Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Small scale power generation

  1. Jan 29, 2014 #1

    I'm a Mech. Eng., so forgive my ignorance to power electronics, but I'm hoping someone can explain this to me.

    Considering the scenario where a small gas turbine (< 500 kW) drives a generator to provide power to the power grid. The power output of the turbine will not be constant.

    1) What type of generator would be most suitable?
    2) What type of power electronics would be required to convert the generated power to a form suitable export to the power grid.
    3) Can the load torque required to spin the generator be modulated in a way that would allow the rpm of the turbine to remain constant while the power output changes?

    Thanks in advance,
  2. jcsd
  3. Jan 30, 2014 #2


    User Avatar

    Staff: Mentor

    Welcome to the PF.

    What country and state/province are you in? Your local power utility should be able to answer most of your questions about this. They will be the ones inspecting and approving your final installation anyway.
  4. Jan 30, 2014 #3
    Thanks for the reply berkeman,

    I'm located in Alberta, Canada.

    I think the local utility would have some advice regarding the output side of #2. I'm looking for some more generic information from a theory of operation standpoint for #1 and #3. Both of which would likely have an impact on input side of #2.
  5. Jan 31, 2014 #4
    since you are going to be connected to a grid I think most suitable would be asynchronous generator: mainly because it is cheap - you dont need excitation system, the grid will provide excitation. it is same machine as motor, all you need is to spin it above its "synchronous speed" and power will flow to grid. you dont need any electronics. downside is that it can not operate without grid connected. also, produced power is function of a rpm. it is fine on small hydro plants, not sure about gas turbine though. these generators need to have variable rpm in order to produce various power output. this variation in rpm is not large, look at page 2:


    if your turbine has constant rpm you will need synchronous generator with its own excitation. such machine exactly fits to your #3 condition. it is the same solution as in any thermal power plant. i guess (!!) most simple (cheap) variant would be synchronous generator with permanent magnets. it needs grid to operate but you have grid.
  6. Jan 31, 2014 #5
    That's very helpful. Thanks. It looks like I've got a bit of reading to do now...
  7. Feb 3, 2014 #6
    In the good ol' days, they'd adjust the current through the generators field winding and the governor on the prime mover. There would be an incandescent lamp between the divided down output of each phase and the divided down output of the line. The lamps would pulse brighter and dimmer as the phase of the generator slid past the phase of the line.
    The field winding and governor were adjusted until the difference in voltage was minimal (lamps went dimmest) and the governor was adjusted to make the lamps cycle very slowly. While all lamps were out, a contactor would be thrown tying the line and generator together.
    With a thump and shudder, the generator would go online, but would not be generating electricity. By pushing the governor up, the prime mover would start delivering torque, which would translate into power. The current and voltage may not be in phase. The current's phase would then be aligned by varying the field current.
    It all sounds like a mess, but it was done manually.
    A simpler, but less efficient method is to use an induction motor with a bank of power factor correction capacitors. This has been done with many home generators, though I've never heard of one over 10kW.
  8. Feb 3, 2014 #7


    User Avatar

    Staff: Mentor

    Errm. I don't think you mean "asynchronous generator" here, you mean "synchronous generator". (Small typo, but a big difference.)
  9. Feb 4, 2014 #8
    Well, I would recommend an asynchronous machine, also called an "induction generator". It is well suited for this. The speed is slightly off sync, but the electrical frequency generated is always synced to the grid. I doubt that 'pocho' made a typo. Asynchronous generator or induction generator if you prefer, is better for tapping onto the grid than a synchronous unit. An async maintains steady electrical frequency synced with grid, regardless of mechanical speed ("slip"). A sync unit must run at sync speed to maintain sync grid frequency.

    The async is better suited for this application.

  10. Feb 6, 2014 #9


    User Avatar

    Staff: Mentor

    You are anticipating being invited to provide power at times of regular peak demand on the grid, and at short notice in emergencies? However, if you install a synchronous generator, you might find that the electricity authority will pay you to run your generator at other times, too, times when they do not need extra power. The synchronous generator can be run so that it appears like a large capacitance, absorbing reactive power and improving the power factor of the grid in your region. This reduces losses in the grid, and slightly reduces the power demanded of the other stations feeding into the grid at that point in time.

    I am recalling this from my student days, developments may have brought changes in how loads are managed these days. :smile:

    As Berkeman indicated, your local power providers are the only ones able to give definitive advice on your proposal.
  11. Feb 8, 2014 #10
    I doubt that it would be economical to export such small amounts of power to the grid, you'll need all sorts of approved switchgear and protection before they'll allow you to hook up and those items plus annual inspections will be costly.
  12. Feb 8, 2014 #11

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    You've not said much about the turbine. What is its optimal RPM?

    You asked about constant rpm - a synchronous generator will do that, but as claude and mike said async will [STRIKE]work also[/STRIKE] approximate it.

    A generator and a motor are the exact same machine, it's just a question of which way power flows - from mechanical to electrical or vice versa. A synchronous one operates at exact integer submultiple of line frequency, ie 60,30,20,15,... etc revolutions/sec. An asynchronous one operates at or near same speed, like within 10% , the speed difference from synchronous being in proportion to power. The integer submultiple is same as number of machine's pole pairs.

    You're talking about >500 horsepower which is a substantial machine and almost certainly three phase. If this is a backyard project, you'll want to check with your electric company - most utilities are picky about how 'foreign' generators get connected to their grid. And there's a lot of safety concerns of course.

    EDIT: added ...........................................................................................................
    For small scale demonstration of principles, a fractional hp squirrel cage motor (old washing machine?) driven by a variable speed motor (abandoned treadmill? ) should suffice.
    Last edited: Feb 8, 2014
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Small scale power generation