Assembly2 assembly meshing of boolion unioned parts min size .1 in or .05" in never seems to finish windows shows not re

About the development of the FEM module/workbench.

Moderator: bernd

User avatar
HarryvL
Veteran
Posts: 1285
Joined: Sat Jan 06, 2018 7:38 pm
Location: Netherlands

Re: Assembly2 assembly meshing of boolion unioned parts min size .1 in or .05" in never seems to finish windows shows no

Post by HarryvL »

so ....

With E= 234422 N/mm^2, OD=25.4mm and ID=20.3mm, EI=E*π/64 *(OD-ID)^4=234422*π/64*(25.4-20.3)^4 = 7.66E6 Nmm^2

All your tubes have the same EI, but you have 2 lengths (pcft_24: L=610mm and pcft_98: L=2489mm).

Finally you have put (only!!) 1 N of load on each element, giving a uniformly distributed load of pcft_24: q=1/610=0.00164 N/mm and pcft_98: q=1/2489=0.00040177 N/mm

Putting it all together, the deflections of your tubes are:

pcft_24: u=0.00164/8*(610)^4/7.66e6=3.7mm
pcft_98: u=0.00040177/8*(2489)^4/7.66e6=252mm

I am a bit rusty, so it is quite possible I made a few mistake here ... but is only to demonstrate that 3D elements make no sense for the full model.
josheeg
Posts: 82
Joined: Tue Nov 12, 2013 3:10 pm

Re: Assembly2 assembly meshing of boolion unioned parts min size .1 in or .05" in never seems to finish windows shows no

Post by josheeg »

Hi I got factory specs that could go into kicad fem. I figured it would help even if I tried a hand calculation.
density ~1700 kg/m^3
modulus ~34 Msi (~230 GPa)
poisson ration 0.2

or higher m
Good Morning Joshua,
Thank you for your email. We appreciate your interest in DragonPlate.
Without knowing which product your are looking at, I can only speculate about the specifications.
Your density seems high compared to what we typically use. We use 0.06 lb/in3 (~1700 kg/m3).
Your modulus also seems high. The raw carbon is typically around 34 Msi (~230 GPa). We do offer higher modulus
options as well (57 Msi and 110 Msi). Depending on the composition of the tube you are looking at, the final modulus
using 34 Msi carbon fiber can vary from 5 Msi (34 GPa) to 17 Msi (117 Gpa). This would be higher if you use the higher
modulus options available.
Your poisson ration of 0.2 seems appropriate.
Best Regards,
Dean
josheeg
Posts: 82
Joined: Tue Nov 12, 2013 3:10 pm

Re: Assembly2 assembly meshing of boolion unioned parts min size .1 in or .05" in never seems to finish windows shows no

Post by josheeg »

thschrader nice photo editing what did you use and thank you all for taking time to educate me.
josheeg
Posts: 82
Joined: Tue Nov 12, 2013 3:10 pm

Re: Assembly2 assembly meshing of boolion unioned parts min size .1 in or .05" in never seems to finish windows shows no

Post by josheeg »

yes it is 5220mm or 204" inches. there would probably need to be 48" in segments instead of 98 as max size so it could screw apart and go in my car. :lol:
kwp20048segforcar.png
kwp20048segforcar.png (147.88 KiB) Viewed 1177 times
josheeg
Posts: 82
Joined: Tue Nov 12, 2013 3:10 pm

Re: Assembly2 assembly meshing of boolion unioned parts min size .1 in or .05" in never seems to finish windows shows no

Post by josheeg »

I do not have much experience making model airplanes and the large size is to test myles loyds crosswind kite power calculations as a technical curiosity.
But I also don't know if the foam over the carbon fiber poles would be strong enough or fold. That is why to try this cost effectivly on a project budget I would need to use fem or hand calculations. To get the confidence to buy the parts and assemble it.

This shows how the cp pressure on wing surfaces can be done with openvsp vspaero. That would let me know how strong the surface has to be.
I think I could be wrong there is something in openvsp to make or model the wing cord pieces that could go on a carbon fiber spar.
fem analisis could let me test some of thowse design options before buying materials and run to falure.
cpofwingsurfacesm.png
cpofwingsurfacesm.png (515.29 KiB) Viewed 1175 times
thschrader
Veteran
Posts: 3129
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: Assembly2 assembly meshing of boolion unioned parts min size .1 in or .05" in never seems to finish windows shows no

Post by thschrader »

According to your FEM help-request in this forum:
https://forum.freecadweb.org/viewtopic. ... 0&start=10

A year ago I have written a FEM beginner manual
https://www.dropbox.com/s/vr6pxaj1bt4ph ... l.doc?dl=0

You can simulate the wind on your glider with the FreeCAD cfd workbench
https://github.com/jaheyns/CfdOF
https://forum.freecadweb.org/viewtopic. ... &start=110
josheeg
Posts: 82
Joined: Tue Nov 12, 2013 3:10 pm

Re: Assembly2 assembly meshing of boolion unioned parts min size .1 in or .05" in never seems to finish windows shows no

Post by josheeg »

Well why won't my model simulate? if your tutorial is on simulating forces on a model?

I would like to try cfd workbench that looks good. But wouldn't I need to have the skeleton work?
I am new to design of something like this so I do not have a checklist.
thschrader wrote: Wed Aug 15, 2018 6:46 pm According to your FEM help-request in this forum:
https://forum.freecadweb.org/viewtopic. ... 0&start=10

A year ago I have written a FEM beginner manual
https://www.dropbox.com/s/vr6pxaj1bt4ph ... l.doc?dl=0

You can simulate the wind on your glider with the FreeCAD cfd workbench
https://github.com/jaheyns/CfdOF
https://forum.freecadweb.org/viewtopic. ... &start=110
thschrader
Veteran
Posts: 3129
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: Assembly2 assembly meshing of boolion unioned parts min size .1 in or .05" in never seems to finish windows shows no

Post by thschrader »

josheeg wrote: Thu Aug 16, 2018 8:25 am Well why won't my model simulate? if your tutorial is on simulating forces on a model?
I would like to try cfd workbench that looks good. But wouldn't I need to have the skeleton work?
I am new to design of something like this so I do not have a checklist.
Your model can be simulated (calculated), as Harry has shown above.
The problem is: your model should be meshed with 2D-shells for the tubes
and 3D-solids for the cube-connectors. ATM this dont work in "normal" FC-FEM
(bernd is working on it). Meshing all in 3D is ineffective and produces errors.
BTW:
Designing an aircraft is one of the hardest things you can do.
You should start with a simple problem to get a feeling for FEM.
CFD is as hard as FEM. My advice: practice, practice, practice...
A little motivation :)
https://www.youtube.com/watch?v=cPk21C0Wpkg
josheeg
Posts: 82
Joined: Tue Nov 12, 2013 3:10 pm

Re: Assembly2 assembly meshing of boolion unioned parts min size .1 in or .05" in never seems to finish windows shows no

Post by josheeg »

The four electric motors are on the end of the two forward most stalks then two tubes are for the main wings then the stalks behind the wings and two tubes are for the tail.

I was considering some kind of foam from wecutfoam.com flying foam has been down.

But once the structure is tested to handle the loads I can then move on to the wing surfaces designing to meet the requirements.
I was considering squair screw on connectors with lock washers to the tubes. That should hold them.

Your saying ribs and stringers is how most planes are done, yea but then you can see how the stress modeling and design gets more difficult.
Maby OpenVSP would be able to help with that but it has such a learning curve.


Beam elements? As thin shell tubes? I will have to try setting up some shells like that...

The lift and takeoff vtol option should probably have the least load and flying its when it is flying crosswind and tethered then the motors would be small generators and the largest forces would be on it. (small flying wind turbine)
HarryvL wrote: Thu Aug 09, 2018 11:15 am Oh it's an airplane frame :shock:

Just a few questions and observations then:

Where do you plan to attach your 4 engines?

How do you attach the wings and what is their composition?

In a conventional design the plane derives strength and stiffness from ribs, stringers and skin (like a thin-wall composite beam) or an monocoque stress-skin. Is it your intent that in your design the stiffness comes from a bending frame of 0.5" tubes?

Anyway, pending a better understanding of how your design works, I would in the first instance model your frame as simple beam elements (available in FreeCAD)

Where to apply static loads and supports depends on where your engines sit and the flight mode (lift off and landing vs horizontal flight). Also in a very skinny design (which you are aiming for (?) dynamic loads need to be considered.

Finally, you may well find that appurtenances (like the wings) provide much greater strength and stiffness than the skinny tubular structure itself.

Hope this helps.
josheeg
Posts: 82
Joined: Tue Nov 12, 2013 3:10 pm

Re: Assembly2 assembly meshing of boolion unioned parts min size .1 in or .05" in never seems to finish windows shows no

Post by josheeg »

I got your calculation Harryvl into a spreadsheet the deflections mm come out very close probably spreadsheet rounding errors.

Is there a point where the deflection shows it will proabably start to show signs of breaking or a point to avoid?
tubecalc.png
tubecalc.png (77.18 KiB) Viewed 1084 times
I got the updated specifications from dragonplate and the 3 loads I wanted to try.

That way I setup the spreadsheet and it is easly shared and tested. Also if I have to make design changes I can.

Here are the tube specifications do I put density in this equation somewhere EL?

"Hi I got factory specs that could go into kicad fem. I figured it would help even if I tried a hand calculation.
density ~1700 kg/m^3
modulus ~34 Msi (~230 GPa)
poisson ration 0.2"

"The force on other systems is 3100 newtons.

1/16 of that 194 newtons

and 620 newtons would be tests I would wonder how the tubes would be effected.

the material properties would be
density
460 kg/mm^3" So the spreadsheet should allow me to test all 3.
HarryvL wrote: Thu Aug 09, 2018 9:51 pm so ....

With E= 234422 N/mm^2, OD=25.4mm and ID=20.3mm, EI=E*π/64 *(OD-ID)^4=234422*π/64*(25.4-20.3)^4 = 7.66E6 Nmm^2

All your tubes have the same EI, but you have 2 lengths (pcft_24: L=610mm and pcft_98: L=2489mm).

Finally you have put (only!!) 1 N of load on each element, giving a uniformly distributed load of pcft_24: q=1/610=0.00164 N/mm and pcft_98: q=1/2489=0.00040177 N/mm

Putting it all together, the deflections of your tubes are:

pcft_24: u=0.00164/8*(610)^4/7.66e6=3.7mm
pcft_98: u=0.00040177/8*(2489)^4/7.66e6=252mm

I am a bit rusty, so it is quite possible I made a few mistake here ... but is only to demonstrate that 3D elements make no sense for the full model.
Attachments
tubecalc.ods
(8.42 KiB) Downloaded 37 times
Post Reply