Adding generators together :/

Discussion in 'General Electronics Chat' started by Josato, Oct 26, 2011.

  1. Josato

    Thread Starter New Member

    Mar 8, 2011
    17
    0
    Hi all,

    I have been asked a question about generators, and it perplexes me. Someone I know has a set of resistive elements that require a large amount of kW, and the only way they have to power them is with a petrol-driven generator. However, it is insufficient. They do, however, have more than one.

    Their question was this: is it possible to 'add' the generators together to provide enough power?

    Obviously, they are not synchronous generators, so my thinking was that they would certainly be out of phase. However, I have no idea what would happen if you connected them in parallel. Any difference in waveform would fry the connection, would it? Or would they force each other into synchronisation?

    The motor cuts the lines of flux => alternating current, one phase. This is connected to another two motors (ignoring the resistive element). Say they are off, the ac going in would then turn the motors (?), and if they are on, they would be forced into synchronisation(?)

    Also, the power ratings for them are possibly not the same, at 5.5, 6, & 8 kW. If what I said is correct, two would be drawing power from the other to keep synchronised at maximum load..?

    The load is just a resistive element thankfully, with ground, neutral and live.

    If this is a bad idea (which I imagine it is), what about industrial-strength diodes to make them all DC (ish), thereby not interacting and easily put in parallel?

    I might update this with more musings/looking through my course notes for later in the year.
     
  2. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,535
    It is done routinely, but for home users it is quite complex. If it is an AC generator the phase of the signal MUST be matched exactly, and kept that way. No options.
     
  3. tracecom

    AAC Fanatic!

    Apr 16, 2010
    3,869
    1,393
  4. Josato

    Thread Starter New Member

    Mar 8, 2011
    17
    0
    Thanks guys!

    Yeah, I was thinking about it, and realised they would force each other into sync (I wasn't thinking too straight the other night). Unless there was some sort of power circuitry between them and the load (is this usual I wonder).

    Wouldn't it be possible then to connect them together and turn the first one on => turns the second one over, turning it on (=> third one too)? This would reduce the strain of when you connect them together and they force each other into phase.

    I suppose you could contact the company to find out whether they think putting them together is a good idea.

    This would cause the rms voltage output to be x3 then, I expect?

    This link is very helpful indeed, I will read it thoroughly. It makes sense to me, although here: (I heard and saw the results of a
    50 MW diesel genset being synched 180 degrees out as the result of
    reversed leads on the synchroscope. Literally ripped the stator out
    of the foundation and twisted the shaft.)

    They are only 5kW; I assume they will not mind as much being forced into phase? Expensive risk I suppose :/
     
  5. crutschow

    Expert

    Mar 14, 2008
    12,990
    3,226
    You can't connect them together before you turn them on. The off one would appear as a near short.

    When properly synchronized, the voltage of the two in parallel is the same as one (assuming they both have the same output voltage when not connected).

    They will mind a lot being forced into phase. (It may not be as severe as the 50 MW generators you mentioned but it will likely damage them.) You must get them in phase before connecting using the lamp approach as noted in the reference suggested by tracecom.
     
  6. GetDeviceInfo

    Senior Member

    Jun 7, 2009
    1,571
    230
    the dark lamp method works well, and one lamp is required per phase. In polyphase machines, you must make sure that your sequencing is correct, which the lamps will indicate. Prior to connecting, you want your alternators to be very close to the same speeds, which allows you to engage at the right moment (dark lamp on all phases). You also want your voltages to be matched, which in most cases is performed with field current adjustments. Once your synch'd, you can transfer the loads from one unit to the other via throttle, and adjust var loading via field strength.

    You will want good throttle control and some form of measuring your frequencies to comply with load requirements. If it's strictly resistive elements, you've got lots of wiggle room.
     
  7. strantor

    AAC Fanatic!

    Oct 3, 2010
    4,302
    1,988
    So are you saying that, once you get them in phase together and you bring them online together, they will indeed "lock themselves together" in phase?

    EDIT: never mind; I read the link.
     
    Last edited: Oct 28, 2011
  8. Josato

    Thread Starter New Member

    Mar 8, 2011
    17
    0
    Yes, I did think it have a low resistance, but felt this would balance out with the impedance of the coils. Perhaps not. This is quite embarrassing, as I have done a semester on this, AND used an oscilloscope to synchronise a synchronous generator with the grid.

    Looks like I have enough info to go on now. I will let you guys know if everything goes to **** and all three the generators get fried.
     
  9. GetDeviceInfo

    Senior Member

    Jun 7, 2009
    1,571
    230
    they will, however simple throttles often 'bear' into the load, meaning that one generator may runaway with the load resulting in a stall. Hunting will quickly bring the sync down as vars are thrown back and forth, exceeding current protection.
     
  10. crutschow

    Expert

    Mar 14, 2008
    12,990
    3,226
    Remember that a generator also can work as a motor, which is what happens if you apply voltage to a stalled generator, and motor stall current is normally very high.
     
  11. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    Divide the "resistive elements" into 2, then drive each separate element with its own generator. That should have been anticipated in the design before you got to this point. ;)
     
Loading...