Need help

Hello everybody,





I need some help to create an assembly of roller chain. For
this I want to create a pattern on a curve (curve that I create in assy) of a
subassembly like in pictures below.


Does anybody know how to solve this problem? Or have a better idea
how to do the pattern?


Thank you in advance.

View attachment 1352View attachment 1353

i have solved this problem a couple of different times (igus chains/belts) by using the same technique.


1. create your 'profile' curve (you have already done this)


2. puta datum point on the curve - most likely at a 'actual lenght' since you know the spacing or pitch of the individual components (in your case links of the chain-also be sure to pick the entire curve and not just one segment of the curve). Pattern the point at a constant pitch.


3. in the individual component (chain link) added datum points at the center of the chain link centers. these points will be used to assemble to the patterned points on your curve mentioned in step 2 above.


4. at assemblyalign the points from the link onto the points of the curve. an aditional 'orient' constraint may be needed, but you chain links will pattern via a reference patter.


incidently-if all aspects are done correctly, the curve profile can change shape and all of the components and points will follow accordingly!


-chris

Hi gvi70000


There is a great tutorial describing step by step to go through this topic.


Look in knowledge database of PTC searching with these criterias:


Check mark theese two choices:Suggested technique, Pro/Engineer help


and select thoses criterais


asssembly, wildfire 2.0, exact


N.B.: you must have a maintenance agreement with PTC to acces this database but it is full of valuable tips &techniques of all sorts.


Beam

Thanks for the info; I have been wondering how to do this tidily for a
while.



Even though the PTC method is APPROXIMATE because the chord length is
shorter than the curved pitch length and would thus require some
pin/bush clearance to avoid interfering solids upsetting hidden line
removal, I expect it shouldn't be too hard to make it exact by
substituting the sketched radius of the datum curve for the segmented
path around each sprocket. There are several ways to accurately get the
sprocket arc segments equal to the chain pitch and the straight lines
between sprockets will also be a multiple of chain pitch which makes
phasing between chain "arcs" easier than phasing the sprockets
correctly.



Changing the path from circular to segmented will lose some parametric capability but
I do not consider that to be a problem as the chains would only ever be
modelled well after the sprockets and centres are finalised because the
modelling cost is so high



DB


Edited by: Dell_Boy

For anyone that is interested, I have done a bit more experimentation
on the subject since we use a lot of chains and sprockets at work and
developed a somewhat different method to model chain runs correctly.



It revolves around having one assembly family table model for ANY
"curved" run of chain and another to model any straight run of chain.
With a few simple relations it is possible to get the tables down to
about 4 or 5 easy-to-manage columns and including the ability to start
with either an inner or outer link.



The alternating straight and curved instances are assembled on to a
sketched datum curve chain run with the following restrictions.



The straight sections are an integral number of pitches long.

The "curved" sections are not sketched but the angle between the ends
of straight runs corresponds with an integral number of teeth on the
meshing sprocket.



Also remembering that straight runs meet PCDs tangentially ONLY in
special cases and that the calculated centres between two sprockets are
ideal rather than exact because chain pitch varies slightly with
rotation.



The link parts also form their own two family tables allowing for any pitch of roller chain.



When it comes to meshing with the sprockets, it is far easier to
assemble the sprocket into the chain rather than trying to set the
phase of each sprocket and wrap the chain around it.





DB





Modelling igus chains is substantially easier than roller chain because,

there is usually only one element to repeat,

you are not trying to mesh it with sprockets

you don't have closed loops

path lengths aren't that critical

and with two sprockets and no tensioner your roller chain centre distance is pre-determined.


Edited by: Dell_Boy