Portable Catia V5 R21
Roman Bayramdurdiyev
i would like the opinion of some Catia V5 surfacing guru. I just had a conversation with a master of class A surfacing on Catia V5 and his assessment was that Rhino to Catia in STP ( forget IGES it s crap we all know that ) the rhino surfaces are not that great. Work needs to be redone in some instances.
From my experience I didn t find it horrible. I was pretty satisfied ( maybe I m not as perfectionist as others or I m completely missing something) But are there any Catia V5 surfacing guru who sees that working with Rhino surfaces is not ideal ? and why exactly ?
Yeah i Guess you guys are right. I also had in the past had to clean up catia surfaces before sending it to a cnc. Too many gaps. I guess i m a bit paranoid. But one thing i noticed is that exported rhino surfaces are broken in multiple surfaces in Catia, whereas Alias surfaces come out really nice in Catia. Yet they re all Nurbs surfaces.
Agreed, i d like to look at the surfaces. I m surprised to hear that some experts in Catia say they need to rework rhino surfaces but are happy with Alias surfaces. They say rhino surfaces are broken down into much smaller ones in Catia.
We used Icem previously moving away from it due to the yearly maintenance costs, instead we purchased the Rhino plugins called VSR before they were purchased by Autodesk. We are also users of Catia, Unigraphics, Ideas and Creo.
The good point on Rhino is the direct and easy modification of a NURBS surface like in Blender and the easy transformation of a curve or surface from one degree to an other.
The good point on Catia was/is the easy analyse of the curveture by a section through the surface. But in those time Catia had problem generate or keep a closed a highlevel non-uniform surface pipe with a kink or without a kink. Every time it broke up the pipe by an Iso-line.
To add even more confusion I found that if I used code to generate my CATIA surfaces they sometimes seemed to be of higher quality with fewer IGES export issues that the same surface made the traditional way.
I remember wHen i imported Catia V 4 to rhino v3 and then rhino v4 it was a nightmare. I would import surfaces broken down into tiny tiny ones and hundreds to clean up as well. Catia v5 improved because it was working on windows platform wHereas catia V4 was Unix based. Airbus had huge problems with the a380 design because the french were using v5 and germans were on v4. I wonder how Alias does it. Never heard people complain when importing in CatiaV 5.
When browsing the mail storage I see filenames containing ":" which is invalid for Windows. Dovecot uses it for adding some sort of status info to the filenames in the maildirs. I want to leave it that way. But Windows cannot read those files.
I've tried mangled names = Yes but that does only clobber up the filenames till unusuability. It converts all filenames with illegal characters into 8.3 format - we cannot use this for recovery. And I don't think it would have an effect on VFS settings anyway.
So why does samba not use the VFS module? I can see no errors about it in the logs. And "locate" reveals the respective library vfs_catia.so installed. The samba version is supposed to support this configuration according to docs:
Now you want to apply the downloaded patch to the file: source3/modules/vfs_catia.c in the samba-3.6.3 folder. I couldn't work out how to apply the diff using 'patch' so applied it manually - it's only 3 lines. Rebuild the samba package, in the samba-3.6.3/ folder:
I knew that it is able to connect to CATIA from VB or even JAVA but I cannot find the "CATIA.Application" object in the ActiveX list in LabVIEW. For instance: -there-a-way-to-exchange-data-with-catia-v5-r20-using-j...
Unfortunately, while COM is the basic protocol ActiveX is built upon it is not enough to make a COM interface ActiveX compliant. As such accessing a COM interface from within LabVIEW is not possible directly. The only way to do that is by creating an external component in C, C++, VB, (even C# although that will be quite painful since it means interfacing to unmanaged code in terms of C# programming) and then integrating the resulting library as DLL or ActiveX control into LabVIEW.
LabVIEW absolutely needs the ActiveX additions to COM in order to be able to determine the exported classes, methods and properties of a COM object. COM itself does not have a clean way of enumerating those things for a visually integrated environment like LabVIEW. If NI would have made access to raw COM objects possible, you would have to configure a configuration dialog for every COM object, method and property, where writing the according C code to access that object would be in fact at least as simple or even easier.
With ActiveX containers you might be able to access CATIA, but it would be something you need to develop on your own. We have the capability to work with SOLIDWORKS from our SoftMotion Module in LabVIEW to control a CAD model's movement, but nothing with CATIA at this point.
I think, as well, you might be able to output the files from CATIA into a more standard format (e.g. STEP files?) which you can then import into a 3D Picture Control. It's been a few years since I was playing around with it, but I think it's possible. I think you would need to manage the linkages between the parts in the assembly yourself.
We are currently using solidworks as our primary design software and mastercam as our primary CAM software. We are looking to buy a seat of CATIA because we are doing a lot of development work with a company that uses CATIA to design, and we wish to maintain Solid histories/feature tree. I have aquired a sample version of CATIA and my initial thoughts are that it looks pretty complicated! I worked with Unigraphics NX for a couple of years and I guess it seems more like it than solidworks. Besides the pricing, can anyone give me accounts of there experiences and how CATIA compares as a CAD package, but also as a CAM package? We will definitely be keeping solidworks as our primary CAD software for now and will keep mastercam as our CAM software too. Any insight would be greatly appreciated!
Thanks for the responses! Have any of you had any experience with the CATIA training courses? I'm not sure of the specifics but I'm sure we will invest in at least the fundamental training courses. Do you think this will be a sufficient for someone with a CAD/CAM background to get a useable handle on how to work the software? Obviously they won't come off the training courses as an expert, but just as far as basic solid and surface manipulation, are the training courses good for learning these aspects? Keeping in mind that we will still have solidworks there as our primary package
I come from an Autodesk Inventor background and took catia classes and had no problem drawing it in. Now I only designed in it and never actually did detail drawings. So the detail part of it I dont know but the actual 3d drawing was no problem after the training courses.
All in all, they both have their strengths and weaknesses. I prefer CATIA now primarily due to the the things I mentioned, but either package is capable on the CAM side. If it were up to me the deciding factor would be the type of work I did. If it is aerospace where models are primarily in CATIA from creation, and it is large multi-axis parts then CATIA I would choose. If it were not aircraft and/or primarily 3 and 4 axis work on smaller-mid sized parts I would choose mastercam.
I agree with most of what Merritt (MLS) said. Catia will do just about everything & is very expensive. We train it here and our instructors consistantly win a yearly competition for Catia drivers. I've taken four of the courses we offer and didn't have too much trouble learning it. I took the machining module in R11 and it was very cumbersome but I've heard there have been many changes for the better in the machining module now at R21. Since we train Catia we get it very cheap but we have never cut a chip with it because we get no break on post price. We have a Mazak Integrex in the engineering Dept. and I heard a post for that cost $25,000.
Thanks everyone for your thoughts on the matter. I should've mentioned that we are working aerospace and large assemblies/complex surfaces in some cases. I guess the picture that is getting painted for me is that CATIA is great at high end surfacing/machining, but for simpler work mastercam handles better. I really do like mastercam for the work that we are currently involved with. I will admit that I have not had proper trainng in multi-axis or 3D surfacing toolpaths in mastercam, but I was quite skilled with them in unigraphics and from what i have witnessed it is not mastercams strength. Thanks again. Brendan
I'd completely agree with gcode's asessment. Complicated as hell on the Mfg. Side, but dayum it's powerful. You can enter and exit a part/cut 1,000 different ways (only slight exaggeration). ON the CAD side I'm a mid-range user. It's probably way overkill for what I do but I sure love it. When I need to design workholding, it makes assembly modeling a breeze.
If I were going to make a recommendation, I'd have to look at what exactly you want to do. From a design standpoint, they are both very powerful. CATIA is a little stronger on the surface modeling side than NX, and NX is a little stronger on the solid modeling side. If you're going to do a fair amount of work with organic shapes, I'd lean towards CATIA. If I were leaning more towards the prismatic shaped things then NX cets the nod. If you need PLM... NX edges out CATIA. If you want simulation, NX edges out CATIA, if you need FEA, CATIA edged out NX slightly IMHO. CATIA... Dassault is VERY proud of their software (as they should be) and NX is too, but they are more aggressive at pricing. If they know they are up against CATIA for a sale, they will discount. NX... they like to hear from their customers. CATIA not so much, their motto os "STFU and pay up byotch!". CATIA has a VERY active user's forum. Not quite as active as eMastercam but it is definitely a great resource. I'm not sure about NX user forums. I'd imagine they have something.
b37509886e