|
[Sponsors] |
July 21, 1999, 17:10 |
turbine question
|
#1 |
Guest
Posts: n/a
|
I have a question about design of high temperature internally cooled turbines.
In the design of these blades the aerodynamic design step is usually done before the thermo-structural design. ie the blade shape is (provisionally) determined and then the internal cooling passages necessary are designed afterward. since the erodynamic design is done (using CFD) before the internal cooling analysis how is the thermal boundary condition on the blade specified. is it adiabatic (at least for the first cut)? my quandary is that the fluid flow constitues only a portion of the thermal field which also includes the as yet undesigned blade internals. but the fluid flow depends on the thermal field. if so is an iterative technique/procedure used? can someone outline it for me? |
|
July 21, 1999, 17:54 |
Re: turbine question
|
#2 |
Guest
Posts: n/a
|
(1). Design by definition is "iterative". (2). Have you heard of the phrase "team work". In this case, some output from the aero codes are specifically formatted for the heat transfer analysis. (3). The internal cooling analysis covers the blade, the cooling passage, with aero boundary conditions from the outside. (4). Aero analysis also can include surface film or cooling holes and the blade wall in the analysis. (5). Currently, it is limited by the turn-around time of each design phase. And in the early phases of the trade study or design, you would like to try out many different arrangement. Therefore, simplified analyses are routinely used with less coupling between aero and heat transfer. But a team is a team, regardless whether you are in a different department. ( considering a summer job at an engine company?)
|
|
July 21, 1999, 18:20 |
Re: turbine question
|
#3 |
Guest
Posts: n/a
|
The aerodynamic performance of the blade is not affected very much by what thermal boundary condition you use, at least as long as you stay within reasonable limits. Hence, you can use an adiabatic condition, a constant temperature or similar to design the blade and simulate pressure distributions and other global aerodynamic parameters. You will have larger errors from turbulence modeling, inlet and outlet boundaries, instationary effect, leakage etc.
But when you want to do a thermal analysis you of course have to use a correct boundary condition, but then your are probably already in the design phase of the internal cooling system and you can simulate the whole thing, including the solid, usually with some kind of simplification of the internal air-system. If you have film-cooled blades things might be a bit different I guess - I have limited experience from that, but IŽd imagine that the film-cooling flow might also affect global aerodynamic performance, anyone know how important this is? |
|
July 22, 1999, 12:36 |
Re: turbine question
|
#4 |
Guest
Posts: n/a
|
(1). I have done some 2-D calculations of flow over a HP blade with a dozen cooling slots on the surface ( rows of holes replaced by equivalent area slots). (2). the boundary conditions of each slot was set at the inner wall side, and the computational domain includes the slot passage through the wall, and the external flow area. (3). The computed flow field indicated that cooling flow through slots can have significant effect on the main flow field, including large separation bubble ( under certain conditions). (4). So, the answer is the flow through the cooling holes on the blade surface will have visible impact on the main passage flow (relative to the non-cooling case). Whether this effect should be included in the next round of iteration depends on the goal of the design. Design is not analysis, it is intended to produce configuration to meet the requirements and resources limitations. Nothing is ideal in the design.
|
|
July 22, 1999, 17:47 |
Re: turbine question
|
#5 |
Guest
Posts: n/a
|
actually John I'm considering a full time job at an engine/turbomachinery company. I see you work for Siemens/Westinghouse. Are you in Florida. Who do I get in contact with about working down there.
|
|
July 22, 1999, 18:00 |
Re: turbine question
|
#6 |
Guest
Posts: n/a
|
(1). It is just an Internet address. I am not actually part of that company. (2). I should be able to receive e-mail at the Yahoo address.
|
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
HA Tidal Turbine Simulation | Andy QUB | FLUENT | 5 | September 19, 2015 06:38 |
Sliding mesh vs MRF in axial turbine simulation | Vito | FLUENT | 3 | December 21, 2011 04:57 |
Setting up gas turbine blade geometry for meshing | sherifkadry | Main CFD Forum | 0 | June 1, 2009 16:28 |
Help! How to find efficiency of hydro turbine? | Sanjay Jain | FLUENT | 0 | May 9, 2008 12:09 |
Modelling a ducted turbine boundaries | t_clark | OpenFOAM Running, Solving & CFD | 1 | March 11, 2008 13:57 |