Dialog Box October 200830 Sep, 2008 By: Cadalyst Staff
Readers have their say.
After reading "Technology for Civil Infrastructure," ("AEC Insight," Cadalyst, May 2008), I wanted to comment. One of the driving principles of Bentley's overall software architecture, and perhaps even more intensely in the geospatial part of the business, is the principle of supporting and enabling interoperability between software programs, file formats and (spatial) data stores. The majority of Bentley's geospatial users in an enterprise environment are using either Oracle Locator/Spatial or ESRI's ArcGIS, and Bentley's geospatial products work seamlessly with both programs at the desktop level with products such as Bentley Map (or Bentley Map-based applications such as Bentley Electric, Bentley Water, Bentley Wastewater, or Bentley Gas), and at the enterprise-GIS level using Bentley's ProjectWise Connectors for ESRI ArcGIS or Oracle. In other instances, Bentley supports industry-specific GIS platforms such as GE Energy's Smallworld, which works out of the box with Bentley Expert Designer Electric. Bentley's approach to GIS is pragmatic and applied. Bentley is not in the generic GIS market; it is all about advancing GIS for infrastructure and that means being able to access, edit, and post GIS data wherever it resides to enable the lifecycle management workflows that support and sustain infrastructure.
Bentley Systems, London, United Kingdom
Link to Nowhere
You changed something about the URL links used in the Tips & Tools Weekly e-mail newsletter compared with what you used at the end of last year or the beginning of this year. My IT folks tell me they strip and block URLs with IP addresses in them to prevent viruses. As a result, it is hard to follow the tips offered in the e-mail. Please consider rethinking this new URL format used in your e-mail newsletters.
Ship Systems, Pascagoula, Mississippi
Cadalyst is making changes to minimize the use of these tracking URLs that are problematic for many readers. You'll notice the change first in Cadalyst Tips & Tools Weekly e-newsletter, then we'll roll it out in the remainder of our e-newsletter titles in the coming weeks.
Kudos to Robert Green
Keep those superb Robert Green articles coming! He seemed to be writing directly to me. Thanks for including his articles.
Innophos, Nashville, Tennessee
Reports of 2D Death Have Been Greatly Exaggerated
Your article titled "Is 2D Drawing Doomed?" prompted me to respond. I have been drawing for a long time. When I started it was drawing on paper. Now working in AutoCAD 2009, I find that drawing parts has not changed much, only the software has. The person drawing still has to visualize in his mind what he intends to draw. We still use 2D drawing at my company with no intentions of going to 3D. 3D is simply not needed for what we do. Our drawings are not rocket science, but the machines we build are. 2D drawing for a lot of small businesses will be here for a long time. It would be nice to read an article about how 2D drawing is still going strong!
Kusel Equipment, Watertown, Wisconsin
I agree that 2D drawings are still necessary today. Perhaps in the future by applying STEP or other technology to communicate design-intent information, drawings will become obsolete, the necessary information will be placed completely within the model, and all downstream users will be using high-level technology that can extract and apply the embedded information, but that day has not yet arrived.
2D drawings are used by the factory, both by the machinist (and optionally by the numerical control [NC] programmer) before they move on to inspection. Today, much quality assurance (QA) is done in real time by the machinist on the shop floor. Inspecting bad parts will not correct the problem, just scrap out the order that was run. This is where quality is built into each process step, and each process must have this information communicated clearly. This is where tolerances must be known, understood, and used. True positioning for holes, for instance, must tolerance their relationship to one another when they are a grouping, and their relationship to datums. These are expressed by their tolerance in location as well as tolerance in diameter. The same is true for all features of a given part. Parts cannot be made perfectly. Tolerances and the use of geometric dimensioning and tolerancing (GD&T) rules and methods are necessary to enforce the original design intent. The design engineer has a huge vested interest in overseeing that this step agrees with this design intent, and the manufacturing engineer, who is best at applying the tolerances, is cost- and process-driven and is seeking to give the factory information and tolerance for each feature that will effect the highest productivity at the lowest process cost. One factory or area in a large factory may be making a part with a hole in it, and another is making a pin to fit into that hole. In the end, when the parts come together, the design intent must converge, and only through explicit tolerancing will this happen correctly and satisfactorily.
At the same time, even before the factory begins to manufacture, solid modeling is best done using the nominal design dimension. A hole might be dimensioned as 0.5000" +0.0005"/-0.0000" diameter and a pin dimensioned as 0.4997" +0.0000"/-0.0005" diameter; a theoretic fit of 0.0003" to 0.0013" will result between the two components. Nominally, the diametric size of the hole is 0.50025" and the pin is 0.49945"; a nominal fit of 0.0008" would result, which will fit the design intent exactly. However, modeling the hole as 0.5000" diameter and the pin as 0.4997" diameter and not using the tolerance in the model creation robs the design intent and implies that a fit of 0.0003" is desired. The factory loses the benefit in controlling production of this additional 0.0010" of fit between the two items. This does not include additional tolerancing constraints implied by -- for example, Regardless of Feature Size or Maximum Material Condition notes -- in considering the relationship of the hole with part datums or other features or the straightness, taper, or maximum out of roundness of the pin.
A gap exists, along with the danger of affecting the original design intent, when a customer supplies only a 3D solid model for a factory to produce. The NC programmer, assuming the part will be produced using computer numerical control (CNC) machinery, can only program the part to the size of the model features. If tolerances and GD&T features also are communicated, then more manufacturing resources can be spent on close-tolerance features and less on nonfunctional or loosely toleranced features. Where one hole might be a through hole for bolt, another may be a hole for locating pin, used to align the part with another part at assembly. Knowing which features to concentrate on, and which are not of great concern, will wring out the lowest production costs and the highest productivity and result in parts that meet the quality and design intent of the designer.
Regardless of the solid model or the actual shop results, the devil is in the details, and these are communicated by the tolerances and the GD&T features, which are best expressed by 2D drawings.
CyberCADCAM, Olalla, Washington