To 3D or Not to 3D

[rhoscadman]

Hello and Happy New year to all.

After much talk and discussions over the past 2 years, the powers that be (management) are having a sit down "thrash it out until we come to a decision" meeting tomorrow with the design draughtsmen in our company.

They have come to the conclusion that after all we really need to move into the 3D world. So of course, now it's rush, rush, rush.

We dabbled in the 3D world about 3 years ago when a manager that had NO concept of CAD, 2D or otherwise, allowed himself to be convinced by a snappy salesman to go down the Inventor route. Due to our type of industry this was a total failure. We found Inventor so cumbersome, such a change in CAD mentality that we discarded it a few months later.

Our company designs, procures and installs Upvc/GRP Odour Control vessels/filters with all the accompanying engineering disciplines: Chemical, Electrical, Structural, Mechanical, ductwork, piping, steel supports etc.  We generally just do outline designs on our G.A.s and the fabrication drgs aren't that detailed anyway. All the more intricate plant, like valves, fans and dampers we buy in anyway and so use the supplier's CAD details.

The general consensus is now to upgrade from LT to full blown Autocad which has 3D on it to start with, and then move to more sophisticated software like Plant 3D (expensive) as our 3D expertise increases.

What has alarmed us though is that we receive a number of 3D drawings from suppliers and clients that have used diverse 3D packages (be it M.E.P., 3D Acad, 3D Plant, Inventor, Revit etc. etc. etc.) and had difficulty in converting it into a format that LT would be able to use. There were all sorts of shenanigans when we discover that not all 3D programs (even amongst the Autodesk family) "see" each other. Does that mean a company needs to have a whole range of 3D Software in it's armoury? Or is there a programme that sees it all?

Your opinions/advise would be much appreciated.

Rhoscadman

[kcobabe]

Hate to tell you, but the salesman was right.  You should be using Inventor.  The problem sounded like your people did not get trained and if they did you did not get trained by an Autodesk certified trainer.  Granted Autodesk Certified trainers often will show you the out of the box training.  What I asked out trainers to do is spend the last day off ciriculum and show designers and drafters how to do a real world design.  In other words take a typical part and design it.  Usually you want it to be something small that can be completed in one day.

Sounds like most of your personnel have only used CAD.  When you switch to a CAM package it's 100% different.  Does not matter if it's Catia, Pro E, Solidworks, Inventor, etc...  All of them are pretty similar and have rules you have to follow.  Trying to move over into full blown AutoCAD is only going to make things more complicated.  Granted it's gotten better over time, but still lacks the ability that Inventor has.  Especially when you add in the engineering software that let's the engineer check the design for issues.

I would take a good look at inventor and this time do training on it with a certified trainer.  You just can't switch from 2D to a full CAM package.  That would be like going from learning how to walk to driving a 700hp Ferrari in a day.  Just is not going to happen.

Kind of surprised the salesman didn't sale up with some training added on for a discounted price.

[dgorsman]

Different industries follow different work patterns, necessitating different software and workflows.  Hence Revit is going to be different from Inventor, Inventor is going to be different from Plant3D.  Not to mention different developers have their own development plans and legacy support requirements so Inventor will be different from CATIA, CATIA is different from PDMS.  Even the AutoCAD verticals like Plant3D and AutoCAD MEP are focused on their disciplines, creating custom objects which require that specific program to work with them; you can view them using the proper object enabler.  There is an AutoDesk product, Navisworks, which is capable of reading many, MANY, different types of 3D files to assist with coordination.  But getting models from different products together in the native environment is very difficult.  Get the software you need (and Inventor may not be the one you need), and get trained up in it.  Moving beyond basic drafting to a design-based workflow requires specialist knowledge from the designers in both the software and the design practices.

[JamesMaeding]

If I was an owner of your company, I would say "show me it can and should be done using whatever software sounds good".

The goal is to make money in a business, and that does cover more than getting the next product out.

It covers keeping future clients happy by providing a product they want to buy.

So someone has to learn that new software and do some production on it to see what skills are needed and if it works for you.

This talk of "diving in" is nuts IMO. The only people that do that in business have lots of money to spare which I doubt yours does if they are fighting about it.

If you do not have access to someone that can prove the workflows needed, you should be very careful about any changes in tooling.

[steve216586]

My company has been in the process of converting to 3D for the last six months. We use Alibre Geomagic as our 3D program of choice and also still use AutoCAD Vanilla, Electrical and Mechanical for 2D and older projects, where it is just not practical to convert to 3D at this stage of the game.

The key aspect I see, is as to what format the drawings can be saved. We can easily use .SAT files in AutoCAD 2D anything. OR export to Geomagic in that form. Our work flow rarely crosses but when needed, there is not much hassle between the two softwares. Bonus for everyone!

As far as training and getting up to speed as a team. We have those old dogs. They stay at AutoCAD 2D all day long. Myself and a few others have picked up Geomagic on our own and helped each other out with our "discoveries". To tell you the truth, after I finished the tutorials, I was able to complete parts and assemblies on my own. I did reference the tutorials often, and still do now and then, but they are not so much a crutch as they were last September. The tricky part was learning some of the nuances of a new program. e.g. How to place restraints onto objects in an assembly and still maintain the flexibility to revise and redesign when necessary. This took some team cooperation by sharing our trial and error successes and failures.

I think if you have a few people willing to learn, and do it on their own time, to add another tool to their work belt, you will have success and not lose man hours on formal, unneccesary training which doesn't fit your real world usage. A few months down the road, those who didn't move forward, will wish they did, and those who chose to learn, will feel gratitude from the whole team and management.

[JamesMaeding]

when you say "convert to 3D", that can mean a lot of things.

If you mean just start drawing your parts using 3d solids, that is one thing.

If you have certain processes automated in 2d, you will need similar automation in whatever 3d world you choose.

So identifying key workflows, and fleshing those out in another tool, before you drop your old workflows, should be done if you care about your job.

The software vendors love us to play it risky and for free.

Let me see it work, is my motto, and I have found I had to write my own tools in many situations to do thise major transitions.

Got C#?

[steve216586]

By using the word "converting", it is intended to mean our parts control/inventory, sales support, manufacturing, distribution and design work will soon be completely 3D.

"If you have certain processes automated in 2d, you will need similar automation in whatever 3d world you choose."

We may live in different drafting and design realms but...

I've found most of our 2D automation processes are obsolete when working in 3D. We don't need the lisp, script, diesel, palette support typically seen in AutoCAD 2D anything. Also it reduces the drafting time by 2/3rds. Especially when it comes to redesigns and revisions. It is too easy to produce complete orthogonal, detail, and section drawings in 3D compared to 2D. Hiding, emphasizing, calling out, BOM creation, design analysis, library management of parts and equipment are all some of the aspects of "converting" to 3D which makes the user/engineer more productive.

I should have included the other important aspect of 3D conversion. Library of parts. There are several really good sites out there to draw upon, like Traceparts. But as in AutoCAD, most of YOUR library is specific to your unique company's need and therefore are generally internally created and stored in your company's library. You will dedicate hours to the creation of those industry specific parts needed to complete your designs.

Mere growing pains in my opinion. Something we've all been through and will go through again.

[ampster402]

I'd tend to follow along these ideas if possible, keeping in mind you have legacy CAD data you will have to carry forward and continue using to some extent.

Keep AutoCAD around, whether it be LT or the full version.  More than likely you will find situations where you just don't need a full 3D model and Acad will do just fine.  Along this same thought, do not believe you'll be able to do all 2D work inside Inventor.  While it's possible to some extent, Acad will provide more tools to make editting 2D data easier than Inventor can.

Do not attempt to move up to 3D in AutoCAD, then decide to move to Inventor or any other solid modeling software.  If you haven't heard it yet, AutoCAD is not Inventor and vs versa.  While you may see some Acad features/tools/options in Inventor and some Inventor stuff in Acad (constraints for example), they are both completely different animals which you have already discovered.  If you spend time ramping up 3D in Acad, then in the future move to something completely different you'll regret spending the time you did doing 3D in Acad.

And like others have mention, do not skip the training.  Everybody should have professional training and not attempt to self-teach themselves.  Bad habits start there.

And the growing pains part, yeah alot of us have been there already.  I imagine after the current "ole timers" retire (50 yr old myself so I am included!), there will be yet another growing pain session when they move into virtual reality solid modeling - or what ever the next wizbang solution is for designing.

[JamesMaeding]

that training subject is tricky, as the people that train may not do production, so teach bad habits also.

I find that the most effective training is on the spot. This essentially means one person apprenticing under another.

That takes time to spread, as you only have so many mentors to start with.

The best production teams I have seen use several tools, not just one.

[JamesMaeding]

You say you cut drafting time by 2/3's?

I'd like to hear what kind of work you do.

I find that the 3d world actually leads to more automation, as there is more information to take advantage of.

It does depend on the subject matter though...

[steve216586]

My current company:

"HmmHmmm" is a worldwide leader in the design, manufacturing, and distribution of high purity and corrosion resistant equipment, systems and services. "HmmHmmm" products and services provide proven solutions for many industries, including pharmaceutical, biotechnology, microelectronics, metal finishing, and chemical processing.

Our Products: Pumps Process Instrumentation and Controls Plastic Valves, Fittings and Tubing Tanks and Mixers Pump Transfer Stations pH Neutralization Systems High Purity Water Generation Equipment Wet Process Benches Chemical Feed Systems Metal Finishing Components and Systems Rainwater recovery Systems UL Custom Control Panels Custom Plastic and Stainless Steel Fabrications Our Services: System Engineering Site Preparation Transportation & Rigging System Installation and Upgrades Controls Integration Start-up, Testing, Training Documentation Maintenance & Technical Support Instrumentation Calibration Pump Equipment Repair

I'm not sure what you mean by "more automation"?? Automation of commands, routines, blocks etc? I'm not even going to try to guess.

[JDMather]

I took some time to look at the questions (and a file example) you posted several years ago while struggling with Inventor.

You actually weren't doing too badly compared to most of the garbage I see. You were at least heading in the right direction in terms of modeling techniques.  It did appear from the example file though, that the absolute, positively most fundemental foundation technique in Inventor was somehow missed in your training.  As a result, it looked to me like you were doing wayyyy too much work.   In my estimation - only about 10% of users of any CAD software really know what they are doing.

Personally, if I were doing your type of work - I would run, not walk over to Inventor.  I would rather go to the dentist then do your type of work in AutoCAD 3D.  But then, I know Inventor, upside down and inside out, backwards and forwards.  I started experimenting with Inventor back around 2001 and somewhere about a year later pretty much abandoned AutoCAD as a 3D (or even 2D) tool except for 2D schematics.

In my experience those who do not naturally gravitate to newer tools (we are already a few years into the 21st century) tend to stick very strongly to what they already know.

I predict that your foray into 3D will be very difficult no matter what software you use.  (probably not so much for you as for the rest of your team)

[pendean]

Hire, part time, someone fully capable in the software package(s) you are interested in converting to (not tired old AutoCAD as others have noted, no future there) and have them prove its viability in production at your office. Validation requires time in production, not training and hoping it may happen.

Once validated, your company will then be ready to talk "converting": and I'll be the one that says it here, phase out those that resist change and promote those that embrace the change (if your company is into making money and is not a charity or work program).

[BeKirra]

There is a similar thread at AUGI forum.

http://forums.augi.com/showthread.php?156503-Who-got-to-choose-their-CAD-software

[rhoscadman]

Thank you all for your prompt and informative advice and opinions.

To be brutally honest, I like Autocad. I have used it for 16 years and have become quick and proficient at it. Because I'm such a lazy bugger I have created many "one-stop-shop" macros that, at a push of a button or click of an icon I can carry out commands which originally were tedious and time consuming. I have created tool palettes that allows me to drag and drop ALL our library of parts. So many short cuts are available in Autocad that, frankly, I don't see in any other package. I'm hoping we stick with it, but am not convinced that just the bog-standard Autocad will satisfy us in the months to come. I have dabbled a bit with Acad Plant 3D and like what is available. Especially in the piping and steelwork sectors.

As I said we generally just do outline designs so I am firmly convinced that Inventor is NOT the way to go. We do not design to that level of detail. It's just not required here. And also there isn't the level of flexibility/versatility. I really do not want to spend an hour arguing with the programme because it won't allow me to fit a square peg into a round hole.

Last year during a prolonged "down" period my design manager had me replicate our whole 2D parts library into 3D with a Autocad 2013 programme. After a few hours it became 2nd nature and bish-bosh I knocked them out in about 2+ weeks. He then asked me to do the same with Inventor. Oh man, within 4 hours I still hadn't completed a single simple GRP elbow. Try it this way; oh no you can't do it that way because of blah-de-blah-de-blah, try it another way same thing............I just had enough by the end of the day I had maybe created 3 parts and we decided that it really wasn't for us.

Anyway, I value your opinions, even though we might disagree, and will certainly keep them in mind when we thrash this issue out later on today.

[ampster402]

You can automate alot of things in Inventor.  Some automation can happen right in the part, assy or drawing files.  More complex automation can be done using iLogic.

An elbow should be rather easy and could have been done in a couple of mins.  You can make iFeatures in Inventor or create your own Content Center library for common parts.

Ref what someone else stated about saving 2/3'rds of their drafting time, back when I first started using Inventor I was given the task of pushing one of their product lines into Inventor.  The goal was to reduce the 10+ week lead time it took someone with AutoCAD to produce the complete drawing package.  Over time, using just the part, assy and drawing files and a rather long excel spreadsheet, we were able to reduce the lead time down to 1 week to 8 days or so to complete the entire drawing package used to manf that multi-million dollar product. 

ah those were the good yrs.  After completing that project in Inventor - I had a couple yrs worth of very good raises, something along the lines of over 2 to 3 times the normal raises given out back then.

I would not be so quick to believe Inventor is not the way forward.

[rhoscadman]

Just a quick update on what's happened in the past few weeks.

We have called in the Autodesk re-seller and thrashed it out with him for a couple of hours.

The end result was that, yes, he agreed with us that Inventor was too steep a learning curve for us dinosaurs, but also that full Autocad would not satisfy our need for efficiency in the nex year or so. So we are going to try AutoCAD 3D PLant. THe price is favourable and it has the right sized bag of tricks. We are seriously (at least management tells us) planning to move into this scenario in the next couple of months.

Should be interesting.

I will keep you posted with developments/news.

[PatrickHughes]

rhoscadman,

It sounds like you've got a good handle on the direction you intend to go with and Plant 3D might (almost) exactly fit your needs.

As to moving to 3D over 2D you probably realize there will still be a need for 2D output but much of that could be driven by 3D models. It also sounds as though you've got a good grasp of creating the libraries and tools you'll need (macros, pallettes, etc) do you also write any lsp routines and such?

Hopefully your "higher ups" are aware and accepting of the fact that for a while there will be a learning curve and a loss of some productivity for a period. Also hopfully you've got a team that is willing to move into the 3D realm without too much of a fight.

While I have no experience with leading a team into a transition, quite some time ago I made the move to 3D and I've written a few of blog posts that you might find useful. I called them Dr. Who and the CAD Manager. Dr. Who and the CAD Manager 

AutoCAD has made vast improvements in 3D work and I'm personally hopeful that it never dies as some people believe it eventually will. It is still a very capable tool.

[ampster402]

---

@PatrickHughes wrote:

AutoCAD has made vast improvements in 3D work and I'm personally hopeful that it never dies as some people believe it eventually will. It is still a very capable tool.

---

Remember DOS?

Where does that fit with todays computing world?  It's there but buried and when an unsuspecting user stumbles apon it now there are usually gasps heard and no one knows what to do with it.

It's here today (AutoCAD) but I don't have strong feelings it will always be maintained.

[PatrickHughes]

I delayed a move to Windows for a long time so I do indeed remember DOS and for the rare occasion when it might come in handy (hey it could happen) I keep a copy of my MS-DOS 5.0 book around for looking up commands. In fact, for some things DOS would still be more efficient, just not as convenient.

You may be right about AutoCAD but there is still a lot of 2D work being done and for some industries likely will for a long time.