Changing CADD Software (Solid Edge to Creo) 1

[JBirner]

I started this discussion because i could not find much on the actual transition period of changing software.
there are several on choosing software or going from 2d to 3d, but not much on implementing a change between 3d systems.
Any Comments or helpful links would be greatfully appreciated.

First my Qualifications to help people understand my skill level.(Im not a complete newbe)
7 yrs Pro/e Extremely Skiled in family Tables (up to about 8 levels deep)
5 yrs Solidworks
VB and VBA.

I work for a company that is Primarily Solid Edge ST3, they have been a S.E. Customer for several years (8+)
They were in the process of changing when the guy (One of my Prior Supervisors) in charge of switching left the company.
He reccomended me as his replacement, as i had been trying to get on with this company for quite some time.

My question is for help, I am trying to put together a roadmap of what need to be done in order to have this change done in the next 2-5 months. I have a basic plan but I know there are pitfalls that others have encounterd that I would like to plan for ##IF## at all possible.

We do have a week of onsite training for our 5 full time users, when we are ready.



Currently I things that are done include:
(no PDM at this time unfortunatly)

Start parts, assemblies and drawings templates made.
Standard Config.Pro Created, containing all search paths
Hardware library updated.
Part revision standard ;process (yeah I'm changing this too, due to bad practies in the past)


I plan on creating assemblies in Creo that are most often used and exporting them as .Iges Files.
(that is how we handle purchased components)
then removing those components and asemblies from the solid edge directories forcing people to use/replace them
with the .iges models. Any New Assemblies/parts will be made in Creo and exported out for use in solid edge.
At such time comes that it makes sense for an entire assembly out of Creo then it will be done there.

The transition time will be approx a year if I had to guess, our products have 700+ piece parts in them and we sell
a variety of machines. They use some simlar parts and assemblies but each style of machine is built different to achieve higher speeds due to more Servo motors or more capacity.

Please feel free to comment and pick apart any portion of this or leave any tips or pitfall you may have. Any links, blog articles, charts are welcome as well.


J.B.

 

[KENAT]

My commiserations. At a previous employer we were meant to transition from Solid Edge to Pro E (1999-2001 time frame), the guy whose bright idea it was left and I got passed the task - less experience at the time than you have though.

There were many and varied issues and it never really worked out. Once simple redraw job ended up costing 3X budget because we'd estimated originally based on using SE that we were all familiar with, but we had to do it on Pro E that we all just had the 8 day or so training on - then to top it off we got contractors in to do much of it who actually had less experience than us in one case and in the other case must have just lied on his resume or something.

By the time I left Pro E had been all but dumped and most work was back on SE. Of course back then Pro E's usability sucked, maybe it's better now.

The impact on productivity - though it should be temporary - is a big deal.

There are a few threads that almost touch on this kind of thing, my current employer occasionally threatens to change from Solid Edge to Solid works and I'm pretty sure I've brought up the topic.

Honestly, there needs to be a pretty convincing argument for the change to outweigh the loss of all the sunk cost in those legacy files. In my case the argument that 'SW if more popular than SE' just doesn't seem enough of a justification.

Posting guidelines faq731-376

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

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

 

[JBirner]

From ""My"" experience, of Pro/E, SolidWorks, and S.E. iI Prefer S.E. the least.

I do however work with people that have the worst modeling practices in the world. (Gold Meadalists)
Very intelegent people, but most of them never had anyone show them bad practices. ( undefined sketched, no dimensions just ext.refs to parts with no Dims.)
I see this as an exciting challenge. The Mech. design stuff gets boring in my opinion. a part is a part is a part.


That may influence my opinion, But I think SW is a better part modeler with a cleaner interface. IMy Favorite is Pro/e, after you understand the POWER
of family tables and how to use them to increase productivity. Yeah S.E has similar capabilities, with tables of parts, but not nearly as intuitive and quick to make and modify.

With what I can forsee, I can easily take 40 man hrs out of engineering time on 50% of the projects in just one area.
Also Greatly speed up our sevice departmens orders.. Not counting the man hours recreating of standard parts and assemblies that
Pro/e family tables due by references and and a few mouse clicks. ( it lends itself to be more like a parametric library).

With that said. before i was hired on. They had stopped S.E. Maintainence and had bought the Creo Seats and a single licence to start implementing the administration type stuff.

 

[KENAT]

JBirner,

I wasn't coming from a point of view of which CAD system was/is better. I've always tried to stay CAD Agnostic, though I'm probably not having used SE the most. If we didn't have a 3D system and were looking to get one SE probably wouldn't be at the top of my list.

Just trying to say that - at least based on the work flow at the 2 places I've used it - that changing CAD systems was or would be daunting. The idea of not being able to just pick up existing designs and simply tweak them without having to completely or partially recreate the model & drawing is a big deal.

Posting guidelines faq731-376

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

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

 

[JohnRBaker]

Note that if you're running Solid Edge ST3 you ARE a couple of releases behind. Note that I'm not a Siemens SE person (my product is Siemens NX), but I would hope that your organization gives a look at the current capabilities of the latest release of Solid Edge.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[KENAT]

JBirner, they may be a bit too SE biased for you but these thread hits on a few issues. thread562-289546 thread562-238351

Posting guidelines faq731-376

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

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

 

[JBirner]

Yeah ,I dont care about the bias of the software (as long as it is cadd software) that they are talking about. Yes the Data translation seems to be a bit of an issue for people.
That is why as a Preemtive measure we start modelng the current and mostly reused models to have ready when we switch over.

In my opinion that is why you keep an the existing versions of the old software. At some point the new software will become the norm and most parts will be re/created out of ease.
Then for legacy purposes when most work is done with the NEW SOFTWARE, then IF needed the older models can be opened and used/exported.
If the older software seems to have issues and need dumped altogether, then the models and bulk of the legacy data can be exported as a dummy type (iges/parasolid) and a PDF of the Drawing and a line drawn in the sand.. IF changes are needed THEN someone can spend the time rexporting/modeling/translating to make it current .

Alot of our parts can be recreated and drafted quickly and easily, but just switching immediatley would be time consuming, which is why i suggested Exporting entire assemblies from the New System as IGES and move the Old System version to the obsolete folder. This allows use of the Old system until it makes sense to use the new system entirely.
Slowly forcing the changeover. Basicly a forced learning curve the people dragging their feet.

Yes we are on a bit older version od S.E. but overall I am not impressed with it. It has a nice interface, but I personally dont like it, maybe my brain is Just Pro/e and SW wired.
and like i said this place THRIVES off of bad modeling practices.


Has anyone actualy Gone throught this implementation before? (3d software to 3d software)
Are there any better or other methodsOther than My idea, or a HARD line in the sand?
(My Idea that the exporting from NEW Software and obsoleting Old Software models and force people to insert the iges in the old software, durring transition period)

Should we start a new Numbering shceme to differentiate the change or skip the next 1000 numbers? (good idea of bad idea)

Should 1/2 the team beregulated to ONLY the new Software to start implementation?

 

[MikeHalloran]

The last place I worked was a mostly SW house.
Many of their parts could be recreated and redrafted quickly and easily.
... but they had at least 36,000 parts.
Every time SW releases a major upgrade, all of those parts have to be converted. It's a simple matter according to SW; just launch their magic robot, let it do its thing, and assign someone to handle the (thousands of) exceptions that the robot reports. There is of course no such person because there is no budget to pay for cleaning up after robots.

In principle, because CAD models are stored as numbers, it should be a simple matter to convert from one modeler to another; just release a robot to go through your parts folders and do the conversions. In practice, it's mostly not possible, because robots' performance isn't all that good going from revision to revision with one product, and going from one product to another is impossible because all of the CAD suppliers actively work to prevent any sort of conversion program from being effective in converting from their format to any other.

Because of the enormity and difficulty of the conversion problem, I become suspicious of the motives of anyone who champions a change from any one program to any other program.


Well, okay, that's not the only reason.
At another place, 'we' transitioned from AutoCAD to 'brandX', after a long and exhaustive selection period that included extended demonstrations from all the potential suppliers. No one I queried on the selection team could understand how brandX won the competition.
Some time later, brandX went bankrupt. ... and within weeks of that, both the drafting supervisor and his boss retired, without notice or warning, very quietly.

Oh. After the bankruptcy, brandX couldn't support us even to the miserable level they had achieved before, so we transitioned to Pro/E.

So, yeah, I have been through a 3D to 3D transition, twice, and I think you are grossly underestimating the cost and the difficulty of doing so.



Mike Halloran
Pembroke Pines, FL, USA

 

[KENAT]

Effectively running 2 CAD software in parallel for an extended period has some cost and other implications itself.

You could eventually hit the issue that the old version of SE is no longer fully operable on the current windows OS.

I asked reps from SW about doing the switch when it came up at my place and they couldn't give me a a good methodology etc. for doing it.

Posting guidelines faq731-376

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

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

 

[JBirner]

Mike That sounds all to familiar of a practice. I do intend that there will be a lot of rework but as the powers that be already decided on changing to my software of choice (pro/e) im developing the best plan i can., No Magic robot, just rework as needed and use S.E. for the bulk of the parts.And have approx a year of transition.

I was told , true or not) when the went 3d here they likes S.E. and went to SW presentation, they drove approx. 100 miles to go see it
and it ended up being at some guys house. when they got their he answerd the door still wearing his pajamas.
Needless to say we have been on S.E. ever since.


And as for my current users, I dont think There is nanyone here that could be considered a "Super User" after my 3 months of S.E. and everyone else has 4+ years, They still come to me for questions and answers.

 

[GregLocock]

I've seen 2d to 3d and then 3d to 3d, all handled inelegantly but without missing any more gateways than usual.

The key seemed to be to start new projects in the new software, and leave legacy parts on the old software, until there is enough experience to translate/recreate the files in the new software.

I'm sure that Pareto has something to say on the matter as well.

Since in my opinion a CAD program is just a CAD program the idea of changing from one competing system to another for any other reason than cold hard cash seems dubious to me, and the cash side should be fairly easy to work out.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

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

 

[JohnRBaker]

Every time SW releases a major upgrade, all of those parts have to be converted. It's a simple matter according to SW; just launch their magic robot, let it do its thing, and assign someone to handle the (thousands of) exceptions that the robot reports.

When someone asks me about how NX handles this sort of issue I simply grab a file that I keep on my system just for when that question comes up. You see this file was last saved in 1994 using UG V9.1. We're currently delivering NX 8.5 which if we were still using the old UG number scheme would be 'UG V28.0', and I open this directly in the latest release without running any sort of 'conversion', it simply opens, as if I had created it yesterday.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[GregLocock]

Fair enough, but a company that didn't even manage to maintain /internal/ version compatability will be subject to endless scorn from its users, eg Mathcad.

The thing is, if we move from IDEAS to Catia (to pick a not very random example), can we (a) translate parts or (b) translate parts and retain history trees etc, or (c) none of the above, ie IGES at best.








Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

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

 

[ewh]

Do you have any other formats to choose from other that iges? Using many custoner translations in the past, I have found iges to be the least user friendly and (IMO) prefer stp or parasolid instead of iges by far.

“Know the rules well, so you can break them effectively.”
-Dalai Lama XIV

 

[JohnRBaker]

Note that you can migrate Ideas files to NX WITHOUT losing Features, Assembly Structure or Drawings.

John R. Baker, P.E.
Product 'Evangelist'
Product Engineering Software
Siemens PLM Software Inc.
Industry Sector
Cypress, CA
Siemens PLM:
UG/NX Museum:

To an Engineer, the glass is twice as big as it needs to be.

 

[KENAT]

Greg, when we tried to do it getting good ROI (or lack there off) numbers was actually quite difficult. There are a lot of factors where you need to try and estimate the impact etc.

I looked up the presentation a colleague & I threw together with a mind to sharing at least some of it but where we'd customized it to SE V SW and condensed it a lot of the issues weren't really given in any detail.

Posting guidelines faq731-376

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

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

 

[cowski]

I worked on a project to implement a PDM system for CAD files for one company and I did some work for another company that was switching CAD and PDM software (NX + Teamcenter to Pro/E + Windchill).

Since you are not currently using any PDM system, I'd suggest starting with a good hard look at your current file system; change it as necessary. Run a program that finds duplicate files (there are many good ones available, some are even free) and weed out the ones you don't need. If your current CAD has any utilities that look for broken links in your assembly files, use them to make the assembly files "current" (no links to files that have been moved or deleted).

The company that was transitioning CAD programs had 2 databases, the CAD database and the manufacturing database. It was understood that the manufacturing database drove what really got built and the CAD database would be revised if/when discrepancies were found. In the new system, they wanted the PDM system to drive both manufacturing and CAD. We spent a lot of time correcting CAD assemblies to match the manufacturing database before the conversion. How far you take this will depend on the needs of your company.

The transition involved multiple projects, dozens of CAD users, and 1000's of files. It proceeded in phases. A project would be frozen while it's CAD data was migrated. Users would go to training, help out other projects, or help the migration project during this time. The bulk of the parts were "just parts", i.e. we only needed the CAD data to drive tool paths and drawings; these were converted to a neutral format (STEP files) and imported to the new system. Some of the parts were "design parts", i.e. they used spreadsheets, expressions, and sketches to capture design knowledge and were used to help design new parts; these were carefully rebuilt in the new system.

Even with careful planning and the input of experts from both CAD systems, there were a few situations where there was no equivalent functionality in the new CAD/PDM system for an existing workflow; workarounds had to be devised as we went.

www.nxjournaling.com

 

[JBirner]

cowski, as you sugessted I have been looking at our cad file system. I plan on changing the way we revise our models and KILL Several bad drafting practices durring the switch. currently we have NO PDM, (Wish i could change that as well) and we use the revision name in the file name.
With that method as anything gets revised, old models still use the old rev. so you have to verify you parts are right and replace parts with every assembly.
There are no tools to search for broken links shor tof opening the assemby,then it only works ((IF)) the person who rev'd the old part moved the obsolete file to our obsolete file location.

Most of our parts are simple, its the drawings that take the time with the tolerancing required.

I plan on using just the plain part number for the current rev and appending the revision on obsolete revs, the Drawing links should remain just fine.

Also modifying sketches and features is more easily found in the new software ( in my opinion/experience), currently our models have a lot of synchonous pars and dimensions and Nobody still here models that way and correcting them is a pain. The lack of references, dimensions and underdefined sketches runs rampid and is frightening to say the least with our product tolerances.

I will be forced to maintain the same nubering system, but i hope to have a newer log, to maintain what system it was created in. We will not have the luxory to be able to freeze projects which is why i plan to export iges/parasolids to use in the interm. while replacements are created in the new software.

 

[KENAT]

"There are no tools to search for broken links short of opening the assembly" -off topic but if nothing else revision manager can possibly help with this. Also when we looked at moving our SE data into a pdm system they had some utility to search for duplicate file names and possibly other things - it was a few years ago and I forget details.

Posting guidelines faq731-376

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

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