Hi there I know there’s a bunch of speed droop forums on this site but I have a few questions pertaining what the pictures I’ve posted mean.

I think I know how isochronous works and speed droop works.

Isochronous is basically speed droop set to 0 to constantly adjust output to keep unit frequency at the desired setpoint.

Speed droop is basically how far the frequency is out of wack to make adjustments to fuel and it’s normally at 5%.

Now what I don’t understand is with the photos I’ve uploaded, speed regulation changes the wicket gates based off of the speed adjustment control knob error and adjusts generation to meet that speed adjustment set point but where is frequency? Does this governor not control the wicket gates when frequency goes up or down it just changes it based off of what the operator sets the speed adjust?

Now their definition of speed droop is looking at wicket gate position instead of mega watts and still looks at speed adjust error but why not frequency I’m very confused. Doesn’t it need to look at frequency instead of speed adjust error that’s literally controlled by the operator?

 

I should add this is a Woodward electronic governor and unit is 60MVA

 

Droop Speed Control is about how much the energy flow-rate input to the prime mover changes for a change in the ERROR between the machine speed reference and the actual machine speed.

When a prime mover and generator set is synchronized to a grid with other prime movers and their generators the speeds of ALL the machines SYNCHRONIZED to the grid are FIXED by the frequency of the grid with which they are synchronized (connected to; parallelled to/with). It's IMPOSSIBLE to have any machine producing any power at a frequency that is different from the grid frequency. There is no "smoothing" or correction device that will turn 59.7 Hz and 60.5 Hz and 59.9 Hz and 60.3 Hz into 60.0 Hz. EVERY MACHINE SYNCYRONIZED TO A (WELL-REGULATED) GRID WILL BE PRODUCING POWER AT THE SAME FREQUENCY AS EVERY OTHER MACHINE ON THAT GRID. PERIOD. FULL STOP. That's the way AC (Alternating Current power generation works.)

If you don't believe me, watch the speed of a droop machine paralleled to/synchronized to a grid with other synchronous machines and their prime movers as the operator changes the load of the machine (USUALLY by changing the machine's speed reference). Since the amount of energy flowing into the prime mover is a function of the difference between the machine's speed reference and its actual speed, and since the machine's actual speed is 'FIXED' by the frequency of the grid with which it is synchronized changing the machine's speed reference changes the ERROR between the speed reference and the machine's actual speed which changes the energy flow-rate into the prime mover which changes the POWER produced by the prime mover which changes the power produced by the generator being driven by the prime mover. And, one of the main considerations of any AC power grid is its frequency--as you have noted. So, the grid keeps the generator and its prime mover spinning at a constant speed (because the frequency is relatively constant) and to change the power being produced by the machine one changes its speed reference.

For MOST governors (but NOT all--and Woodward is guilty of this with several of their governors--when the speed reference is 100% and the grid frequency is 100% of rated there is ZERO difference between the machine's speed reference and its actual speed. So, therefore, even if the generator breaker connecting the machine to the grid is closed the electrical power being produced by the machine is zero, because the error between 100% and 100% is zero.

Increase the machine's speed reference while the generator breaker is closed to, say, 100.25%, and the error between the machine's speed reference increases to 0.25% (100.25% minus 100%) and the electrical power being produced by the generator increases because the energy flow-rate increases into the prime mover.

For your machine, with a 5% droop regulation setpoint being rated at 60 MVA (presuming unity power factor; zero MVAr), 0.25% of 5% is equal to (5% divided by 0.25%) to 20, and 60 MW divided by 20 equals 3 MW. So, increasing the machine's speed reference by 0.25% will cause the machine's electrical power output to increase by 3 MW.

Another way to say the same thing is that for every 1% change in the error between the machine's speed reference and its actual speed the electrical load of the machine will change by 12 MW.

THAT'S WHAT DROOP SPEED CONTROL IS PRIMARILY ABOUT. One CANNOT change the machine's speed/generator's frequency--its being "controlled" (fixed) by the grid frequency (great big--HUGE--magnetic forces at work inside the generator are literally limiting the speed as a function of frequency).

Be VERY CAREFUL when reading Woodward Governor Control instruction manuals and white papers--they are full of errors and OFTEN DO NOT PROPERLY STATE ALL OF THE MACHINE OPERATING CONDITIONS WHEN ATTEMPTING TO DESCRIBE DROOP SPEED CONTROL. It can be VERY MISLEADING. Even textbooks and reference books neglect to describe IMPORTANT operating conditions when attempting to describe Droop Speed Control--which is where SO MUCH of the misunderstanding of Droop Speed Control originates.

It's actually a very simple concept, but because there are TWO variables involved (machine speed reference AND actual machine speed) and because Droop Speed Control is also a very important part of grid stability (maintaining power output when grid frequency is abnormal or unstable) people just get SOOO confused about Droop Speed Control. Again--PRIMARILY Droop Speed Control is about how much the energy flow-rate into the prime mover changes for a change in the error between the machine's speed reference and its actual speed. EVEN WHEN Droop Speed Control is contributing to grid stability when grid frequency is abnormal (when grid stability is oscillating, EVERY MACHINE on the grid is going to experience similar frequency excursions--because they are all SYNCHRONIZED together on the same grid. Synchronism is a VERY important and overlooked word in AC power generation; yeah, we all say the machine is being "synchronized" but IT'S NOT just about the act of closing the generator breaker to connect the machine to the grid--it's what happens AFTER the generator breaker is closed, also.

I'm not going to read all the Woodward verbiage; I've done that for too many years trying to explain Droop Speed Control tp people. Woodward has a different philosophy and a DIFFERENT understanding and way of explaining Droop Speed Control and of implementing it into their governors (prime mover control systems). It's not wrong, per se, but the way they describe Droop Speed Control in their instruction manuals and documents is very misleading, and in a couple of cases, it actually is flat out wrong.

MOST references and textbooks and governor manufacturer written materials try to say that when a machine is "loaded" that it's actual speed decreases--and that's NOT correct for synchronous machines operating in parallel with (synchronized to) a grid with other prime movers and their generators. IT JUST DOESN'T HAPPEN UNDER NORMAL CIRCUMSTANCES. PERIOD. FULL STOP. They SHOULD be saying that a single generator powering a load or loads independent of any other prime mover and its generator will decrease in speed (frequency) if the load on the machine increases--meaning IF THE NUMBER OF LIGHTS AND ELECTRIC MOTORS AND TELEVISIONS AND COMPUTERS AND COMPUTER MONITORS AND TEA KETTLES AND AIR CONDITIONERS INCREASES WITHOUT THE ENERGY FLOW-RATE INTO THE MACHINE'S PRIME MOVER INCREASING. They are NOT properly describing what happens to a single machine NOT SYNCHRONIZED to a grid with other prime movers and their generators. It's misleading.

Now, if a hydro turbine manufacturer knows very well how much the power output of its turbine increases as the wicket gate position changes (opens or closes) that is the same as how much the electrical power of the generator will change when the generator breaker is closed and the machine is producing electrical power. I see a CT and a PT--that's what a load transducer (MW transducer) needs to calculate MW and produce an output to the governor/control system to use in controlling load when there's a load reference (and in most cases a load reference is converted to machine speed reference when its operating in Droop Speed Control). Grid regulators and grid operators COUNT ON and EXPECT machines operating at Part Load (less then rated power output) to be using Droop Speed Control--because that "other aspect" of Droop Speed Control is VERY IMPORTANT to the regulators and operators. They know what your droop regulation is and they can use that to calculate how much power the machine can increase (or decrease) when the grid frequency deviates from rated (high or low). They need to be able to estimate/calculate how the machines on the grid will respond to a grid frequency disturbance and when they CAN'T count on the responses of machines they will have trouble controlling the grid frequency and ultimately the load which the grid can provide. (This occurs because when the machine is operating at a stable condition at Part Load while synchronized to a grid with other prime movers and their generators the machine's speed reference is stable, at say, 103.5% and under normal conditions the grid frequency is 100%, so the error between the two is stable at 3.5% meaning the energy flow-rate into the machine is stable at that error. When the grid frequency changes (let's say it decreases to 95.6% of nominal grid frequency) then the error between the two increases (103.5% minus 99.6%) to 3.9% which will increase the energy flow-rate into the machine to maintain the load the machine is carrying even though its actual speed has decreased. This is how the "other aspect" of Droop Speed Control works. Again--it's ALL about the error between the machine's speed reference and the actual speed of the machine. If either changes the error will change--and that's what is supposed to happen and that's what the grid regulators and operators are expecting on and counting on, for machines that are not already at rated load (they are at Part Load and can therefore increase the load they are producing).)

That's all I can add to this. Droop Speed Control is about how much the energy flow-rate into the prime mover changes for a change in the ERROR between the machine's speed reference and the machine's actual speed. Under normal operating conditions when the grid frequency is at or very near rated, that value (the machine's actual speed) is at or near 100%, and to change the load the machine is producing one changes the machine's speed reference to change the error between the two signals to change the load the machine is carrying. If the machine speed reference is constant and the grid frequency changes the machine's actual speed changes with respect to the machine's speed reference and the error changes which causes a change in the energy flow-rate into the machine. That's it. That's all. Nothing more and nothing less.

[NOTE: I have presumed everyone reading this understands the very basic formula of AC frequency with respect to generator rotor speed: F=(P*N)/120 (Frequency is equal to the number of magnetic poles of the generator rotor multiplied by the speed of the rotor and that product (quantity) is divided by 120. For example, for a 2-pole machine running at 3600 RPM the frequency of the generator output would be 60.0 Hz ((2 times 3600) divided by 120 equals 60.0 (Hz)).

That's all folks! That's the Droop Speed Control lesson for this thread in Control.com today.

 

thanks for the reply but I’m still confused but I’m gonna write my questions and responses in bold.

Droop Speed Control is about how much the energy flow-rate input to the prime mover changes for a change in the ERROR between the machine speed reference and the actual machine speed.

I don’t understand the machine speed reference part. I worked on an old Woodward mechanical governor from the 60s and it had a speed adjust dial just like almost any other governor, It was -15 to 5, adjust the dial counter clockwise or clockwise and the wicket gates open or close to produce power I completely understand this I even added an encoder to it’s shaft and created a MW control “AGC” to control the unit and you can see the unit going up and down as grid frequency changed but the speed adjust stayed the same. Now Are you calling the speed adjust dial the speed reference? If not where is it getting this speed reference? I don’t understand this, I’m assuming the actual machine speed is either coming from a PMG or a tooth magnetic pickup sensor but idk where this speed reference is coming from I’m also very confused on how Woodward is explaining with there 0 to 50 to 100% speed adjust setting where 100% is 63hz and 50 is 61.5 and 0 is 60hz none of this makes sense. Simply isn’t it at 63hz wicket gates are shut and 57 they’re wide open


When a prime mover and generator set is synchronized to a grid with other prime movers and their generators the speeds of ALL the machines SYNCHRONIZED to the grid are FIXED by the frequency of the grid with which they are synchronized (connected to; parallelled to/with). It's IMPOSSIBLE to have any machine producing any power at a frequency that is different from the grid frequency. There is no "smoothing" or correction device that will turn 59.7 Hz and 60.5 Hz and 59.9 Hz and 60.3 Hz into 60.0 Hz. EVERY MACHINE SYNCYRONIZED TO A (WELL-REGULATED) GRID WILL BE PRODUCING POWER AT THE SAME FREQUENCY AS EVERY OTHER MACHINE ON THAT GRID. PERIOD. FULL STOP. That's the way AC (Alternating Current power generation works.)

I completely understand this

If you don't believe me, watch the speed of a droop machine paralleled to/synchronized to a grid with other synchronous machines and their prime movers as the operator changes the load of the machine (USUALLY by changing the machine's speed reference). Since the amount of energy flowing into the prime mover is a function of the difference between the machine's speed reference and its actual speed, and since the machine's actual speed is 'FIXED' by the frequency of the grid with which it is synchronized changing the machine's speed reference changes the ERROR between the speed reference and the machine's actual speed which changes the energy flow-rate into the prime mover which changes the POWER produced by the prime mover which changes the power produced by the generator being driven by the prime mover. And, one of the main considerations of any AC power grid is its frequency--as you have noted. So, the grid keeps the generator and its prime mover spinning at a constant speed (because the frequency is relatively constant) and to change the power being produced by the machine one changes its speed reference.

For MOST governors (but NOT all--and Woodward is guilty of this with several of their governors--when the speed reference is 100% and the grid frequency is 100% of rated there is ZERO difference between the machine's speed reference and its actual speed. So, therefore, even if the generator breaker connecting the machine to the grid is closed the electrical power being produced by the machine is zero, because the error between 100% and 100% is zero.

Increase the machine's speed reference while the generator breaker is closed to, say, 100.25%, and the error between the machine's speed reference increases to 0.25% (100.25% minus 100%) and the electrical power being produced by the generator increases because the energy flow-rate increases into the prime mover.

For your machine, with a 5% droop regulation setpoint being rated at 60 MVA (presuming unity power factor; zero MVAr), 0.25% of 5% is equal to (5% divided by 0.25%) to 20, and 60 MW divided by 20 equals 3 MW. So, increasing the machine's speed reference by 0.25% will cause the machine's electrical power output to increase by 3 MW.

Another way to say the same thing is that for every 1% change in the error between the machine's speed reference and its actual speed the electrical load of the machine will change by 12 MW.

THAT'S WHAT DROOP SPEED CONTROL IS PRIMARILY ABOUT. One CANNOT change the machine's speed/generator's frequency--its being "controlled" (fixed) by the grid frequency (great big--HUGE--magnetic forces at work inside the generator are literally limiting the speed as a function of frequency).

Be VERY CAREFUL when reading Woodward Governor Control instruction manuals and white papers--they are full of errors and OFTEN DO NOT PROPERLY STATE ALL OF THE MACHINE OPERATING CONDITIONS WHEN ATTEMPTING TO DESCRIBE DROOP SPEED CONTROL. It can be VERY MISLEADING. Even textbooks and reference books neglect to describe IMPORTANT operating conditions when attempting to describe Droop Speed Control--which is where SO MUCH of the misunderstanding of Droop Speed Control originates.

It's actually a very simple concept, but because there are TWO variables involved (machine speed reference AND actual machine speed) and because Droop Speed Control is also a very important part of grid stability (maintaining power output when grid frequency is abnormal or unstable) people just get SOOO confused about Droop Speed Control. Again--PRIMARILY Droop Speed Control is about how much the energy flow-rate into the prime mover changes for a change in the error between the machine's speed reference and its actual speed. EVEN WHEN Droop Speed Control is contributing to grid stability when grid frequency is abnormal (when grid stability is oscillating, EVERY MACHINE on the grid is going to experience similar frequency excursions--because they are all SYNCHRONIZED together on the same grid. Synchronism is a VERY important and overlooked word in AC power generation; yeah, we all say the machine is being "synchronized" but IT'S NOT just about the act of closing the generator breaker to connect the machine to the grid--it's what happens AFTER the generator breaker is closed, also.

I'm not going to read all the Woodward verbiage; I've done that for too many years trying to explain Droop Speed Control tp people. Woodward has a different philosophy and a DIFFERENT understanding and way of explaining Droop Speed Control and of implementing it into their governors (prime mover control systems). It's not wrong, per se, but the way they describe Droop Speed Control in their instruction manuals and documents is very misleading, and in a couple of cases, it actually is flat out wrong.

MOST references and textbooks and governor manufacturer written materials try to say that when a machine is "loaded" that it's actual speed decreases--and that's NOT correct for synchronous machines operating in parallel with (synchronized to) a grid with other prime movers and their generators. IT JUST DOESN'T HAPPEN UNDER NORMAL CIRCUMSTANCES. PERIOD. FULL STOP. They SHOULD be saying that a single generator powering a load or loads independent of any other prime mover and its generator will decrease in speed (frequency) if the load on the machine increases--meaning IF THE NUMBER OF LIGHTS AND ELECTRIC MOTORS AND TELEVISIONS AND COMPUTERS AND COMPUTER MONITORS AND TEA KETTLES AND AIR CONDITIONERS INCREASES WITHOUT THE ENERGY FLOW-RATE INTO THE MACHINE'S PRIME MOVER INCREASING. They are NOT properly describing what happens to a single machine NOT SYNCHRONIZED to a grid with other prime movers and their generators. It's misleading.

Now, if a hydro turbine manufacturer knows very well how much the power output of its turbine increases as the wicket gate position changes (opens or closes) that is the same as how much the electrical power of the generator will change when the generator breaker is closed and the machine is producing electrical power. I see a CT and a PT--that's what a load transducer (MW transducer) needs to calculate MW and produce an output to the governor/control system to use in controlling load when there's a load reference (and in most cases a load reference is converted to machine speed reference when its operating in Droop Speed Control). Grid regulators and grid operators COUNT ON and EXPECT machines operating at Part Load (less then rated power output) to be using Droop Speed Control--because that "other aspect" of Droop Speed Control is VERY IMPORTANT to the regulators and operators. They know what your droop regulation is and they can use that to calculate how much power the machine can increase (or decrease) when the grid frequency deviates from rated (high or low). They need to be able to estimate/calculate how the machines on the grid will respond to a grid frequency disturbance and when they CAN'T count on the responses of machines they will have trouble controlling the grid frequency and ultimately the load which the grid can provide. (This occurs because when the machine is operating at a stable condition at Part Load while synchronized to a grid with other prime movers and their generators the machine's speed reference is stable, at say, 103.5% and under normal conditions the grid frequency is 100%, so the error between the two is stable at 3.5% meaning the energy flow-rate into the machine is stable at that error. When the grid frequency changes (let's say it decreases to 95.6% of nominal grid frequency) then the error between the two increases (103.5% minus 99.6%) to 3.9% which will increase the energy flow-rate into the machine to maintain the load the machine is carrying even though its actual speed has decreased. This is how the "other aspect" of Droop Speed Control works. Again--it's ALL about the error between the machine's speed reference and the actual speed of the machine. If either changes the error will change--and that's what is supposed to happen and that's what the grid regulators and operators are expecting on and counting on, for machines that are not already at rated load (they are at Part Load and can therefore increase the load they are producing).)

That's all I can add to this. Droop Speed Control is about how much the energy flow-rate into the prime mover changes for a change in the ERROR between the machine's speed reference and the machine's actual speed. Under normal operating conditions when the grid frequency is at or very near rated, that value (the machine's actual speed) is at or near 100%, and to change the load the machine is producing one changes the machine's speed reference to change the error between the two signals to change the load the machine is carrying. If the machine speed reference is constant and the grid frequency changes the machine's actual speed changes with respect to the machine's speed reference and the error changes which causes a change in the energy flow-rate into the machine. That's it. That's all. Nothing more and nothing less.

[NOTE: I have presumed everyone reading this understands the very basic formula of AC frequency with respect to generator rotor speed: F=(P*N)/120 (Frequency is equal to the number of magnetic poles of the generator rotor multiplied by the speed of the rotor and that product (quantity) is divided by 120. For example, for a 2-pole machine running at 3600 RPM the frequency of the generator output would be 60.0 Hz ((2 times 3600) divided by 120 equals 60.0 (Hz)).

That's all folks! That's the Droop Speed Control lesson for this thread in Control.com today.

 

I hope I've gotten all the mathematical calculations correct; like other contributors to Control.com (and many other World Wide Web forums), I'm not the best proof-reader of my own writing, and I don't have the luxury of a proof-reader to catch and correct my errors (mathematical AND grammatical). But, here goes, I have a day or so to correct any errors anyone points out; after that, my ability to edit a response is taken away. So, if you see any glaring errors--describe them quickly and I will try to get them fixed before I'm cut off from editing the response.

@ekhananya, you wrote: "I don’t understand the machine speed reference part. I worked on an old Woodward mechanical governor from the 60s and it had a speed adjust dial just like almost any other governor, It was -15 to 5, adjust the dial counter clockwise or clockwise and the wicket gates open or close to produce power I completely understand this I even added an encoder to it’s shaft and created a MW control “AGC” to control the unit and you can see the unit going up and down as grid frequency changed but the speed adjust stayed the same. Now Are you calling the speed adjust dial the speed reference? If not where is it getting this speed reference? I don’t understand this, I’m assuming the actual machine speed is either coming from a PMG or a tooth magnetic pickup sensor but idk where this speed reference is coming from I’m also very confused on how Woodward is explaining with there 0 to 50 to 100% speed adjust setting where 100% is 63hz and 50 is 61.5 and 0 is 60hz none of this makes sense. Simply isn’t it at 63hz wicket gates are shut and 57 they’re wide open"

I have a few questions but I think I have addressed most of your concerns--maybe not directly, but indirectly (like the part about confusing graphs and charts). Again, I'm not really fond of Woodward documents, though they ARE usually better than many other manufacturers' documents in most respects, but they are not perfect by any stretch of the imagination.

I don't understand the speed adjust dial thing-a-ma-bobber. It's scaled from -15 to +5. So, what the unit was being synchronized, what was the setting of the speed adjust dial? After synchronization when one wanted to increase the load of the prime mover and generator which direction did the operator turn the speed adjust dial--towards the +5 or towards the -15? When the machine was at 50% of rated power output what was the setting on the speed adjust dial? When the operator wanted to unload the machine which direction did he turn the speed adjust dial, towards the +5 or towards the -15? (I've seen a couple of Woodward governors that had what I would call reverse-acting Droop Speed Control--at zero MW the machine speed reference was 5%, and at full load the machine speed reference was 0%.) And, yes, while I don't understand the -15 to +5 scaling as it relates to electrical output, I do think that's the equivalent of the machine's speed reference device/method.

I'm only familiar with SPEED/LOAD RAISE and SPEED/LOAD LOWER buttons on an HMI or a bat-handle switch on an operator's "board" that, when turned one direction increases the machine speed reference and when turned the other direction decreases the machine speed reference (as you described). It could be a pot (potentiometer) or an analog value (voltage value calibrated to some value depending on the parameter (volt per percent speed, for example)), or it could be a digital value in a programmable control system. On the equipment I commissioned the machine speed reference was always 100.3% when the machine reached synchronizing speed, which means that the error between the machine speed reference and the actual speed was 0.3% and the SPEED/LOAD INCREASE button or switch was how the machine speed reference was increased. (This positive speed error was used to ensure that when the generator breaker closed during synchronization that positive kW or MW was being produced to prevent reverse power from tripping the machine's generator breaker. It didn't have to be 0.3%, it could have been 0.2%, or 0.1% or 0.05% or even 0%--but by having some positive value it ensured positive power was flowing out of the generator immediately after generator breaker closure. That was GE's default for the equipment I was commissioning; other manufacturers/programmers often use different amounts to achieve the same results.)

The actual speed could be from a passive- or an active speed pick-up--OR it could be from the PTs on the generator terminals that provide voltage indication to the synchronizing circuit and the incoming voltmeter. REMEMBER: F=(P*N)/120, or written another way N=(120*F)/P, so the frequency of the voltage from the generator PTs is DIRECTLY proportional to the speed of the generator rotor (which is usually directly coupled to the hydro turbine, but not always) which is directly proportional to the prime mover's speed. The error between the machine's speed reference and the machine's actual speed could be converted into a wicket gate position if the relationship between wicket gate position and load was well known. That's done a LOT in Droop Speed Control--the error between the machine's speed reference and the machine's actual speed is converted to a control valve position, or even to a load (kW, MW) value. BUT, still there is almost always some kind of speed reference being compared to the machine's actual speed--whether the actual speed comes from speed pick-up(s) OR from the generator terminal voltage doesn't matter it's still the machine's actual speed (for a 60 Hz machine 60.0 Hz is 100% of rated frequency, and if the generator rotor speed at rated frequency was 60 RPM then when the machine is at 60.0 Hz it's speed is 100% of rated.

Droop Speed Control relies on the fact that when a prime mover and its generator is synchronized to a grid with other prime movers and their generators that it's speed--AND frequency--is at or very near system rated frequency and can't be easily changed. So, the machine speed reference--even though it might be "telling" the machine to run at 103.5% of rated speed IT CAN'T ACTUALLY RUN AT 103.5% OF RATED SPEED (because it's synchronized to a grid running at or very near grid frequency, so the generator rotor speed is fixed by the grid frequency (per F=(P*N)/120)). So, the ERROR between the machine's speed reference and the machine's actual speed is used to change how much energy flows into the generator's prime mover when grid frequency is at or very near normal.

I, and another contributor to Control.com, like to say this is the origination of the term "Droop Speed Control." By increasing a machine's speed reference knowing the machine's actual speed WILL NOT change (appreciably) when the machines ARE synchronized to a grid with other prime movers and their generators (ALSO operating in Droop Speed Control!) increases the error between the two signals which is then used to increase the energy flow-rate into the generator's prime mover (which increases the electrical output of the generator). So, in other words, because the machine's actual speed is lagging behind ("drooping" below; less than) the machine's speed reference--Droop Speed Control. Droop Speed Control makes use of the fact that under normal grid conditions (meaning a stable frequency that is at or very near the nominal grid frequency) the machine's actual speed WON'T change. The machine's actual speed is allowed to remain constant (because it's synchronized to a grid with other machines)--it's being allowed to "droop" below the machine's speed reference in order to create the error which is used to change energy flow-rate (or kW, or MW, or whatever variable is being used as the control setpoint and feedback). This is one of the PRIMARY benefits of Droop Speed Control.

The other primary benefit of Droop Speed Control is that machines operating in Droop Speed Control when synchronized to a grid with other prime movers and their generators will not be trying to control frequency--and because they aren't trying to control frequency they will "share" load (produce stable power (kW, MW)). Think about it this way: When the grid frequency is stable (and hopefully at or very near nominal grid frequency) and a machine's speed reference is also stable and not changing the ERROR between the machine's speed reference and the machine's actual speed is STABLE--one of the definitions of "sharing" load with other machines. If a machine with 5% droop regulation has a stable machine speed reference of 103.5% and the grid frequency is 100% the ERROR between the machine's speed reference and the machine's actual speed is a stable value of 3.5%--which the energy flow-rate into the machine is stable and not changing, which means the electrical power output of the generator (kw, MW) is also stable and not changing.

If one had two or more machine with their governors operating in Isochronous Speed Control and synchronized to a grid with each other (and possible other machines) they would be fighting to control grid frequency--which would mean the outputs of the two machines would be changing a lot and very quickly, oscillating and when this happens the grid usually experiences serious problems (frequency and load) and usually one or both of the Isoch machines will trip off line resulting in further grid problems which may lead to a brown-out or black-out of the grid. Droop Speed Control is the governing mode that allows multiple machines (prime movers and their generators) to be synchronized together on the same grid and produce stable electrical power as if they were all one single prime mover and generator. That's what grids do--they allow multiple machines SYNCHRONIZED together (producing electrical power at the SAME frequency!)--to act as a single prime mover and generator supplying a load or loads MUCH larger than any single machine could produce. And do so in a stable fashion (meaning grid frequency is stable and power outputs of the machines are stable). All because of F=(P*N)/120 AND Droop Speed Control. [NOTE: I'm referring to pure Droop Speed Control, not some version of Isochronous Load Sharing Control.]

Droop Speed Control is the most common method of governor operation for machines synchronized to a grid with other machines because if all the machines use a common, or at least similar, method of stable operation when synchronized to a grid with other machines the grid will be stable and more easily managed by grid regulator and grid operators. If machines all use Droop Speed Control for the majority of their operating below rated power output the grid, knowing the droop regulation percentages of the machines and the capacities of the machines the grid regulators can anticipate how the grid will respond to frequency excursions, and in some cases they can actually control certain machines (using AGC, Automatic Governor Control) to help support or even return grid frequency to normal if the machines are not already at rated power and are operating on Droop Speed Control.

The other benefit of Droop Speed Control is that for machines NOT operating at rated electrical power output synchronized to a grid that is experiencing a deviation from the nominal grid frequency (a decrease or increase in grid frequency) those machines not operating at rated power output and operating in pure Droop Speed Control will respond to the change in grid frequency and change the machine's electrical power output.

For example, if the grid frequency decreases to 49.7 Hz (on a 50 cycle grid) that means there is insufficient power being generated by the machines synchronized to the grid to maintain rated speed and frequency. (This usually occurs when one or more machines suddenly trip and are disconnected from the grid for some reason.) Those machines operating in pure Droop Speed Control (sometimes called Primary Frequency Response, or Free Governor Mode) and NOT operating at rated electrical power output will automatically increase the energy flow-rate into the prime movers to increase the electrical power output of the generators to maintain grid load and to help support grid frequency. They WON'T return the grid frequency to normal/nominal--but the action they provide gives the grid regulators and grid operators time to make appropriate responses to the grid issues and hopefully prevent a grid brown-out or black-out. For example, a machine with 5% droop regulation and operating with a machine speed reference of 103.5% will be producing approximately 70% of rated power (3.5% divided by 5% time 100), and when the grid frequency decreases by 0.3 Hz it means the grid is lacking approximately 0.6% of capacity (the power needed to maintain grid frequency at nominal), which means the machine's actual speed has decreased by 0.6% (49.7 Hz divided by 50 Hz times 100), which means the speed error for that machine (and ALL the machines operating at less than rated electrical power output, since they are all synchronized together!) has increases. So the speed error between the machine's speed reference of the machine in this example, unchanged at 103.5%, and the machine's actual speed, which is now 0.6% lower than it should be at 99.4% of rated (49.7 Hz divided by 50 Hz times 100) has increased to (103.5% minus 99.4% equals 4.1%, 4.1% divided by 5% times 100 equals 99.4%). So, the power output of this machine would increase to (4.1% divided by 5% times 100) 82% of rated from 70% of rated. This is yet another of the benefits of Droop Speed Control!

If another machine with 4% droop regulation were operating on the same 50 Hz grid with a machine speed reference of 103.5% with the grid decreasing to 49.7 Hz it could NOT go to full rated electrical power output--because although the speed error between the machine's speed reference and the machine's actual speed increased to 104.1% (103.5% minus 99.4% equals 4.1%)--putting the speed error greater than 4.0% which is above the rated electrical output of the machine, so it's output would be limited--but it would STILL be contributing to the power available on the grid (even though it couldn't increase its power output to the full amount dictated by the speed error between the machine's speed reference and it's actual speed (which is N=(F*120)/P, from F=(P*N)/120). And, to remind you--EVERY machine synchronized to the grid would see it's frequency decrease (or increase) by the same amount during a grid frequency disturbance. They are ALL SYNCHRONIZED together, acting as one machine...!

In it's most basic form, Droop Speed Control is about how much the electrical output of the machine will change for a change in the difference (the error) between the machine's speed reference and its actual speed, which, when synchronized to a grid with other prime movers and generators will be a function of grid frequency, N=(F*120)/P. If either of the two variables in the Droop Speed Control equation changes--the machine speed reference OR the machine actual speed--the error between the two will change, which will change the energy flow-rate into the prime which is directly related to the electrical power output of the generator. Change the energy flow-rate--change the electrical output (kW, MW). As long as the grid frequency is stable at or near rated grid frequency and the machine speed reference is stable the electrical power output of the machine will be stable. If the machine speed reference changes but the grid frequency remains stable the electrical power output of the machine will change (up to the rated electrical power output of the machine). If the machine speed reference is stable but the grid frequency changes the error between the two will change which will change the electrical output of the machine. (THIS is the part MANY people--operators, technicians, operations supervisors, plant managers, engineers--don't understand. They (wrongly) believe the power output of their machine should remain stable during grid frequency disturbances even though the grid frequency is changing. And they get very ... angry (sometimes) when the output of their machine(s) is unstable when the grid frequency is unstable. For machines operating on pure Droop Speed Control at less than rated electrical output, when grid frequency is unstable the grid regulators and grid operators expect and actually rely on machines changing their power appropriately to help support grid operation. And, believing or wanting anything else to happen is simply wrong. It's NOT how Droop Speed Control operates.

Now, you keep referring to a Woodward document written for hydro turbine governors/control. Again, I have consistently found errors in Woodward documents particularly when describing Droop Speed Control. AND, Woodward sometimes used what I refer to as inverse Droop Speed Control, meaning when the unit is initially synchronized the machine's speed reference is 104% or 105% (corresponding to what the droop regulation percentage of the machine is), and as the machine is loaded by the operator the machine speed references decreases to 100% when the machine reaches rated electrical power output. (Confuses the hell out of me, but then I don't think like Woodward; I think like GE--which is good when working on GE controls and not so good when trying to understand other manufacturers' controls (like Woodward, or Siemens; I can get very confused very easily.) But, no matter how it's actually done, Droop Speed Control must accomplish the things described above--change the load by an amount related to the change in an error in the control loop; participate stably in the production of electrical power when synchronized to the grid; respond as expected to grid frequency deviations. And, I think some manufacturers use the phrase "Droop Speed Control" even though they aren't using machine speed--but they still have some method of generating an error signal that responds appropriately to changes/errors, because for the machine to be synchronized to a grid with other machines it needs the functions that Droop Speed Control provides.

And, one can factor in the machine load to the control loop, but it's not really necessary, and, depending on how it's done it can confuse people who are already confused by Droop Speed Control. The error amount can be used to change energy flow-rate (if the energy flow-rate is being monitored to detect actual energy flow-rate and changes in energy flow-rate), or energy control valve position (if the relationship between control valve position and energy flow-rate is known (often it's a linear, straight-line relationship, though more and more these days it's non-linear--but a non-linear relationship can still be used in the control loop if it's known), or it can be used, in the case of hydro turbine control, to control wicket gate position if the relationship between wicket gate position and energy flow-rate through the wicket gates are known (again, linear or non-linear position versus flow-rate), or, if wicket gate position versus electrical power output is known that can be used also. It can even be used to directly control load (electrical power output of the machine) by using an accurate means of detecting electrical power output. Droop Speed Control can be adapted to many types of machine and control schemes. It's very versatile.

But in the end it's still always about how much the energy flow-rate into the prime will change for a given change in the error between two properly chosen parameters. Speed is directly related to frequency (N=(120*F)/P), so machine frequency can be used in place of machine actual speed. The reason speed is most commonly used is because when power generation started more than 150 years ago there was no way to use actual machine load in the mechanical methods (flyball governors and links and levers) used to control energy flow-rate to control load. The operator did--and still does--monitor load when changing load manually even though the operator is actually changing the machine's speed reference to affect a change in generator electrical load! That's what was done more than 150 years ago, and that's what's still done today! Old school? Yes. But because electrical power generation, transmissions and distribution systems were designed that way in the beginning doesn't make it any less useful or "bad" today--it works, and it works AMAZINGLY well. And, there's no external communications protocols or methods required (employing wires or radio signals) for all the machines to sense changes in grid frequency--since they are all operating at speeds proportional to grid frequency when synchronized to the same grid they ALL know what's happening to the grid as grid frequency deviates from normal! No external methods of sharing load or changing load as grid frequency deviates from normal--it's all based on grid frequency, and the fact that every machine synchronized to a grid runs at the same frequency and can sense the frequency (speed) changes just like every other machine. (Sometimes I write same speed/frequency when describing machines synchronized to a grid--and while the frequency part is true, the speed part isn't really true. It IS true that every machine operates at a stable speed (usually referred to as synchronous speed) when the generator is synchronized to a stable grid at or near nominal frequency, and that speed only changes when the grid frequency changes (because the number of poles of a generator is not, normally, something which can easily be varied during operation!).)

And, when the machine is NOT synchronized but is running, changing the machine speed reference WILL change the machine's actual speed--it just can't do that when it's synchronized to a grid with other machines (operating in Droop Speed Control).

Sorry; I am still amazed how this was all derived and put into practice before computers and communication networks. It's simply genius.

Don't overthink it. Don't believe EVERYTHING written about Droop Speed Control--because, again, most of the time it's written by ivory tower intellectuals with little or no ACTUAL machine operating experience. They don't properly define conditions like they should, and it leads to lots and Lots and LOTS of confusion. They use terms without properly defining them ("sharing" load, for example). They like to use graphs and charts, but don't properly explain them--they think everyone else can understand the graph or chart because its intuitively obvious to the author (a fatal assumption). Even the Wikipedia description of Droop Speed Control is also extremely obtuse and difficult to understand.

How much will the load of a machine change for a change in the error signals being used to control it (speed; load; position; flow-rate--whatever!)? That's the basic principle of Droop Speed Control. How will the machine respond to grid frequency deviations? Will the machine produce stable power output when synchronized to other prime movers and their generators? All of these things are related and an integral part of Droop Speed Control. Which makes it difficult to explain--but it's really very simple.

As previous contributors to Control.com said, read this. Put it aside for a day or three, then re-read it. Start with one section (sorry; I should properly identify sections--but I'm not writing a book I'm just trying to describe what is a very misunderstood concept that is actually very simple but has so many bits and pieces to it; maybe in a future winter when I have nothing better to do) and make tables or charts as you need. If I caught all my mistakes, I think it's pretty self-explanatory--but I will fix or clarify as necessary if you tell me what isn't making sense. I don't solve doubts, though. I do provide clarifications.

 

its starting to make sense to me although i am pretty dumb, but i have a few questions for you. what's the maximum speed reference percentage you have seen? you also mentioned that you commissioned a unit and had its synchronized speed at 100.3% but what happens when the lake the head pressure changes? that 100.3% is not going to be the same, you also mentioned wicket gate position and speed adjust position is linear but it is not because there are multiple variables that come into play and one of them is head pressure. in another one of your examples you mentioned speed reference of 103.5%, now lets say the grid is at 60hz so its 100% machine speed and 103.5% machine speed reference, now why doesn't governor open its wicket gates to maximum output to try to achieve 103.5%? what if the wicket gates are open at 90% and speed reference is 120% and the grid frequency is 60hz now there's a 20% error I know its physically impossible to physically change the speed of the unit because we're connected to an infinite bus but why doesn't it try to open wicket gates to achieve this? why does it just sit there fat dumb and happy with this error? this is what is confusing me.

 

With respect to Droop Speed Control I used to think I was pretty dumb, too. I read EVERYTHING I could find, printed and on-line. Some of it was so contradictory and confusing I almost gave up hope of ever grasping the concept much less the formulae and the operation. But, I would put it away for a while, come back to it and set aside the stuff that was confusing and contradictory to most other writings, and slowly I started putting it together piece by piece.

You asked about Droop Speed Control. I provided information about Droop Speed Control. You mentioned working on a hydro turbine control system (a governor). I have almost no experience with hydro turbine control systems; the bulk of my work was on GE-design heavy duty gas turbines, a little on GE steam turbine control (mostly Combined-Cycle Steam Turbine applications which are a little different than other steam turbine applications), and a very few GE-design aeroderivative power generation applications. I worked on some hydro turbine excitation control systems and hat a little exposure to the hydro turbine governors—but they were more than 60 years old—and still operating well—and I wasn’t able to get very deep into the governor controls.

If you want help with a specific hydro turbine application and its idiosyncrasies and nuances, I am not the person. I have no experience with the type of control parameters you are describing. I DO NOT believe the operational problem you are describing are attributable to Droop Speed Control. I would suggest the problems you are describing are related to faulty or intermittent inputs, such as head pressure, for example. But that’s a SWAG (Scientific Wild-Arsed Guess). And nothing more.

It’s entirely possible Woodward has integrated the parameters you described into some kind of bias or limit that affect Droop Speed Control. But I am not the person to be diving into Woodward manuals; every time I’ve done that I’ve not had good success, and almost every time I’ve contacted Woodward Technical Support for assistance, I’ve been told, “Oh, that bit of written text is not exactly the way we implemented it.” Or, “That block of code doesn’t operate the way it is described in writing in the manual,” and they don’t have any graphical descriptions of the block operation—it’s just a rectangle with inputs and outputs (on BOTH sides of the block so no one knows if it’s an input or an output). And, their programming interface is called GAP—for Grapical Application Program!!! Yeah, there’s some blocks (rectangles) and lines that can be used to interconnect rectangles with a mouse, but that’s about the extent of the graphics. It’s really pretty pathetic. All one has is what’s written in text in the manual—which must exactly the version used to create the program and which often DOESN’T reflect what’s happening inside the rectangular block—but because one can’t see what is happening inside the block one has to call Woodward Technical Support only to be told you have the wrong version of the manual, or as was the case when I calmly, “It doesn’t really operate that way,” and they can’t offer a written description of the way it does operate.

I know they make some really good equipment; I have configured and programmed a couple of 505E’s which still had some documentation glitches but nothing really bad. But, their MicroNet and other programmable systems are not my cup of tea, or coffee or kambucha.

So, if you need help with a specific hydro turbine and control system, I am probably not the right person. Sorry.

 

Thats okay I really appreciate your help and taking the time to write this, you are my hero thank you Sir!!!!

 

I said, explicitly, that Droop Speed Control can be adapted for non-linear position control mechanism changes, but has the governor that you're working on capable of adaptation, and if it is capable, has it been adapted? In my experience with Droop Speed Control it generally doesn't have a load feedback, but one iteration I worked with does--the machine speed reference is both a speed- and load reference, and there is load feedback for this version, so if, for example, fuel supply pressure drops below rated the Droop Speed Control will open the fuel control valve to maintain the speed/load reference based on actual load feedback.

You didn't answer my question about the -15% and +5% scaling.

In the application you are talking about is there an input to the governor about lake head pressure? If so, how does it modify the machine "speed" or load reference?

I've NEVER seen a speed reference of 20%. The control systems I've worked on had an upper limit of 107% on the speed reference for machines with 4% droop regulation. 20% is just not a realistic number, in my personal opinion. If a machine speed reference was 120% for a machine with 5% droop regulation something seems amiss with the configuration/programming.

If the machine speed reference of a machine with 5% droop regulation, when the machine speed reference is 103.5% the fuel control valve will open to the position that corresponds to 70% of rated power output. If the fuel pressure and/or flow drops causing the electrical output of the machine to drop then the control system--without any other inputs/feedback. Operating in pure Droop Speed Control the load will decrease. If the machine is operating in load control (called Pre-Selected Load Control in the GE Mark* system) it would increase the machine speed reference--but only up to 107%. That's to prevent a sudden load jump if the energy pressure/flow-rate suddenly recovered.

I neglected to mention pure Droop Speed Control is straight proportional control--there's nothing to correct the error between the reference and the control variable. There are no integral or differential components. That's probably ONE reason why the governor isn't making any changes like you think it should--it is straight proportional control. If the speed reference is 103.5% and the actual machine speed is 100%, the error between the two is 3.5% and that translates into a wicket gate position reference, which might be modified (biased or limited) based on lake head pressure if there's an input for that and the governor is configured to do that properly. Proportional control won't make any effort to correct the error--because it NEEDS the error to function properly. It just doesn't try to correct for the any error to the error, or to the error. That's for pure Droop Speed Control. And that's what you asked about.



Anyway, best of luck with your issue. I just don't have the time or the desire to start reading Woodward manuals again. Sorry.