Weld chain with flexible-positioning of the ends

[tleati]

Hello,

I am attempting to model a weld chain in order to then assemble it in an assembly.

As you know, a chain has a fixed lenght and fixed number & size of its links: I would ambitiously like that when I assemble it, I can position it by making its ends (first and last link points) coincident to two points on the guesting assembly, and then the chain, conveniently designed for the purpose, adjusts its shape in assuming a parabolic (or parabolic-wise) structure as subject to the gravity.

anyone has ever done it and/or has some suggestions on how to?

(I know that for sure it is necessary to use flexibility.)

many thanks

Bye

Tommaso

This thread is inactive and closed by the PTC Community Management Team. If you would like to provide a reply and re-open this thread, please notify the moderator and reference the thread. You may also use "Start a topic" button to ask a new question. Please be sure to include what version of the PTC product you are using so another community member knowledgeable about your version may be able to assist.

[Patriot_1776]

THAT'S a tall order.....  Good luck! 

[mbonka]

Hello Tommaso,

l have searched my notes about chains and found following.

- Roller chain --- is attached

- Welded chain --- is attached

Youtube links:

Chain model with Pro/ENGINEER - YouTube‌ ---- in my opinion it´s the best solution

Chain modelling in ProE - YouTube

Sprocket mechanism with Pro/E WildFire 4.0 - YouTube  --- l like this video .

Hope all attached files are freeware.

Regards

Milan

[tleati]

Hello Milan,

thanks a lot for your suggestions and your files, I have downloaded them and they will be extremely useful.

My problem however, (I have not pointed it out clearly maybe) is not much how to do the chain (in my case it is a weld chain this time, but I will certainly have to build a roller chain too in the future), because I already have an idea how to do it (that is quite much what Kenneth wrote in the next post and how reported on your pdf instruction and part).

My problem is that I would like the first and the end links of the chain to be assembled to two whichever points on the guesting assembly and make the chain adapt to those two points, whichever relative position these two points have from each other. I know that this may be done by flexibility, but how exactly shoud I have to do it I don't know, because with flexibility you can adapt a dimension of your flexible component to a length/distance/angle in the guesting assembly, but not directly adapt two points to two points on the guesting assembly. (I hope to have been sufficiently clear).

At this point I am wondering whether may I use the cables application too to do this.

thanks

bye

[tleati]

by the way, I cannot open gph files as I have Creo Parametric 2.0, do you know how to convert them into .prt or .asm or could you kindly provide them (as a .stp too at most, just to see them) ?.

thanks

bye

[KenFarley]

The only way I can think to do this is

(1) Get a good understanding of the mathematical definition of a hanging chain (a catenary). It's not a parabola; here's a possible source of info: Catenary -- from Wolfram MathWorld .

(2) Use the mathematical equations to define a curve that will "drape" between your two connection points.

(3) Define points along the catenary curve you've defined to position the chain links.

(4) Assemble the links onto the points.

As Frank stated, this is a very "tall order". You are going to have to arrange things so you end up with an integral number of links. Assembling the links will probably not be an easy task either, since they have to alternate as you go along. It'd be a fun project, if it was my job to do it.

[tleati]

Hello Kenneth,

thanks for your suggestions and your link, yes indeed I was wrong it is not a parabola but a catenary.

As I wrote to Milan indeed I already have an idea (and I realized it also) of how to perform the steps 1-4 you mentioned, the main issue is in fact the assembly of the points.

Probably with a fixed point and an axis leading to the other point, with the distance and the angle of the axis as parameters of flexibility.

Yes it's intriguing, even if very much time consuming as I imagine!

bye bye

[Patriot_1776]

I've been thinking about this as a fun little aside.  You didn't mention orientation of the sprockets, horizontal I take it?  How accurate do you NEED (not want) it to be?  I might do one for fun, and see if I can't mechanize it with sprockets.  For the trajectory, I'd simply use a conic top and bottom to connect the arcs on the left and right, and use a perimeter dimension equal to pitch X number of links to control it.  Adjust the rho values to get the sag you expect.  Would sag be PARAbolic, or HYPERbolic?  An ellipse maybe?  Whatever it would truly be from a math standpoint, the rho value could do it.

[KenFarley]

Oh, are you thinking it's a drive chain? I was thinking it was a welded chain, like on an anchor, or a safety chain. Drive chain would be fun and you might be able to assume minimal or no sag due to the chain being under tension. Although, that is not reality.

It's not parabolic, but defined by a catenary curve. I put a link above for the equations.

[tleati]

Hi Frank,

Indeed it is a weld chain that I am trying to make, and the main problem is the assembly of the two first and end points, to the guesting assembly. Concerning the equation, it would be a catenary as Kenneth said.

thanks

bye bye

[TomU]

I don't think I'd worry about the individual links but instead focus on creating a controlling assembly (with maybe a couple of parts and a skeleton model) that would automatically adjust to the conditions you put it in when assembling it to the end point locations.  I can imagine using a curve that has it's length automatically adjusted to yield the required number of increments, and then patterns of points or coordinate systems along the curve to control actual link assembly.  It really seems pretty doable.

[tleati]

Hello Tom,

thanks for your reply. I repeat that my problem is not the assembly itself, I have already done it. The links automatically adjust to the catenary curve connecting the end points: but, the problem is, how do I easily and rapidly assemble these two points in an assembly between two points that are not at the same distance and at the same height of the points of the original chain that I have to assemble? (this chain, in my case, is hanging between two points just like those some head-bangers keep on their trousers between the wallet and the belt, just to say one).

I need a chain which can be, once set its lenght and number of links, assemblable between any two points at any distance and at any height difference (in the limit of the chain length obviously)...this is the main obstacle as in Creo, as I see, it's not possible to assemble models making two points directly coincident and set a special flexibility which allows all the rest of the chain to adapt to it...the only flexibility parameters I see are by value, lenght, distance, angle, area, diameter...

bye bye

[TomU]

In this particular case, I'm not sure flexibility is what you want.  Instead you want to actually assembly portions of the chain assembly to your upper level assembly and then let the chain adjust (based on whatever logic you've built in the chain assembly).  Basically, assembly the two end links wherever you need them, then let the rest move accordingly.

[tleati]

Yes a form of flexibility is needed for sure (not the one actually existing in the program), otherwise how could it adapt to the couple of points on the guesting assembly.

Not portions indeed, we work with plm Windchill so we have to keep the component as it is -> for example, chain 30 cm long, 10 links: I design it with these characteristics and that remains in the assembly, even if the chain will be cutted out when purchased from supplier from a roll.

This same chain, if used in another product, will maybe have the two end links at a longer distance from each other when assembled: so when I put it in, the software tells me constraints are not valid, as that chain had been designed (distance bewteen end links) for the other assembly...that is the problem that makes me think about flexibility...

[BobNJ]

‌tommaso:

I am not sure what a weld chain is but I thought this link may be of help to you.  It describes what I think you may be trying to do, however Leo's model uses hoses instead of chain.  If you think the method may work, then the challenge would be to model it as a chain instead of a hose.

http://www.e-cognition.net/pages/FlexHose.html

Bob

[tleati]

Hi Robert,

thank you so much of your link, indeed if I had a vague idea of making it by defining a curve and its angle with a datum plane, now your clip clarified it and says it can be possible to do!

At this point yes, the most challenging step would be to implement the catenary curve.

I hope to get into it as soon as possible and being able to post the results if anyone may be interested.

With weld chain I mean exactly this one:

bye bye!

[mbonka]

Great tip Robert.

l found different method, but also used for hose model.

Dynamic Analysis in PTC Creo Parametric - YouTube

l think roberts idea looks better.

As Robert sad: "...the challange would be to model it as a chain instead of a hose."

Maybe try to combine my previus input about welded chains together with Robert Whiteman idea about hoses model.

You are so near to slove it...

[tleati]

thanks for your video, indeed another way to build the chain part.

Yes in fact I will try to merge the chain-building suggestions with Robert's input to get to the result...hoping it will not be much time consuming especially for the equation of the catenary! ...will keep you up-to-date anyway!

bye bye