FIRST Inventor lib?

[Tottanka]

Hey, i just started learning Inventor, and i was wondering where could i find readt First-kit-of-Parts cads.
I need CIM's, Globe motors, Battaries, RC, CMUcam2 etc...
Is there any website where such cads are hosted?
Thanks, and sorry if there is already such a thread.
(BTW, if thee isn't such a host, I'm willing to make one on our team's website).

[Taylor]

http://www.firstcadlibrary.com/

[Tottanka]

http://www.firstcadlibrary.com/

Thanks!
Is there anywy i can upload there my own files?

[Billfred]

Thanks!
Is there anywy i can upload there my own files?
I imagine that if you've got some KOP or other common robot part that Ed Sparks (same name here on CD) doesn't have, he'd be up for uploading it. As for your own creations (such as robots or gearbox drawings), you can always post the screenshot on CD and share the actual files through other means.

edit: read Chris' post below.

[ChrisH]

I imagine that if you've got some KOP or other common robot part that Ed Sparks (same name here on CD) doesn't have, he'd be up for uploading it.


Actually as I understand it, Ed does not generaly upload parts from others. He wants the parts in the library to be RIGHT. So he only uploads parts he has measured and modeled himself.

This might sound like a pain, but I can think of one year we used an "off-brand" model of a Fisher-Price gearbox to design our robot. It turned out that the models were slightly off from the real parts. We figured this out because we couldn't install the gearboxes without cutting apart and re-welding the frame. The frame members we were using for mounts were too close together, even though the model showed clearance. We measured and found the frame was built according to plan but the FP models were off.

Fixing it by reworking the frame would have taken time we didn't want to spend and would have required re-designing other subsystems. The "easy" way out was to change the manipulator design concept and build a custom 1000:1 gearbox, in two weeks. We took the easy way

I appreciated Ed's approach a lot more after that.

[AdamHeard]

Actually as I understand it, Ed does not generaly upload parts from others. He wants the parts in the library to be RIGHT. So he only uploads parts he has measured and modeled himself.

This might sound like a pain, but I can think of one year we used an "off-brand" model of a Fisher-Price gearbox to design our robot. It turned out that the models were slightly off from the real parts. We figured this out because we couldn't install the gearboxes without cutting apart and re-welding the frame. The frame members we were using for mounts were too close together, even though the model showed clearance. We measured and found the frame was built according to plan but the FP models were off.

Fixing it by reworking the frame would have taken time we didn't want to spend and would have required re-designing other subsystems. The "easy" way out was to change the manipulator design concept and build a custom 1000:1 gearbox, in two weeks. We took the easy way

I appreciated Ed's approach a lot more after that.


Indeed.... we used some bad models this year, but luckily we caught the error before production started when someone found a .pdf with different dimensions.

Also, be carful with the motors.... Some can be slightly different year to year and not exactly match the cadlibrary website. I know this happened this year when I compared the model to what I was calipering; I just don't remember what motor.

Either way, FirstCADLibrary is awesome. Don't forget though, IFI has CAD files of their wheels (and sprockets... which can be useful even if you don't buy theirs) and AndyMark has most of their products' files.

[JesseK]

On the same note you can trust the parts from firstcadlibrary very very well. Trying to CAD a custom gearbox the very first day you learn CAD is a terrible idea though I was somehow able to accomplish it with some fairly abstract thinking. Luckily I had the motors and motor mounts to guide me to ensure the holes lined up perfectly.

[Tottanka]

Is there ayway to work with those parts in Solidworks?
If not, is there any Soliwirks available first library?

[AdamHeard]

Is there ayway to work with those parts in Solidworks?
If not, is there any Soliwirks available first library?

Download the step files of the parts. Those will open in any program.

[gblake]

Download the step files of the parts. Those will open in any program. True - But some might reword that to be "Those will work equally poorly in any program" :-)

[Tottanka]

Download the step files of the parts. Those will open in any program.

Thanks!

[AdamHeard]

True - But some might reword that to be "Those will work equally poorly in any program" :-)

How so? I've never had a problem with step files... Even exporting whole assemblies of robots.

[gblake]

How so? I've never had a problem with step files... Even exporting whole assemblies of robots.I am a novice at this; but what I have read/heard is that STEP files basically define the shape of an object by modeling it's skin as zillions of small triangles, and they don't convey important information about things like the radius of a circular arc, or the density of a material being modeled, or that a couple of concentric circles/holes are supposed to stay concentric (as opposed to just happing to be concentric because of random luck), etc.

If I have developed an accurate sense of how STEP files can be used, the effects of using them instead a tool's native format, or some other more compact and sophisticated format, should be fairly predictable.
Portability = A good thing
Increased CPU & RAM consumption = A potentially bad thing
An inability to have the "meta-info" like face and vertex definitions, or concentricity added to them by the using program = A thing that limits their usefulness in many situations.
I might have the details of this a little off, but I think I have the general picture correct. If I don't, I'm all ears. :)

Blake
PS: See http://www.chiefdelphi.com/forums/showpost.php?p=507920&postcount=5 also.

[ChrisH]

I am a novice at this; but what I have read/heard is that STEP files basically define the shape of an object by modeling it's skin as zillions of small triangles, and they don't convey important information about things like the radius of a circular arc, or the density of a material being modeled, or that a couple of concentric circles/holes are supposed to stay concentric (as opposed to just happing to be concentric because of random luck), etc.


What you are describing sounds more like an .stl file than a STEP file. STLs are used by additive fabrication (sometimes called Rapid Prototyping) machines to generate build files. STLs are a standard format used by all RP machines I am aware of including stereolithography, selective laser sintering, fused deposition modeling, e-beam melting, fab@home (a DIY RP machine for approx $3K!), and a couple of others that are rare and difficult to find. They can also be used to generate a conventional CNC tape if you have the right software.

STEP on the other hand does contain true geometry information. STEP is the 3D version of IGES, a 2D standard for graphics data transfer. IGES has been expanded in later versions to handle 3D data. Today either can be used for transfering complex surface information from one CAD system to another. Neither is entirely bullet proof and when I have to transfer parts from one system to another I will typically request both a STEP and an IGES version. I'll typically start working with the STEP file because it is a true solid model but if it doesn't translate properly I can always go back to the IGES and work from the surfaces there. IGES always works, within its inherent limitations, but is generally more work for me in overcoming thse limitations.

Both STEP and IGES will loose some of the additional data like density, drawing notes, colors, etc. This has more to do with the lack of standards in how such information is treated by the CAD systems than technical capability.

[AdamHeard]

Both STEP and IGES will loose some of the additional data like density, drawing notes, colors, etc. This has more to do with the lack of standards in how such information is treated by the CAD systems than technical capability.

Yeah, that is the only issue I have ever had with .step files.... And I accept it because of the nature of the different programs. If I have to send a step file in for production, I'll send a drawing with it to show threads, tolerances, etc...

[JasJ002]

The inventor library is a good source for a number of generic items found on just about every robot. But to the judges and fellow students it shows a lot when you're will to take those generic images and add more detail, essential taking the object to the next step.

[JD Mather]

read/heard is that STEP files basically define the shape of an object by modeling it's skin as zillions of small triangles,


As noted you are probably confusing STEP with stl or 3ds.
Step files are neutral files that should accurately translate geometry between any CAD program. There is no feature history tree however.
The feature history tree can be rebuilt in Inventor using the free Feature Recognition add-in from http://labs.autodesk.com SolidWorks also has a feature recognition add-in for STEP files

Autodesk Inventor Certified Expert
Certified SolidWorks Professional

[gblake]

As noted you are probably confusing STEP with stl or 3ds.
Step files are neutral files that should accurately translate geometry between any CAD program. There is no feature history tree however.

OK - After one of these I was itrritated, now I'm irritated enough to write a response - Either you guys need to read more carefully or I need to write more clearly....

No. I do not mean stl or 3ds files.

I do mean STEP files.

The feature trees are exactly the sort of meta-information that I was saying missing from the STEP files. The visible and hidden information in a feature tree or similar set of data is crucial to the efficient operation of most CAD programs.

I pointed out the that the STEP files does have the right shape, but without the feature tree meta-information, they are no fun to edit, and they seem to be much more complicated for my SolidWorks program to manipulate (chewing up more CPU and RAM). I asserted/speculated that this is because instead of storing and internally manipulating a relatively small and relatively easy to analyze meta-information description of the object(s) in question, the program has to deal with a big ole verbosely-defined blob of shape, and that forces the software to do many more operations and to do many operations "the hard way".

Again, I might have the details just a bit muddled, but the symptoms are real; and I do mean STEP files.

In the sense I outlined in my post below, and again here, STEP files can be said to work equally poorly in most CAD tools. This is a side effect of their portability and leads me to remember the old "free lunch" adage.

I didn't say that they fail to work or are inaccurate. I said that in other important senses of the word "work", they don't work as well as the tools' more specialized formats.

Blake