Hello everyone,
Is it possible to assembly built parts along a spiral path? There is no "sweep" feature in assembly. The feature I want to build is all parts sweep along a spiral path. The reason I built all parts separately and assembly them all in assembly is that I want to change the shape/path of the spiral and all parts will automatically change their shape too. Otherwise I have to change all part one by one, it would be inconvenient for simulation. Any ideas?
Hello,
To build the spiral curve needed for the curve driven pattern, try "Equation Driven Curves".
Woflram math has some spiral formulas:
Logarithmic Spiral -- from Wolfram MathWorld
I hope this helps.
Cheers,
Kevin
Thanks for both Kevin.
My situation probably is unique. Since all parts have been built (as components) and are used in different assembly cases. My case is that I want twisted these parts (change their shape ONLY in my case) under assembly and all parts will remain the same as what they are, so others can use them for their specific assembly.
Do you think whether it is possible?
Best regards,
Tracy
Hello,
If it's a twisting of the final parts on themselves, then look into the "Flex" feature.
You can create a twisted config so others can use the "regular" parts untwisted.
Cheers,
Kevin
Thank you so much, Mr. Chandler.
I checked the "Flex" feature(never used it before, I am new to SW). I guess this feature is still under Parts. The tricky thing is that SW doesn't have all features, which "Parts" do, under Assembly. My case is that firstly, assembly two parts together (simple just using the built parts); secondly, twisted it (I think I have to do it under Assembly), then assembly another layer on the top(no twisting needed).
Any further suggestion?
Best regards,
Tracy
Hello,
If I'm understanding your situation correctly, you need to "stack" three "layers" of two parts with the parts in the middle be twisted.
If the twisted layer is made of the same parts, just each twisted individually, then you can use Flex and create configurations in each part.
The feature in the tree is the Flex feature and it's suppressed in the "regular" configuration.
Then in your assembly, change the configurations for the parts in the middle layer to the twisted versions.
If you can, upload parts or a sketch of what you're looking for.
This will help my brain function better.
Cheers,
Kevin
Thank you so much again, Mr. Chandler.
Attached is a single sub-assembly (I have already assembled two parts together). Under assembly, you may insert components twice to get two components of the single sub-assembly, construct them as tangent (so the layout diameter of the general sketch would be double size of that of the single sub-assembly), and then twisted them. If the turns of twisting could be controlled ( turns can be changed and two sub-assembly always remain tangent) , it will save my life...
Best regards,
Tracy
Hello,
Attached is another approach to create the twisted pair as a multi-body part using only unmerged Sweeps.
I used three Sweeps so I could assign easily separate jacket and wire materials and colors.
I only did the twisted portion but you can loft and sweep from there.
One of the downsides to using Sweep turns is that they're not exposed in Manage Equations so linking the number of turns in one place isn't possible (at least not through Equations).
So you have three separate turns that aren't linked for easier adjustment.
Cheers,
Kevin
Hi Kevin,
Thank you for your kind help. Could you help me to figure out the solution for these issues I posted with a word file. I was not able to upload my SW file, it said my file is too large even a zipped file.
Best regards,
Tracy
Here is a method you could try, although I must tell you it was made exponentially more difficult by the requirement to use separate parts for each component (conductor1, jacket1, conductor2, jacket2) and using a subassembly for each wire. Anyway, it is all modeled in context of teh top level assembly and uses a Master part to drive the part sweep sketches. You can change the parameters of the helix in the master part and all fo the individual parts will update.
Hi Jamil,
Thank you for your kind help. Could you help me to figure out the solution for these issues I posted with a word file. I was not able to upload my SW file, it said my file is too large even a zipped file.
Best regards,
Tracy
Thank you so much Kevin and Jamil. Sorry for late responding, since I was struggling with other twisting issues.
Both of your solution are helpful for my design. I hope you both have some time to look into the issues current I have. I post a word file which describe the issue I encounter and the SW file too.
Sincerest regards,
Tracy
sweep a single strand - then use a circular pattern of the BODY to get the rest of the strands. keep your sketches as simple as feasible.
you will likely need to add an axis to define the circular pattern.
Thank you Jeremy.
I did follow your instruction, swept a single strand, and I got what I want. I did create an axis for the circular pattern. However SW doesn't allow me to choose Bodies to Pattern. I only can choose Features to Pattern, and I also click the Geometry pattern option (that SW current support only). then SW runs so long but nothing I get.
Do I need to create a solid body for circular pattern? How? I am pretty new to SW.
Best regards,
Tracy
you need to check this box to switch to a body pattern (you can pattern surface bodies or solid bodies). and depending on what version of SW you are using, you can't switch after the pattern is made - need to delete and make a new pattern feature.
Yes, I did! Thank you so much, Jeremy!
Currently I am working on a braid shield model. It's similar with this one Tutorial: Making a realistic wire-braided texture in modo but it was made in modo. Do you think we can do it in SW purely?
Best regards,
Tracy
correctly modeling the 3D geometry of a braided shield would be extremely challenging - and is probably pointless. why are you trying to do so?
Hi Jeremy,
I would simulate the performance of wire shielding under RF. Do you have any suggestions?
model a simple tube and adjust material properties in the simulation until you get correlation to the measured results.
Thank you so much Jeremy.
It makes sense. The thing is that where to get the measured results of this simplified models since there is no product like the simplified models.
generally any simulation method will need to be validated - you need to show that the simulation method represents the real world. this is done by simulating a known case and showing that you get the expected results. so you really should simulate an existing product first - preferably one you already have test data for.
Tracy,
You can try a curve driven component pattern or sketch driven component pattern. Not sure if it will work with a spiral but worth looking into.