Dezignstuff SolidWorks Blog

This is the book that has been brewing this summer, and is scheduled to become available in November. A little editing and last minute additions remain. It is based on SolidWorks 2009 and 2010, and is a reference for new administrators, SW users who act as their own administrators, and IT people pressed into SolidWorks service.

I spent a couple of years as the Engineering Services Manager for a high tech manufacturer in Rochester, NY, and I learned a few things, but all of it the hard way. I was responsible for SW admin, document management, and the product development process admin in addition to a team of SW users. I spent some time as a software implementer and advanced troubleshooter with a SolidWorks reseller, which gave me experience with a wide range of companies and their business, engineering and documentation processes.

I asked Ed Cyganik, long time administrator for a company with structured processes to be my technical editor for this book precisely because his experience was very thorough and very different from mine. I was more familiar with writing my own standards for use in the consumer products and plastics design industry, and not as familiar with the highly structured auto/aero/military world of standards that Ed knows. Ed added a lot to the book, particularly in standards and file management areas.

This was a book that I felt needed to exist. So many people ask for this type of information, and basically, if you can’t find it on the SolidWorks website, and you can’t get it in bits and pieces from the SW forum, you’re not gonna find it.  The other CAD Admin books available on the market are either directed at Autocad, or are almost entirely about schmoozing management, not at all about any actual CAD functions, or rather old.

The big topics are installation in its various forms, Toolbox and Standards/Best Practice. The goal is as always to arm the reader with enough information to make informed decisions. The book is ~540 pages.

Chapter list:

1. Defining CAD Administration
2. Preparing Information Technology
3. Preparing Document Management
4. Operating System Tools you need to know
5. Training Users
6. Installing the Software
7. Licensing
8. Standardizing Settings
9. Maximizing use of Templates
10. Implementing Toolbox
11. Establishing Libraries
12. Writing Standards for SolidWorks
13. Custom Properties
14. Establishing Best Practice
15. Automating Tasks
16. Supporting Users
17. Collaboration
18. Managing Change
19. Rolling out new Service Packs
20. Rolling out new versions
21. Introducing SolidWorks Workgroup PDM
22. Installing and Configuring the Vault
23. Establishing users, groups and permissions
24. Using Toolbox and other libraries with Workgroup PDM

App A on-line help resources
App B Sources for Standards
App C Tools, Options

Here is what it’s not:

• Not a tutorial
• Not for teaching new users
• Not for classroom use
• Not to be taken internally

Toolbox is a piece of SolidWorks software that I have been in the habit of giving very poor reviews. In the SolidWorks releases prior to 2007, these reviews were definitely warranted, but in 2007 they (partially) solved the Huge Screws issue, and since then they have solved a number of smaller problems.

Due to all of these changes, I’ve recently done a bit of research on Toolbox again. It has been a couple of years since I went through the software thoroughly, and it was time again, mainly because of preparations for the upcoming SolidWorks Administration Bible (available this winter?). Anyway, Toolbox is a huge administration topic, and using the book as a platform to rant with the same old tired rants is not really an option. That’s what the blog is for ;o). How huge of an administration topic is Toolbox? Well, here’s a list of functionality that you will find either inside or associated with Toolbox:

• Fasteners: screws, bolts, nuts, washers,
• Bearings
• Bearing calculator
• Bushings
• Keys
• O-rings
• Grooves (O-rings and circlips)
• Pins
• Sprockets, Gears and pulleys
• Retaining Rings, Circlips
• Structural Members
• Beam calculator
• Cams
• International standards including ANSI (USA), AS (Australia), BSI (UK), CISC (Canada), DIN (Germany),  GB (China), IS (India), ISO (International/Europe), JIS (Japan), KS (Korea), and MIL (US Military).
• User-defined standards
• Uses mate references
• Brand name parts including Helicoil, PEM, SKF, Torrington, Truarc, Unistrut
• Mold components including DME, Hasco, PCS, Superior, Progressive
• Smart Fasteners: auto-populates assemblies with fasteners, also uses semi-auto mode
• Smart Fastener configuration with fastener types, hole types and size tolerance
• Hole Wizard: database of hole sizes to fit standard hardware including 3 fit conditions
• mass-populate holes through selection techniques
• Link hardware and hole sizes
• Special recognition as Toolbox parts in PDM
• Limiting selection by standard, type and size
• Automatic application of custom properties
• Customized part numbers and descriptions
• Specify drive type and thread display for screws
• Option to use configurations or create parts for separate sizes
• Administrator lock down of library settings with password
• Sharing Toolbox between multiple users

Wow. That’s a list. The problem is, does it all work? And if it does, why don’t people love this stuff? Is it really just that we have become so accustomed to hating Toolbox? My new take on Toolbox is that the main things wrong with it are that 1) Smart Fasteners delivers much less than its potential in fully automatic mode, and 2) that the documentation really doesn’t even begin to be adequate and 3) Toolbox is not a plug and play library – contrary to the “dumb-it-down” philosophy at work in the rest of the software, Toolbox is too complicated – it requires implementation setup for any set up other than a stand alone user who never shares assemblies and doesn’t use Toolbox meta data.

They have improved the documentation significantly, but the problem is still the “why” question. I think SW fails to realize that knowing what every button does is not enough. You have to somehow be able to apply that information to a real world situation. There has to be some sort of interpretation.

Again, I have mixed feelings about the documentation inadequacies, and maybe this is all by design, who knows. If they had great documentation, I’d still have a 9-5 day job. So as a writer, I’m glad the dox suck, but as a customer advocate, I wanna rail at it a little bit.

What were the major arguments against Toolbox? Has the situation really changed that much to warrant a whole bloc of people changing their minds about it? Really the big thing for me was the huge screws problem, where configurations would go missing due to changes or reinstallation, and you’d lose all your size info in assemblies. But that has been fixed, it now (since 2007) recreates missing configurations. If you look at Toolbox now, you see that it is very complex, at least the set up for it.

Why is there no option for just a simple library? You can actually use TB to create a simple library.

 Loading ...

 Loading ...

 Loading ...

(Image from ASME Y14.41-2003 publication)

Does anyone use the 3D standards as laid out in ASME Y14.41-2003? Have you used them as a reference for your own company standard? Are you having any success with transmitting 3D standard type data to manufacturing? Do you find advantages or disadvantages?

Ed posted this as a comment, but I thought it works better as a top level post. I’ve seen ed post this before on the SolidWorks forums. Thanks to Ed for posting this. I know I’ll be reading it to pick up some good info.

Again, thanks to Ed for contributing this valuable info!

=====================================

The information listed below is the “short version” of a document used to train SolidWorks users on specific requirements and procedures. The “long version” contains contains hyperlinks to another ten associated documents. I apologize for some of the formating that did not copy properley. If you are interested in the original document, e-mail me at “edward.cyganik@itt.comREMOVE”

“SolidWorks Reference Guide”
This document should be utilized during creation and as a final review of all SolidWorks objects, Parts, Assemblies & Drawings. For ease of use, the Check List below contains links to expanded or example information found throughout this document and to related external documents. For additional information, refer to the SolidWorks User Guide, On-Line-Help and the Engineering Data Index that contains links to wide variety of related documents.

Check List

1. GENERAL: Go to “GENERAL” Section
1.1. Are your SW options and external references correct and up to date?
1.2. Are you using and maintaining proper directory structures?
1.3. Is local company data backed up on a network drive?
1.4. Are you working locally in your own directories?
1.5. Do you know your working environment? Are you comfortable with all interfaces?
1.6. Is the computer doing all the work? Are you an active participant?
1.7. Are you maintaining your computer properly?

2.ALL OBJECTS: (Parts, Assemblies & Drawings) Go to “ALL OBJECTS” Section
2.1. Is the proper template being used for …
2.1.1.… Parts?
2.1.2.… Springs?
2.1.3.… Assemblies?
2.1.4.… Drawings?
2.2. Is the naming convention correct?
2.3. Have proper values been filled in for all Custom File Properties? If they don’t exist have they been created?
2.3.1.… Parts & Assemblies?
2.3.2.… Drawings?
2.3.3.… Springs Drawings?
2.4. If equations have been used; Are there any errors? Are they correct? Do they make sense?
2.5. Check the FeatureManager Tree (FMT) to verify that the object has been rebuilt.
2.6. Are all items the proper color? Everything is not Black & White.
2.7. Orient to a descriptive view, for all SolidWorks Objects.
3. SKETCHS: Go to “SKETCHES” Section
3.1. Have the default planes been used where ever possible?
3.2. Are Dimension & Geometric Relations attached to the correct entities?
3.3. Have you minimized requirements for sketching and reduced the probability of errors?
3.4. Has design intent been captured?
3.5. Are sketches geometrically correct and properly dimensioned?
3.6. Is the sketch proper for feature to be created?
3.7. Have construction aids been used to define sketches?
3.8. Have base features been utilized for sketches?

4. ALL MODELS: (Parts & Assemblies) Go to “ALL MODELS” Section
4.1. Is more than one component fixed?
4.2. Does the FMT contain minus signs, plus signs and/or question marks?
4.3. If configurations exist, were they created properly? Are they correct? Have Design Tables been incorporated?
4.4. Are exploded views and/or 3D views available?
4.5. Has the model been reviewed?
4.6. Has the model been visually inspected?
4.7. Have external references been reviewed?
4.8. Would a temporary drawing aid in the design process?

5. PART MODELS & FEATURES: Back to top of Check List or Go to “PART MODELS & FEATURES” Section
5.1. Has the part been created to suit the manufacturing process? Also see: ENI_DFM.doc
5.2. Has the part been created to suit assembly needs? See: ENI_DFA.doc
5.3. Have features been aptly named? ()
5.4. Is an “Alter Item” or “As Manufactured” part model required? See AlteredItems_BaseParts.doc & AM&A_MODELS.DOC
5.5. Has the part been constructed smartly?
5.6. Has the Hole Wizard been utilized where possible?
5.7. Have Library & Palette Features been employed?
5.8. Have the following items been created or setup in the part mode …
5.8.1. … Material Density?
5.8.2. … Reference Dimensions?
5.8.3. … Tolerancing?
5.8.4. … Annotations? (Notes)
5.8.5. … Cosmetic Thread Callouts?
5.8.6. … Surface Finishes?
5.9. Are there suppressed or hidden features in the FMT?
5.10. Is the display correct for all sketches, planes, axis, points, etc.?
5.11. Have requirements for Part Configuration been considered?
5.12. Do you have a robust Part Model?
5.13. Have Text Features been minimized?

6. ASSEMBLY MODELS: Go to “ASSEMBLY MODELS” Section
6.1. Is the assembly modeled properly?
6.1.1. … Proper Base or First Component?
6.1.2. … Use of Neutral/Symmetry Plane?
6.1.3. … Proper use of Sub-Assemblies?
6.1.4. … Required Assembly Features?
6.1.4.1. Is an “Alter Item” or “As Manufactured” assembly model required?
6.2. If required at drawing level, have the following items, been created in assembly mode for later use…
6.2.1. … Isometric or 3D Views?
6.2.2. … Exploded Views?
6.2.3. … Weld Symbols?
6.2.4. … Surface Finishes?
6.2.5. … Reference Dimensions?
6.2.6. … Tolerancing?
6.3. Are there suppressed, hidden or lightweight, “out-of-date” components in the FMT?
6.4. Are there any suppressed mating conditions?
6.5. Do the terms or appear as suffixes on any components in the FMT?
6.6. Have all requirements for Assembly Configurations been satisfied?

7. DRAWINGS: Go to “DRAWINGS” Section
7.1. Has the drawing been setup properly?
7.2. Has the display been set for all items in all views?
7.3. Are all proper/required model dimensions displayed?
7.4. Have driven (reference) dimensions been kept to a minimum?
7.5. Have true model dimensions been captured in hole call-outs and general & FOD notes?
7.6. Has the annotation “Hole-Callout” been utilized where possible?
7.7. Have all drawing items been created in the proper view, sheet or template?
7.8. Have the general notes been created properly?
7.9. Have the proper Blocks been used?
7.10. Have View Inserts been utilized?
7.11. Is the Design Table correct? See DT_SW&ST_Contents&Appx.ppt
7.12. Is the Bill of Material correct?
IF IN DOUBT ABOUT ANY OF THESE PROCEDURES, ASK FOR ASSISTANCE!

 GENERAL: Back to Check List
 Support Files;
 Properly maintain your Option File. SWOP2006.reg & sw_options_2006.doc
 Organize & manage your data & external references.
 Proper Directory Structure;
 Create appropriately named directories.
 Do not place any files or create any directories (working or personal) in any application directory.
 Segregate files with adequate sub directory structure.
 Backup all current or in process data to the proper network drive.
 Work in YOUR directories ONLY. DO NOT work in any of your neighbors’ directories.
 Perform maintenance regularly. (Review/Remove temp files, delete back up files, etc.)
 Get to know the environment & the GUI’s for WINNT, SolidWorks, SmarTeam, etc.
 Be aware of what is required; a menu pick, an input to a pop up menu, a screen pick, etc., use Help menus.
 Message windows are exactly what their names imply, delivery of messages, some good & some bad. In any case, you should pay attention to this portion of the GUI:
 A feature is created but does not display or you do not see it;
 Was it created at all? Was it aborted? Was it created with an error?
 Was a failed feature inadvertently suppressed during a rebuild or in the feature creation process?
 A component is assembled but does not display or you do not see it;
 Was it assembled at all? Was it aborted? Was it assembled with an error?
 Was a failed component inadvertently suppressed during the rebuild of the assembly or during the insertion of a component in the assembly process?
In ALL of the cases mentioned above, the message window would have displayed a response or status. Always be aware & review accordingly.
 Actively watch or review the regeneration process of parts, assemblies & drawings. This can be done…
 … during initial retrieval. (all objects are regenerated)
 … after a modification is made. (from point of modification to the end)
 … when you roll back the feature manager. (from the top of the FMT to the Rollback location.)
 Review the feature manager tree, then justify or resolve the following;
 Are all minus signs (-) acceptable?
 Over defined items, indicated with a Plus sign (+), must be resolved.
 Items that cannot be solved, shown with a Question mark (?), must be solved.
 First component in an assembly can be fixed (f) all other fixed components must be scrutinized.
 Resolve all Red Exclamation Marks (!), follow the Red Arrows ().
 Review All External References () and Resolve if necessary;
  Normal External Reference
 ? Out of Context, External Reference not Up-To-Date. (Load Model)
 * Locked External Reference (Not a normal practice. Why is it locked?)
 x Broken External Reference (Unacceptable.)
 Review your computers and personal working directories prior to start & at the end of a session.
 Delete all unnecessary files. Keep your directories clean, including:
 Using the SmarTeam Local Files Explorer, keep “C:\Work” clean.
 After a crash delete all SW temp files, files = “~$FileName.sdlxxx”.
 Periodically cleanup files in the temporary SW backup directory, files = “Backup of FileName.sdlxxx”.
 Clean up the “Temp” directory, “C:\Documents and Settings\login-name\Local Settings\Temp”. Delete all “*.tmp” files and all temporary files generated by using SolidWorks, SmarTeam, the SW Viewer from SmarTeam.
 Clean up the “Temporary Internet Files” directory.
 For detail on all of the above, see: Maintenance&Performance.doc

 ALL OBJECTS (Parts, Assemblies & Drawings) Back to Check List
 Start objects as follows: SeedObjects.doc
 Create parts using: Enidine Start Part.prtdot (or “ISO_” part templates)
 Create coil spring parts using: Right-HandCoilSpring or Left-HandCoilSpring.prtdot
 Create assemblies using Enidine Start Assembly.asmdot (or “ISO_” assembly templates)
 Create drawings using ASIZE, BSIZE, CSIZE, DSIZE, CoilSprg, CoilSprgS, CoilSprgCG, or CoilSprgSCG. (or ““ISO_” or “CMC_Enivate_”.drawing templates)
 Create all parts, assemblies & drawings using the proper naming convention by referring to the Enidine DRM, ES-3000, the Part Number Code Book (PNCB) and to any approved group or department specific documents.
 Fill in values for Custom File Properties as follows:
 For Parts & Assemblies: cust_part+assy_props.doc & ModelProps.txt SW_CoilSpring.doc
 For Drawings: cust_drawing_props.doc
 For Coil Spring Drawings: CoilSprgDrwInst.doc.
 Use TOOLS/EQUATIONS/EDIT ALL to verify that there are no errors. Have dimension names in equations been given understandable & meaningful names? If not, consider adding equations to ensure proper rebuild and minimize the number of manual edits required. “Equations – Useful Math Functions & Operators”
 If there is any RED in the FMT of any object, errors must be resolved. Rebuild (CTRL B) or forced rebuild (CTRL Q) to fix all problem areas. No RED (Bleeding) exclamation marks (!) or arrows () are acceptable.
 Everything is not black or white, colors have the following meanings: SW_Color_Codes.doc
Check the color of Dimensions, Notes, Sketch Entities, etc. to verify proper status, only Black is acceptable.
 Black. = Satisfied
 Blue = Under Defined
 Red = Over Defined
 Brown = Dangling
 Pink = Not solved
 Yellow = Invalid
 Driven = Gray
 Prior to saving databases, orient to a descriptive view, shade and then zoom to fit. For drawings, zoom to fit is all that is required. This technique provides for proper viewing when using the File Open dialogue box, Quick View and the SmarTeam viewer.

 SKETCHES:
 If possible, use the existing planes (Front, Top & Right) for sketching & orientation. Dimension to these planes and the origin whenever possible.
 Dimension & Add Geometric Relations to what is required and not what happens to be picked or assumed to be picked. Perform these tasks;
 while in an isometric view, rotate to an advantageous view and/or use HLG display.
 using select other.
 by selecting from the FMT.
 To avoid unnecessary sketching and reduce the possibility of errors, use Predefined, Copied or Derived sketches.
 To capture design intent, use edges, offset edges, dimension to edges and apply geometric relations to edges where possible.
 Modify sketch by Dragging an Entity or Vertex to verify or determine proper movement/design intent then Add/Change the Dimension Value for verification.
 Use “Tools/Sketch Tools/Check Sketch for Feature” to verify proper sketch to feature type.
 Use centerlines, construction arcs, points, etc., to aid in solving/defining sketches.
 Relate as many new sketches as possible to the models’ base features. Using this method will reduce the number of parents involved and therefore reduce modification and rebuild errors.

 ALL MODELS: (Parts & Assemblies)
 The FMT shall contain only one fixed (f) component. If more than one component is fixed, there must be justification.
 Check the FMT. Are Minus signs (-) acceptable? Plus signs (+), denotes over defined. Question marks (?), mean something could not be solved. Investigate all cases and resolve as necessary.
 If a model contains configurations, check for proper names of all configurations. All families of parts and assemblies must be Design Table driven. Items in the internal MS EXCEL table should be given meaningful/understandable names, these item include, dimension symbols, feature names, parameters, etc.
 Provided for exploded and/or 3D views by creating, naming & saving properly.
 Review the creation process by using the Rollback Bar. Look at the modeling techniques that were utilized, possible errors & sound practices.
 Visual interrogation of a model can reveal a wide variety of problems. Shade the model & view from numerous directions. Many times a misplaced item (feature or component) will show up as a flaw in a shaded image. Conflicting volumes can also be detected in HLG/HLR views. Reviewing the surface topology of any model can sometimes result in finding errors that have gone undetected.
 Any model that contains external references or is an external reference for another part must be reviewed by critiquing relationships, dependencies & references. The Parent/Child pop-up menu and the File Find References command can be used for investigation. If any of the following items are found in the FMT , ?, * or x, then RMB on the item and select List External Refs, then investigate as necessary.
 In order to perform measurements and/or to better visualize a design, it may be beneficial to create a temporary drawing. If a temporary drawing is created, it can easily be modified to satisfy requirements for production drawings.

 PARTS MODELS & FEATURES:
 Map out part modeling strategy prior to start. Do this by determining…
 … what manufacturing process will be (lathe, mill, cast, forged, molded, sheet metal, etc.). If a part is created on a lathe, start with the proper bar stock. An extrusion should start out with the proper extruded shape & size. The Overall Finished Dimensions are all that is necessary to simulate stock size. Extra material shall not be added.
 … the base (first) feature type (extrude, revolve, swept, blend, advanced, solid, thin, etc.). It should provide for the most bang for the buck, it should be the most stable.
 … if a neutral plane or axis feature would be helpful.
 If this is the case, refer to the Part Number Code Book, the Enidine DRM, ES-3000 and associated materials pertaining to “Altered Items” or “As Manufactured” models.
 As a minimum, key features should be given meaningful names for quick and easy identification.
 When necessary, take advantage of creating construction features such as planes, axes & points and/or use sketch entities to aid in a design process. Any construction aids should lend themselves to capturing design intent. As new features are created & added, dimension & align them to surfaces, axis, planes & edges which make sense. Perform these tasks while in an isometric view and/or by using select other.
 If possible, create all holes using the Hole Wizard for addtional benefits at the drawing level (Hole Callouts).
 Library & Palette Features should be used whenever possible. To decrease modeling time, consistently dimension common features and reduce the possibility of errors, see LibFeat.doc & LibraryFeatures.xls
 Determine & then create or setup the following part model items;
 Material Density (For precise weight calculations.) See the “Density Tab” in ENI_STD_NOTES.xls
 Reference Dimensions should be created in the part model where required but should be kept to a minimum. Excessive use of reference dimensions is a sign of poor modeling practices. Consider changing the dimensioning scheme of sketches to reduce the number of reference dimensions.
 Add all tolerances in the part model. This is where the definition of form, fit, design intent, manufacturing requirements are determined. Setup all tolerance types, nominal, plus/minus, limits, symmetric, GD&T, datums, basic dimensions, etc., in the part model.
 Annotations or Notes can be created at the part level also. If holes are created using a variety of features (versus the Hole wizard) then use the Annotations command to capture the required information so that parametric notes can be used downstream in drawings.
 During the process of creating Cosmetic Threads, be sure to fill in the Thread callout box to provide for the required notes on drawings.
 Like all other annotations, Surface Finishes should be added at the part level.
 All suppressed (gray) or hidden (wireframe) features in the FMT must be investigated and resolved. Suppressed and/or hidden features are only acceptable if there are configurations present and the state of the features are controlled in a design table.
 Investigate and set the proper display for all sketches, planes, axis, points, etc. This is a simple matter of determining whether an item is to be Shown or Hidden.
 Part Configurations should consider the following;
 Create Design Table with column headings for ease of use.
SW Help Pages: Summary of Design Table Parameters, User Notes & Comments.
 Properly name the default part and file or database name.
 Make sure your default/generic part model is robust, correct & flexible.
 Identify items to be added to the design table. Modify dimension names suitably, aptly name features, etc.
 Verify all configurations upon creation.
 A robust part can handle drastic changes. To determine if your models have what it takes, “Flex the Part”, do this by realistically modifying a random number of values on the part. If failures occur, determine the cause & correct.
 Keep Text Features to a minimum, embossed or engraved text carries a lot of overhead. SW_Txt_Feat.doc Consider creating text at the drawing level by associating any text/identification requirements to drawing views.

 ASSEMBLY MODELS:
 Map out assembly modeling strategy prior to start, by determining …
 … the base component. If a damper is to be created, do not start with a bearing, start with the shock tube or cylinder. The first component should be the most stable/dependable component in the assembly.
 … if a neutral plane or axis feature would be helpful. For assembly purposes, assembly planes and/or axis have an advantage over part features in that the Parent/Child relationship is less likely to become an issue.
 … if sub assemblies need to be created. Caution: Sub-Assemblies are not to be created for the sake of SolidWorks. Sub-Assemblies, separable or inseparable, are created to provide for logical manufacturing processes, assembly processes and spares or for the procurement of items.
 … requirements for subsequent configurations.
 … If assembly features are required, will they be created at …
 … assembly level? There are limitations that may prevent the use of Assembly Features, namely, assembly features are restricted to removal of material. To add or deform material, refer to the Part Number Code Book, the Enidine DRM, ES-3000 and associated materials pertaining to “As Manufactured” models and “Altered Items”.
 … part level? For simulation of Assembly Features that are created at the part level, again, refer to the Part Number Code Book, the Enidine DRM, ES-3000 and associated materials pertaining to “As Manufactured” models and “Altered Items”.
 … both levels, Assembly Features combined with features created at part level? Ask for assistance.
 Determine & then create or setup the following;
 Isometric or 3D Views should be created, named and saved.
 Exploded Views should be created, named and saved. This would require the creation of a assembly configuration that would also need to be named & saved.
 Weld Symbols requirements should be handled during the design stage. Identification of weld joints and types of welds should be done in the assembly model.
 If machining takes place at the assembly level and a machined area requires a Surface Finish, then the annotation should be added at the assembly level.
 If required, Reference Dimensions should be created in the assembly model and similar to part mode, they should be kept to a minimum. Excessive use of reference dimensions in an assembly may have an undesirable effect.
 For the most part, tolerances added in an assembly model would only apply to the dimensions of assembly features. When required, setup all tolerance types, nominal, plus/minus, limits, symmetric, GD&T, datums, basic dimensions, etc., in the assembly model if they will be required at the assembly drawing level.
 All suppressed (gray), hidden (wireframe) and/or out-of-date/lightweight (feathers w/stripes) components in the FMT must be investigated and resolved. Suppressed and/or hidden components are only acceptable if there are configurations present and the states of these components are controlled in a design table. Out-of-date components must be brought up to date by opening and resolving.
 All suppressed mating conditions must be investigated and resolved. Expand MateGroup Icons in the FMT and look for Grayed-out Paperclips, these denote suppressed mates. Suppressed mates are only acceptable if there are configurations present and the mate suppressions are controlled in a design table.
 It is unacceptable to have any components with the suffix or . If either one of these appear in the FMT, they must be resolved by setting the components configuration to a properly defined configuration name. Management of any configurations shall be done using a Design Table.

 ASSEMBLY MODELS: (continued) Back to Check List
 Assembly Configurations should consider the following;
 Create Design Table with column headings for ease of use.
SW Help Pages: Summary of Design Table Parameters, User Notes & Comments.
 Properly name the default assembly and file or database name.
 Make sure your default/generic assembly model is robust, correct & flexible.
 Identify items to be added to the design table. Modify dimension names suitably, aptly name features and verify that all component names are valid. Additionally, identify mates and mates with dimensions (angular and/or offset) and properly rename them for easy identification and use.
 Verify all configurations upon creation.

 DRAWINGS:
 Map out drawing strategy prior to start, by determining…
 … what views are necessary. Are they available in the model?
 … what scale is to be used. The “overall scale” is determined by the drawing’s “sheet scale”. A “View’s” scale is defined in the “Drawing View Property” and is used as an exception to the overall drawing scale only! The drawing’s “sheet scale” is the predominant scale.
 … the size of the sheet or format required.
 … if multiple sheets are required.
 … is a Design Table to be input on the drawing?
 … is a BOM to be input on the drawing?
 To best depict models, determine the proper display mode to be used for all views. In addition to view display, component & edge display mode must also be considered and set for all views.
 Verify that all required dimensions are displayed. The vast majority of a drawing’s dimensions are to be “Model Dimensions”. To verify dimensional requirements, use Insert-Model Items and select all annotation types. Use the Undo command as necessary to display properly once dimensions have been checked. (Good idea to save the drawing prior to checking.) NOTE: The proper technique for annotating a drawing is to insert model items and “hide” those that are not required. By using this technique, verifying required dimensions is easily accomplished using the Hide/Show Annotations command.
 Add reference (driven) dimensions only when necessary. If excessive dimensioning is taking place at the drawing level, there is likelihood that a dimensional scheme should be changed at the model level.
 If holes are created using a variety of features, then use the Annotations/Note command and select model dimensions to parametrically capture the information. This process must also be used to properly capture information for General & FOD notes.
 If holes are created in a model using the Hole Wizard or Insert/Feature/Hole/Simple, then use the Annotations/Hole Callout command to generate the proper notes.
 Associate items to the proper View, Sheet or Template. Do this by activating the required view or by specifically editing the drawing sheet or template as follows:
 Views (captions, unattached dimensions, notes, lettering, labels, cosmetics, etc.)
 Sheet (tables, BOM, etc.)
 Template (general notes, title block and revision block information)
 Note: To save all drawing changes, save the drawing file only. Never save the template!
 Check for the proper display of Cosmetic Threads, Datums, Datum Targets, Feature Dimensions, Reference Dimensions, Geometric Tolerances, Notes, Surface Finish & Welds. Do this by using the Annotations Folder Options or the Menu Options under Insert/Model Items.
 Use the proper/approved company notes for all company documentation and keep notes as one item, not separate entries for each note. ENI_STD_NOTES.xls Also, utilize the Unicode Character Map for non standard characters, see ExtendedCharacters.doc
 Use the proper/approved company blocks for all company documentation. SW_Blocks.doc
 Use the proper/approved company view inserts for all company documentation. ViewInserts.slddrw
 Design Tables should be sized and placed correctly. Be sure that the MS Excel file has been sized to display all required information, WYSIWYG. If design tables become large and unruly, see ManagingLargeDesignTables.doc
 Use the proper SolidWorks “Table” based Bill for BOMs. (Replace MS EXCEL BOMs when possible.)
 For most BOMs, use “EnidineBOM.sldbomtbt “ (MS EXCEL not for new design.)
 For BOMs containing “bulk items”, use EnidineBulkItemsBOM.sldbomtbt (MS EXCEL not for new design.) Also, refer to BOM_Settings.xls for instructions.
This is just an example, click HERE to go back to the Check List.

Ed Cyganik, or “Eddie” as many folks know him, is a fellow I’ve known for a number of years, going back to my days working for SolidWorks resellers in Western NY in the late 90s. I met him while trying to troubleshoot problems in a SolidWorks/SmarTeam installation. Over the years I’ve learned a lot of stuff from Ed and people like him. You may have run into him if you spend any time on the SolidWorks Forums. Besides the forums, Ed also speaks at SolidWorks World, which is where this picture of him came from. He has a lot of accumulated wisdom not just with SolidWorks, but with other CAD packages, 2D and 3D, as well as dealing with people.

Eddie has been active in the SolidWorks community probably as long as I have, and is probably responsible for a lot of great functionality, particularly in drawings. He’s a good guy, knows how to speak his mind, and has the knowledge and experience to back it all up. Basically, he’s the kind of guy you want to have on your side when you have questions about stuff.

Eddie and I will be working together on an up-coming writing project, and I just wanted to introduce him here. I don’t think he’s active on any of the blogs, but that’s ok. There are many ways to communicate. Blogs are just one of them.

Eddie, tell us a little about where you work, what they do and what your part in it all is, if you can…

My current position is with ITT Enidine Inc., Controls Group. We design & manufacture products that provide solutions for energy absorption and vibration isolation. I am a Senior Designer by trade and CAD Administrator when I don’t have anything else to do. Seriously though, my CAD duties go well beyond administration as I am also responsible for; Teaching CAD and Training in General to meet Company & Industry Standards.

Software packages used prior to SolidWorks were; Anvil 5000, AutoCAD & Pro/E. I have very little experience in the two former packages but Pro/E and I go back to its inception. I used, implemented and taught Pro/E over a period of 11 years at 4 different locations. I loved Pro/E, it always did everything I needed it to do.

Right now, I guess the simple reason is that we are heavily invested in SmarTeam. In-a-nut-shell, we have a broad and experienced user base that is using a product that has been extensively customized to suit our needs. Switching to anything else would be a major undertaking.

Sadly, the recent version are nothing more than a tease. I have 2008 and 2009 loaded, customized and ready to go but we are still using SW06 as our production version. There are some tools (functionality & capability) that would be beneficial to the company but for the most part we are able to do what we need to get our jobs done.

Based on this, if I were to answer your original question, I’d have to say; Recent versions of SolidWorks, as well as all of them, come along to often. We do not have the time to beta, pre-production and production test SolidWorks with our OS and related software packages (CAE, CAM, PDM, ERP, etc.) not to mention our computers, networks, printers, the list goes on and on.

That’s a really good question, especially since I just filled out the request for our yearly maintenance. Interestingly, If we had dropped our subscription service in 2006, we’d be looking at a break even point if we had to repurchase all of seats. We have a good relationship with our VAR (CADimensions, if you can mention them) and I have a great relationship with SolidWorks. So, up until now it has really been about maintaining relationships and I’ll leave it at that for now.

·         Fundamentals

o       Double-Click to Zoom Fit

o       Magnifying Glass w/Zoom Selection (G Key)

·         Sketching

o       Ghost imaging of missing sketch geometry (Helpful for working with models created by others.)

o       Zero & Negative Value (About time!)

·         Features

o       Ghost imaging of missing reference entities. (Again, helpful for working with models created by others.)

o        

·         Parts

o       Measurement w/Dual Dimensions

o       Assigning Custom File Properties from Task Manager (Have yet to review but in theory, this could be a time saver.)

o       Convert to Sheetmetal

·         Assemblies

o       Ghost imaging of missing mates. (Again, helpful for working with models created by others.)

o       Performance (Any performance gains but especially assemblies and their drawings are most welcomed.)

o       Unloading hidden components.

o       BOMs in Assembly Documents (Once again, About time!)

o       Hinge Mates

o       Sensors

o       Instant 3D in Assemblies (Should have been part of the initial roll-out of Instant 3D. This is where the tool has the most benefit.)

·         Drawings & Detailing

o       Customized Drafting Standards (Still need the capability to “lock” settings to prevent users from changing.)

o       Dimension Leader for Same-Sized Features (This tool, that is long over do, will allow for clearer documentation.)

o       Multiple Jogs for Dimensions & Callouts (Another tool, that is long over do, will allow for easier creation of clear documentation.)

o       Format Painter (This basic MS Word tool has the capability to really speed up the process of detailing. However, if they would have asked me to write the Product

Specification, functionality would have been implemented differently. I would have eliminated “Favorites”, now known as “Style” by combining them with the Format Painter tool. Basically, Favorites provide for predefined settings/requirements, while the Format Painter allows for the selection of a setting or requirement that is currently displayed. The interface would be activated by the selection from a pop-up menu. As an example, a dimension is selected, then “Style-Select”, then pick a dimension with the desired style. Or select a dimension, then “Style-List”, in this case, another menu or window would pop up with all “Dimension Styles” that have been saved. The latter selection, Style List, would load all “favorites” automatically. Well that’s the way I’d do it but what do I know?)

Now that is a tough one! My short list consists of about five items but if you are holding me to just one, then I’d have to say, …Provide the ability for every single input in SolidWorks to be internalized/maintained once a save is performed. What I mean is this: There is a file know as the “gtol.sym” file that we have customized. This file must be present in its default SolidWorks location on every computer where your files are to be opened, otherwise the symbols you have defined cannot be interpreted. This problem has existed forever and version after version, nothing changes.

I am a big fan of the capabilities of “pop-up” menus. I think they are not only fast and easy to get used but the also allow a user to stay focused on the task at hand. When you put all of these advantages together, I believe they improve user proficiency and increase throughput. So with the advent of the “S” shortcut menu, you have even more tools at your disposal to stay focused. No more hunting, searching and moving your mouse all around. If employed properly, pop-up menus can eliminate most other menus, toolbars and icons.

My least favorite things about SolidWorks are unfinished/incorrectly implemented capabilities that continue to be left by the wayside year after year. I won’t go into anything particular because anyone reading this can come up with a list of their own.

Well I could write a book about that and probably should have before you did, but that’s water under the bridge. For the long version, I did a presentation at SWW 07, titled “Getting Off on the Right Foot” that cover the major issues. (I/you could make this available if you like.) For a shorter version, get bye-in from management, get users involved, get other departments involved, document your current process to meet that as a minimum but also look for ways to improve your processes by leveraging SolidWorks and any of it products. Look into everything, test everything and then set up your CAD environment for clean and consistent use by all users.

One other thing: “Never upgrade for the sake of upgrading!” Even if there is something new that you absolutely need, you had better test the living daylights out of the software before you even consider it.

Think in 3D, all the time. If people are coming from a 2D background, they still had to have the ability to picture things in 3D but they continually interpreted or documented in 2D. In my mind, there is a big difference making the complete transition.

Teaching “What’s New” is another matter.

Sorry for the cliché but teaching old dogs new tricks best decribes some of my frustrations. Now before the old dogs start barking, let me clarify. When I say “old dogs”, I’m referring to those who have been using the software 4 to 10 years. They learned and got use to drop-down menus, separate icon toolbars for all categories, a limited set of keyboard shortcuts. Now, even with a wide variety of keyboard shortcuts, the Command Manager, RMB Menus, the “S” Menu, I still have users stuck in the past using the software as it was when they were originally taught.

Well not administration but teaching CAD. I had one guy who started out on the board and then used a 2D CAD package (electronic drafting). He had about 20 years experience. Anyway he came to a class I was teaching and was full of enthusiasm but by the thrid day he was so lost that he came to me and said if he didn’t make any progress by the next day that he was going to quit. Unknown to me, he had not, could not understand a sketch plane with reference to the 3D world and how one would orient to work on a design. I worked with him after class, before class, I brought in technical books with illustrations and various training aides from the web. I asked one and then another employee to sit with this guy and try to explain the concept, all to no avail. The guy just never got it and quit the class. After working at the same company for another 2 years, he eventually quit and moved to Atlanta where he and his wife opened a restuarant. The restaurant lasted less than a year and the guy was forced to go back to what he always did.

Fast forward, about 4-5 years later, I’m now out in Denver and I get a phone call and the person says’ “Hello, I bet you don’t know who this is?” Believe it or not, I recognized his voice and it was none other than this guy I had tried to train so many years ago. Next he says to me; “I bet you cannot guess what I’m doing right now?” All I said was I know your not working at the restaurant. He goes on to tell me he’s still in Atlanta but he is now at Lockheed. I was kind of afraid to ask but I said, What are you doing? He tells me he’s been working in Engineering using Pro/E for about 4 months. I was dumb-founded and didn’t know what to say right away. So he says, I got hired and right away they send me to training and he says, “All I’m thinking is, Here we go again”. But as soon as the instructor started talking about Sketching and Datum Planes and Orientation, …Bam, everything clicked, the proverbial lightbulb lit. And now you know the rest of the story.

Although it may not look like it now, I think we’ll be there in October!

What types of stuff does your company standardize in 3D? Do you make your users name their planes in a certain way? Do you require all users to make the first sketch related to the origin in some way? Do you require users to take advantage of symmetry when it is avaiable? Do you want them to morror sketches? features? bodies? assemblies? Do you require features to be named in a certain way? Do you require fillets all to be at the bottom of the tree? How much detail do you require or are users required to leave off certain types of detail (such as extruded text or knurls).

Everyone has heard of “drafting standards” where you control layers and line types for drawings. Do you control the same sort of stuff for models, or you let your users freewheel this kind of thing? Do you ever have problems with one user picking up models that another user created? Are there any types of features that are strictly off limits for production parts? Do you disallow multibody techniques? surfacing? flex or deform features? splines? 

I’m not suggesting [yet] that you should or should not be doing any of these things, I’m just curious about what your opinions are on the topic? Leave comments and let’s discuss this issue.

 

Devon Sowell and I first met on the old comp.cad.solidworks usenet newsgroup several years ago. Devon is an independent contractor/consultant who does primarily machine design projects, and CAD implementation consulting. Devon has two blogs, one just for general SolidWorks topics, and the other for PDM topics. I believe his is one of very few blogs dedicated to PDM issues. 

In some spare time, Devon has written and self-published a book titled SolidWorks PDM. Take a look at his site, you will be able to purchase the book directly from Devon. 

In keeping with part of my new emphasis on CAD Administration, I thought that an interview with one of the PDM gurus would be in order, and Devon accepted. My questions are denoted with M: and Devon’s responses with D:.

M: Devon, could you tell us about your new book on File Management and your
new website PDM Solutions? What has interested you so much about this
field that you have taken on these projects?    

D: Sure, first thanks for contacting me, I always enjoy talking to you. The check’s in the mail.

My 2009 SolidWorks PDM book is really a manual; Ring bound, 8.5×11, color. In the manual, I show detailed overviews of 2009 SolidWorks Explorer, 2009 SolidWorks WorkGroup PDM, and 2009 SolidWorks Enterprise PDM. Using screenshots and Tips & Hints, I show the default settings and the most common options and edits. Also, I include a matrix showing the differences between the applications. Three different sample PDM systems are shown, Manual, WorkGroup PDM, and Enterprise PDM. Finally, if you save up enough coupons on the back cover, you can get your ProE decoder ring.

M: What is the name for your new book?

D: I call it SolidWorks PDM (Product Data Management); “Managing” the CAD files from SolidWorks, and other CAD applications, and all the supporting document files such as Engineering Changes Orders, BOMs, Materials, Approved Suppliers, e-mails, Renderings, catalogs, mp3 files, Led Zeppelin songs…

M: What do you consider to be the most important tasks that a CAD Admin is responsible for?

D: Good question; I say there are two elements to CAD Admin today, specific management of the SolidWorks software application and the specific management of the CAD files and the supporting documents; PDM.

I like to focus on PDM. PDM is not just like watching paint dry, it’s more, much much more, like the sweet smell of a rose, or a beautiful sunrise. Also consider this, the FDA and ISO require specific “document controls” that must be adhered to, and validated. These processes must also be audited by an independent party.

M: How would someone go about becoming a PDM Administrator?

D: Contact me, I’ll train them. http://www.pdmsolution.com A shameless plug ha!

Again, if PDM floats your boat, and you’re a SolidWorks User, start with SolidWorks Explorer and then investigate WorkGroup PDM, followed by Enterprise PDM and SmarTeam PLM. Agile PLM and SAP ERP are used in my area. Finally contact the NSA to view the largest data set in the world. Good luck with that one.

M: Is there one single area of responsibility where a good CAD Administrator can make the single most significant positive impact on a business?

D: Well, how about stopping SolidWorks users from detailing SolidWorks drawings in AutoCAD, ha!

A CAD or PDM Admin should focus on streamlining the design and documentation process. During the past 10 years, I’ve seen companies with poor productivity and companies with great, rapid productivity. It should be obvious which is preferred, unless you’re still using AutoCAD 13.

M: How is administering SolidWorks different from administering AutoCAD?

D: AutoCAD was first used by cavemen chipping stone tablets with flint points. SolidWorks was created by Jon Hirschtick in 2025, who then used it to design and build a 3D printer. He then printed out a time machine, from H.G. Well’s drawings, and traveled back to 1992 to create the SolidWorks Corp. we know today. Really, he told me while we played blackjack.

AutoCAD 2D files contain all their data in the single file (except for the weird “x-refs”, whatever they are). SolidWorks, of course, is 3D centric. It is very important to understand SolidWorks External References. Now “InContext External References”, they have a similar name, but they come from another planet, the same planet that Mork came from.

AutoCAD 3D files; their photons spin backwards, so they don’t work very well at all.

M: Why is File Management so important?

D: File management in my office is not important at all, just ask my wife. Drum roll, please.

Here are some questions I ask my customers:

Do your Engineers spend more time searching for files than they do editing the  files?

Do you have 5 CAD files called Front Cover, all in different folders, some on the  Network, some on Joe’s hard  drive, you know, the one that just crashed?

Do you “make” your Designers work off the Network? So, when they Save their work, they can go to Starbucks for a Latte?

Did the FDA Auditor come to your company and put Police Tape across the front  door?

Do you believe that everyone should use PDM software?

Yes, resistance is futile. The Borg use it, they use WorkGroup PDM, planning to migrate to Enterprise PDM this summer.

M: Do you have recommendations for independent contractors (such as
yourself, or me) when it comes to file management and PDM?

D: No, only you and me, that’s it, right? That’s what we talked about?

M: What is your favorite PDM software, and why?

D: Enterprise PDM, it does it all, really, everything. You just click once and that’s it. Then go home, you’re done. Just like this interview.

Devon, thanks so much for sharing your thoughts on SolidWorks PDM. Best of luck to you with the new book, you’re going to have to show me how to do that self-publishing thing.

I have to admit, I’ve been kind of dreading writing the article on Toolbox. It will involve installing the software several times, breaking and fixing Toolbox, going through all the settings (again) and trying to see the + and – of all of the different controls and possible usages. Several of you said you’re looking forward to it, so that’s even more pressure to get it right. 

So of course I’m totally avoiding it. Honestly, I’ve had a lot of other work to do in the past week, I just haven’t had the time to dedicate to this research project that I need to do it justice. But I am kind of energized by this CAD Management issue.

So I read a book. I picked up Robert Green’s Expert CAD Management. I inquired about Ralph Grabowski’s book on the same topic, but it was 6 years older than the other book.

I’ve seen Robert Green’s name thrown around on the AutoCAD sites as a CAD Management guru. I have never really read anything else by him, nor visited his website prior to researching his book. 

The book gets a lot of praise on the Amazon site. The first thing I did was to read through the table of contents. It became immediately apparent that the book was primarily concerned with management rather than about the technical side of the issue. There is literally only 2 pages of information on managing a SolidWorks installation, and none of that is relevant for anyone with more than a passing familiarity with the software. 

Green’s perspective is decidedly old-school. In most scenarios he proposes, SolidWorks is a big, hard to use 3D system, and the easy to use stuff is AutoCAD, which always has more users. He seems to assume that CAD management happens in a company with a bevy of “drafters” whose main skills include knowing how to draw lines with AutoCAD, and the word “design” doesn’t seem to occur to him. He doesn’t acknowledge that engineers sometimes use CAD or that not all CAD users report to a CAD Admin. 

The book doesn’t deal AT ALL with technical details such as part numbering, file management, libraries, installation details or licensing, but is very concerned with “managing management’s expectations”. After reading all of this I expected a chapter on “dress for success”. This is very much a  book about the philosophy of managing a group of technicians. It will help you look good to upper management, but it won’t teach you much about the real nuts and bolts of your job.

I know the type of information covered in this book is important, but is it the ONLY thing that’s important to CAD Managers? I had hoped that it would be evenly spit between management and technical data. I don’t believe you can be a good technical manager unless you are a real student of the actual technology. Nor do I believe that Green’s assumption of a 1950s sweatbox drafting department is the most valid option.

With all of that said, his suggestions on how to figure out what your job is and how to communicate with management, while often obvious or common sense, are presented in a very methodical way.

So I’ve got some polls for you. It’s been a while since we’ve done any polls here on Dezignstuff blog.

 Loading ...

 Loading ...

 Loading ...

 Loading ...

 Loading ...

 Loading ...

 Loading ...

This is the view from the Roanoke airport this morning. A low fog and a gate with no plane. Story of my life. 

There are at least two consolations about being stuck in the Roanoke airport: it isn’t Atlanta, and it has a free internet connection. So there are plenty of things to blog about, and now I’ve got plenty of time to do that.

Let’s talk about CAD Management today. One of the jobs I’ve had that I liked very much was working for an optical instrument company as the Engineering Services Manager. In this job I was primarily responsible for a group of 5 people who would get loaned out to various departments who needed CAD help. Manufacturing, Engineering, Quality, we were the overflow workforce for all of these groups and had primary responsibility for the document change process. In addition to leading that group, I was also the CAD Admin, responsible for installation, standardization, optimization and training. I helped establish and implement the product development process, codifying it into an actual flowchart so that everyone could understand and follow it.

The challenges I had as a CAD Admin for a 3D CAD process were typical; implementing libraries, enforcing part numbering, getting people to use custom properties correctly, bug fighting, end user support, creation and enforcement of standards, file management, stuff like that. 

It’s always easiest if you can build a CAD department from the ground up, but who ever gets the opportunity to do that? You usually have to step into someone else’s world, and determine what needs to be fixed, and what you can learn to live with, and possibly what you can learn from your predecessor’s way of doing things. 

In addition to the surfacing challenges I’ve been hosting, I also want to begin hosting a discussion on CAD Admin topics. You can find several blogs on AutoCAD admin, and this is a topic that doesn’t get enough attention. CAD Admin types typically have to learn on their own how to deal with the situations that come up. 

One of the topics that people seem to avoid is Toolbox. My blog gets tons of hits on that keyword because I’ve posted about it a couple of times. On my old site (which is now gone) I had a white paper I had written on Toolbox from before the time of blogs. 

So the next post I do on CAD admin will be about Toolbox. It has changed a lot since I wrote about it last, and I could use a little refresher research on the topic. In prep for the discussion about Toolbox, do you have any questions or comments on it? How do you use it? Do you use an alternative? Let’s talk Toolbox!

This post is mainly my summary and commentary based on comments from the last 5 posts. I’ll start by summarizing what has already happened. It is hugely long because there are so many facets to the problem.

Goal

The goal here is to propose a model for revamping SolidWorks subscription (maintenance, support and new versions) offerings. SolidWorks says they are a customer driven company, and this will put that assertion to the test. My guess is that when they run the numbers they have more to lose by allowing people to choose what they need instead of forcing them to buy what they don’t need or want than they have to gain by perceived good will toward all customers. So we will see if this is a customer driven company or a bean counter driven company. Time to put your money where your mouth is, SolidWorks.

Background

This started as a simple notice that a late fee that few people ever actually pay is being increased almost 50%. It turns out that if you want to skip a year of maintenance (support and new versions), it winds up costing you more money than if you stay on maintenance. This late fee is clearly meant as a deterrent, not an additional fund raising scheme. SolidWorks uses the threat of assessing the fee to keep you from lapsing on maintenance payments. My guess is that once weened from that maintenance teat, former customers tend to not go back.

I’m guessing that what prompted SW to increase this fee is that people are not re-upping for subscription after the 2008 release. I still believe that the overall average perception of SW08 is that it was among the worst releases ever - full of half completed changes, the most extreme changes used as defaults, miscalculated new features, too much visual fluff, obsoleted not-so-old hardware at an alarming rate, required too much retraining, and overall shows that the SolidWorks Corporation is not in touch with the majority of real users.

My opinion here is that SolidWorks should not charge its users for its own mistakes. SW08 was a mistake. I think it’s a mistake big enough for someone to lose their job over, although I’m not in a position to say who. They took a risk, and were bitten by the downside. I don’t think charging customers for the failed attempt is the right thing to do.

Rewarding loyal customers 

Last SolidWorks World, I started asking people (Wilkinson, Hiss) in interviews what SolidWorks was doing to reward loyalty from long time users. I don’t think I got any clear answers. Between the new late fee and the interface changes that erased many of the gains SW has made over the years, I think we have our answer. I need to say that some of the 2008 interface changes were positive, but I can show quantitatively for most of the  so-called improvements, there are mitigating factors that offset their effectiveness. Two steps forward, two steps back. In several cases, the new functionality is unquestionably inferior to the old functionality (removing CommandManager options, removing slider and splitter bars, removing the ability to remove face colors, the entire 2008 colors/appearances scheme, using left click context bars from the FeatureManager, removing interface images from the documentation, and I’m sure there are many more).

Hit a nerve

I knew when I decided to write this short post that it was going to create a stir. Some people like Rob Rodriguez and Devon Sowell have subscription as a bit of a pet peeve. I wasn’t really prepared for what happened, though. I had 75 comments over the course of 3 days. My average is probably 3 comments per post, and I post 2-3 x per week. I was getting 1200 hits a day for a couple of posts and the main page. My usual is 300 – 400 hits per day for the main page. I got 6 notices about new forums that people wanted me to contribute to. Usually, this doesn’t happen at all. I’ve never had a SW VP comment on my blog before. I had a couple of SW employees contact me about the blog, which only has happened a couple of times before. A couple of people who are upbeat about almost everything commented that they didn’t like the looks of this one (Ricky J and SolidSmack).

So yes, I hit a nerve. Subscription is the 800 lb gorilla at most user group meetings. People want to express outrage, but they don’t do it because it feels futile, and because it has been done so many times before. This is how a stupid fee that no one ever pays turned into the biggest train wreck Matt Writes has ever seen.

Lipstick on the pig

So SolidWorks is trying to attract new users through a sexy consumer product-ish “lipstick on the pig” approach, and yanking the rug out from under loyal customers. The sexy new colors, visual effects, backgrounds, shadows, reflections, Appearances, RealView and what not combine I’m sure to look spell binding in demos to Inventor users, but in practice for existing users are very much a distraction. Why have resources been squandered on this kind of thing instead of improving functionality? These issues are cosmetic, and treat the SolidWorks software as if it is a new way to package Barbie dolls instead of an engineering tool. To me, this is incredibly misguided, a waste of time and resources, and displays an out of touch mentality. Of course you can find people who find things like this useful, but you can say the same for Toolbox, Deform Surface Push and the Undercut Detection.

The match

When I was told that Rich Welch was going to make a public statement that would clear up the misconceptions about the late fees, I was momentarily appeased, but kind of wondered what misconceptions there were. The notice that went out was pretty clear except that it allows for first a $500 fee to be charged, and then an additional $850 fee. 

Unfortunately, Welch simply threw a match into the building with some incredibly poorly chosen words, and then attempted to put out the fire by dousing it with gasoline. This self immolation was an incredible thing to behold. Here we have a guy at VP level who aparently has never seen a real user before, and doesn’t stay in touch with what real users have to say about anything. I personally don’t think Mr. Welch can effectively continue at a job in Customer Services at SolidWorks Corp after this train wreck. Really, I mean I can’t avoid laughing out loud thinking about the situation. Not the fact that he could and should lose his job, which would be personally tragic for him, but just exactly how naively stupid could he be, talking in a condescending fashion to users during a minor crisis? The Vice President of User Services?!?!? Who could possibly have any confidence left in a guy like that in such an important leadership role? If you haven’t read it, you owe it to yourself to do so. Read the 5 posts immediately preceding this one.

What users want

I took a poll that suggests that most users who answered the poll are paying their subscription, but that they don’t think they get good value for that money. Most users find their resellers are not all that they should be. People seem to be split on the service pack issue, whether the SPs are of much value, if they should be free, or if they should not exist at all.

A previous poll seems to suggest that people on average think that a new version every 18 months is about right. This would mean that a yearly subscription in many cases would not allow a user to get a new version of the software. Users are not thinking in terms of annual subscription because it is an artificial model forced on them by accountants, not something that naturally suggests itself from the practical realities of using, supporting and maintaining software. You can build a building while ignoring the realities of gravity, but that story will have a bad ending. I think SolidWorks needs to pay more attention to natural forces than imposed goals achieved by any means available.

Users want options. Many users expressed a distaste for paying for service from a reseller who simply is not capable of providing the service. Others expressed distaste for paying for bug fixes which should be free anyway. Some users think the software should be closer to actually complete when it is released.

SolidWorks is in the business of forcing customers to buy stuff they don’t need. This is the only believable explanation for eliminating SW Office several years ago. You can’t get Photoworks without also buying PDMWorks. Your choices are stripped down SW or the Office suite with PDMW.

I think the days of the one-size-fits-all, all-or-nothing subscription are long past its usefulness. These days there is a wider range of SW users than ever before. Some users are far more sophisticated than even the best resellers. Some are new or part time users, late comers to parametrics, 3D and the entire concept. You cannot use the same model to support the entire range.

Users want SW to test their software more thoroughly, rather than forcing customers to actually pay to help debug released software. Its PREPOSTEROUS! All software has bugs, sure, but SolidWorks releases software with maybe thousands of known bugs. Why do we pay to do beta testing if you aren’t even gonna fix the freaking bugs we find in beta?  Why do we volunteer to do alpha testing if you don’t fix the conceptual errors in the software?

Divide and conquer

Overall, most of the comments from users were well thought out, and only a few qualify as emotional rants. The same themes keep appearing again and again. The one thing that has happened here is that users see that even though they may use the software differently, we have very similar needs on the business side of things. Once people start talking and realizing that they have ideas in common, it became a common cause.

I’m sure this is why SolidWorks uses the “divide and conquer” psychology in their enhancement request process. They can conquer each of us individually, but if we ever banded together, we could beat them. Throwing data down a black hole on the SolidWorks site never results in a community reaction like what has happened here. It just results in frustration and resignation.

I did have a couple of reseller employees comment, and interestingly enough they waited until the very end to say anything.

The problem of resellers

Resellers can be good or bad. In fact, within a single organization you can have individuals that do a good job and individuals tht do a bad job. So it’s not always a simple thing to say that “resellers suck” because it’s not always true. And even in a sucky reseller organization, there are usually good individuals.

Anyway, I’m not so concerned with whether resellers suck or not. They either do or do not. I’m more concerned that SW can not do business without resellers. So while I may agree with some of the reasons for users wanting to eliminate resellers, it’s just not a realistic part of the solution.

On the other hand, resellers get fat on the SW subscription money, and if I were a betting sort, I would bet on that several resellers saw their subscription money going down after 2008 and complained to SW Corp to do something about it. I’ve seen good resellers hit rock bottom when it comes to the quality of technical support they provide. There is no incentive to provide good support because in order to do that, they would have to hire top notch users instead of beginners to do tech support. Many…(most?)… resellers hire people who don’t even know the software to do tech support. Again, this is not universal, because some resellers do provide quality service that is worth paying for.

Quite honestly, the best support comes from other users on the forums. Rarely a reseller will have insight into a problem that other do not. SolidWorks could hardly insulate its users from any more technical information than they do now, but users find a way, and the information gets out. Yes, I am implying that SolidWorks holds information hostage trying to prop up the value of the reseller chain in general.

Does this mean hard times for the resellers instead of just milking that ol’ cash cow and never feeding it? For some, yes. Evolve or perish. You can’t keep working off of the old PTC model to constantly bilk customers.

Answer?

To me it seems clear. In order to incentivize resellers to provide great support, we should stop subsidizing bad support. People who don’t want support still want new versions and service packs. So what’s happening is that SW is subsidizing resellers by forcing users to pay for support that in many cases doesn’t exist, and the customer doesn’t want or can’t use.

Separate goods from services. If you want technical support from your reseller, pay for it. If you don’t, don’t. Could anything make any more sense? If you want new versions, pay for them. If you don’t, don’t.

Service packs should be free.  I will stick to that one. You can’t bundle new functionality with bug fixes and act like that justifies forcing people to pay for bug fixes. Common work ethic demands that you are responsible for fixing your own mistakes, especially if your customers are the ones who identify them for you.

How much?

Say the software is $4000, because you can still get a version for that price. If you want full boat support, you can get it for $1295, just like now. If you want just tech support, that’s at a rate you can negotiate with your reseller, say $500 annually per license with a 30% progressive discount for each additional license, or on a per incident or hourly basis, say $50-100 per incident depending on how involved it is.

If you want just upgrades, say that’s fixed at $800 for the next consecutive version, $1200 to upgrade after skipping a version, until maybe after skipping 5 versions the upgrade fee caps out at maybe $3000. You always get some benefit from having been a customer, instead of the current plan, where they have a fair price for new customers but rape old customers.

Service packs are free. Do you hear me? So if people don’t want to upgrade until SP5, that’s fine.

You also have to determine what kind of support is expected just from buying the software:

- help with getting a serial number and whatever it takes to activate the license

- bug reports

Resellers should not be responsible for answering “how to” questions for people who are not on support. Anything beyond that is paid support, training or consulting. We have to stop subsidizing resellers who don’t provide good support. If a reseller offers tech support, training, and consulting, where does he put his top people? Of course, in consulting. That is the one area where competence matters. Tech support is a throw away, because, what if your support sucks? There are no consequences. If your training sucks, well, at least you still have the corporate manuals, which are worth something. But even with training, SW keeps info from users so well that they almost have to go to official training to learn how to use the software. So that is half subsidized as well.

If you stop subsidizing bad support and mediocre training, those areas will improve. Stop subsidizing by allowing users to choose which they want to pay for.

One of the favorite reseller tricks about tech support is to say “this is beyond the scope of technical support, may I suggest Advanced Assemblies training class next month?” What really is tech support? Is it for stupid “how to” questions? Is it for new release stuff? Are hardware, OS or network problems in the realm of tech support? Is it only for bugs or workarounds? Should it cover complex solutions that might be considered consulting? It’s a tough question to answer, and I don’t think this is really defined anywhere.

And here’s one that just from me personally. SolidWorks does not MAKE me productive. I make myself productive. Rich talks about the software as if you could set a toilet in front of the computer and it would be productive if the computer had SolidWorks on it. I really resent this stupid attitude. It is only the combination of a user with know how and a particular project need that makes productivity. And frankly, I think the user is a better judge of productivity than someone sitting in an office in Concord contemplating navel fuzz.

And another one from me personally. Long time users and advanced users are almost completely overlooked by SW. SW seems to be focussed on converting new users rather than keeping old ones. I read somewhere that it costs 8X as much to regain a former customer as to retain an existing customer.

If SW tries to charge for its forums, they will die and everyone will go to Eng-Tips or somewhere else. The forums are currently the best place for support, as comp.cad.solidworks was for many years before.

Product Quality

This topic is a perennial favorite. The last time it was addressed formally was with Richard Welch, and of course there were all of these assertions that SW was doing everything humanly possible to make sure the product was the best it could be. Of course nothing really improved.

My biggest gripe about the quality of the SW software is that it is perpetually half done. The biggest problems are the little things that never get fixed. You have to be on subscription because if you aren’t you’re going to have a big hassle getting service packs, and you need service packs because the software has so many bugs and there is such a rabid inclination to upgrade as soon as the software is released.

SW should at least provide what has become known as SP3 quality at initial release. SP3 is the “best practice” adoption date that  many companies go by. You should be able to use released software right out of the box. Common work ethic, guys. Have a little pride in what you do. You don’t even bother fixing beta bugs that you know about. This is inexcusable, especially if you want to charge people extra for support.

Mr. Welch added a comment to the discussion, but I thought what he has to say warrants a new post:

Hey guys
Let’s address the two issues raised here individually. First, the issue of the late fees.
There is a very small handful of customers who let their subscription support lapse, but continue to take advantage of the benefits of paid support for weeks, months, or even years before renewing their service. These late fees have existed since SolidWorks was founded in 1995 as a way to recoup some of the money that we lose by providing ongoing support to these customers while the vast majority of customers continue to pay for support on schedule. Again, these late fees are nothing new—they’ve existed since the subscription support program began, and have historically been applied only to those very few customers who opt to stay on a specific version of software for an extended period of time. If you take a look at the document Matt posted, you’ll notice that it’s not discussing a new policy, but that the late fee is simply increasing by $350 for people who go past a new time threshold. Like Matt has said himself, this only affects people who have been using subscription support but have not paid for it for more than 120 days/3 months.

As to the second issue regarding the increase in the fees themselves, there are a few reasons why the new tier and price was added. There are major improvements in the product itself that warrant ongoing investment in both core functionality (performance and quality) as well as new functions.  These new functions require more support in order to make our customers as productive as possible.  And, all of the revenue generated from these subscription program fees is 100% dedicated to improving and enhancing the software’s functionality so that our customers can be more productive and efficient.  It costs SolidWorks money to continue to provide subscription-level support to people who aren’t paying for it. There’s also the fact that things just cost more over time.  And again, this late fee increase doesn’t affect the overwhelming majority of our customers—only those who continue to take advantage of support past that 120-day/3 month period.
Does that make sense to everyone?

If not, let me know.

Rich Welch – VP Customer Services SolidWorks

I’m kind of flattered that a SW vp took the time to try to clarify a few things, or at least to adjust the spin. I have a couple comments on Mr. Welch’s comment:

According to the email I got, the first penalty of $500 shows up at 61 days after renewal date. If you buy the software without subscription and after 120 days want to get subscription, or allow your subscription to lapse more than 120 days, that costs whatever the subscription was plus $850. If you lapse for 2 years you have to pay 2*$850 + 2*(annual subscription). As Grabowski says, you are paying for more for not being on subscription than you would pay for being on it. The clear intention here is as a deterrent. Recouping costs? Really? hmmm.

My impression is that most people want to go the other way. This is why I think people who this doesn’t apply to are upset. If you look at the poll from the last post, most people subscribe even though they don’t think they are getting much value for that money. I think people want the option to drop the support, and  have an economical plan for just buying upgrades.

Plus, frankly, I don’t buy the explanation that resellers want to recoup money for supporting people who don’t pay for it. I worked for resellers, and I know how many non-support people get support. Very few. Besides, there are much better places to get support than resellers.

I think this is an intimidation tactic to prevent people from allowing support payments to lapse. It is not a way to raise money from the fees. So few people will pay the fees that the additional income won’t offset the grief of making the decree. The main goal of the fees is to make sure that people never pay them, but they pay the subscription religiously. It’s a deterrent.

I think it was necessary because 2008 was a disaster and has cost SW and its resellers a lot of maintenance contracts.

If you haven’t heard from your friendly neighborhood SolidWorks reseller yet, you will. The terms of SolidWorks subscription services are changing. It seems the main force of the change is that the penalties for missing a year of subscription are going up. Here is a screen shot of the letter from my reseller:

This is a dangerous gauntlet for SW to throw down. They are betting that this is just going to scare people into never letting their subscription lapse. But what if it confirms that the subscription is simply not worth the check you have to write? For me personally, the reseller provides nothing of value whatsoever, not to say that this is everyone’s experience, it’s just mine. The service packs are of dubious value. In 2008, how many *.1 releases did we have? 0.1, 1.1, 2.1, 3.1? *.1 releases mean that there were errors that no one counted on. Too many sloppy mistakes.

It’s a little maddening to me, because I pay the subscription mainly as a formality. I have a license by virtue of being a partner (in the Publishing category) and a commercial license for which I have to pay maintenance. I still do commercial work alongside my writing activities, so technically I need a commercial license for that work. Of course I haven’t had the commercial license installed for a couple of years.

With the SW08 release, I’m sure more than a few companies have bailed out of the subscription treadmill. This new policy may be aimed at intimidating these customers or former customers into hopping back on. With such a junk release (08) immediately after a reasonably usable release (07), its easy to see why people would just freeze at 07 and choose to forego updates.

What do you think of the “subscription treadmill”? Does this new policy look like corporate bullying to you? Do you think you get your money’s worth out of your subscription money?

Ok, I’m pulling your leg. There is no such thing. The funny thing is that some people think there is one. Some people crash and crash and crash. I don’t doubt that there are people who have real problems with the software continually crashing, but there are also a lot of people, me included, who don’t have crash problems. It doesn’t have anything to do with a different edition of the software, it has most to do with “computer hygiene”.

Just so there is no misunderstanding, what I’m saying is that if you are crashing more than say a couple times a week, the fault is most likely your own. Doesn’t it make you angry when people attack the victim? How can it be my fault? I install the software, the software crashes. Period. End of story. I didn’t write the software. It’s SW’s fault!

No, it’s your fault, and here’s why. It’s not because I’m a SW corporate shill, or because I’m trying to make nice with someone at SW I want a favor from. I haven’t sold out or caved in or given up. I’ve just seen enough of real problems and real solutions to know that that there are things that really make a difference. And if you don’t listen this time, well, that’s not my fault.

Crashes will find YOU personally

The number 1 trait of people who don’t crash SW continuously is that we are a little bit anal retentive. If you are sloppy, or assume that because you can do things it means that you should do them, or if you don’t pay attention to warnings or best practice or recommendations, you are a good candidate for a slew of SW crashes. Basically, if your computer hygiene habits are lax, you run a greater risk of crashing.

Buying a computer that won’t crash SW

I know a lot of people are going to disagree with all of this, but that doesn’t matter to me. If you are crashing and can’t fix it, you need to hear this.

The first thing you need to do is to buy a computer from an organization that offers a warrantee. An organization that tests configurations and has a burn-in procedure, and will take the whole thing back if it doesn’t work. That means that you shouldn’t build your own. Build-your-own computers are a big source of unsolvable crashes. Underpowered power supplies, incompatible components, IRQ conflicts between obscure components, overclocking, and lots of stuff you or I don’t understand. Yes, I know there are people who are very successful with building their own computers. If you are crashing all the time and you have built your own, you are not one of them. Sorry.

There are two kinds of mistakes you can make when getting a computer to run SolidWorks: buying cheap crap and buying expensive crap.

Be Conservative

Seriously. You can still vote for Obama, but don’t push your selection of hardware and other software to the bleeding edge. Back off a little bit. The newest stuff may sound really cool, but choose the tried and true over the latest thing you saw in a gaming blog. It’s so tempting, and you gotta be cool, but if you are having crashes that you can’t solve, you need outside help, and being conservative with hardware selection is safe advice.

Do you really need RAID 0? It’s 3X more likely to fail than a normal hard drive set up, and costs more. The most expensive doesn’t mean the best. 16 GB of DDR3 RAM? talk to someone who configures computers for a living. It may be a big expense that doesn’t buy you much. SLI? Bluetooth? Are you really gonna use that?

Video Cards

This is the last time I’m going to say this. ATI Radeon cards are crap for SolidWorks. Really. I don’t care what the frame rate specs are, they are crap for SolidWorks. They are meant for DirectX not OpenGL. I know they are supposed to be compatible with OpenGL, but there is a difference between theory and practice. I know it is just corporate greed that makes you pay more for a Quadro than a GeForce, but you gotta take the hit on this one. Are you crashing with your Radeon? See. Told you.

Get an nVidia Quadro. Don’t go cheap (NVS = a 2D card), but you don’t have to double the cost of your new PC with a 4600 or 5600 either. A 570 works. 1700 is better. You may not see any benefit between the 1700 and a 5600. A lot of people try to low-ball video cards. Trust me, this is not a good place to save a couple bucks. On the other hand, I think a lot of people waste money on $2000+ graphics cards that don’t deliver that much benefit. Expect to pay $300-$500 for a decent video card. It’s a business expense.

Operating system

At this point, XP is still a safe bet, but it may not age well. New hardware that comes out a couple years down the line may not have drivers written for XP. On the other hand, your existing hardware may not have drivers for Vista, especially if you go 64 bit. From a SolidWorks standpoint, either XP or Vista should work.

Do you need 64 bit? Are you on the verge of needing 64 bit? Everything is moving eventually to 64 bit. If you choose Vista, Vista requires at least an additional gig of RAM just to make it work. Vista 32 almost doesn’t make any sense. If you have data sets that drive your computer to use over 2 GB of memory, you should move to 64 bit. There is still some software or hardware that doesn’t run on 64 bit, which is a bummer. You may be forced to dual boot, so you have a machine that can run with large data sets, and will also use all your software and hardware (just not at the same time).

Processor

The big question here is dual or quad? Depends. If you use Cosmos or Photoworks a lot, go quad. If you do complex sheet metal parts, go dual. If you do surface modeling, quad might help, but I would go dual. In general, clock speed is important, and quads generally have slower clock speeds than duals. If you are doing machine design, quads are probably not going to help you much. Complex drawings of large assemblies may benefit from quads. Much of this is still black art, because SolidWorks refuses to say anything really definitive here. I know that multibody models take advantage of multicore processors, but don’t have any direct experience with quad core or double dual core.

PC or Mac

Don’t be a moron. Would you buy a Volkswagen Beetle to haul lumber? No. You wouldn’t. Sure it’s cute, but it’s not made for the job. If you buy a Mac to run SolidWorks, you might as well just throw away $2000 and get something that will work better. Mac does not offer an adequate selection of hardware. They break most of the recommendations I’ve made here. If you want a Mac, that’s fine. Go get a Mac, but I wouldn’t recommend it for anything other than casual use of SolidWorks. I know, some people claim they work great, but people claim to get good results with Radeon cards, and some people actually voted for Jimmy Carter. Lots of things sound like a good idea at the time, but what it does in practice is what’s important. Mac gives you very limited options for video card and processor. The options they do give tend to be ridiculous, such as dual quad cores and the 5600 video.

Other software

I always keep 2 computers. An older one for messing around (iTunes, internet, email, installing junk apps, etc) and one for real work (SolidWorks, MS Office). Don’t install crap on your SolidWorks computer. Really, just don’t do it.

System Maintenance

First, make sure you have a back up image of your computer set up. Really, do it. Make sure the image is taken when things are fresh and new. If you have to reformat your hard drive and reinstall the OS to get this, it will be worthwhile. If you are crashing a lot, a reformat may be in the cards for you anyway. Remove software you don’t use. Use C-Cleaner to clean the crap out of your registry, and your temp file junk. Run defrag once a week. It is best if you can run without real time antivirus, but make sure you are behind a good firewall, and make sure to scan all your data at night.

Just in general, treat your computer as if your job depends on it. Be sober and serious about what you put on it or in it. Be a little anal about this because it will prevent you from crashing a lot. Go home and play World of Warcraft, turn the radio on or use your iPod instead of iTunes, use an old retired CAD box to put junk apps on, with a KVM switch to change between CAD box and junk box.

Dealing with corporate IT nazis

If you have to deal with corporate IT nazis, sorry. Sometimes they put stuff on your system that is the cause of problems. If they make you use Novell, it’s time to find a new job. If they put monitoring software on your computer, I’d scan my ass and fax it to the CEO, tattoos and all. Companies that don’t trust their employees need to stop hiring off the bottom of the barrel or fire their paranoid IT director. Computers are tools, not priviledges.

I understand that none of this is really helpful advice, but you often can’t win the fight against bad IT policy, so I recommend taking any revenge you can. I busted an IT department at a large company when I was brought in as a consultant and the IT fellow completely wasted a day of paid consulting time. He was apologetic and cooperative for my second visit (after my report to the boss).

At one job I was able to negotiate semi-IT status as the CAD admin. This company had a great bunch of guys in IT. They truly helped rather than hindered the effort. You’ve often got to make the case to management as to why SolidWorks users need more control and access than word processor or spread sheet users.

Summary

Anyway, I hope you get the picture. Having no crashes is no accident. It really does make a difference when you are more careful about what you do.

In software where the end product (CAD data) is highly process-based, and where there are almost innumerable possible ways of doing any particular task, it is a given that the accuracy of the final product is of utmost importance, but it also becomes important how you arrived at that final product. When you are working with data that may be revisited later or working in an environment where multiple people are contributing, consistency in the process of making SolidWorks data is key.

A lot of people seem to be looking for best practice rules to implement at their company or to support their point of view in an argument about 3D CAD standards. Some of them act as if they think Moses came down from the Mount with CAD best practices carved into silicon.

Part of the reason people are looking for this is that SolidWorks has played fast and loose with best practice from the beginning. Many users were sent down the road with a pat on the back, a smile and a “look how easy this is”. Like unprotected sex, that’s gonna come back at you one of these days.

Fast and loose is all fine when the software is relatively unsophisticated, but times have changed. When selling SolidWorks, no one says any longer “look how easy this is compared to Pro/E”, now they say “look how powerful this is compared to Inventor”.

As the software has become more sophisticated, the only thing that has kept pace with that sophistication is the reseller implementation services. Resellers realized by answering (or not answering) a lot of tech support questions that customers were coming up with questions that weren’t covered by training, and they weren’t really in the scope of tech support. The new field of questions had to do with process. 3D CAD really is a different beast from 2D, and it really does offer us new ways of re-using data.  And it really does change our engineering process.

Anyway, SolidWorks made the software so easy to use in part by shortcutting a lot of general CAD best practice ideas. Now that a lot of users are entrenched in bad habits, some of them are looking for better ways of dealing with this more sophisticated tool. “Best Practice” has this air of panacea about it – a cure-all developed by the collective wisdom of top users from all disciplines. It’s partially true and partially untrue. Sometimes you have to separate the wisdom from the wives tales.

First of all, best practice suggestions are just that – suggestions. There is no single set of hard and fast rules that can be applied to any two different companies. Every company has a different way of doing business, and there are different individuals involved in every situation. Each of those factors will change what works best in that situation.

Second, best practice suggestions are often self-contradictory. When I suggest practices for a company, some times I will suggest one thing, and then I may turn around and suggest exactly the opposite. Some situations call for one tactic and a different situation may call for something entirely different.

Several factors can influence the best practices or standards that you use at your company, and let’s not be shy about saying it like it is. One of the factors is training or overall skill level. If all of your CAD users are wizards, and they are all ultimately flexible, then you should be able to make absolutely anything work.

The other situation is more common and more of the condition that best practice rules are meant to deal with. Users who are not experts can sometimes get themselves in deeper than they can dig themselves out. Best Practice suggestions help guide users like this to help keep them productive, even if they are venturing into a new area of the software where things are unfamiliar.

Finally, best practice suggestions can never be implemented blindly. You and your CAD operators will still have to think about things before they do them regardless of how many CAD standards you write down. There is no substitute for the human mind, even in engineering processes.

The difference between CAD data and engineering design data is that CAD only communicates shape and dimensions. To make your CAD data into engineering data takes a lot of extra thought, planning, research, inventiveness and communication.

As an example of some “best practices”, several years back, I wrote some of these “Rules of Thumb” lists, some of which are still valid.

Did you ever see that 2005 kids movie called “Robots”? Wow. What a great, and I think completely underrated movie. I could sit and watch that with my 4 year old cousin over and over and over again. Robin Williams, Mel Brooks, Ewan McGregor, Greg Kinnear, Paul Giamatti. The premise is that this crazy inventor robot inspires kids to be inventive. But the company that he starts from his inventions is taken over by a greedy robot, who turns the whole company into an evil empire. It shows how even in a fantastic mechanical world, there is still this caste system between old and new. It’s funny in a mechanical sort of way, but it also has a bit of a lesson about modern culture. The main evil element in the movie is a mega corporation making and selling “upgrades” for robots, making you feel bad if you aren’t upgraded. To me, this movie is saying that upgrades (symbol for materialism/consumerism) are just being pushed on you so that someone else can profit, and the benefits to you are pretty hollow, when it comes down to it. Heady stuff for kids, but I think they need to be able to think for themselves rather than letting marketing and advertising (or even cool animated movies) think for them.

Anyway.

I got burned recently  by “upgrade-itis” – that uncontrollable urge to upgrade software for no apparent reason – with my recent WordPress (blog software) upgrade. I actually did hesitate for a moment, and asked myself “is this really prudent?” Frighteningly, I heard the answer in my head say “No”, but I went ahead and did it anyway. Having little conversations with your inner self is one of the scarier parts of being an independent, and not working with an office full of other people all day.

I always lament the new versions, but that’s because I’m driven to try them out as soon as possible. Upgrade-itis? Yeah, I got it bad. New versions always mean a lot of work for me. Still, I can’t avoid them, I’m drawn like a moth to the flame. 2009 is a bit of an exception. I’m so anxious to bury the stinking corpse of 2008 that 2009 could not come soon enough, and after having played with it for a couple of months by now, I think it is overall pretty good.

Still, I think the beta period is WAY too short. Rhino has beta periods that last months. Rhino is successful and powerful software, and they have serious competitors, but they are only on version 4.0. I personally don’t run into bugs in Rhino, but I don’t use it as much as SW either. I don’t think Rhino puts much stock by what panic stricken self promoting marketing people say. Neither do I.

Back when SW first started, they claimed they were doing new releases every 6 months. The release cycle gradually slipped, and now I believe we are on a 12-14 month cycle. How often do you need a new version? Is every year too frequent? Release cycle has a direct relation with product quality (meaning reliability, stability, predictability). Is getting new features quickly more important than getting new features that work?

Brian Benton over at CAD-a-blog set up a little poll and asked AutoCAD users how often they think Autodesk should bring out a new version of their software, and this is what they had to say. Granted, these are AutoCAD users, and are probably more afraid of change than most.

So let’s take a couple little polls, and try out my new poll plugin (that SodaHead crap didn’t work very well).

 Loading ...

What part of CAD Management/Administration interests you the most?

• Personnel management (hiring, evaluation, training, etc.)
• Technology management (software evaluation, implementation planning
• General management (budgeting, scheduling, process development, meetings with upper mgt, etc.)
• Software installation, usage, configuration, maintenance
• Creating, enforcing CAD standards
• IT issues (hardware, network)
• Process development (documentation process, product development process, etc.)
• Other (please leave comment)

View Results

 Loading ...