Just Because You Can . . .

I work for a small design engineering company who occasionally farms out work in order to maintain our focus on the higher end work. Last week I received an approval package from one of our suppliers for a tooling package. I usually blow these off because it’s “just tooling” but this time I decided to take a closer look at the detail. What I found prompts me to write.

The tooling is for a composite lay-up mold. Our parts are usually very large and require extremely accurate molds to lay-up in. We don’t need a closed cavity like an injection mold or die-casting mold. I began by comparing my original model file to the cavity in the tooling. The part was most definitely in the tool, which would have ended my analysis, but I decided to check out how this tool was put together. The split lines for this tool were perfectly placed on the exact split lines of the part, however being an open tool with no closed cavities it would have been easy to make a straight split line outside of the silhouette line since we’d be removing those pieces to release the part, anyway. By having a curved split line the machine shop now has to contour the flange instead of making nice flat planes. This is an example of the power of 3D CAD creating this beautiful curved mating flange that deviated from a straight line by only 1 inch. What this designer didn’t account for was how much more work it was to machine that surface. Now, this supplier has sophisticated 5-axis machine tools so it’s their own nickel on the line to produce this geometry. One thing is for sure, those sides will only go on one way, but then a well-placed dowel or notch would also make assembly a no-brainer. Matching contoured split lines is harder than matching flat split lines.

When I got to the drawings I saw another example of automation gone amok. This same part has a line of symmetry that could be used to mirror mold pieces along a diagonal line. It’s not immediately apparent but certainly there when I designed the part. So, I checked to see if opposite parts were, in fact mirrored. Nope, the mold surface was mirrored (by design) but all holes, pins, fasteners and clips were unique to each mirrored part. Most 3D CAD will maintain associability between a mirrored part and its parent. That’s good automation. Even assemblies can be mirrored to good advantage. How about creating arrays of fasteners instead of individual instances?

When memory was scarce, rendering was done overnight and 3D was a fraud we took care to conserve resources not the least of which is our time to create efficient computer models and there was a sense of pride in getting a model to fit on a floppy disk.

Have we gotten lazy or are we just not paying attention to the basics of good design practice because we have the resources at our fingertips to create huge 3D models with no intelligence or economy? I’m not talking about trivia like schematic vs. modeled threads or weld callouts vs. actual beads. I’m talking about planning up front and thinking about that guy out on the shop floor that actually has to flesh out your work. Cut them some slack and knock off the fancy curves and give them 90degree angles not 90.22 degs (I’m not kidding), straight lines and round holes.

By Marinus B. (Ben) Bosma
Engineer
IAM Systems
Email Contact

Rating:
Reviews:
Review Article
  • Re: Just Because You Can . . . October 11, 2006
    Reviewed by 'Norm'
    I think you will find three key reasons why designers continue to have bizarre parting lines.
    1. Not talking to the manufacturing people and truly understanding the mfg process. Particularly in this case an open tool.
    2. Industrial Design drives the look of the design and ID likes curves, especially cool blends
    3. Designing without ever even trying to "define" a partling line/plane first.
    I feel your pain, but don't blame it on the CAD. There are just as many and probably more examples where CAD saves the day in todays lifecycles. If it didn't, we still be on mylar.



      Was this review helpful to you?   (Report this review as inappropriate)


  • 'Just Because You Can . . .' December 09, 2005
    Reviewed by 'tomL'
    I hear ya Marinus and totally agree. I have also seen design engineers spend literally weeks to get a blend on a part when the client is chomping at the bit for parts and toolmaker sits idle. All for what, since the tool maker states the blend is not needed since it is a natural feature that is generated by the tool making process anyway.
    As a product designer I have to say we are fall prey sometimes to make the parts as realistic as possible and sometimes loose site of the big picture. Thankly I keep my feet on the ground 99% of the time. But yo know you get that new release of your software with that cool feature and you just gotta try it on something. Its a case of the WOW! cool new feature but totally counter-productive.
    So make enhancement requests to you software company be careful what you ask for, you just might get it.
    ;-]
    TomL

      4 of 5 found this review helpful.
      Was this review helpful to you?   (Report this review as inappropriate)


  • Re: Just Because You Can . . . October 11, 2006
    Reviewed by 'Joe'
    This was the point I was trying to make in the “Increase Solid Model Value with 3D GD&T” topic. Making a drawing lets more eyes see the design.

    I remember a time when I got a layout at Pratt and Whitney on a protective shield for the afterburners. It was basically designed the wrong way with the split on the radius instead of on the outside edge. I made the Chief draftsman aware of this and they sent it back to the design engineer.

    The parts I am seeing from the design staff without drawings are so expensive and difficult to make that I can only surmise the no one is really scrutinizing the parts.




      3 of 3 found this review helpful.
      Was this review helpful to you?   (Report this review as inappropriate)


  • Re: Just Because You Can . . . November 30, 2006
    Reviewed by 'Bseitz'
    Its not unusual for people to "play" with the technology and do things just because you can. I recall one of my first jobs that got me into engineering consulting. I was acting as the Mfg - Engineering liasion. A truly wonderful sheet metal part was transfered over to my Cad/Cam system for a manufacturability check.

    It was a masterpiece of bending and welding. Unfortunately, it couldn't be manufactured, at least not in our factory at the time. I sent the drawing/model back and received a storm of critics from the design department. I escalated it to the engineering department director, who promply dismissed my critic also. He saw the bends and folds were beautiful, a materful use of the CAD tools at hand.

    I finally escalated it to the VP of Engineering, there I was surrounded by a heavyweight engineering staff and I only a first year Mfg Eng. and that was pushing it, as I had just come out of the Air Force from being a Flight Engineer/Jet Mech.

    We all went to the big conference room, I had printed out the drawings and an estimate of costs to actually build the series of components; a few fuel tanks. While the sheet metal costs and fabracation process was cheap, the end result of how it had to be build would be very expensive I told the VP of Engineering. I had estimated the eight tanks we could make would cost about 10 million.

    The design engineer bristled at such an estimate and asked how could I justify such. All during this time the VP of Engineering was stairing at the drawing and smiling. He knew where I was going.

    I told the design engineer that actually the first seven would cost less than a million to produce however the last one would cost the company nine million in insurance and fines, because someone was going to ask.

    The Engineering Manager shouted ask what? I responded simply, "where the sheet metal mechanics went? We only have eight and when the last one is gone someone is going to ask questions. He asked again what do you mean?

    In order to build as designed I'll have to seal a mechanic inside the tank of each one. The design was such that you could only assemble and weld the assembly from the inside and the fuel port was not large enough to allow a man to exit it once sealed.

    At that moment the VP got up and asked the Design Manager did he actually check this design? The manager himed and hawed a little and said he'd use the system to chech it. And again the question did you check the design yourself rather than a report that showed all the edges mated cleanly.

    Needless to say, just because you can...



      Was this review helpful to you?   (Report this review as inappropriate)


  • Fancy Schumancy December 07, 2005
    Reviewed by 'me_nc'
    II too have seen this problem and often need to send a design back to have it simplified. Many times the design engineer will challenge these request because if they have done it once before to them it is “open season” and they should be able to continue. Generally when explained to them the burdens the shop has with the fancy or not so simple features that do not add value, they understand and try to comply.

    The other side of the coin is the design engineer has customers and sales that continue to ask for more than they really need, they want a design a certain way, they don’t care how it’s made, and want it yesterday for next to nothing. They are often given little or wrong information to work with and are under pressure to get it done fast.

    I don’t think people want to do a bad job but do overlook some details for others. I do think that the squeaky wheel does get fixed, so instead of being lazy and blowing it off why don’t you (we) go back and get it changed and tell them why without assuming they are lazy.

      6 of 6 found this review helpful.
      Was this review helpful to you?   (Report this review as inappropriate)


For more discussions, follow this link …

SolidCAM - See For Yourself

Featured Video
Editorial
Latest Blog Posts
Sanjay GangalMCADCafe Lens
by Sanjay Gangal
NVIDIA GTC October 2020 Keynote
Jobs
Failure Analysis Engineer for Flextronics at Milpitas, California
Mechanical Engineer for Flextronics at Milpitas, California
3D Designer and Developer for Los Alamos National Laboratory at Los Alamos,, New Mexico
GIS ANALYST I.8030200 for Dallas County at DALLAS, Texas
Structural Engineer for Kiewit at Lenexa, Kansas
Design Engineer (Engineer 1/2) for Los Alamos National Laboratory at Los Alamos,, New Mexico
Upcoming Events
DEVELOP 3D Live 2020 New Date November 3rd at Sheffield University ctagon and INOX Building, Durham Road Sheffield United Kingdom - Nov 3, 2020
Digital Twin 2020 - Now in November 2020 at Melbourne FL - Nov 4 - 5, 2020
Sensors Expo & Conference at McEnery Convention Center SAN JOSE CA - Nov 16 - 18, 2020
Spatial 3D Insider Summit - Free VR Event! at United States - Nov 16 - 17, 2020
Kenesto: 30 day trial



© 2020 Internet Business Systems, Inc.
670 Aberdeen Way, Milpitas, CA 95035
+1 (408) 882-6554 — Contact Us, or visit our other sites:
AECCafe - Architectural Design and Engineering EDACafe - Electronic Design Automation GISCafe - Geographical Information Services TechJobsCafe - Technical Jobs and Resumes ShareCG - Share Computer Graphic (CG) Animation, 3D Art and 3D Models
  Privacy PolicyAdvertise