Could anyone please explain the two above systems and there differences or refer me to some text that will
In a nutshell, Isochronous Speed Control refers to the prime mover governor speed control mode that controls the frequency (speed) of an AC generator (alternator) and Droop Speed Control refers to the prime mover governor speed control mode that allows multiple AC generators (alternators) to be operated in parallel with each other to power large electrical loads, or to "share" load.
The frequency of a synchronous AC generator (the type most commonly used in AC power generation) is directly proportional to the speed of the rotating electrical field(s) F = P * N / 120, where F = frequency (in Hz), P = number of poles of the rotating electrical field, and N = the speed of the rotating electrical field (in RPM).
In Isochoronous Speed Control mode, the energy being admitted to the prime mover is regulated very tightly in response to changes in load which would tend to cause changes in frequency (speed). Any increase in load would tend to cause the frequency to decrease, but energy is quickly admitted to the the prime mover to maintain the frequency at the setpoint. Any decrease in load would tend to cause the frequency to increase, but energy is quickly reduced to the prime mover to maintain the frequency at the setpoint.
In Droop Speed Control mode, the governor of the prime mover is not attempting to control the frequency (speed) of the AC generator. The term "share the load" causes much confusion, but just refers to the ability of the prime movers of AC generators to smoothly control the production of torque when connected in parallel with other generators supplying an electrical load.
Droop Speed Control, in fact, refers to the fact that the energy being admitted to the prime mover of the AC generator is being controlled in response to the difference between a speed (frequency) setpoint and the actual speed (frequency) of the prime mover. To increase the power output of the generator, the operator increases the speed setpoint of the prime mover, but since the speed cannot change (it's fixed by the frequency of the grid to which the generator is connected) the error, or difference, is used to increase the energy being admitted to the prime mover. So, the actual speed is being "allowed" to "droop" below its setpoint.
On a small electrical grid, one machine is usually operated in Isochronous Speed Control mode, and any other (usually smaller) generators which are connected to the grid are operated in Droop Speed Control mode. If two prime movers operating in Isochronous Speed Control mode are connected to the same electrical grid, they will usually "fight" to control the frequency, and wild oscillations of the grid frequency usually result. Only one machine can have its governor operating in Isochronous Speed Control mode for stable grid frequency control when multiple units are being operated in parallel. (There are Isochronous Load Sharing schemes in use in various places around the world, but they aren't very common.)
On very large electrical grids--commonly referred to as "infinite" electrical grids--there is no single machine operating in Isochronous Speed Control Mode which is capable of controlling the grid frequency; all the prime movers are being operated in Droop Speed Control mode. But there are so many of them and the electrical grid is so large that no single unit can cause the grid frequency to increase or decrease by more than a few hundredths of a percent as it is loaded or unloaded.
Very large electrical grids require system operators to quickly respond to changes in load in order to control grid frequency properly since there is no Isochronous machine doing so. Usually, when things are operating normally, changes in load can be anticipated and additional generation can be added or subtracted in order to maintain tight frequency control.
One method many electrical grid operators use to control grid frequency is called AGC, or Automatic Grid Control. Units being operated in AGC get their Droop Speed Control speed setpoints adjusted remotely in response to commands from the system operator(s) to maintain grid frequency.
There really is a dearth of material on Droop and Isochronous governor control, and much of what is written is very difficult to understand and, frankly, is explained in unrealistic terms. One of the best authors on the subject is Charles I. Hubert; his 'Preventive Maintenance of Electrical Equipment' and 'Electric Machines: Theory, Operation, Applications, Adjustment, and Control' both contain some good sections (though brief) on the two speed governor control modes.
This author is hopeful that someday soon, an article on Wikipedia will be started and that the collective thoughts and efforts of many knowledgeable people will be combined into one thoughtful and insightful-and most importantly, understandable--definition of the concept of governor speed control modes.
We have a situation with 2 x 15MW generators and a 14MW load. We want one to take the maximum load (becasue it is running on cheap gas) and the other to run on diesel, which is expensive. It is suggested that we can do this my running one on droop and one on isoch. We have a PMS system. Any ideas how to do it ???
If one trips out we want the other one to be able to accept the load
The term "PMS system" is not familiar; can you explain the acronym and describe what you are talking about, please? Please try to remember that to many people the acronyms bandied about a plant on a daily basis may not be familiar to others so when posting a question with an acronym it's always a good idea to explain the acronym.
Presuming you have a small AC (Alternating Current) electrical grid/load that is supplied by one or both of the two 15 MW generators and is NOT connected to any other source of electrical energy or electrical grid, it would be hard to understand how the frequency is controlled and maintained if one of the units is not run continuously in Isochronous mode. Is there some special kind of frequency control which is acting through Droop speed control to maintain the frequency on the grid? Do the operators control the frequency manually? Does the load not vary much, so that Droop speed control can be used to maintain grid frequency?
When a generator is being operated in Isochronous mode and another generator whose prime mover is being operated in Droop speed control mode is paralleld with the Isochronous machine, the load on the Isochronous machine is a function of how much load the Droop machine has taken/accepted from the Isoch machine.
For example, if the load is 14 MW and only one unit is supplying the load, its governor is usually being operated in Isoch mode to maintain grid frequency. If another unit is paralled to the load with the Isoch machine it is usually done so with the second machine's governor in Droop mode.
As the Droop machine is "loaded" the tendency would be for the grid frequency to decrease--but the Isoch machine reduces its power output (and generator output) to maintain the frequency. (Remember: The electrical "load" on a grid is NOT a function of the number, rating, or output of the generators connected to the grid--it's the combined total of the "consumption", the motors, lights, transformers, etc., which are connected to the grid.) So, if the Droop machine in this example was loaded to 7 MW, the Isoch machine's load would automatically drop to 7 MW. If the Droop machine was loaded to 12 MW, the Isoch machine's load would automatically drop to 2 MW.
One cannot "load" an Isoch machine by pushing or holding the RAISE SPD/LOAD button or handle--if one does so, what he/she will be doing will be increasing the frequency setpoint and in doing so the grid frequency will increase.
The load on an Isoch unit is "controlled" by the amount of load on the Droop machine, in this simple example of two generators one of which is being operated in Isoch mode and the other in Droop mode.
Now, there are Isochronous Load Sharing Schemes (is that what's in operation at your site?) whereby multiple units can "share" the task of maintaing grid frequency--but these are not very common and require tuning and testing, both of which make many people very nervous. And, as the size and nature of loads tend to change, the schemes need to be re-evaluated and re-tuned and re-tested to ensure they are working as required--another expense and effort many companies aren't willing to expend once a plant has been commissioned and put in service....
Can either machine be automatically switched to Isoch mode (by, say, a breaker status contact)?
It's up to you to decide how you want to operate your machines: Load the unit running on cheap gas fuel to 13 MW by running it in Droop mode and the load on the other unit which is running on distillate fuel (which should be operating in Isoch mode) will drop to 1 MW. If the Droop unit trips, the Isoch unit will very quickly increase its output to maintain the load AND the grid frequency.
OR, put the unit running on distillate fuel in Droop mode and operate the other unit running on cheap gas in Isoch mode and only load the Droop unit to 1 MW. However, in this scenario, if the Isoch unit trips, the Droop machine will maintain the load BUT the grid frequency will drop until the operator either raises the unit's Droop speed control reference OR puts the unit's governor is Isoch mode.
The choice is yours. But we'd still like to know what a "PMS system" is, and how the units are currently being operated....
There was a previous reference to "PMS system" (http://www.control.com/1026218262/index_html). From that reference, it can be inferred that the PMS system can be used to control the load of multiple units by supplying load control setpoints (probably in the form of 4-20 mA signals to the External Droop Speed Control inputs) to the prime mover governors.
It doesn't seem that this system could be used to control the load of multiple generators if one of the generators were operated in Isochronous control mode. The Isochronous control mode would probably not recognize the External Load Control signal since it would be applied to the Droop control mode.
It would be most helpful if one uses an acronym in their post that they define the term at least once (the first usage). PMS can be Property Management System, Post-Meiotic Segregation, Permanent Magnet Synchronous, Popular Music School, or, the most commonly found usage on the World Wide Web, Post-Menstrual Stress--but none of these descriptions seem to be applicable to this usage of PMS....
In short, a PMS system as we (Marine Industry) use the term is Power Management System (PMS). This system typically control the Diesel (or LNG (Liquid Natural Gas))engines on a higher level (over the govenors) and the system will also control all heavy consumers on the plant (load sharing, load shedding, blackout preventions etc). Also opening/closing of breakers is typically controlled from a PMS system. The PMS system consist of a PLC (Programmable Logic Controller) with I/O (Input/Output) cards and an HMI (Human Machine Interface).
Very nice introduction to Droop/Isoch by the way!
Thanks for the information!
So, PMS can also mean "Power Management System"... Aren't acronyms wonderful? Especially TLAs (Three-Letter Acronyms)!
It's possible to see how such a system which controls all large consumption, circuit breakers, etc., could be used to provide a Droop speed control reference to a prime mover governor which was "tuned" for desired response (i.e., loading/unloading rates a little faster than typical).
It does seem though, that the PMS system would have to be "disabled" if one of the units were to be run in Isochronous mode. Or at least "de-tuned" to prevent interaction and instability.
PMS is the Power mangement system, which is basically a PLC program switcing different modes of generator according to grid breaker or electric network breaker. If grid is not connected, it decides the ISO machine selection through some logic.
I was doing some research because of load sharing problems I am having with my plant powered by 4 2MW gen sets all running in Isochronous mode. The load is not being distributed evenly and I am getting some very low power factors with a large amp draws. We are running through a ASCO switchgear. In the fifth paragraph of your post you talked about this set up briefly and mentioned that constant tuning and testing. Can you expand and give me a bit of an idea of where I should look first thanks.
>As the Droop machine is "loaded" the
>tendency would be for the grid frequency
>to decrease--but the Isoch machine
>reduces its power output (and generator
>output) to maintain the frequency.
>(Remember: The electrical "load" on a
>grid is NOT a function of the number,
>rating, or output of the generators
>connected to the grid--it's the combined
>total of the "consumption", the motors,
>lights, transformers, etc., which are
>connected to the grid.) So, if the Droop
>machine in this example was loaded to 7
>MW, the Isoch machine's load would
>automatically drop to 7 MW. If the Droop
>machine was loaded to 12 MW, the Isoch
>machine's load would automatically drop
>to 2 MW.
Edward... Is it correct to state that your facility's alternator's are supplying power to the facility's electrical load? Furthermore if connected to an "External Grid" they do not normally export or import power!
if you are having high amp draws and poor power factor, then you need better droop current compensation installed on each unit's AVR (automatic voltage regulator). Besides speed control, generators require voltage control. Like the other authors suggest having one generator operate in isochronous mode (frequency control) and the other gens in droop speed control (followers) The voltage regulators must be set up in the same way. One of the four 2MW generators should have the voltage droop mode off (this unit sets the voltage of the system grid). The other 3 2MW units should operate in voltage droop mode (following). Using this method you will not get circulating current between generators because the three units paralleled in droop mode will not "push" the voltage up or pull it down. A simple current droop compensation circuit requires a CT (current transformer), a resistor (a load which builds a voltage when current is passed through it, and a transformer (basically an amplifier) to boost that ac voltage formed on the resistor. This circuit is installed on each generator and a switch can short the CT current effectively disabling droop compensation. Also the output of the transformer is connected in series with the voltage sensing of the AVR voltage regulator. Depending upon the direction of the current the voltage sign either adds or subtracts from the nominal sense voltage. Thus this negative feed back circuit protects each generator against high circulating current (over current tripping)and poor PF power factor etc.
Hope this helps.
Len Howe AET Avionics.
My e mail is lenhowe at cintek dot com.
In your example scenario, i.e. one machine in iso at 7MW and other machine in droop at 7MW, what will happen if the total load demand drops to say 5MW? What will be the response of iso machine because the system frequency will increase because of droop machine's response?
A good article and it clears many doubts. Generally people get confused droop mode to load mode and isochronous mode to speed mode. Further to elaborate the droop mode consists of load mode and speed mode.
Even though i am replying on your old thread, i feel you are the right person to answer my query.
I am briefing my problem in details as below,
We have 2 nos STG sets of 25 MW and 15 MW capacity, and we are trying to operate in parallel with SEB grid both STGs. Both STgs have woodward governor 505E.
The present arrangement is, we have given tie breaker signal to both the governor from SEB breaker to operate in droop mode. When both STGs are operating in parallel with each other and SEB, we are operating both STGs in load control mode.
When SEB breaker opens both STGs are coming to isochronous mode of operation, even though they are parallel with each other. When this happens the shifting of load between two STGs takes place on continuous basis.
Now my query is: i want to keep both STGs in droop mode of operation and should come out of load control mode when SEB breaker opens.
i hope you have understood my query. please share your experience to solve the issue. When contacted woodward they are marketing isochronous load sharing relay, which is not acceptable to me until unless i am convinced about this relay requirement.
This is an old thread.... And this topic has been smothered to death on (speed)control.com.
Whenever two governors are in Isoch mode (without some kind of Isoch Load Sharing scheme) they will "fight" each other to control the frequency. That's their nature, and they're not disposed to sharing the load. Each governor wants to control frequency and so each will respond to any change in frequency (which results from a change in load, system load).
An Isochronous Load Sharing scheme will allow the two units to "split" or "share" the load between them depending on how the scheme/controller is programmed. It will allow more than one governor to be operated in Isoch mode and will reduce and possibly even eliminate frequency swings as a result of load changes. Whether or not two governors need to be operated simultaneously in Isoch mode is a function of many factors, and to cover all the possibilities would take the equivalent of several chapters in a textbook.
Imagine two people riding on a tandem bicycle, each with pedals to contribute to the torque required to move the bicycle and any load they might be carrying. Imagine they have to maintain a constant speed (which in an AC system speed is directly proportional to frequency). The two riders are each prime movers providing torque to the bicycle, just as turbines provide torque to generators. It takes a certain amount of torque to maintain a certain speed (on a flat and level road) when there is no load on the bicycle (packages, etc.) and more torque to maintain that same speed when there are packages and possibly a passenger on that same flat and level road.
Let's further say that the number and weight of the packages can be variable, that they can be dropped off the bicycle and added to the bicycle at random times and intervals while the bicycle is moving, meaning that the torque required to maintain a constant speed is going to change as the number of packages and the weight of the packages change.
If both riders are each working to independently control the speed of the bicycle as the load changes, it's very likely that the speed of the bicycle won't be very constant. If they both apply more torque as a package or packages are added to the bicycle the speed might increase above the desired rate, and if they both decrease their torque at the same time to slow the bike down then the speed might decrease below the desired rate.
But, if they decided that one rider will pedal with a fairly constant torque and the other rider will vary his torque output to maintain the desired speed as the number and weight of the packages changes, then it's more likely that the speed of the bicycle (and it's load) will remain fairly constant.
We'll call the rider who's maintaining a relatively constant torque "Droop" and the rider who's varying his output as a function of load "Isoch."
Let's say that Droop is supplying a constant torque to the pedals equal to 50% of his ability while the bicycle is moving and packages of varying weight are being dropped off and picked up. This means that Isoch's output is varying as necessary to maintain the desired speed as the load on the bicycle is changing. BUT, if the load exceeds Isoch's ability to produce torque, then the speed of the bicycle will slow down, because Droop is only putting out 50% of his output. So, Isoch can ask Droop to increase his torque to help get the bicycle back to desired speed. And in so doing, as Droop increases his torque the amount of torque that Isoch is producing will be decreased as the desired speed is reached. Droop should increase his torque output such that any anticipated load increase can be handled by Isoch so that the speed will remain constant as expected load changes occur.
Now let's say that the load on the bicycle has dropped significantly (a lot of the packages have been dropped off the bike), and the amount of torque that Isoch has to provide is down to near zero (he's just "coasting") to maintain the desired speed. If even more packages are removed from the bike, then the speed will start to increase. And, so Droop will have to decrease his torque output to reduce the speed of the bicycle. Droop should reduce his torque output such that Isoch has to increase his torque output above that required for the loss of any additional packages in order to maintain the desired speed.
AC generators (alternators) with prime movers are identical to this example. They have to operate at a constant speed (frequency) and they have to vary their torque output in order to control speed (frequency) when load is changing when they are being operated "off-grid", or in "island" mode. But, if two governors are both trying to control the speed (frequency) they will generally fight each other which will cause the frequency to be unstable, and the loads on the two generators to be unstable.
We have no way of knowing what the load is when the two units are separated from the grid. Let's presume the two units are running at 10 MW and 20 MW in parallel with the utility. And suddenly, the tie breaker opens, and the "island" load drops to 12 MW. Let's further presume that one (only one) of the 505E controllers will switch to Isoch mode when the tie breaker opens, and let's say it's the unit running at 20 MW (which we'll call "Big"). So, when the tie breaker opens Big will switch to Isoch mode and the other unit (we'll call it "Little") stays in Droop control mode.
Because the power output of the two units at the instant the tie breaker opens exceeds that required for the "island" load, the frequency will start to increase very quickly, and Big will reduce it's load (hopefully very quickly!) to about 2 MW, which in combination with the load on Little (10 MW) should result in a stable frequency. (What will really happen is that the frequency will increase and Little's load should decrease because of Droop action and as Big's load comes down Little's load will increase back to 10 MW as frequency returns to normal.)
Hopefully, when Big closes the control valves very quickly it doesn't cause a sudden spike in inlet steam pressure and maybe even lift a safety valve or two! (The boiler control also has to reduce firing rate pretty quickly to try to help reduce steam pressure because of the reduced steam flow to Big.)
Big, being the Isoch unit, will now respond to any load changes in the "island" load. (This presumes Little is in "straight" Droop speed control with no load setpoint.) However, if the island load decreases by more than two MW, Big can't reduce its output below 0 MW, and so what will happen is that the load on Little will decrease but the frequency of the island will increase.
Now, if someone wants to increase the load on Big the operator needs to decrease the load on Little.
Presuming Little's 505E is operating in "straight" Droop speed control, it's load will remain fairly constant as long as frequency remains fairly constant. If Little's 505E is trying to maintain a specific load setpoint and Big has any difficulty responding to load swings and the frequency starts to vary, then Little won't help try to maintain frequency--it will actually make the frequency problem worse, because instead of changing it's output to help maintain the frequency setpoint by increasing or decreasing load as necssary, it will by trying to keep its load constant which will make the frequency problem worse.
If you want to operate both units in Isoch mode (and there may be compelling reasons to do that, we don't and can't know that from the information you have provided) you should seriously consider investing in an Isoch Load Sharing module. But, it shouldn't be necessary, unless your operators can't respond to load changes and adjust the output of Little to keep Big in a range of output that would allow it to control frequency as load changes.
Steam turbines don't generally make great Isoch machines, because the firing rate of the boiler can't usually be changed very quickly if load changes quickly. It's not that they can't be tuned to be good Isoch units, it's just that it usually takes tuning of both the governor and the boiler control to achieve smooth frequency control, and then only for expected load changes. If the load changes much more than expected for some reason, then they generally won't respond as quickly as necessary.
Very useful and clearly explained information on this thread. Now I have a question: how this theory applies in the case we have inverters (powered by photovoltaic panels or by batteries) in parallel with diesel and/or steam generators?
The concept of frequency control in droop and isochronous mode is the same? What about the isochronous load sharing? Can 2 inverters be in parallel and do load sharing (the same as 2 generators in parallel both in isochronous mode)?
Thanks in advance!
Thanks a lot for your informative article. I have a query about the difference between speed droop & KW droop. A STG's (3MW) governor having KW droop of 4% and another STG's (3MW) governor having speed droop of 3%, what does this actually mean when these two generators are running parallel in island mode? I think you can help me.
Its good you have a query, and not a doubt....
I have encountered some smaller prime mover-generator units, and even some very large steam turbine-generators, that use rated load for droop control rather than speed. This is mostly on units that are never intended to be operated in Isochronous Speed Control mode. The reference would a percentage of rated load (the percentage related to the amount of Droop), and the feedback would be actual load. If the error were to change then the power input to the prime mover would change. This does very little to support grid frequency, but for small units connected to a larger ("infinite") grid it's probably fine.
If the two generators in your example are the only two generators being operated in Island mode, then either the load is relatively stable and the operators can react quickly to change load as frequency varies, or there is some other control system supplying some kind of over-riding input to one unit or the other. When only two generators are paralleled, then the usual method would have one in Isoch Speed Control mode (to control the frequency by adjusting load as necessary), and the other in Droop control mode (Speed- or Load Droop would be fine) producing stable power in parallel with the Isoch unit.
Thanks a lot for your quick response. and you are right that in my example, load is almost stable & operators are very much conscious about any load change.
When 3 generators are running in parallel and if we adjust the droop setting for one particular generator, what are the changes occurs in this situation?
I mean does it applies to all running generators or to that particular unit only?
thanks for the elaborate description about droop and isocho that was really helpful.
Under your scenario with three gen-sets operating in parallel, one in Isoch and the remaining two in Droop, if you change the Droop setpoint of one of the Droop machines the load of that Droop machine will change AND the load of the Isoch machine will also change by an amount equal and opposite to the load change on the Droop machine.
Let's say this is the configuration:
Unit 1: 10 MW machine, operating in Isoch control at 5.0 MW
Unit 2: 4 MW machine, operating in Droop control with 4% Droop, at 2.0 MW
Unit 3: 5 MW machine, operating in Droop control with 5% Droop, at 5 MW
So, the total load on the system is (5+2+5) 12 MW and the load is relatively stable at this time.
Now, this means that the Droop speed reference for Unit 2 is 102%, and the Droop speed reference for Unit 3 is 105%. (Unit 2 is operating at half of rated capacity, so the Droop setpoint is half of rated, which is half of the possible 4% Droop, or 102%. Unit 3 is operating at full rated capacity, which is the full 5%, or 105%.) the Isoch machine, Unit 1, is operating at 50% of rated capacity but the speed reference is 100.0% because it is controlling frequency.
Now, if the operator lowers the speed reference of Unit 3 by 1%, from 105% to 104%, the load of Unit 3 will drop to 4.0 MW. Unit 1's governor, sensing the drop in frequency as the load of Unit 3 is lowered will Increase its output to 6.0 MW, and the frequency will remain stable at rated. Unit 2 will remain unchanged at 2.0 MW. The load on the system is unchanged at 12.0 MW, but now Unit 1 is supplying 6.0 MW, Unit 2 is supplying 2.0 MW, and Unit 3 is supplying 4.0 MW.
If the operator then increased the Droop setpoint of Unit 2 to 103%, the load of Unit 2 would increase to 3.0 MW, and the load of Unit 1 would decrease to 5.0 MW. The load of Unit 3 would be unchanged at 4.0 MW, and the total load on the system would still be 12.0 MW at rated frequency.
Does this help?
Almost forgotten--this website:
has excellent technical information on electrical power plant fundamentals.
This page/pdf file in particular has good information about Isoch and Droop Governor Control!
And, they're costless (i.e., gratis, free)!
I have been following your statements, replies and comments on queries about generation plants control practice and it seems to me that you are a perfect teacher, making thing and statements as clear and simpler as posible, without loosing depth.
Moderator's Note: Markvguy no longer posts on this forum. We do however have some great people who post here.
I am a Spanish speaking teacher (Venezuelan) on electric power systems and I am sure any electric power engineering student can follow your explanations without too much effort. Than you. If you have a blog, or a webpage where you post your experience and knowledge on various issues on Power Systems or power plants I will like to have access to it and I am sure my students, too. Please send any information.