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

Paralleling Generators

  1. Jan 28, 2016 #1
    Looking for some guidance on the effect of long term paralleling of different generation sets in a private network as follows:

    2 x 2.5MVA diesel generators stepped up to 11kV through DYn11 (in reverse to obtain a neutral while islanding)
    3 x 1.0MVA Gas turbines stepped up to 11Kv through Yyn11 (standard running when synchronised to the supply)

    The diesel sets are only ran when network is islanded from supply and can take up to 10 minutes to synchronise with the gas turbines before closing in on there individual breakers, upon closing the Gas turbines breakers open and run down.

    My query then is could all sets run together providing there is sufficient load as we do need to load shed to get ready for the diesels? or export during peak demand? as standard none to the transformers are paralleled (only connected together on the HV side)

  2. jcsd
  3. Jan 28, 2016 #2

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    A question well stated is half answered.

    Why not? You've not said much about your system - what makes you doubt they could ?

    That's a very confusing sentence.
    Do i understand you have difficulty synchronizing your diesels with the gas turbines, and after you've finally succeeded the gas turbines shut themselves down ?

    Why? Is there some interlock on your system that you've not described?

    A block diagram of your system would help.
    Does each generator have its own stepup transformer or do several generators share a transformer?
    If you're paralleling generators onto a bus without a transformer between, voltage regulator adjustment becomes critical. A transformer per generator will encourage the generators to share reactive load amongst themselves even with some regulator mis-adjustment.

    We don't know a thing about your system . Don't ask us to speculate . We'd feel awful if you wrecked something as a result.
  4. Jan 28, 2016 #3
    We would load shed because the load is too much for the diesels to support on there own.
    There is interlocks to ensure the sets diesel and gas can run together only when synchronised, to clarify there is no loss of supply as the diesels take the load the gas turbines (CHP) will come off line.
    Each set has its own transformer (415/11kV)

    Its widely known that gas generation does not like sudden load increases/decreases, but diesel generation is more tolerant those changes.So to reiterate leaving gas and diesel running together would it cause problems when increases/decreases (ie process equipment starting stopping ect) remember both sets have there own voltage and frequency protection as would be normal. to answer the question its probably appropriate to give a scenario. (interlocking is removed to facility)

    Say the load is 5MVA and the sets contribute as above (2 x 2.5MVA on diesel) and (3 x 1.0MVA on gas) a sudden load increase to 8MVA occurs to in the system, this will effect the gas turbines more than the diesels? possibly tripping the protection on the gas turbine sets? which would be undesirable and possibly result in a black start, hence the load shedding as I mentioned earlier
  5. Jan 28, 2016 #4


    Staff: Mentor

    The secret is to make sure that all the GT and DG have speed governors. Then set the speed droop on the DGs to a lower percent than the GTs. Then the DGs will take a bigger share of load changes.
  6. Jan 28, 2016 #5

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Sounds to me like OP's operators need to read up on how their governors work.
    Surely there's a "load limit" knob that limits throttle opening to some setting less than full open.
    That'd let you set GT's for some base load and allow diesels to float with the fluctuations
    which i think is what you asked

    Concept of "Spinning Reserve" says you never place all eggs in one basket. Keep several generators at part load so there's always enough headroom to absorb a load change or loss of one generator.
  7. Feb 12, 2016 #6
    Firstly apologies for not returning to this earlier and thank you for your answers thus far however they did not touch on the problem statement, Initially in the post we had a embedded HV network using a combination of LV DG and GT sets stepping up through transformers with different vector groups.

    DG Dyn11 .415/11kV (connected in reverse to obtain neutral)
    GT Yyn11 .415 (standard running when synchronized to the supply)

    These two transformer groups cannot be paralleled correctly due to phase shift when fed from the one network, my question therefore is:
    Suppose the DG and the GT reference their LV voltages as L1 240https://www.physicsforums.com/file:///C:/Users/VOLTCO~1/AppData/Local/Temp/msohtmlclip1/01/clip_image002.png [Broken], L2 240 https://www.physicsforums.com/file:///C:/Users/VOLTCO~1/AppData/Local/Temp/msohtmlclip1/01/clip_image004.png [Broken], L3 240 https://www.physicsforums.com/file:///C:/Users/VOLTCO~1/AppData/Local/Temp/msohtmlclip1/01/clip_image006.png [Broken]
    Upon putting these phasors through their retrospective transformers there would seem to be a phase shift between the voltages of the two sets. Will this create a problem for closing the breaker?, remember the sets would be synchronized at 50Hz but the voltage angles would be different by up to 30 degrees...
    Last edited by a moderator: May 7, 2017
  8. Feb 12, 2016 #7

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    clearly you don't parallel transformers on both their primary and secondary sides with 30 degrees phase difference.

    What is not clear to me is - are you proposing that ? Are generators paralleled on low voltage side ?

    Why cannot the generators be run 30 degrees out so the phase matches on secondary ?
  9. Feb 12, 2016 #8

    the generators are not paralleled on the LV side but would deliver 415v at 0 angle if required, thus if one wanted to you could parallel on to a LV network quiet simply. The problem is the step-up transformers!!. to run the sets out by 30 degrees would require measuring the voltage angle on the HV side and relaying the correction to the generator no?, no such equipment exists on site!!

  10. Feb 12, 2016 #9

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    and i can't see your images with Firefox browser - can anybody else ?
  11. Feb 12, 2016 #10

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    now i'm really confused.

    Where do you measure the voltages for synch when closing the breakers now? On transformer high side? Is that where your synchroscope senses voltage?

    If there is no connection between your generators they can run at any angle
    and you should see that on your synchroscope when matching up their speed for synchronizing them.

    Here's a picture from a random 'net search on 'two generators one line diagram'
    note G3 must be 30 electrical degrees out of phase with the other two , for its transformer is YY but other two transformers are ΔY .

    Pardon me if i missed your question
    but the above picture , shows how i interpret your problem statement :
    "Can i parallel two generators through different transformers?"
    Sure. The only thing locking them in phase is the interconnecting wires. Each machine is free to assume its own power angle .
    If you synchronized a stroboscope to bus C and looked at the shafts of all three machines,
    you'd see G3 in synch , zero degrees,
    G1 and G2 out by ~30 degrees,
    each of course plus its power angle.

    So it seems to me you have assumed some absolute reference point for phase, maybe one generator
    Make that reference point Bus C above instead of a generator and see if it changes anything.

    What have i missed?

    old jim
  12. Feb 13, 2016 #11
    Thank you Jim
    I have been looking at this incorrectly, provided the DG transformers are backwardly soaked then there is no drought that the could sync to the GT...

    You have filled the hole in my understanding of correctly syn gen sets, again thanks....

  13. Feb 13, 2016 #12

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Thank you for the feedback. Helps an old guy feel still perhaps a little useful.

    old jim
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook