplying a load and the generator's prime mover governor is operating in Droop Speed Control Mode.

Droop Speed Control is about how much the load changes--how much the energy flow-rate into the prime mover changes because load and energy flow-rate are directly proportional--when the error between the prime mover speed reference and the prime mover actual speed changes. It's a linear relationship and it can be explained using a simple y=mx+b function, where "x" is the error between the speed reference and the actual speed (Speed Reference minus Actual Speed in most cases).

So, those references and texts which say the speed decreases when the load increases are not being completely honest about the conditions when they say that. As you read on, you will see they are technically correct but they aren't properly stating all of the conditions for their charts and graphs.

In reality when multiple synchronous generators and their prime movers are synchronized together they operate as one "giant" synchronous generator and prime mover--again, because the magnetic forces inside each generator keep every generator spinning at its synchronous speed. And, all of the motors and televisions and lights and computers and computer monitors connected to the grid appears to be one "giant" load to the one generator and its prime mover.

When the amount of torque being provided by "the" prime mover is exactly equal to the amount required by the load and the frequency is at rated (50 Hz in your example) then the power provided by "the" generator is sufficient to supply "the" load at rated frequency.

Let's take a grid with many generators and prime movers supplying a load and the frequency is 50.0 Hz. Now, let's say that someone starts a group of very large AC motors (driving pumps moving water somewhere). And, further let's say that the energy flow-rates into all of the prime movers remained constant when the group of AC pump motors was started. The grid frequency would decrease in this case below 50.0 Hz--how much? By an amount equal to the relation of the increase in load to the total load on the grid.

It might be a lot, it might be very little--but the frequency (speed) <b>will</b> change. And, this DOES match what the texts and references say happens when load is increased on "a" synchronous generator with it's prime mover operating in Droop Speed Control Mode.

So, technically, the texts and references are correct, but the speed of one generator and prime mover doesn't change--the speed of all generators and their prime movers changes. And the speed changes as a function of the relative change in load to the total load.

Now, let's say someone synchronizes another generator to the grid and the load hasn't changed. As the new generator is loaded (the energy flow-rate into the generator's prime mover is increased) if nothing is done to any of the other generator prime movers synchronized to the grid the frequency of the grid will increase--because there's too much energy for the
Dear Sir
You have explained the facts very nicely. Thanks a lot.
My question is now extended to, if grid frequency is lower than 50 Hz, it will keep on loading generators till generator reaches base load. After reaching base load if still frequency is lower than 50 Hz of grid, what will restrict further loading on generator.

Regards
 
AshutoshMendiratta,

There is a limit to the amount of torque the synchronous generator's prime mover can produce and transmit to the generator via the load coupling (and load gear box if used). That's the absolute limit of how much load the generator can produce--the prime mover rating.

In the prime mover control system, it is Droop Speed Control that increases the energy flow-rate into the prime mover when the grid frequency is less than the prime mover's speed setpoint. And most machines all have a Droop Characteristic, or Droop Setpoint, or Droop Regulation value, that coincides with the prime mover's rated power output. For many heavy duty gas turbines, the typical Droop Setpoint is 4.0%; for many steam turbine the typical Droop Setpoint is 5%. What this means is that when the difference between the prime mover speed reference (setpoint) and the actual speed (which is a function of the grid frequency when synchronized to a grid) reaches the Droop Setpoint the machine (turbine and generator) are producing rated power output (so-called "Base Load" for many machines). For example, if the machine were operating at Part Load (less than rated (Base) load), let's say at a prime mover speed reference of 102%, and the grid frequency decreased by 2% Droop Speed Control would increase the energy flow-rate into the prime mover such that the prime mover would be producing rate power output (Base Load) because with the prime mover speed reference at 102% and the actual speed at 98% (because the grid frequency was at 98%) the difference between the two (102%-98%) would be 4%--which is the point at which the prime mover's power output (and enefgy flow-rate input) would be at rated ("Base Load").

To be completely open and honest here, the above is purely theoretical when it comes to heavy duty gas turbines. That's because as the actual turbine (and axial compressor speed) decreases the air flow through the machine decreases (the axial compressor is spinning slower). This has an undesirable effect: It causes the gas turbine exhaust temperature to rise faster than it would otherwise increase if the machine were running at 100% speed which the fuel was being changed. This tends to limit the total power output of the heavy duty gas turbine when grid frequency decreases, which is not what one wants to happen but because of the way gas turbines are made and operate it can't be avoided.

The simple answer to your question is: Droop Speed Control limits the power output when frequency drops while synchronized to a grid. For gas turbines, there is another limiting factor, the maximum allowable exhaust temperature, which also has a (negative) effect on power output.

Hope this helps!

EDIT: When a machine is operating at it's Droop Characteristic (or Droop Setpoint; or Droop Regulation) value it is at it's rated power output (the rated power of the prime mover--not the generator nameplate power rating, the prime mover nameplate power rating). So, for example, on GE-design heavy duty gas turbines when the actual turbine speed is 100% (when the grid frequency is 100%) and the Turbine Speed Reference (TNR) is 104% (and the machine is in a new and clean condition and the ambient conditions are per the nameplate rating) the unit will be at approximately the turbine's nameplate power output (from the generator). The difference between the Turbine Speed Reference and the actual turbine speed (TNH) will be (104%-100%) 4%--which is equal to the Droop Characteristic (or Droop Setpoint; or Droop Regulation).

As ambient conditions change the power output when operating at Base Load (exhaust temperature control) at rated speed, the power output of the machine will also change. As the ambient temperature decreases below the turbine's nameplate rating, the power output of the machine will rise slightly above the turbine's nameplate rating. As the ambient temperature increases above the turbine's nameplate rating, the power output of the machine will drop slightly below the turbine's nameplate rating. Also, as turbine (and axial compressor) parts wear and get dirty (clean compressors are much more efficient than dirty compressors), the efficiency of the machine will decrease which will cause the power output of the machine to decrease for approximately the same fuel flow-rate and comparable ambient temperatures. So, there are a lot of factors that affect turbine power output, including, as mentioned above, turbine (axial compressor) speed (and air flow).

The topic of Droop Speed Control has been covered so many times on control.com, I often think the name has changed to droopspeedcontrol.com... ;-) Seriously. If you want to learn more about Droop Speed Control, control.com is the place. Many textbooks and reference manuals do not properly describe Droop Speed Control, and only talk about it in strictly theoretical terms without properly stating that's how they are describing it (that's because many of those authors have never actually sychronized a generator and prime mover and don't have any practical experience doing so or actually observing what happens before, during and after synchronization). They have their maths--and that's it.
 
AshutoshMendiratta,

There is a limit to the amount of torque the synchronous generator's prime mover can produce and transmit to the generator via the load coupling (and load gear box





Thanks alot Sir. You have explained to full satisfaction.
Next obvious question is if GTG is at 16 MW only, and we do not want to cross this limit to 18 MW or achieve base load, how to stop machine at 16 MW only. Suppose grid frequency is 48 Hz.

Thanks for extending support.
 
AshutoshMendiratta,

Human operators.

Trained, attentive human operators using RAISE- and LOWER SPEED/LOAD buttons.

If the unit was at 18 MW before the frequency declined, and it went to 21 or 22 MW after the frequency decline, the operator needs to reduce load using the LOWER SPEED/LOAD button to reach the desired load. And if frequency is unstable the operator is going to continue to have to use the RAISE- and LOWER SPEED/LOAD buttons to maintain the desired load.

To my knowledge, there is no "automatic" method of controlling load when the grid frequency is unstable.

And, Pre-Selected Load Control is NOT the proper way to control load for extended periods of time while the unit is operating. Pre-Selected Load Control is the way to change load smoothly and without having to manually use the RAISE- and LOWER SPEED/LOAD buttons. But one the desired load is reached, Pre-Selected Load Control should be disabled (with one click on either RAISE- or LOWER SPEED/LOAD).

There is an option sold for GE-design heavy duty gas turbines called 'Primary Frequency Response' which can be used for this purpose, but it still has to be manually adjusted occasionally. The option is expensive, and there have been several versions, some of which were not properly tested in the factory or in the field during implementation and do not work as intended.
 
Top