This is the DFDC source deck dfdc-dist, *************************************************************************** DFDC 0.55 1/31/05 HHY This is the first source release of the interrim code. The program is not complete by any means and will be undergoing constant revision for the next few months. Users are advised to stay "flexible". This version is based on the old (incorrect) theory documents of 12/04 and assumes that the wake vorticity is set by a rotor local wake advance ratio (WBGAM in the code). This version implements that theory but has shown issues for incorrect wake loading when the wake radius changes significantly. This version implements a duct and centerbody geometry with single rotor in the duct. The rotor can be an actuator disk or a bladed rotor. An actuator disk specifies a disk circulation (which corresponds to a loading defined by an enthalpy jump (total pressure jump) at the rotor disk. For the time being the most stable numerical scheme analyzes the actuator disk cases. Note that the initializer for the rotor-only cases is not implemented. These cases may require the user to start them with an assumed thrust (OPER command RINT with a thrust level) A very initial version of the rotor drag model using line source elements is implemented but these elements (source lines to represent rotor section drag blockage) do not automatically set their strength. For the moment they can be set using the SING command to directly set the source strength (see theory manual). The raw beginning of the DRAGOBJ is implemented but these elements (source lines to represent drag blockage) do not automatically set their strength. For the moment they can be set using the SING command to directly set the source strength (see theory manual). This version implements file load and save for either a DFDC duct file format using keywords to separate data items, or optionally you can read a simple XFOIL multi-element airfoil file (for the duct/CB walls) using the GLOA command (from the main menu). *************************************************************************** DFDC 0.56 2/2/05 HHY Minor changes before release. Added quick hack for FOM display to oplots.f Fixed SAVE option to dfdc.f (did not save comment lines for ACTDISK). Added FOM to global common. Fiddled with rotor analysis convergence params (needs rewrite for more solid convergence). *************************************************************************** DFDC 0.57 2/3/05 HHY Changes to rotor blade initializer (sets BGAM from geometry for imported rotors). New command RINI to re-initialize from blade itself. Added automatic source setup for drag objects and rotor blade drag elements. They are now included in the system w/o intervention. Note that CDA input for drag objects now corresponds to CDave = CDA/A (need changes to theory manual). Fixed PLOT 12 plotting bug (overwriting NP!). *************************************************************************** DFDC 0.58 2/8/05 HHY Changes to GDES for setting reference point for elements. Added options to set reference point to LE,TE,user or cursor specified points. Added option to OPER for geometry plotting box, GBOX command. This takes an Xmin,Ymin and Xmax,Ymax box to use for detail area plotting of Cp and other rotor data. *************************************************************************** DFDC 0.59 2/18/05 HHY Changes to oper.f for CPWR command to properly write x,y,cp,q data for elements to file. *************************************************************************** DFDC 0.60 3/7/05 through 7/10/05 HHY Added multi-plane cascade lift effects to aero.f (changes lift slope as function of blade station solidity). Fixes and clean-up on old solver. More robust criteria for SOR relaxation. Added option to drive blade pitch to match thrust for defined rotors. Fixes to UVC flow survey option. Added streamline trace and display options. Reduced written output with LDBG flag for added output diagnostics Bug fixes for duct and CB forces with TE closure panels (gave wrong forces for both, but CB forces were mostly affected. *************************************************************************** DFDC 0.62 8/23/05 HHY New solver, explicitly satisfies zero pressure on wake streamlines. BC on body wakes set to match vorticity at duct or CB TE. Fixes to streamline tracer. *************************************************************************** DFDC 0.63 9/13/05 HHY Remove almost all traces of old solver. Added Q (velocity magnitude) display as alternative to Cp displays. Added stagnation point finder and display. More trace options (fwd/backwards/both) to streamline tracer. *************************************************************************** DFDC 0.64 9/21/05 HHY Cleaned up considerable bugs in paneling routines, much more robust default paneling, user specified paneling now read at same time as case loading GDES functions now all work on complete airfoils (however some do NOT work on centerbodies for now) Multiple drag objects now allowed (up to 10 by current dimensioning) Fixed OPER menu listing of commands Fixed SAVE bug in DFDC main *************************************************************************** DFDC 0.65 9/28/05 HHY Where to start...many changes All plotting displays in OPER are now cleaned up immensely, new XY geometry display added Solver changes: many... Problems with setting VMAVG and calculating GTH (gamma on wakes) were fixed these caused the centerbody wake to slowly collapse if wakes were relaxed. Bugs with setting TE closure panel strengths were found as well, now fixed VMAVG limiter added to avoid transients with zero velocities Bug in UVC flow survey velocities, Cp's and flow data fixed. Bug in Cp calculation fixed for added enthalpy, entropy and circulation (was not properly setting these for out-of-duct points). Tip gap model now implemented (command TGAP in OPER). Matches some test data very well, may need additional input parameter for zero-loss tip gap limit. The tip gap is modeled by modifying the grid at the shroud end of the rotor to add two radial points. The circulation in the actual gap is set (and held to) zero (BGAM=0). This produces a very realistic behavior for large gaps, some test data seems to show that a minimum gap should produce less loss, stay tuned... Thin surface ducts (zero thickness, like a bellmouth) now work, and appear to work (in my test cases anyway) in QDES for design as well. Some kinks remain in QDES for regular thick ducts, however, and this needs further work. The solver issues may prompt more changes to solver to eliminate the GAM-GTH duality, issue not settled yet... *************************************************************************** DFDC 0.66 10/8/05 HHY This version adds BL analysis to duct and centerbody walls. A new command has been added to the OPER menu, BL, which adds capability to calculate BL's from the inviscid velocity on the walls. Note that this is not an interacted BL (as used in XFOIL). The BL analysis here assumes that the wall velocities are unchanged by viscous effects. The BL submenu includes options for setting transition parameters and does plotting of BL parameters. Viscous forces are now calculated (integrated Cf on walls) and added to inviscid forces. *************************************************************************** DFDC 0.70pre1 10/26/05 HHY This version adds the long-awaited multiple blade capability (rotor+stator, etc, in all the endless possibilities). The code is changed quite a bit for this release to add these capabilities. The multiple blades are handled by an index (1,2,etc) that can be typed in at the OPER menu to change the working disk/blade row. Subsequent commands that refer to a blade row (like RPM, PITCH, AERO, or plot commands that display data for a blade row) then use the index for the working disk. The index of the working disk is displayed only if more than one blade row is defined, and is shown as: .OPERi(1) <- refers to disk #1 as the working disk to change working blade to disk #2 type 2 at the prompt and you will get: .OPERi(2) <- disk #2 is now the working disk A stator that is designed to deswirl a rotor with a BGAM=10 has a BGAM=-10, i.e. you are designing a stator to remove the circulation. This has some implications, a negative circulation implies a negative CL (this depends on how you define CL but to remain simplistic in DFDC a BGAM<0 will be generated by a CL<0. Note that the AERO data for the disk (which is defined separately for each disk/blade row) will need to be manually "inverted" (CLMAX,CLMIN, CLCDMIN) adjusted so that it corresponds correctly for the CL<0 required. Note that DFDC will not allow you to design a stator (or rotor) with negative chord so it tries to do the right thing, the correct input for designing such a stator/rotor is to specify -CLdes. Note that RPM can be positve or negative, to indicate rotation direction. Beware that RPM>0 with BGAM>0 implies power>0, as does RPM<0 with BGAM<0. RPM>0 with BGAM<0 implies negative power. Keep this in mind... The code now can plot the dCL/dAlf in the PLOT 12 option for each disk (as well as a lot of other new stuff, like solidity, loading, etc). Docs are needed to explain all this new stuff... The blade outlines are now plotted on the meridional views as soon as they are designed (actuator disks are plotted as simple lines, as before). You will find, for example, that a stator really needs lots of blades to avoid driving the chord (solidity) through the roof (this impacts the loading through the CLalf which reduces as solidity is raised. Note that a stator design seems to show blades that are oriented almost streamwise. DFDC displays the blade outline defined by the chordline to the zero lift angle. A real section designed for the high turning that is required for a stator will have very high camber (i.e. LE drooped into the swirl direction, TE lined up with the streamwise direction). Options to design stator blading will be added soon and will display candidate sections. Note that "regular" airfoils will probably not work well as stator sections. *************************************************************************** DFDC 0.70rel 11/29/05 HHY This is a cleaned up version of the 0.70pre1 code. This has a few bug fixes but differs principally with addition of the src-dll directory. The src-dll directory contains source and Makefiles for a DLL version of DFDC for Win32 (though it compiles code, a non-dll version library and test code under Unix for test purposes). The Win32 make process in this directory uses Makefile.NT to produce a Win32 DFDC DLL (library with core solver functions). Note that the Win32 DLL is intended for use with a Win32 GUI (developed separately). A normal Win32 executable for DFDC can be made from the ./bin directory using Makefile.NT.