Angles & Parametric parameters

[jsteinhauer]

Hello All,

I am trying to build an "Extractor Arm" Family. These are used in labs, and soldering stations. I am having trouble with angles as parameters. What I would like to do is have an instance parameter where I could plug in the length of each pipe segment and the angle between the pipes. I am not currently able to achieve this. Has anyone had success with this? Am I approaching this challenging family incorrectly?

This would be a ceiling mounted family, where each instance could be a little different. I found a family online, but it doesn't have the functionality I'm looking for. I've attached an image file of the equipment, a screen capture of what I would like the parameters to be & the existing .rfa I found online.

Any help you could offer me would be greatly appreciated. If you need further refinement of my questions, please let me know.

Thanks,
Jeff S.

[phyllisr]

For me, creating reference lines first, creating dimensional constraints to the end points of the reference lines and then creating angle parameters between the references planes and lines is the most successful method. I flex everything every step of the way to insure it works before I create any geometry.

Check the link to some shelf and pole families I recently shared. These are less complex than your family but it may help you understand my approach. This is likely not the only way - just works for me. Hope this helps. Good luck!

http://forums.augi.com/showthread.php?t=72869

[guitarchitect7]

You may even consider looking at door families with parameters for the angle of the door swing. This is close to what your trying to do only in another application.

[aaronrumple]

Do you really need to make this thing move? Seems to me if you are trying to show clearances - symbolic lines or a mass showing the volume it can move in would do.

[phyllisr]

...really need to make this thing move? Seems to me if you are trying to show clearances...

Personally, I might opt for a solution such as you suggest and I also prefer simpler approaches whenever it meets my ultimate goal. However, the original post said nothing about clearances and there is no way to know whether this the design intent.

Since joining AUGI several years ago, I learned that it is impossible to know why someone chooses to pose a question. I frequently do not take the time to write a lengthy explanation about why I want to solve a problem in a particular way. Frequently there are other factors (client expectations, rendering goals, QC issues, trying to understand how Revit "thinks," learning skills that might apply in another circumstance and more) too involved to explain in an AUGI post.

When I pose questions, I welcome work-arounds to accomplish my original goal. I am always interested in alternate approaches. I find these responses so much more helpful than responses that challenge my goals and objectives, particularly if the challenge assumes goals or objectives I did not have.

Mr. Steinhauer, you posted an interesting question. Let me know if there is anything further I can do to help.

[crispin.schurr]

I think the easiest way is to nest 3 layers of families within one another, each dealing with a single pivot situation, in a single plane.

(I tried to rough out a framework of reference lines all in one family and didn't get far)

C

[aaronrumple]

The user must also know the ramifications of his problem.

You want this to move any which way with instance parameters. Expect very major software slow downs as you place more and more of these objects.

If you want to animate the object - Revit won't do it. Parameters can't be easily animated. Another software would do the job better.

If you are a manufacture making Revit parts for others to use inside Revit – then you have to ask the end user – do you need all this stuff? No, because of the performance issues.

You need a few of these just to do an illustration? It will be far faster to model a couple of different configurations than create parametrics. Sure you can make this thing move, but it will classify as at least an intermediate level task. Other than a gee wiz factor, it will gain you little.

This thing is associated with a lab hoods (…a Nederman unit?). My experience with school labs and manufacturing labs tells me I really don’t need to make this thing move to design a good lab. Rather I need good information that helps me place it.

As more new users move to Revit – I see more and more things being attempted which the software is really not designed to do yet. Just answering the question doesn’t really help the final productivity of the user. Just pointing to a way to do it doesn’t really help the user solve the real world problems and may be sending them off on a wild goose chase.

I didn’t post solutions – because they were already posted. No one yet asked “Why?” Challenging the end goal and mention that this is going to have issues does much more to further the users knowledge of how to productively and successfully implement Autodesk software than just saying, "Here – do this..."

[jsteinhauer]

Forcing this user to think about what they are trying to accomplish is one thing. You may not have wanted your response to seem mean, but it did to me.

Since this is a very flexible device, it is important in my mind to be able to model the different ranges of motion. I do not want to animate this family. I just want to be able to change the angle, if the family is sticking into a ceiling. I know that I can model the range of motion with line work. That is very easily done following the product cuts sheets. My original post was asking how users with more experience then me would approach this situation. I tried for several hours yesterday, and wasn’t able to get very far in my family. I searched the forums for “parametric angles” and came up with 496 hits. Most of which are not helpful. So, if this is has already been posted, I guess I need to ask what key words to use when I search? Maybe you could post a link in your response, so I have a direct link.

If I was to use line work to show the range of motion, I would still want to be parametric. If a pipe length where to increase or decrease the lines should adjust to correctly represent this. I have done this with an NMR family. I constrained the radial and axial gauss lines to the dimensions, so they would display correctly for each of the different models specifications. This didn’t seem to cause any issues in the building model that we’ve noticed. I don’t see why adding an angle parameter would cause issues down the road.

I will take a look at the suggestions offered by the other to try to solve my situation.


Thank you,

Jeff S.

The user must also know the ramifications of his problem.

You want this to move any which way with instance parameters. Expect very major software slow downs as you place more and more of these objects.

If you want to animate the object - Revit won't do it. Parameters can't be easily animated. Another software would do the job better.

If you are a manufacture making Revit parts for others to use inside Revit – then you have to ask the end user – do you need all this stuff? No, because of the performance issues.

You need a few of these just to do an illustration? It will be far faster to model a couple of different configurations than create parametrics. Sure you can make this thing move, but it will classify as at least an intermediate level task. Other than a gee wiz factor, it will gain you little.

This thing is associated with a lab hoods (…a Nederman unit?). My experience with school labs and manufacturing labs tells me I really don’t need to make this thing move to design a good lab. Rather I need good information that helps me place it.

As more new users move to Revit – I see more and more things being attempted which the software is really not designed to do yet. Just answering the question doesn’t really help the final productivity of the user. Just pointing to a way to do it doesn’t really help the user solve the real world problems and may be sending them off on a wild goose chase.

I didn’t post solutions – because they were already posted. No one yet asked “Why?” Challenging the end goal and mention that this is going to have issues does much more to further the users knowledge of how to productively and successfully implement Autodesk software than just saying, "Here – do this..."

[jsteinhauer]

Thanks Phyllis,

Your approach seems to me to be working with reference lines so far. Please see the attached .rfa. I found that if I assign all of the angle labels at the same time it doesn’t work. But, if you do one at a time, then click on the “family types” & "apply", it will work just fine. I will be testing this with the geometry soon. Cross you fingers for me.

Thanks for your help.

Jeff S.

For me, creating reference lines first, creating dimensional constraints to the end points of the reference lines and then creating angle parameters between the references planes and lines is the most successful method. I flex everything every step of the way to insure it works before I create any geometry.

Check the link to some shelf and pole families I recently shared. These are less complex than your family but it may help you understand my approach. This is likely not the only way - just works for me. Hope this helps. Good luck!

http://forums.augi.com/showthread.php?t=72869

[phyllisr]

...I will be testing this with the geometry soon...

Would love to see your final result!

As a caution, I have found that symbolic lines (I am a fan because it speeds performance in views) sometimes "confuse" Revit in complex families. Has to do with the specific order when aligning and locking things - have not uncovered the underlying principles yet. Also, when you place this in your project, your staff may have better luck dimensioning to the correct "thing" if you make most of your reference planes and lines "Not a Reference." I generally use Not a Reference for anything I know I will not need for alignment or dimensioning in the project. Including model and symbolic lines.

With credit to Mr. Schurr, I have been tinkering with a combination of my method and his nested method and have been getting good results. In some cases, nesting components can really slow performance but there are times when it works.

As a example, attached is a variation on the Door Swing method Steve Stafford originally introduced. I wanted to create variable swings for the right and left leaf in a double door. So I used the same Swing Angle family I had been using all along, created two instances in the door family then assigned the instance angle differently for right and left.