TFI - CADTips Challenge
@etfrench wrote:The dimension shown (9.0000mm) is point to point while it should be the length of the arc segment which is currently 34 nanometers longer than it should be.
You can drive an arc length in a roundabout way as follows:
User Parameters:
Desired_Arc_Length = as desired
Radius = as desired
Included_Angle = (Desired_Arc_Length * 360 deg) / (2 * PI * Radius)
Sketch:
Needs an arc dimensioned with the Radius parameter, and two construction lines from the endpoints of the arc to the center point of the arc (basically forming a V shape containing the arc), dimensioned between them with the Included_Angle parameter.
@chrisplyler wrote:
So the bent volume could be a bit off from the flat volume, right?
Sure, however the one and only reason I provided the flat and curved details was to elude to how it's modelled in Inventor... it's not a sheet metal part and hasn't been done using behaviours associated with that. To cut a long story short, this was originally going to be an Inventor only challenge but obviously people are more than welcome to try using other packages... I just can't guarantee that those programs would solve the geometry in the same way as source. But that drawing was put together not as a manufacturing detail but to elude to how it could be modelled in Inventor, I appreciate Fusion doesn't have the same tools which indeed causes some confusion.
It's the first challenge of many and I'll learn some things from this one, mostly that it's not a good idea to produce a curved model with two different forms! It'll be impossible to avoid using tools which are exclusive to Inventor but I'll try and make them more cross-platform friendly moving forward.
Having said that, nearly 30 people have got it right so far so it's certainly not impossible!
@chrisplyler wrote:
So...you're allowing some very minute variance in the Fusion property values relative to your Inventor property values? Or are the successful Fusion entrants nailing it exactly?
Thanks man, much appreciated. Fortunately this challenge series isn't too much of a time sink, compared to a workstation performance benchmarking video that I'm currently doing which has taken up 20+ hours this weekend already ... the challenge videos should be a nice relaxing break!
Regarding variance, I think for this one test I will allow for an absolutely minor amount of variance given that Fusion doesn't have the modelling feature that this challenge was almost entirely built around... and the detailing was done to support that, so yes. I'll see how close most people get with Fusion first, but it should be pretty easy for me to see who's got as close as possible within the limits of Fusion!
@chrisplyler wrote:
So...you're allowing some very minute variance in the Fusion property values relative to your Inventor property values? Or are the successful Fusion entrants nailing it exactly?
I understand how it could be difficult setting up such a challenge without expecting plenty of critical feedback. I still think it's a very cool idea, and even like this particular one. I would consider all that feedback as a benefit though...the discussions lead to all of us learning stuff from the attempt and from the discussions.
I appreciate you putting in the time to do it, and especially to participate in the discussion here. I look forward to more challenges!
I could not agree more and hope @Neil_Cross will put out more challenges like this. It is much more interesting when people participate who use different CAD systems.
From comparing my Fusion 360 geometry and the ZW3D geometry I can see where there are minor differences that account for the 10mg or so differences in weight and the other properties. I would almost believe that the actual geometry created by different softwares is much closer. Creating geometry is mainly handled by the geometric modeling kernel of CAD applications and we already have three different ones included in just this thread.
ASM (Autodesk Shape Manger) used in Fusion 360 and Autodesk Inventor
Parasolid used in Solid Works
Overdrive used in ZW3D
Here is a little screencast where I simply do a visual comparison of the geometry created in Fusion 360 and ZW3D.
For both components the Display Detail Control was set to fixed/high so viewport tessellation would have limited effect.
Still it is apparent that the geometries are indeed very close in most areas indicated by the visible Z-fighting. It also shows as explained earlier in this thread that the fillets show most of the deviation.
Will you be posting your Inventor master file to this thread at the conclusion of the challenge?
ETFrench
@etfrench wrote:
Will you be posting your Inventor master file to this thread at the conclusion of the challenge?
I hadn't planned on it, I didn't create this thread!
I'll have to have a think about how hosting the finished file would potentially affect people who don't come across the challenge until later in the future. YouTube seems to be pushing the video out quite hard so it'll make its way to people for the first time long after this month is over, so I'd need to think about whether there's any down sides to hosting the finished file, and where to host it. The original plan was to show the exact steps to remodel it so anyone could do it, but in light of it being rather difficult to do that in Fusion, there might be some benefit to just giving it out at the end of the month. It'll probably not be on this thread though, I'll likely host it in Dropbox or Autodesk Drive and link to it within the original video so it's visible to everyone who finds the challenge.
Without giving too much away for those who are still tackling this - starting from the flat part I got 4 different results in Inventor and SolidWorks depending on order of operation.
The difference is more pronounced further from the center moving out to the corners.
I will post video of my findings nearing end of the contest.
@TheCADWhisperer I am looking forward to that and will add my own video of the process in ZW3D.
I've found and watched the video @Neil_Cross referenced in the challenge.
In ZW3D once the flat models done the rest is just one operation.
However that can also be done in a sequence of two an I'll now have to go back and compare those results.
I don't think Neil quite realized what the started. All the more I appreciate this seemingly innocent challenge 😉
Just wait until you see the design for Invusion 002, it's a cracker, difficulty has increased significantly and it's 100% do-able in Fusion 360... it's 11 times more difficult to do it in Fusion but I've managed it and I'm super excited for September! I almost want to just release the challenges every week but I just know after a few weeks I'd struggle to keep up with that!
Hi guys, it's been a while since I was around, but this challenge series looks promising.
My attempt is simple. A flat quarter of a "grill" bent twice, with use of SM.
In my opinion, making this a curved immediately is an overkill.
We don't have bend tool so the best is to use SM.
The next challenge is more Fusion friendly than this one, my Fusion knowledge isn't as extensive as my Inventor is so I wasn't sure if the tools in Fusion to get this one done are missing or if I just didn't know where to look.
There's no restrictions on using SM though. Can you not measure the second angle in the Autodesk Viewer? The first angle on the R300 bend was only provided as a validation for something that I'll explain when I wrap it up for this month.
