The ISO T24 STEP-Manufacturing team met as part of the larger TC184/SC4 meeting in Stockholm during the week of June 11-15. On June 14, the team met at the KTH production engineering labs for an Industry Day and demonstration of machine tool accuracy calculation.
We discussed the modeling of semantic tolerances for distributed assemblies. The key issue is how to represent a semantic tolerance when the constrained faces are in different file(s) to the tolerances. Two solutions were discussed and will be tested in the coming months.
We began the discussion of CAM to CAM data exchange that continued on Friday.
The meeting was held in the XPRES Manufacturing Research Laboratories (www.xpres.se) of KTH. Many thanks to Professor Torsten Kjellberg, Dr. Gunilla Sivard and their team for letting us use their excellent research facilities.
The scenario planned at the beginning of the project was demonstrated. Multiple CAD and CAM systems were used to build the STEP data defined by the STEP and STEP-NC standards. Videos showing how the data was made have been placed on the demonstration web page and are also listed below. Custom systems were used to "fill in the gaps" where a commercial translator was not yet available. They created PMI data for the tolerances, kinematic and accuracy data for the machine, and catalog data for the cutting tools.
The machining accuracy demonstration showed how to predict the accuracy of a machine tool for a particular part in a particular setup by measuring the loaded and unloaded accuracy of a machine tool. The unloaded accuracy was measured as per the ISO 230 specifications and represented both as ASME B 5.59-2 and STEP data. The loaded accuracy was predicted using a loaded double ball bar test developed by KTH and tool force load software developed by Boeing.
The results were good with a strong correlation between the accuracy predictions and the measured part tolerances, but further testing is necessary because of differences in the predicted and measured accuracy of the Y axis.
The definition of tool changer motions was shown by exporting the kinematics from NX using one of the custom translators developed by KTH. The data was loaded into a viewer to verify and execute the STEP kinematics.
The definition of PMI for a cutting tool assembly was demonstrated by assembling components from two tool vendors (ISCAR and Sandvik Coromant). Tolerances were then added to the assembly using NX. A tool developed by KTH associated those tolerances with the properties defined by ISO 13399. The result was then exported as an AP242 file that was viewed in ST-Machine and a STEP viewer developed by LKSoft.
The above is at the top of the STEP food chain because it contains geometry, assemblies, semantic tolerances and externally defined properties (ISO 13399/PLIB). It will now be used as the STEP-Manufacturing test case for STEP file distribution. The geometry components will be divided and returned to the two tool vendors (Sandvik Coromant and ISCAR). The team will then test whether the semantic tolerances and PLIB properties in the master file can be maintained using the new Part 21 external references.
We continued the discussions on CAM to CAM data exchange. We agreed to make a presentation to the CAM vendors at the IMTS show on Thursday September 13th. The presentation must emphasize the value proposition for vendors and users. An invitation letter was drafted and reviewed.
We discussed the next steps. Mikael Hedlind suggested that a STEP-NC Notepad should be developed to enable the development and editing of toolpaths. We reviewed the STEP cloud architecture and made changes to reflect the harmonization of AP242 and AP238.