Try as hard as I could I can't get that to work. I get it all the way to the run and the solver stops and errors because it has numerical difficulties. I had the beam in two 250mm sections (like before) with a little square above for the force but when I go to run it then it poops out. *shrug*
Fixed on each end of a solid 1 piece 500mm V-Rail with the 1 sq mm pad in the middle. Treat as beam was used on the v-rail and I did everything as in that video and put the pressure on the top pad that is in the middle and it will not run. Says...
I think torque on extrusion is a bigger threat. One end attached and 100N.m of torque at the other end caused somethin like 5mm displacement. It is a lot of torque though. Looking at the photo you can tell why profile is less that happy twisting. The outer webs don't contribute much to the overall structure under torsion. I'll have to put a distributed load off the front center to see if it gets ugly. Edit: If anyone is curious. @25N.m (about 18.5 ft-lbs) the displacement was about 1.3mm(.05"). I believe this would be a function of the greatest dia. and not the cord length (I'll try to look into this). I also believe that If we calculated a final cord angle then the total amount of vertical change (say for a z-axis) would drop further. Edit: The total drop was something like .021" I'll upload pics and the steps on how I did the beam deflection sim later.
Please do as I would like to learn how to get this beam deflection to work using this technique and in that picture that is twisting and I expected that just by looking at them because extrusion only really has a pushing straight out and nothing really to take care of twisting so it basically just collapses into itself like an old rope. If that were solid then there would be a difference though I am unsure how something would twist in our uses.
Yes. I took a photo of the torsion study. Note: This was just a quick test to see how the extrusion webs behaved. I had an idea and since I was running the force study I decided to do a quick torsion test to see the distribution of stresses. I thought there would be more through the outer web, but the study says otherwise. The majority of stresses didn't go past the inner web as much as I thought. Makes sense since it is closer to the axis of rotation, the outer webs are independent of each other, and have a much smaller cross section opposing the force. NOTENOTE This study does NOT represent how a 20 x 40 acts when it is mounted between two gantry plates by screws, has a fully loaded carriage with point contacts from wheels, distributed over a length. That's a study for another day! Maybe our resident solidworks captain could attack this someday. I'll post that tut. in a bit.
Yep, you did well my friend and that study was about what I expected out of extrusion and if you look at say Misumi's extrusion I expect there to be a few differences from your findings in V-Rail too. I am just not sure if it would be majorly different but there does seem to be a whole lot more meat to them than the V-Rails.
View attachment 11091 Funny you mention that. I was just thinking about their extrusion. My "untested" conclusion was that the Openbuilds rail has more function with its form that the very slight differences in torsional ridgidity (spell that right?lol) of another extrusion probably doesn't pass a cost to benefit analysis. Had some internet/modem/router issues (no not that router, though there's issues there atm! I'm back in pieces. :/). Nets back up now. Be posting those pics asap. I OB from my phone and I don't know if it's possible to post pics in a post like on a pc. So I may have to make several pic attachments and try to caption each one in a post. I'll send a test post first.
This is the contact patch. It is smack in the center of the beam. The idea is to simulate a rope attached to the center of the beam with a weight of force attached. The contact patch extends for the extremes of both flats and is .005" wide (Sorry. Imperial!). The sketch was started on one of the flats and brought across to coincide with the furthest edge of the other flat. It was then extruded as a .001" solid. Surfaces didn't end up working for this. I figure .005" width was suitable for the weight and any round over of the twine, rope, string was negligible. Now click on the simelation tab. Then study advisor! On the right. On the right. On the right. Then on the left, after you choose models modeled units, click okay. On the right again Right. Now the left should look like this. Right click fixtures...click fixed geometry. Select both end faces. Click the infamous Okay. Now Right click external loads. left click force. Select the top of that small extrude. Click Run this study. Wait. Wait again. Voila. You owe me a beer.
The line explain what ya did there please. What plane did you use? Mesh failed on that one piece so something I am doing is not right.
Yep, I tried that and no go. I tried to extrude as one or have separately and it just will not work. Care to share the file and I can match it with what you described?
Maybe tomorrow. I can add the creation of the initial force section flat. That's really all it is, is a thin extrusion off of the top of the beam. The sketch is started on one of the flats and extended to the other.
Oh, I got it to work the simple way and the issue was I needed to add a load which is Gravity. In your auto way I noticed that was the only difference. What I am noticing is that if I do 400N or 1N the numbers are barely changing so something is not kosher.
I believe it may be an issue with the two fixed ends. Maybe add a roller support to one end or go simply supported. I'll try this later.
I am seeing -0.001101mm maximum deflection using 1lbf and that would be directly in the middle and when I changed that to 100lbf I see a maximum deflection of -0.1101mm. I am getting changes but 100 pounds versus 1 pound the 20x40 should buckle and the 100 pound weight make a much larger dent than 1/10 of 1 mm of deflection in 20x40, or at least I would think so.
The shorter the beam the less the deflection. There is a shorter distance to the supports. If one hangs a 14lb spindle on 1500mm length there will be more deflection than on a 250mm beam. The load is closer to its supports. If you eliminate the beam then the load is directly on the supports.
If I threw a thousand pounds on it it should not only deflect but squash too. If I had my nephew stand on one at 100 pounds it would not be able to support that weight is my estimation and the real world deflection would be magnitudes more than what SW is saying above.
