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

Steam Turbine (SST-9000)

  1. Jan 28, 2013 #1
    Has anyone experienced with SST-9000 siemens steam turbine here? I am much astonished to see the steam parameter and the output.

    Main steam parameter- 290 Degree C/ 75 Bar
    Reheat steam parameter – 277 degree C / 9 Bar
    ST power Output is 1700MW.

    Can anyone explain, how is has been made?

    Thanks in advance.
  2. jcsd
  3. Jan 29, 2013 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    Sounds like a high output turbine for a large central station power plant.

    Nothing to see here. Move along.

    I take it you don't have much experience with steam or gas turbines.
  4. Jan 29, 2013 #3


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

  5. Jan 29, 2013 #4
    Thank you to your reply. I know its high power turbine. But it frequency is only 25/30 HZ. in addition, my doubt is how they are producing this much power with lower parameter without compromising the efficiency? May be i am not gone through many turbines, but i know somewhat about the basics and eager to learn if i missed something. Seeking for your help regarding the same.
  6. Jan 29, 2013 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    I'm not following what frequency you are talking about. If the 25/30 Hz is some sort of electrical power frequency, that is more a function of the generator which the turbine is driving. The number and arrangement of poles in the generator will be determined, in part, by the operating speed of the turbine and generator.

    Notice that these units are intended for use where the steam is generated by means of a nuclear reactor. The HP unit receives saturated steam, and the 3 LP sections are all on a common shaft with the HP section. I don't think the loss of a few percent in turbine efficiency matters much when you have a nuke making steam.
  7. Jan 29, 2013 #6


    User Avatar
    Science Advisor
    Homework Helper

    The basic reason for the "low" speed is to reduce the centrifugal stress in the rotating blades etc. The bigger the diameter of the turbine, the lower the speed.

    After you have selected the running speed (in this case, half of the 50 / 60 Hz electrical frequency) you design the aerodynamics of the system to be efficient at that speed.

    The data sheet gives the "last blade profile length" as 1.83m. Looking at the diagram, the blade tip radius is about 3m. So at 30 Hz, the blade tip speed is 2 pi x 30 x 3 = about 560 m/sec. That's plenty fast enough to design an efficient turbine.
  8. Jan 29, 2013 #7

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    25/30 hz says it's for a four pole generator, 1500 or 1800 rpm.
    Turbine designers like to speak in hz rather than rpm because when machines get that big mechanical resonances become important.

    Indeed it's a central station turbine.
    That temperature and pressure is about right for a nuke plant.

    Figure centrifugal force on that blade tip - one ounce weighs many tons.
    They're made from exotic steel.

    I made a small knife from a piece of old turbine blade, but have yet to figure out how to heat treat it to restore hardness.
    Is there a metallurgist in the house?
    Last edited: Jan 29, 2013
  9. Jan 31, 2013 #8
    25Hz or 30Hz is the usual rotation frequency for a steam turbine in a big power plant.

    One limit to the tip speed is the material's strength, but it's more relevant to gas turbines where power vs mass counts, that is on aeroplanes. In a power plant, efficiency is more important, and it tells that tip speed must be way lower than sound speed - this is a harder limit than material strength, especially as temperature isn't as high as in a gas turbine. It also means that an efficient turbine needs many stages; materials would allow fewer.

    With a given tip speed - which relates strongly with vapour speed, sound speed and material capability, hence is the determining factor rather than angular speed - one can trade diameter for frequency, that is, at 50Hz the turbine would have half the diameter.

    Though, vapour needs a huge exit from the turbines. At 1700MWe the throughput is huge, and for efficiency the exit pressure is tiny. The remaining vapour speed is lost power as well, so (1) mass throughput (2) tiny pressure (3) small speed combine into (4) huge exit section.

    This tells why such low-pressure stages use to have 2*3 or 2*4 exits (symmetrical to compensate the axial load) of D~8m each. At 50Hz the diameter would be halved, so 4 times more units would be needed, oops.

    For the same reason, the turbine is built directly over the condenser, just one floor above. The 2*4 huge pipes, which must withstand the atmospheric pressure, would better not be longer.


    Funny things:

    You can have a conversation close to this 1700MW machine (2,000,000 HP). This tells how smooth turbines and alternators run as compared with piston engines.

    The alternator is nearly 100% efficient but is much smaller than the turbine. This would be true at a gas turbine as well. Electric machines are small, and even light, if allowed to run at 100m/s. Electric motors would easily replace gas turbines at aeroplanes if electricity were available.

    Each low-pressure turbine unit has its own shaft. Coupling is made at the plant after installation. This eases manufacturing and resonant frequencies, but is also needed because the shaft bends under gravity. The bearings are not aligned, but slightly tilted to accommodate the bent shaft.
  10. Jan 31, 2013 #9
    The steel for blades of steam turbines is a martensitic stainless. It's similar to the common X20Cr13 or 1.4021 or Aisi420 (0.2%C, 13%Cr) with slight improvements. I've forgotten the details, but expect a little bit of Mo and V - possibly Ni as in Pelton water turbines but not necessarily.

    Maximum hardness is very good (like 1400MPa yield strength) but not huge (the alternator has 2000MPa Maraging steel because of unfavourable shape). Due to the tip speed reduced for the turbine's efficiency, such a strength suffices, enabling this nice alloy that also brings resistance to corrosion by saturated vapour, impact strength, and long-term resistance to heat because its martensite is nearly stable.

    Heat treatment is not subtle: solution, quenching (easy speed thanks to much Cr and little C, in air for a knife) and temper at 280°C. This makes 1400MPa YTS instead of 650MPa. Temper between 280°C and 560°C is not done , as it would lose both strength and resilience. Added V might need further steps. Temper diagrams were available online at Böhler for European alloys 4021, 4057 and the like.

    This is nearly a classical knife alloy (X40Cr13). Not as hard as a razor blade (X90CrMo17) but more resilient: good knife. Possible V and Mo would improve it as a knife.

    Fun: as you mentioned, steam comes from a nuclear reactor, at a low temperature. To exploit as much core's temperature as possible, steam is saturated at the HP turbine's inlet, and this gets only worse over the stages. Turbine blades get the impact of water droplets at ~400m/s - that's why I mentioned impact strength... In fact, resistance to impact erosion is the most important choice parameter here. It determines the alloy - this one is excellent - and the time between blades replacement.
  11. Jan 31, 2013 #10

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Thank you enthalpy ! That heat treat information is a LOT of help ! Thanks to you, I found some information at Allegheny-Ludlum also.

    You mentioned water impingement - leading edge of this blade has a stellite insert that is extremely hard. It was oblivious to my file.

    You also mentioned resonance.
    We measured torsion in our turbine shaft while in operation. It twists about 3.2 degrees from no load to full load(750 mw). Once we captured recording of a generator trip, which immediately unloads the shaft. It showed a torsional natural frequency of ~7 hz that is lightly damped. That's why SSR is a concern - one cannot afford to excite that resonance.

    Thanks again

    old jim
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook