Will HP do for 3D Printing what they did for Pen Plotting back in the early 80's... 1

[JohnRBaker]

I just saw this item today...

http://seekingalpha.com/pr/16491237-hp-delivers-worlds-first-production-ready-3d-printing-system

...and while a new 3D Printer is not news, I couldn't help but think back to the days when the only direct 'physical' output from a CAD system was a Drawing.

A standard part of virtually all CAD systems has been a 'plotter' so that Drawings could be 'printed'. Back in the early days of the CAD industry (I've been part of that history since 1977) plotters were large and often rather expensive pieces of equipment. When the company I worked for (back in Saginaw, MI) installed our first CAD/CAM system in 1977, included was what was then the cutting-edge in plotting, a Calcomp 960. Now this was an E-size, two-pen, belt-plotter which cost over $50K (and this was in 1977 dollars). Now that was a deal. Across town a the GM Steering Gear plant, they were using a Xynetics flatbed-plotter that cost them well over $100K and our Calcomp was much faster. However, they both suffered from the same inconvenience of having to clean pens, keep them filled with ink and if you needed more than two line widths, you had to change pens and re-plot, which was a real pain-in-the-###, but that was state-of-the-art at the time and still an amazing thing to watch your perfectly lettered Drawings being produced at the push of a button.

Well a few years later, after I had changed careers and went to work for McDonnell Douglas selling CAD systems, we were still limited pretty much that same pen-plotter technology. By then the large-format (D- and E-size) plotting was dominated by brands like Calcomp (Flatbed and Belt-plotters), Benson (Drum plotters), Xynetics (Flatbed), Gerber (Flatbed and Photo plotters), etc., all of them large and expensive.

Then in 1981 HP introduced the HP 7580. This was a D-size, eight-pen, so-called 'grit-wheel' plotter (that simply pinched the drawing sheet between a set of grit-wheels to move it thus eliminating the need to tape down your drawing sheet) which was very fast and cost less than $16K. Now it's true that HP had been in the plotter business for years, making smaller lab-sized X-Y plotters (I used them in college back in the late 60's early 70's), but when they moved into the full-size engineering drawing business, nothing was ever the same. And it wasn't just the price point, but also the shape factor and with the support for true multi-pen plotting it opened up things more use of colors and such. But one thing else it did was to make the technology more practical for everyone.

Case-in-point: When we sold our first CAD system with one of those new HP 7580 plotters I can remember when our installation group (remember in those days CAD systems were generally sold turn-key with CAD Terminals, CPU's, Disk Drives, Printers, Tape Decks, Paper-tape punches/readers, often digitizers and of course a plotter) calling up HP and asking when they could schedule a service tech to 'install' the plotter at the customers site (since this was the first HP plotter we had sold, our people had no experience with them). Anyway, the HP people thought it was a joke or something since all you had to do was plug it in, connect it to the CPU, turn it on and go. This was one of the first high-tech devices where the 'installation instructions' were what you'd call a sort of 'comic-book' like format with mostly pictures showing what plugged into what and almost no text whatsoever.

For anyone out there who doesn't gp back as far as I do in this industry, here's a few images of those original plotters we had to contend with:

Calcomp 960 belt-plotter

Xynetics flatbed-plotter

HP 7580B plotter

Anyway, the point I was trying to make is that perhaps HP will have the same effect on where 3D Printing is going as it did on Plotting. After all, anyone who can remember the early 3D Systems and Stratasys Stereolithography machines can see how these newer generation devices have already come a long ways and when someone as large and significant as HP, with their lonmg history of '2D printing' moves into the next 'dimension', well it will be fun to watch.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

 

[JohnRBaker]

There will also be breakthroughs with respect to being able to 'manufacture' parts that would be impossible to do using any other 'conventional' technique. For example, since the is no need to be able to reach a surface with a cutting tool or have to able to 'release' the part from a mold (although you might be able to do an investment casting) it's going to allow for very exotic shapes. Also, the number of parts to accomplish something could be reduced since again since there is no need for parts with 'draft' or being able to machine certain areas of the part and then assembly pieces together afterwords. So one should not look at this as simply a replacement for current manufacturing processes but rather as an opportunity for going back to the designers and freeing them from having make sure that their models are 'manufactureable' from a classic point of view.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

 

[KENAT]

John, I think that's one place it is making inroads for high performance applications which aren't as directly cost sensitive to manufacturing methods.

For aerospace etc. the ability to have parts that are almost hollow except for an internal latticework has major potential for instance.

In fact, it has major potential for my field, I just haven't been able to persuade anyone yet - and the available DMLS materials have some issues for us in other material property aspects.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[BiPolarMoment]

The latest DMP machines are printing titanium that is almost as good as wrought 6-4 and meets the material properties in F136--as long as it's been heat treated (stress relieved or HIP'd). I'm using the process to create some implants near-net shape that are actually cheaper than their original PEEK counterpart (mostly due to proprietary material costs) but impressive nonetheless. Similar story for SS17-4.

 

[JNieman]

Speaking to design without regard to traditional manufacturability - I find it very difficult to break my mind out of the constraints. At this point, it is almost subconscious and instinct that I design with traditional processes in mind.

I've read about some significant design changes that DMLS or SLS type processes have allowed. Speaking of KENAT's note toward hollow structures with intricate interior structures, this article from NASA comes to mind:

http://www.nasa.gov/marshall/news/nasa-3-D-prints-first-full-scale-copper-rocket-engine-part.html

“To circulate the gas, the combustion chamber liner has more than 200 intricate channels built between the inner and outer liner wall. Making these tiny passages with complex internal geometries challenged our additive manufacturing team.”

With hints toward both cost reduction and performance improvement for a production mindset rather than one-offs or prototypes:

“Our goal is to build rocket engine parts up to 10 times faster and reduce cost by more than 50 percent,” said Chris Protz, the Marshall propulsion engineer leading the project. “We are not trying to just make and test one part. We are developing a repeatable process that industry can adopt to manufacture engine parts with advanced designs. The ultimate goal is to make building rocket engines more affordable for everyone.”

Obviously a niche outlier, but it is a proof of success/potential, certainly.

 

[BiPolarMoment]

To iterate another use--assuming we can establish material traceability and properties--producing a near-net (or net) instrument in a day rather than 12 weeks is a big deal for us. But again, I'm speaking for DMP/DMLS, not "Multi Jet Fusion". If HP figures out how to 'print' Ti or SS, I'll be more interested.

Not everyone is trying to make 10,000 parts; I have more examples of <10 part runs than >10 part runs (okay, that's hyperbole, but it feels true). But up until recently even if I prototyped in DMLS I had to traditionally manufacture the parts released to the field (at horrendous cost)... but that's a regulatory story and I wont cry you a river.

 

[MacGyverS2000]

LPS for MM... that's my thinking, also. Manufacturing will eventually catch up / surpass current methods... until then, we're stuck with endmills and billets.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[itsmoked]

I've been stymied a few times by carrying a design into several machine shops who all turned me away because I only needed one or two pieces. A great nitch problem for 3D printing to slay.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[MadMango]

There's definately a place and purpose for 3D printing, no doubt. The original article was specific about "production-ready". As we all know, one man's "production" might be 20 a year, another's 20 a minute. I look forward to the 10x speeds and 1/2 cost claims to be real.

"Art without engineering is dreaming; Engineering without art is calculating."

Have you read faq731-376 to make the best use of these Forums?

 

[harolddavy]

That look so nice.