Best way to translate Catia Files into SolidWorks? Use Inventor.
I caught some flak for saying that the best way to get a feature into SolidWorks was to ask Inventor for it (last paragraph). Well, it turns out there was more than a little truth to that. SolidWorks users have long been able to import files from SolidWorks competitors like Pro/E, Solid Edge and Inventor. But when it comes to importing data from within the Dassault cartel, well, that’s a different story. It’s nothing short of really disrespecting your customers, and another nail in the coffin of the “customer driven” moniker. The idea is that if you can read the format, you can convert from that package. Dassault is apparently scared that people may figure out that they could pay less for a modeler, and get SW instead of Catia.
If you ever hang out on the forums, the first question newbies ask is if they can read SW09 files in their pirated SW05. The second is what is the best backdoor to get Catia data into SW. The Catia => SW translation is one of the messiest around.
So now, Grabowski points us to the Autodesk site, where you can now get a Catia translator for SolidWorks – er – Inventor, I mean. That’s probably all it will take for SolidWorks to suddenly say that users have decided they want this, and finally add it in. What do you think?
Loading ...
Loading ...
Loading ...

Matt,
Had to smile as I read this post (shortly after reading RalphG’s).
The parallel or irony is that Inventor – or more importantly, Autodesk, have their own translation troubles that, in some ways, match the Cat-what_SW combination.
The missing component – in trying to solve these issues – is the USERS; thankfully blokes like you and others I make the appropriate noises but not enough users stand up for themselves or simply, as I have seen, simply swallow the crap their vendors/dealers dish out about the so called ‘problems’.
ie, I had one customer tell me, in great detail, how Inventor could not create an AutoCAD solid and why it was not Autodesk’s fault. He did all this whilst I took a SW file into Inventor, then into an AutoCAD solid -using MDT – that he then used to create the detail drawings he was being asked for; go figure, he still believed he was being told the truth after watching what I had just done for him!
I advocate customers/CAD software users should make upgrading (or switching) in the future dependant on improvements in compatibility.
In Solid Edge V20 you can also import Catia V6 and you don´t need any plugin!
I’m one who hopes we do get the v5 translator into SolidWorks. There is no technical reason at all why we cannot get a GEOMETRY translation. Here’s the thing. Spatial Technologies (creator of the ACIS modelling kernel and owned by Dassault) have for many years offered a CATIA v5 translator package that bolts onto a range of kernels. In fact they offer the CATIA v4 translator package to virtually all ACIS licensees for peanuts (which is why when dealing with CATIA I usually ask customers to backsave the v5 file into a v4 format and open it in Ashlar-Vellum Cobalt, then take it into SolidWorks from there…..as the CATIA STEP and IGES export is terrible in the hands of most CATIA users).
Now if Autodesk do pull this off and include it in all versions of Inventor what is an automotive or aerospace supplier going to do? Thing is all Tier 1 suppliers need to talk the native language so they will have invested in CATIA (same for Aerospace – Airbus suppliers for example can purchase CATIA licenses at vastly reduced prices). But there are a hell of a lot of Tier 2 and 3 suppliers supplying the suppliers who use SolidWorks, Autocad, Inventor, SolidEdge etc.
Many of them will also buy maybe 1 seat of CATIA v5 and deal with the file translation internally. Now if Autodesk do ship this with all products then SolidWorks have no option in my opinion, but to integrate a v5 translator with SolidWorks – all versions.
Then of course we are back to the that old “will SolidWorks change to the CATIA kernel” routine. Who knows?
Wow, this is interesting. The last place I worked, we spent days trying to locate and track down SOME type of translator for CATIA to be able to open native SolidWorks files. Our parent company was using catia, and us SW. Basically my boss didn’t want to hassle with us having to spend time converting our SW files to IGES/STEP/etc. and just have us send them our files. Seems we couldn’t ever get a straight answer from the Catia reps as to what was available and for how much.
@R. Paul Waddington
RPW,
CAD software consumers do need a Ralph Nader. Corporate arrogance is out of control, and gets worse as companies get bigger. They just expect to get away with it without being opposed.
@gale
Matt you talk about adding a nail to the coffin……
Big brother NX and SE can work in the same environment. Even if there is two separate programmers team they can talk to each other.
Configuration we can see, depending how the organization is setup;
At the top we have five NX stations, just under we have 15 Solid Edge seat (classic, Foundation and Design&drafting) all mix under the same roof. If you push further add free Solid 2D draft for red lining/viewer
@gale
Sorry I thought CATIA V5, not V6.
Matt, I don´t understand the excitement about the Catia V5/Inventor-converter from Autodesk. There are commercial products around for a long time which can translate Catia V5 directly into SolidWorks-files including Catia-features. Datakit is selling such products within their CrossCAD-product line. Another strong and well-priced translation solution is offered by SpaceClaim within the optional translator packages including viewing and editing capabilities in both directions. The problem with translator products is that many prospects think translators are for free ignoring the huge efforts for the developers to understand both worlds of CAD-systems. The real mess is the poor quality of a lot of CAD-data so that translators will fail to do the translation job. An acid test for translators is translating real-world CAD-data including numerous traps which require good understanding of the basic architectures of the different CAD-systems. A translator is only as good as it can handle these day-to-day field data. In many cases healing and repair functions for corrupt geometry are mandatory otherwise a translator is not worth the money.
Matt,
You are correct; the shape of CAD software has, for some time, been dictated by the marketing goals of CAD vendors more than customer requirements.
In an Autodesk presentation several years ago, it was stated (by)Autodesk’s CADD software was a consumable not an asset. Whilst, at the time, this was said in the context of a discussion to do with the cost of ownership, new product cost and up-grades it did show how Autodesk had moved its thinking away from the users’ (requirements) elevating the managements desire to ‘earn more money’ by what ever means, regardless of the effect that may have on its customer base.
‘Verticalization’, the lack of compatibility of products from the same company and the ability to make data transportable between competing systems was, and is, about dividing the CADD market into ‘artificial chunks’, forcing customer to buy – often underutilized – differing products and add-ons etc. simply to increase vendor revenue.
You mentioned arrogance: CAD vendors do not see themselves as the tool suppliers they are; as this definition requires they accept they MUST take their lead from their customers. Their preference is to – incorrectly – elevate themselves into a dictatorial market controlling role, the main aim being self-fulfillment.
This is why I say users have to be more assertive; demanding, from the vendors, products that more closely resemble the tools they deserve and, to stop buying, or upgrading, until they get what they want!
With the global, financial, climate the way it presently is, I venture to suggest most, if not ALL, existing users of CADD products could ‘park’ their software purchasing and upgrading plans.
In doing so, they would be improving their ‘bottom line’ and would be forcing CAD vendors back into their rightful place in the supply chain. It would also mean, to get sales, CAD vendors would need to develop new ‘features’ that ACTUALLY earned more money for their clients – not just themselves.
CADD users will pay for something that will genuinely earn them money or additional business.
[L=D6-51991.pdf]http://www.boeing.com/companyoffices/doingbiz/supplier/D6-51991.pdf[/L]
The first question one should ask themself is whether the IGES data from CATIA accurately reflects the internal CATIA. To this end some Boeing sites use a program (written by Boeing) which creates a second CATIA model from any input CATIA model. This program runs as a batch job on MVS and uses Dassault evaluators to grid numerous points on all the curves and surfaces in a model. We then IGES both CATIA models out. The assumption is that CATIGE can surely do points correctly. Using IGES_CHK, we can now determine that the CATIA IGES file of curves and surfaces lies at the same locus as all of the points in the CATIA IGES file of points. This proves the validity of the IGES curves and surfaces. The second question one should ask themself is whether a given CAD system hears the CATIA IGES well. To this end we ask our suppliers to import the CATIA IGES and to echo it back as a second IGES file. The second IGES file is compared back to the CATIA IGES file. If the check is favorable, we assume the CAD system translated the CATIA IGES well. These checks are to be executed by the supplier.
In any case, it would be prudent for anyone using IGES as a translation tool to run some sort of validation check to verify the integrity of the data coming from the sending CAD systsem.
The SW Import diagnostics tool should NOT be used on imported files with out customer approval because it would defeat the D6-51991 spec.
The STEP format has virtually replaced the IGES format but Boeing does not have a validate program for STEP files.
The STEP AP203ed2, 210, 214 224, 238 Application Protocol will accommodate GD&T data so hopefully this will be the new standard soon…CATIA will export the STEP 203 ed2 format.[L=ISO_10303]http://en.wikipedia.org/wiki/ISO_10303[/L]
[L=SC4 online]http://www.tc184-sc4.org/SC4_Open/SC4%20Legacy%20Products%20(2001-08)/STEP_(10303)/[/L]
I do appreciate a detailed description of export from Catia from a Catia user. I’ve never had a good experience with Catia IGES data. STEP is the format of choice. Even if you don’t have an automatic verifier, can you just reimport and compare the step file? There is no wonder in my mind why things take so long in a big company.
Kindly excuse if the question is very elementary. Am not a designer or a CAD engineer, but am interested lately and would like to understand if 2D Autocad files can be read in the Catia V5 drafting output directly without the aid of a translator. If this could be possible, would it be also possible to directly derive solid models of the same in Catia V5, just as 2D views of a Catia model can be had directly in the Catia drafting/2D output?
Thanks & regards
Vinod
This is another question and perhaps more related to the hardware than Catia V5 itself. Upon Catia V5 installation on a machine with the below configuration……on a DG33FB motherboard, Quad 2 core 2.4 Ghz processor with a 4GB ram and has an Nvidia quadro fx580 graphics card on which I am trying Catia V5 installation. It has been successfully installed on another machine of the same configuration, but on this one it appears to give me an error message stating ‘Invalid data structure’. Couls someone please advise firstly on the suitability of the configuration of the machine for high end CAD applications….such as say die design where file volumes could reach anywhere between 400 Mb to 1200 Mb and secondly what is the ‘invalid data structure’ error due to? What could be the easiest remedy to it?
Thanks & regards
Vinod
@Vinod
I don’t have any answers to help Catia users, and Catia users don’t seem to frequent this blog in large numbers. I know a lot of people come here looking for Catia V6 stuff, but I really don’t have anything other than the research I’ve done from other sites. Good luck in your search for more info.
If CAD data is poor quality, who is to blame? The user or the translation software? If the software, is it the input or output? I think trying to shift the blame around is a dirty old obfuscation game that software vendors are fond of playing, and I think it makes things less rather than more clear.
Certainly users should always round trip data to make sure they can at least read what they write. CAD vendors are always trying to make this issue as murky as possible because the sky would fall if users were suddenly able to do what they wanted with their data. It is the customer’s data, after all, not the vendor’s data.
People see that SolidWorks and Catia are owned by the same company, and assume that they should be able to talk. But they don’t. The fact that they don’t is due to an active business decision that someone somewhere made, not a technical problem. Same as with the backward compatibility issue. I don’t see what is so difficult to understand about that. If you buy software, why should you need to buy additional software to do part of the main mission? Why does SW read Pro/E and not Catia? If you want to say that it would cost SW money to program a Catia translator, I’m cool with that. They could drop any one of about 30% of a given year’s enhancements which are pure fluff and write the translator. I’d be willing to bet it’s already written, just not shipped or enabled.