==== AFT Arrow 4.0 (Version Date 2012.05.04) Release Notes ===== This file contains last minute information not included in the version 4.0 manual. Topics. 1. Information about tokens. 2. Errors in and updates to the AFT Arrow 4.0 manual 3. Sample and verification files included 4. Notes for network installations 5. Setting up User Log tracking 6. Graphic Server conflicts 7. Changes and bug fixes ============================================================== 1. Information about tokens ============================================================== Information on issues related to protection tokens is given in two places. The TOKEN.HLP file located on setup CD has basic information about managing tokens. The ARROW.HLP file has additional information about setting up pointers to tokens located on LAN's or WAN's. Look for the CONTROL.AFT topic. ============================================================== 2. Errors in and updates to the manual ============================================================== First printing - none ============================================================== 3. Sample and verification files included ============================================================== When you install AFT Arrow, a folder named EXAMPLES is created in your AFT Arrow folder. Several input files are provided for your review. See the help file ArrowExamples.hlp for additional information. In addition, a directory called VERIFICATION is created. Here there are numerous files that give information on calculation verification. See the help file A4Verify.hlp for additional information. ============================================================== 4. Notes for network installations ============================================================== See the ARROW.HLP "Information on Software Installation" topic for important information about network installation. AFT Arrow 4.0 requires a number of auxiliary files to run properly. Because of how 32-bit Windows operating systems work (Windows 9x, NT/2000/XP), certain files must be registered in the Windows registry. For software installed directly on the client PC, this situation works well. However, it creates a dilemma and an inconvenience for software installed on a local or wide area network and the installation becomes two-step process. The first step is to install AFT Arrow on the network using the "Network (auxiliary files on client)" in the setup program. This performs a typical network installation except for the auxiliary files are not installed or registered. The second step is for each client to run the Client Setup program installed into a subfolder below AFT Arrow named "Network Client Setup". This process uses standard installation protocol whereby the auxiliary files are copied to the client's SYSTEM folder only if they do not exist, or are newer versions than those that already exist. The Client Setup will also install the example and verification files on the client and will create shortcut links on the Start Menu referencing the applications and files on the network server. Finally it will update the Arrow4.ini file with details of the setup. ============================================================== 5. Setting up User Log tracking ============================================================== Summary of User Logs Each time a user starts AFT Arrow information about the user’s computer log-on name, computer name and time AFT Arrow was started is automatically logged in a file called ArrowUser.log. This file is located in a folder called Logs. The log can be viewed using the AFT User Log Viewer that is installed in the Application folder. This will show all people currently using AFT Arrow. Chempak data token usage is logged in to a separate file that can be viewed using the User Log Viewer. Note: This log file is only created when software tokens are used. No log file is created if a hardware key is used. Using the AFT User Log Viewer When AFT User Log Viewer is started, it will automatically determine where the application token is located and try to open the user log file. The User Log Viewer can be run independently and does not need to have AFT Arrow running to work. A table at the bottom will show the current people using AFT Arrow. It will show the log-on name of the user, the computer they are using and the time they started AFT Arrow. If there is contact information available (see Requiring Contact Information below) it is also shown. Choosing Refresh from the File menu can refresh the list. Other information about the log file and token is also shown above the user table. This includes the location of the log file, the date and time the log was created and last modified, the location of the token, the token serial number and the maximum number of current users. You can switch to view the user log for the Chempak data by choosing Chempak Log from the File menu. This will show the people who are accessing the data files from any application. For example, if one person is using the Chempak data in AFT Arrow, another in AFT Fathom and a third in Chempak Viewer; all three users will be listed in the table of users. You can switch back to the application user log by choosing Application Log from the File menu. Control.aft and Remote Tokens The AFT User Log Viewer relies on the Control.aft file located in the application folder to determine which application log file is to be viewed and where the application and Chempak tokens are located. If this file is not found an error occurs because the User Log Viewer cannot get the required information to run. The default location for the application token is the application folder. However, the tokens can be deployed at a different location for various reasons. The Control.aft file describes where the token is found. The User Log Viewer will use this information to determine where the log file is located. The log file is always placed in a folder called Logs, which is a sub-folder of the location set in the Control.aft file. The Control.aft file also describes where the Chempak data files and the Chempak data token are located. If no license or token is found for the data, an error occurs and no information will be given. Requiring Contact Information It can be helpful to have more information about the person using AFT Arrow (or Chempak data) than is standard. In this case contact information can be required. To enable this feature, open the log file (located in the Logs folder) and find the heading [FILE CONTROL] and set Require Contact Info = TRUE If contact information is required but not available, the user that launches AFT Arrow will be asked to provide his/her name, location and telephone number. This information is saved in the user’s Arrow4.ini file, located in the Window folder. This information can be changed by the local user by editing the Arrow4.ini file with a standard text editor such as Notepad. Disabling the User Log Feature The feature of logging users can be disabled by editing the log file (located in the Logs folder). Find the heading [FILE CONTROL] and set Enable Log = FALSE Multiple Entries and Other Problems Occasionally a user may be entered in the log file even though he has closed the application. This can occur for a number of reasons, including not exiting the application normally. If this happens, the next time the user launches the application he will be asked if he wants to overwrite the entry in the file with this new log-on information. Typically the answer is "Yes, overwrite the information". If the user is actually running two incidences of the application then answer "No" to this question and both incidences will be logged. ============================================================== 6. Graphic Server conflicts ============================================================== AFT Arrow uses a third party tool for graphing called the Graphics Server. There are four files that are used to load and run the Graphics Server: GRAPHS32CX GSW32.EXE GSWDLL32.DLL GSWAG32.DLL Because of how the Graphics Server is written, multiple copies of GSW32.EXE or GSWDLL32.DLL in your Path will cause problems when your try to load AFT Arrow. In practice, the most common problem is a network installation where all AFT Arrow files have been installed into the network folder along with AFT Arrow. Thus, GSW32.EXE and GSWDLL32.DLL will exist in the AFT Arrow network folder. A client PC which happens to have copies of either or both of these files on the local hard drive (usually in the WINDOWS\SYSTEM folder on Windows 95/98 and WINNT\SYSTEM32 on Windows NT) will fail to load properly. The most likely reason copies exist on the local hard drive is another application installed locally uses them. A complete list of your options is given below, along with an explanation of the impact of each option. The recommended option is number 5. In the following explanations, the term conflicting files means the GSW32.EXE and GSWDLL32.DLL files. Your WINDOWS\SYSTEM folder is just called your SYSTEM folder. 1. Delete the conflicting files on your hard drive. This will solve the problem for AFT Arrow, but may cause other software you use to stop working. This is not a permanent solution in that if you install other software in the future that installs copies of the conflicting files into your SYSTEM folder, the current problem with AFT Arrow will occur again. 2. Change the names of the conflicting files on your hard drive. This will solve the problem for AFT Arrow, but may cause other software you use to stop working. However, when you try to run the other software and it fails to load because of these missing files, you can change the names back to the original names. This will then cause AFT Arrow to stop working until you rename them once again, but will allow you to run your other software that needs these files. This will also require you to remember which files you renamed. This is not a permanent solution in that if you install other software in the future that installs copies of the conflicting files into your SYSTEM folder, the current problem with AFT Arrow will occur again. 3. Move the conflicting files on your hard drive into the application folder that uses them. This will solve the problem for AFT Arrow and still allow the other software you use to run properly. However, you will need to know which software uses the conflicting files so you can copy the files into their folder. If you remember what you do, you can move the files out of the SYSTEM folder (or wherever conflicting files exist) and into a temporary folder. When a software application that used to work fails because these files are missing, you can copy the files into its folder at that time. This is not a permanent solution in that if you install other software in the future that installs copies of the conflicting files into your SYSTEM folder, the current problem with AFT Arrow will occur again. 4. Allow AFT Arrow to automatically rename the files for you. This is not a perfect solution, but it does have advantages. This option enables a feature whereby AFT Arrow searches for the conflicting files in both your SYSTEM folder and WINDOWS folder before the full AFT Arrow application is launched. If it finds them, it automatically renames them and thus only the copies in the AFT Arrow network folder are found. When you exit AFT Arrow, the files will be renamed back to their original names. The advantages are, 1) you can continue to run your other software that uses these files and AFT Arrow with no conflicts (as long as they are not run at the same time) and, 2) even if you install other software in the future, this option will continue to work for you. The disadvantages are, 1) you are permitting AFT Arrow to change files on your local hard drive every time it is run, which understandably makes some users uncomfortable and, 2) you cannot run AFT Arrow at the same time as the other software which needs these files. 5. Reinstall AFT Arrow with the option to run auxiliary files on the client (recommended). This option is the best long-term option. To do this, first uninstall AFT Arrow from the network and reinstall it with the option to use auxiliary files on the client. See Information on Software Installation for more help. ============================================================== 7. Changes and bug fixes ============================================================== *** Changes for Version Date 2006.03.30 (none: initial release) *** *** Changes for Version Date 2006.04.13 *** >> Changes to computation logic that could impact results << 1. None. >> Changes to data management that could impact results << 1. Added check on Cost Groups for the CST module. 2. Fixed problem with the Area Change Specification Window losing the user specified loss if it was being based on the downstream pipe area. >> Changes that do not impact results << 1. Fixed problem where the tokens for the modules were not being checked properly which cause the module to not find a valid token. 2. Added Sentinel hardware key support for modules. *** Changes for Version Date 2006.05.03 *** >> Changes to computation logic that could impact results << 1. When a Control Valve is fully open (not controlling) and uses a Cv loss model it was not taking into account sonic choking. This was fixed. >> Changes to data management that could impact results << 1. The last two parameters at the end of the Compressor data were reversed. This caused Compressors to not remain attached to database items. The database item needs to be deleted and readded. 2. Fixed problem with the mass unit 'g' for gram. >> Changes that do not impact results << 1. When using Global Junction Edit, if a the data is coming from a database item all the data is now copied to the junction first and then any additional data selected on the grid is copied. 2. When an Arrow 3.0 model that uses One-Way-Valves specified as a Relief Valve is brought into Arrow 4.0 the icon will be changed to Arrow's default. 3. Changed the Valve specifications form to better reflect the required input when a Handbook valve type is selected. Fixed problem were the handbook valve type was not being saved. *** Changes for Version Date 2006.05.05 *** >> Changes to computation logic that could impact results << 1. Fixed problem with Control Valves that keep them from failing fully open when the dP was greater than zero but the fully-open K factor was not sufficient. The valves would report 100% open and full set flow, but this situation was not feasible. >> Changes to data management that could impact results << 1. None. >> Changes that do not impact results << 1. Fixed problem entering data in the Control Valve Open Percentage grid. The data could be pasted in from the clipboard but not entered by typing directly on the grid. *** Changes for Version Date 2006.07.06 *** >> Changes to computation logic that could impact results << 1. Control valves which were fully open and used a Cv value were not calculating choked flow correctly. This was fixed. >> Changes to data management that could impact results << 1. Fixed problem showing Chempak mixture concentrations for only some pipes or junctions in the output. 2. Fixed problem when a model was reopened and the output was unarchived the fixed flow compressor nominal efficiency was not updated and remained 1 until the model was rerun. 3. For relief valves which were exit type and choked, the pressure drop across the valve was not being reported correctly. This was fixed. 4. For control valves which were failed open and used a Cv value, the Cv value was not reported properly in the output. This was fixed. 5. The displayed values of flowrate for different components of a Chempak mixture were reported incorrectly in some cases when normal or standard units were requested such as "scfm" or "Nm3/h" >> Changes that do not impact results << 1. Added a check for pressures and temperatures above critical values when using ASME properties to prevent an erroneous error message regarding steam property calculations from being generated when writing output to the output window. 2. Fixed problem with a Note that was changed in a child scenario but had the same number of characters as the parent. The changes would be lost when the scenario was changed. 3. Fixed problem with the output parameter descriptions not being displayed (wrapped) in the Output Control. 4. Fixed problem where the junction type specific output tabs would not print. 5. Changed behavior of Pressure Drop Control Valve special condition setting. If the user choose the Open Fully - No Control option it now treats the PDCV as a lossless valve. 6. Fixed a problem when reading in Arrow models prior to version 4 where a relief valve was being displayed with a check valve icon. 7. The reported gas constant for Chempak mixture components had an incorrect unit conversion. This only affected displayed values in the General output and not any calculations. *** Changes for Version Date 2006.07.27 *** >> Changes to computation logic that could impact results << 1. Fixed a problem whereby Pressure Reducing Valves (PRVs) with backwards flow and hence loss of pressure control were not calculating properly in some cases, particularly when there was branching pipework before the valve. >> Changes to data management that could impact results << 1. None. >> Changes that do not impact results << 1. Model Data was not printing out when there were no fitting losses. This was fixed. 2. The Fluid Database window was ot allowing graphing of certain parameters. This was fixed. *** Changes for Version Date 2007.03.05 *** >> Changes to computation logic that could impact results << 1. Fixed GSC handling of pressure sustaining valves with the pressure setpoint as a GSC variable. The setpoint was not being adjusted by GSC. 2. Fixed a GSC problem where GSC variables with flowrate setpoints changed but this did not get properly communicated to all pipes in the system. 3. Pressure sustaining valves which used a static pressure setpoint were having difficulty converging on the static pressure. This was fixed. 4. For convective heat exchangers, made some changes to how the heat rate converges and how it uses inlet vs. average physical properties. Should be slightly more accurate now. 5. Pressure drop control valves being used as a GSC variable did not vary the pressure drop during goal seeking. This was fixed. 6. Fixed some problems where pressures were not being set in closed sections, including when a compressor was running in a topping tank or a never-fail control valve. 7. Fixed a problem in some cases when a fixed flow compressor is in series with a control pressure pump where the flow between the two was set to zero. 8. Fixed a problem in some cases where the pressure drop across a FCV was not correctly displayed because it was being calculated before the downstream conditions were set. >> Changes to data management that could impact results << 1. When using a single compressor/fan to represent multiple parallel compressor/fans (using the # of parallel compressor/fans feature,) the flow was not being handled properly when creating a compressor/fan vs. system curve. This was fixed. 2. Fixed a problem in summing the total flow in some cases where the flow is backwards. 3. Added the control valve failure time to the output archive so it can be displayed in the output when the model is reopened. 4. When the setpoint of a Control Valve was changed it would disconnect the valve from the database. This was fixed so that changes to the setpoint and failure actions would not disconnect the valve from the database. >> Changes that do not impact results << 1. Fixed graphing problem that occurred where the user has not selected a proper pipe sequence for a profile plot, but was still allowed to create a profile plot thereby generating a strange graph. 2. Made a change to graphing colors so default colors do not overwrite user specified colors. 3. Made change to Chempak whereby the maximum number of components in a mixture could be more than 10. 4. Fixed issue with GSC where venturi loss values used as GSV variables were not being allowed at all. 5. Fixed GSC issue for screen junctions where a unit conversion problem occurred when area was used as a GSC variable. 6. Valve junctions which use handbook K values were never intended to allow them to act as GSC variables. If they were chosen as such, the variable would not be varied. Now the user is informed that they cannot be used as GSC variables at all. 7. Branch junctions with 3 or more connecting pipes should not be able to use static pressure as a GSC goal because it is poorly defined. This was removed as an available goal. 8. Fixed a confusing error message that could occur when there were fixed volumetric flow compressor/fans in closed pipe sequences. 9. Fixed GSC goal display error in the Output window. When using differential goals with junctions, the label for the second junction was displaying the wrong junction number. But it used the correct junction in the goal seeking. 10. There was no check to make sure a user with a controlled flow or pressure compressor/fan had actually entered a setpoint value. Now this is checked. 11. Added automatic relaxation information to the output results. 12. Fixed problem not copying the Junction Delta from the parent scenario. 13. Fixed problem where the Visual Report Title font became extremely large. This occurred with international settings where a comma is used as a decimal separator. 14. Fixed a problem with the Visual Report Title font being reset to the default when it was the same as the parent. 15. Fixed the display and transfer in the Same As Parent menu on the Output Preference Window. 16. Fixed problem displaying the Control Valve Fully Open Loss in the Model Data using certain combinations of display parameters and changed the heading to better reflect what data is being shown. 17. Made clearer information in Model Data for the Special Conditions and heat transfer for several junction types. 18. Fixed problem with the Ignore Blank Columns when the Model Data was printed. The action was reversed. 19. Added scenario notes to the Model Data. 20. Fixed problem displaying the Base Area on junction specifications windows when the pipe was a rectangular duct. It was displaying the hydraulic diameter. This was only a display problem. 21. Fixed problem displaying the pipe roughness data in Model Data using certain combinations of display parameters. *** Changes for Version Date 2007.06.05 *** >> Changes to computation logic that could impact results << 1. Fixed a problem when a model had two check valves in series that both fail. Previously only one was closing. 2. Fixed a problem with ASME Steam when trying to get a saturation value in Region 3. It was not able to determine the saturation line and would return an error. >> Changes to data management that could impact results << 1. Fixed a problem unarchiving the flow and resistance results when international number delimiter settings are used. 2. Fixed a problem with saving the segment length and Mach step size in Solution Control when international number delimiter settings are used. >> Changes that do not impact results << 1. Removed the message that the ASME Steam calculations could not determine the region. 2. Fixed a problem with Workspace background pictures were in one scenario and removed from a child. It would remove it from the parent as well 3. The Additional Efficiencies grid on the Compressor Specifications window was not editable. This has been changed so data can now be entered. 4. Fixed problem accessing a Chempak Data license using the Dinkey Dongle NetKey. *** Changes for Version Date 2007.08.24 *** >> Changes to computation logic that could impact results << 1. Added No Solution to the Pipe Heat Transfer output for pipes in closed areas. 2. Fixed problem not updating the Orifice Cd during a GSC analysis. 3. Changed the PSV failure state when the valve has to open beyond its fully open setting to be treated as excessive back pressure and fail accordingly. >> Changes to data management that could impact results << 1. Fixed a problem where the wall temperatures were being reported as 0 deg. R in the pipe output. The calculations were correct, but the values were not being reported correctly. 2. Fixed problem when the pipe heat transfer data was globally edited. >> Changes that do not impact results << 1. Fixed a problem with the density and enthalpy curves entered for different pressures when all 5 were not supplied and when they are curve fit. 2. Fixed problem showing the fully open losses in the Inspection Window and in Global Edit. 3. Added separate copy protection type for Chempak Fluid Database. 4. Fixed problems with display of status bar and grid on computers with certain screen properties. 5. Fixed problem with Renumber Increment feature. If junctions were renumbered Arrow would crash. 6. Fixed problem with enthalpy balance when Chempak mixtures were used. It was possible to lock up Arrow when loading a scenario. This was fixed. *** Changes for Version Date 2007.12.31 *** >> Changes to computation logic that could impact results << 1. Fixed problem when Compressors were used as a GSC goal where no units were available during the calculation of the goal resulting in an error. 2. When junctions displayed the mass flowrate in STP-dependent type of units (e.g. MMscfm) the incorrect STP values were not updated correctly and would result in an incorrect value in the output. The calculations were OK and the pipe results were correct. 3. Made a slight change to logic of when the length march method should switch to mach number march. This change should rarely have any effect. If it does have an effect, it should improve convergence. 4. When using compressor/fans with a flow controller and the flow units set to standard or normal volume for rate (e.g, scfm or Nm3/h) the flow rate conversion was not being performed properly and an incorrect flow rate was used for the control setpoint. This was fixed. 5. Changed transient tee resistance calculations by adding a lower flow ratio cutoff to avoid extremely low flow ratios which can cause unstable and inaccurate results. >> Changes to data management that could impact results << 1. Fixed problem where using multi-scenario output when doing a batch run where one or more, but not all scenarios use Chempak fluids would cause the output file to become corrupted when a scenario was rerun. The regular output view was not affected. 2. Fix problem in the Compressor Summary where the dPstatic was reporting the dPStag for Controlled Pressure compressors. 3. Fixed problem unarchiving out results where GSC was being used to vary Fixed Volumetric Flow Compressors. The results would not match the original run. 4. For a fixed flow compressor/fan, the nominal efficiency entered by the user was not being used in the output results. Arrow was always using 100% efficiency. >> Changes that do not impact results << 1. Fixed problem with models that have a large number of closed pipes and junctions taking an extremely long time to load a scenario. 2. Fixed problem where the Pipe Fittings & Losses were not printing in Model Data. 3. Fixed problem that caused long delays when multiple pipes and junctions were deselected on the Workspace. 4. When a module is activated when a model is loaded in the Workspace, the model is now exhaustively checked to insure all data that is newly required for that module is present. The pipes and junctions that have missing data are set undefined and the checklist is unchecked. 5. Added logic to handle output prediction of compressor effective polytropic constant when temperature rise is very small. Previously it could cause an overflow error. 6. Changed how control valves with Cv vs. percent open tables report the Cv in the Output window Valve Summary when the valve is controlling to setpoint. Previously the Cv was reported in the Cv column and now it is reported in the Cv Estimated column. 7. Fixed Workspace Undo problem when trying to undo the stretching of a single pipe. *** Changes for Version Date 2008.01.17 *** >> Changes to computation logic that could impact results << 1. When using assigned flow junctions and the GSC module, the assigned flow junction temperature variable was not working. This was fixed. 2. Fixed a problem which was amplified by a change made in the 2007.12.31 version related to detailed tees. The problem exhibited itself when pipes connected to a particular tee had reference flow directions opposite of the physical flow direction. The tee calculation routine should have taken absolute values when getting flow split ratios but it did not. This has now been fixed. 3. Fixed a problem created in the 2007.12.31 version which caused flow imbalances in models with fixed or known flows in branching areas. The flows were overwritten with initial guesses and never changed. >> Changes to data management that could impact results << 1. Fixed problem when an fixed loss value was entered directly in the Pipe Specifications Window when the regional settings used a comma as a decimal separator. The Fittings and Losses window was not affected. >> Changes that do not impact results << 1. When one moves a junction which has connected pipes, and the pipes are not moved at the same time as the junction, then when one undoes the move and then redoes it, the redo would not work properly. This was fixed. 2. Changed how the Solution Balance Summary reports mass and energy balance. Before it tried to always sum junctions to zero and show imbalances in the In/Out field. Now the In/Out field will only display for branch junctions with a flow source. And the total flow for junctions like reservoirs will show an imbalance because reservoirs are imbalanced by their nature. *** Changes for Version Date 2008.05.01 *** >> Changes to computation logic that could impact results << 1. When using dynamic mixtures and junctions with fixed flow rates such as Assigned Flow junctions or Compressor/Fan junctions with a fixed flow, if the fixed flow rate is a mass flow rate that uses normal or standard volumetric units such as scfm or Nm3/h and it is in a dynamic mixture flow path, the flows cannot be calculated accurately. Previously Arrow would display results without warning. Now a Critical Warning is given. 2. Fixed a problem where the losses for a detailed tee/wye that had flow coming in the branch and going out both straight-through pipes (or the opposite flow directions) were being set to zero. This was introduced in the 2008.01.17 version. >> Changes to data management that could impact results << 1. If a dynamic mixing calculation is performed and units in the Output window are changed for a column by double-clicking the column header, and the units being changed to or from are standard mass flow units like scfm or normal mass flow units like Nm3/h, then the mass flow rates were not being converted correctly for all pipes or junctions which are in a flow path which has dynamic mixing occurring. This was fixed. 2. Fixed potential problem in Global Junction Editing where if only the database name was selected, the data was not applied, only the name which caused inconsistent data. 3. When a goal used a density-based mass flowrate (like scfm) errors would be displayed and the incorrect actual goal was displayed when the model was reopened. The flows in the pipes and junctions were correct. 4. Fluid property data in Arrow.dat was updated. Updated Thermal Conductivity and Specific Heat for Air. Fixed curve fit type for Air Viscosity. Updated Thermal Conductivity for Helium. This affects AFT Standard fluid properties. >> Changes that do not impact results << 1. Fixed snap to grid problem when pipes or junctions were near an edge. 2. Fixed Workspace drawing anomaly when pressing the escape key while drawing a new pipe. 3. Added millibars (mbar) as unit of pressure 4. When performing global editing using the sonic CdA parameter for several junction types it was displaying in the Optional area of the Global Editing window. It was moved to the appropriate area. 5. Fixed printing problem for Graph Results which could cutoff part of a graph in certain cases. 6. Added a Caution message to the Output when user disabled either mass balance or energy balance in Solution Control. 7. Fixed display problem in Graph Results if a graph is made using the "Selected Solutions" tab and then another graph is made for a different tab. The x-axis labels were not being updated. 8. When entering a custom fluid into the fluid database the molecular weight is required. However, AFT Arrow was not verifying that this required data was in fact entered by the user. Hence a model using this custom fluid would not solve and give an error. This was fixed. 9. Allowed empty database files to be the target file in Database Manager. 10. Fixed problem where the model data setting for pipes would not be set as the user default unless the Model Data Control window was first closed and reopened. 11. Fixed a problem in the Fittings & Losses database the use of a comma in the type or subtype name would cause the data to become corrupted. 12. Fixed a problem where detailed pipe data for non-cylindrical pipe in Model Data was not being displayed properly. *** Changes for Version Date 2008.08.08 *** >> Changes to computation logic that could impact results << 1. There was a weakness in the solution method for valves modeled with a Cv and Xt value and an assigned flow junction upstream. When such Cv methods are used, AFT Arrow has to iterate on the pressure drop across the valve. This iteration is a complex function for which a Newton-Raphson iteration method is used. For certain conditions it solves to an incorrect negative root which makes it think it is sonically choked. This can cause the upstream pressure to be unrealistically high in the solution. An improvement was made to the iteration logic such that incorrect negative roots will cause the method to try again with a different starting value for pressure. 2. When a model uses all isothermal pipes then AFT Arrow can avoid the energy balance to save computation time. When a model started with all flows zero then AFT Arrow was assuming that all energy balances could be bypassed. It did this by setting the isothermal flag. However, if an event such as a relief valve opening caused the flow solution to change to non-zero values, the isothermal flag was not being rechecked. This caused the subsequent solution to not perform an energy balance and hence give incorrect results. This only affected models where the first solution has all pipes with zero flow. This was fixed. >> Changes to data management that could impact results << 1. In Solution Control if the user selects to disable the global mass and/or energy balance for troubleshooting purposes, and then clicks the "Arrow Default" button the global balance options are not cleared. This was fixed. 2. There was a data handling problem with the parameter "Residence Time" when creating Graphs and the graph values were always zero. This was fixed. 3. If using a compressor/fan input in volumetric flow rate units, when the compressor/fan vs. system curve was plotted the compressor/fan curve was not correct due to using non-updated data for density. This was fixed. 4. Curve fit configurations that use temperature would allow users to change the temperature units without redoing the curve fit. 5. In the Output window junction table the inlet and outlet static and stagnation pressures displayed confused some users in certain cases. Specifically, when junctions had multiple pipes connected the definition of inlet and outlet, as well as the difference between static and stagnation pressure was ambiguous. No calculation logic was changed, but the display logic was changed to make the displayed values less confusing. This was done for Spray Discharge, Assigned Pressure and Relief Valve junctions. 6. Corrected data in Arrow4.dat file. Corrected wall thickness for Stainless Tubing, 2-1/4 inch,11 BWG to be 0.12 inch. Corrected wall thickness for Stainless Tubing, 2-1/2 inch,11 BWG to be 0.12 inch. Corrected the degree position for the Idelchik Ball Valve from 80 deg. to 75 deg. with a loss of 330 (see dig 9-13, pg 565). Changed K factor for 10% open ball valve from Miller data from 350 to 400. >> Changes that do not impact results << 1. When using the batch run on multiple model files and including Model Data in the output report content, and sending output directly to PDF files, the PDF file Was not created because of a file name issue. This was fixed. 2. When using the GSC module, if the user transfers both the GSC variables and the hydraulic results back into the initial guesses it could cause matrix errors on future errors. This was fixed. 3. When double-clicking a pipe or junction label on the Workspace, the user is offered the option of relocating the label or opening the specifications window. And the user has the option to remember the selection for the future. This final option was shown inadvertently and did not do anything. The option is now hidden. 4. A user created fluid has gas thermal conductivity as an optional field. If the user tries to model pipe heat transfer with internal convection the thermal conductivity is required. When a model was run with this condition an error occurred. Now Arrow checks and warns the user if this happens. 5. The thickness of insulation was not being applied to the pipes if a default insulation setting was added to the pipe thought the Material Database. If the name of the insulation data set was selected again in the pipe specification window and closed, the thickness would be applied. 6. If only one data point was used for a pump curve the resulting curve fit was not done and the graph looked like random noise. This was changed to be a flat line. 7. When user selects "No" when prompted to update the curve fits in the Curve Fit Configuration window after changes are made, control is returned to the configuration window instead of closing the window. *** Changes for Version Date 2008.12.04 *** >> Changes to computation logic that could impact results << 1. Fixed logic for GSC when a check valve closes during the goal seeking. The check valve stayed closed rather than re-opening when conditions allowed. 2. Improved logic for when multiple control valves cannot control and check valves, relief valves, and controlled pumps exist. In the past these other components could open, close or lose control when Arrow was trying to sort through which control valves could not control. Now Arrow waits until all control valves are first determined before looking at the other components. 3. Fixed weakness in solver for control valves whereby a valve with fully open loss data can fail to control because of choking. Previously Arrow was not properly checking this possibility. 4. Fully open control valves with Cv data was not using the pipe diameter correction input by the user. This could cause small differences in results. This was fixed. >> Changes to data management that could impact results << 1. Fixed a problem when you have a compressor/fan in your model with variable speed control to control on suction or discharge pressure and the control point is conditional in that the compressor/fan only controls when the control point is exceeded. The results are correct when you run the model, but if you save the model and re-open it the speed is displayed as 100% and the pressure rise is displayed as if the compressor/fan was running at 100%. This was fixed. 2. In the Output window junction table the inlet and outlet static and stagnation pressures displayed confused some users in certain cases. Specifically, when junctions had multiple pipes connected the definition of inlet and outlet, as well as the difference between static and stagnation pressure was ambiguous. No calculation logic was changed, but the display logic was changed to make the displayed values less confusing. This was done for exit Orifices and exit Valves. 3. When running and displaying multiple scenario output some of the junction output could be displayed incorrectly. Specifically, a tank junction at the end of a pipe run which is sonically choked could have an incorrect inlet pressure. The outlet pressure would be displayed for both inlet and outlet. This was fixed. 4. Fixed problem reading in older models with a user specified additional loss valve. Unit conversion error would most likely occur when the decimal delimiter was a comma. >> Changes that do not impact results << 1. Fixed bug in Valve, Control Valve Specifications windows where the Open Percentage table looked for a maximum open percentage of 100% and treated decimal values like 100.0% as not equal to 100% and gave a warning. This no longer happens. 2. Fixed bug in Valve and Control Valve Specifications windows where pasting data into the Open Percentage table for Open Percent and Cv only without area led to the Cv data showing up in both the Cv column and area column. This no longer happens. 3. Changed how Cv and open percentage data is reported in the Output window Valve Summary when there is a Cv vs. open percentage table provided by the user. Previously the Cv was estimated with a nominal Xt. The new method interpolates directly from the table based on flow rate. 4. If part or all of the Workspace where a hidden annotation is located the annotation would be selected. If the selected objects were deleted the annotation would also be deleted. Now hidden annotations cannot be selected and are not deleted. 5. Fixed problem with the pop-up inspection window on the Workspace where it would lock up or crash depending on what mouse action was done while it was open. *** Changes for Version Date 2009.08.04 *** >> Changes to computation logic that could impact results << 1. When a control valve is in a flow path that is closed off it cannot maintain its setpoint because there is no flow. The logic to handle this in some cases could cause premature determination of other control valve states and hence incorrect results. 2. There was an error for junction pressure drop that could occur. Three things had to happen. First, the user modeled the pressure drop using a resistance curve. Second, the upstream and downstream pipes had to have different diameters. And third, the user had to change the "Base Area for Loss Model" from the default "Upstream Pipe" to something else. This affected all junctions with resistance curve options except for relief valve junctions which were already calculating properly. 3. When entering compressor/fan efficiency data, if the data is entered as interpolated and the compression process is selected as determine from efficiency data, the efficiency is 100% regardless of user's data. This was fixed. 4. Fixed problem with an isolated loop that was attached to a tank using a pipe outside of the loop (like a topping tank). The initial guess flows were never updated and the loop would have a flow when none should exist. >> Changes to data management that could impact results << 1. When using Global Junction Edit to changing from a controlled to a fixed flow compressor the control setting was not cleared which left the compressor in the controlled state. This was fixed. >> Changes that do not impact results << 1. When using GSC it was possible, probably through morphing, to confuse the GSC Manager about what type of junction was in the variable or goal. This would cause an error when the GSC model was run. A new check was added to tell the user what happened and what to do about it. 2. Fixed an issue on the Workspace when undoing a move with a pipe which has vertices 3. Fixed some problems running batch jobs when an error occurs in one of the models or scenarios. The batch run could get hung up. 4. Fixed problem with the Fittings & Losses grid when it was collapsed it sometimes could not be expanded. 5. Fixed an issue where carriage returns were not handled properly when adding notes to existing notes using Global Edit. *** Changes for Version Date 2009.10.07 *** >> Changes to computation logic that could impact results << 1. None. >> Changes to data management that could impact results << 1. When using Global Edit and changing curve-fit data, the data was changed even if the edit was canceled. >> Changes that do not impact results << 1. Added CrypKey support. *** Changes for Version Date 2010.02.01 *** >> Changes to computation logic that could impact results << 1. Fixed a problem reading a Bend type Fitting & Loss with a r/D = 1.5 when the decimal separator is a comma. This caused the actual loss for this type to be too large. >> Changes to data management that could impact results << 1. On pipe and junction specifications windows the user can push the F5 function key to jump to the next pipe or junction. If the user happens to have entered a value into a grid on these windows and then pushed the F5 key in some cases the new value was not kept. This was fixed. 2. When defining a new fluid the reference enthalpy and reference temperature are required data. AFT Arrow was not checking to make sure the user entered this data. Now it does in the Fluid Database window. 3. Changed display logic in valve summary for Cv data. The open percentage table was displaying zero and not working at all. This was fixed. When a control valve was choked an incorrect Cv was determined because it was based on unchoked flow. No calculation results such as flow or pressure were affected. Only the Cv output. 4. If doing dynamic gas mixing with Chempak, if the pipe or junction name field is removed from the default output report using Output Control the pipe or junction composition output display was shifted by one column. This was fixed. 5. Fixed problem unarchiving results for Assigned Pressure junctions that use static pressure as the setting. The stagnation pressure would be set to the static pressure value. >> Changes that do not impact results << 1. Fixed a problem when annotations are selected on the Workspace and then deleted using the Annotation Manager Arrow would crash. *** Changes for Version Date 2010.06.08 *** >> Changes to computation logic that could impact results << 1. If a Tank or Assigned Pressure junction was used and a single pipe was connected to the junction and the Energy Balance at Junction option was selected, nothing special was supposed to happen. And usually it did not. The Balance Energy at Junction feature is only supposed to have an effect if two or more pipes are connected to the junction. But if the user on a previous run had completed a run and used the "Transfer Results to Initial Guesses" feature, and then after that changed the input temperature of the junction, the original temperature would still be used at the junction. This was fixed so that whatever the new temperature is, that is used. 2. In systems with exit valves in lines with closed junctions or pipes the stagnant pressure from the exit valve was not being applied properly to the stagnant pipes and junctions. So the pressure results and only pressure results were incorrect only for the stagnant pipes and junctions. And only when there was an open exit valve at the end of the stagnant line. This was fixed. 3. Improved calculation of orifice subsonic pressure drop when using the CdA model. The improvement related to using the current flow rate vs. the previous iteration flow rate when calculating dynamic pressure and converting a pressure drop to a K factor. In the vast majority of cases this will have no noticeable effect on results. However, especially in cases where the orifice would choke but the user did not provide a sonic CdA, the subsonic case did a poor job of calculating pressure drop. This was to be expected since the flow was sonic and the subsonic model was not appropriate. The biggest impact of this change is to make models converge which previously would not. And when they converge there are warnings because of the sonic issues previously discussed. >> Changes to data management that could impact results << 1. Fixed problem importing pipe size changes from Excel where the pipe ID being used was not changed until the pipe specifications window was opened and OK was clicked. 2. Added a check to verify that the fluid being used in the model matched the data in the connected database. If it is different the user is prompted to update or keep the fluid in the model. Note that some fluid properties in the database were changed 2008.05.23. Air properties, for example, were updated and when a model that that used the older properties is opened, the user will be prompted to update the fluid properties or continue with the original values. 3. Fixed a problem adding a compressor configuration to a compressor with existing configurations. >> Changes that do not impact results << 1. Certain windows that created small graphs such as the Edit Fluid Database window had a print feature that did not work reliably. This was fixed. 2. Added warning about compressor/fan efficiency being greater than 100% due to user's compressor/fan power data issues. 3. Fixed a Workspace and Visual Report minor printing problem. If the user directly changed the pipe line pixel thickness on the Pipe Specifications window, it displayed properly on the Workspace and Visual Report but did not print properly. This was fixed. 4. Fixed a problem with Graph Sets. In some cases when an existing set was modified for styles the change did not display until the model was reloaded. This was fixed. 5. Added an error message when the .out file cannot be opened because the model is run from a folder where the user does not have write permissions. *** Changes for Version Date 2010.07.20 *** >> Changes to computation logic that could impact results << 1. Fixed a problem in the case of a branching leg that has zero flow but is not closed. Previously the pressures were not being updated from the initial guesses. 2. In certain circumstances, when a separator junction is in a leg upstream from a chokepoint, Arrow was not calculating the upstream mass flow rate correctly. This has been corrected. >> Changes to data management that could impact results << 1. None. >> Changes that do not impact results << 1. The Separator junction was erroneously set to ignore the special condition setting. This has been corrected. *** Changes for Version Date 2011.02.01 *** >> Changes to computation logic that could impact results << 1. A change made in the 2010.06.08 release to fix another problem created a new problem for Relief valves modeled as inline exit valves. The discharge flow and pressure was not being calculated properly. This was fixed. >> Changes to data management that could impact results << 2. When using static Chempak mixtures, after the model was run the check for saturation of each component was not done properly. However, if the scenario was reloaded the check was done properly. This was fixed and the check is always properly performed now. 3. Fixed a problem with Solution Control parameters affecting Lumped Adiabatic or Lumped Isothermal. In some cases a flag for the marching methods was being saved and used improperly on the lumped methods. 4. Fixed a problem using Copy-Paste where the pipes and junctions were not correctly connected. This could lead to problems when copying from one file and pasting into another. The connections would be corrected before the model was run. 5. Fixed a problem where flow imbalance warnings were generated when a model/scenario was reopened When the relaxation and the relative flow tolerance were automatically lowered. This was cause by checking the solution against the user-specified values and not the lowered values which were used during the run. 6. Fixed a problem with the CST Cost Database editor where the scale table data was being corrupted when another cost database file was opened while showing the scale table. >> Changes that do not impact results << 1. Fixed some display of false error messages that could occur after running a model but would not display when loading a previously run scenario. The messages were related to messages that said things like "exceeded maximum local iteration limit" or "failed to continue towards convergence". If the messages continued to display when subsequently loading a scenario they were real. If not they were false. 2. When you put in a compressor/fan curve and then added variable speed control, and then changed the compressor/fan curve, the variable speed input data would be lost. This was fixed. 3. When working with pre-existing compressor/fan junctions with multiple impeller configurations, if the engineering units used in the original impeller specification have since been removed from being active units (in the Unit Preferences window) then future editing of the compressor/fan configuration will result in a loss of the engineering units and loss of original data. This was fixed. 4. When performing curve fits on compressor/fan data users can set a maximum x-axis value for flow. However, on subsequent curve fits for this compressor/fan the value is ignored. This was fixed. 5. Clarified error message if the user somehow had a solution tolerance of zero. 6. The Renumber Wizard allowed you to change the pipe or junction numbers but it did not erase old output. So it could display output for pipes or junction numbers that no longer exists. This would be resolved the next time the scenario was loaded or the model was opened. At the moment of renumbering it could cause confusion. 7. Fixed problem when jumping from one pipe spec window to another when using general convection heat transfer, an extra layer of insulation would be added and the pipe would be undefined. Jumping again would remove this extra layer. 8. If a valve was copied from another valve using the SameAs feature and it was the same handbook valve type, the K factor would be erased. This caused the junction to become undefined and the handbook valve type would have to be reselected. *** Changes for Version Date 2011.02.09 *** >> Changes to computation logic that could impact results << 1. None. >> Changes to data management that could impact results << 1. Fixed a problem introduced in the 2011.02.01 release that caused the mass flow rate, velocity, etc., to be zero when a model/scenario is reloaded. 2. Fixed a problem saving the compressor controlling state and reading it in from the model file. >> Changes that do not impact results << 1. None. *** Changes for Version Date 2011.02.23 *** >> Changes to computation logic that could impact results << 1. None. >> Changes to data management that could impact results << 1. Fixed a problem when the CdA was entered for an area change but was ignored when the solver was run. >> Changes that do not impact results << 1. Fixed a problem introduced in the 2001.02.01 release where the convective heat transfer grid would not be displayed when a pipe was reopened. The data was there but not shown. If the Convective Heat Transfer option was reselected the grid would appear. *** Changes for Version Date 2011.03.31 *** >> Changes to computation logic that could impact results << 1. None. >> Changes to data management that could impact results << 1. None. >> Changes that do not impact results << 1. Changed help file format to .chm for Windows 7 compatibility. *** Changes for Version Date 2012.01.06 *** >> Changes to computation logic that could impact results << 1. Tightened tolerance when converting Cv values to K values. This may affect the mass flow rate in the pipes when Cv values are used in valves etc. >> Changes to data management that could impact results << 1. When using a junction that has a larger flow area than the pipe itself at an exit to a pressure (such as a relief valve) then the model may not converge if sonic choking exists. This was fixed. 2. When a scenario other than the current scenario and then the Scenario Manager is closed without loading the current scenario the number of pipe and junction objects from the parent was being reset to what ever the parent of the saved scenario was. This caused the data flow from the parent of the current scenario was corrupted and some objects which were deleted in the current scenario but exist in its parent were being shown. This was fixed. 3. When a scenario was promoted it would acquire the GSC Type of the new parent. This would cause problems graphing if the new parent was used a different state. This was fixed so that the original Analysis Type was retained. 4. Updated the arrow4.dat file with minor corrections to PVC, Ductile Iron and Copper Pipe. See notes in the file for specific details. 5. Fixed problem when global editing fan/compressor that are from a database. The current configuration was being reset to the first one. It is now the configuration that is selected in the global edit list. >> Changes that do not impact results << 1. Fixed problem graphing physical properties in the pipe material database. 2. If ESC was pressed when a pipe was drawn between two junctions while in multi-pipe drawing mode, the pipe would be deleted. This was fixed so only the multi-pipe mode was exited. 3. Added a caution when the current scenario is not displayed with the Multi-Scenario output. 4. Moved metric ton units (m-ton) from the English to SI system of units. *** Changes for Version Date 2012.05.04 *** >> Changes to computation logic that could impact results << 1. A potentially serious error was corrected in user custom database fluid property calculation. When using the option for table look up density equation of state, an incorrect compressibility factor (Z) was calculated based on the data. For fluids operating near Z values of 1 (i.e., ideal gases) the error was negligible. For fluids farther from Z = 1 the error was larger. Any models previously built using custom fluids with table look up density should be reviewed and re-run in this updated release of AFT Arrow. >> Changes to data management that could impact results << 1. There was a problem with Enthalpy and Density tables for a Custom Fluid that potentially could impact results. This only is an issue with fluids that were user-entered that used a table to specify the properties. First, if X-Y Interpolation was selected, the number of data curves was not being set and data could be lost. If, however, Polynomial curve was set first and changed to X-Y Interpolation, the number of curves was set correctly. Secondly, if there were less than 10 data curves of Enthalpy, the density table lookup would shifted. Potentially, a lower pressure would not be found in the table and two higher pressure points would be used to extrapolate density from the table. >> Changes that do not impact results << 1. Fixed a problem with the Toolbox Zoom-in feature where Arrow would crash if the zoom in was too small.