Comparison of 4130 and 2024 Tubing, Composite Tubes too.

Discussion in 'Tube and Fabric' started by wsimpso1, Nov 4, 2019.

Help Support HomeBuiltAirplanes Forum by donating:

  1. Nov 7, 2019 #21

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    5,893
    Likes Received:
    3,119
    Location:
    Saline Michigan
    Look at that. Nice table design. We must have gotten our data on strengths from different places...
     
    AdrianS likes this.
  2. Nov 8, 2019 #22

    AdrianS

    AdrianS

    AdrianS

    Well-Known Member

    Joined:
    Jul 5, 2014
    Messages:
    378
    Likes Received:
    228
    Location:
    Australia
    I started with data from https://www.makeitfrom.com/material-properties/Cold-Finished-4130-Cr-Mo-Steel

    I've reverse-engineered your materials, I think :
    4130 : Sy = 63,000 ; Ss = 36,400 ; E = 30,000 ; D = 0.282
    2024 : Sy = 39,000 ; Ss = 22,500 ; E = 10,500 ; D = 0.110


    Sorry to nitpick, but re your latest update:
    4130 bending moment (1558 in.lb) appears to be at yield, not yield/FOS

    2024 1x065 :
    weight appears too low (matches 1x035)
    I get slightly different tensile (4964 vs 4953) and nearly double the 60" column (402 vs 237) : but that may be a bug in my ss
     
  3. Nov 8, 2019 #23

    AdrianS

    AdrianS

    AdrianS

    Well-Known Member

    Joined:
    Jul 5, 2014
    Messages:
    378
    Likes Received:
    228
    Location:
    Australia
    Latest version, with lots of ally sections and Billski materials props.

    When Billski and I get the same results, I'll move on to square sections
     

    Attached Files:

    Vigilant1 and cheapracer like this.
  4. Nov 8, 2019 #24

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    5,893
    Likes Received:
    3,119
    Location:
    Saline Michigan
    I will check.
     
    wanttobuild likes this.
  5. Nov 8, 2019 #25

    AdrianS

    AdrianS

    AdrianS

    Well-Known Member

    Joined:
    Jul 5, 2014
    Messages:
    378
    Likes Received:
    228
    Location:
    Australia
    I'm not meaning to hassle you - I'm just enjoying tinkering with spreadsheets, and learning stuff as I go. The lookup table implementation was an experiment that worked.

    Onwards
    I can see how to do most of the square calcs, if it's worth doing, but haven't looked at compression yet.
     
    wanttobuild likes this.
  6. Nov 8, 2019 #26

    wanttobuild

    wanttobuild

    wanttobuild

    Well-Known Member

    Joined:
    Jun 13, 2015
    Messages:
    597
    Likes Received:
    57
    Location:
    kuttawa, ky
    You guys need to take a break and have a couple of beers. Beer helps everything.

    Thank you guys for giving your time and brain power to assist the community of experimental aviation, in the construction of Safer aircraft.
     
    Last edited: Nov 8, 2019
  7. Nov 9, 2019 #27

    AdrianS

    AdrianS

    AdrianS

    Well-Known Member

    Joined:
    Jul 5, 2014
    Messages:
    378
    Likes Received:
    228
    Location:
    Australia
    I'm afraid I got carried away with square tubes:

    Assuming H = outer size, and h = inner size = (H - 2 * Wall)
    A = H^2 - h^2
    I = (H^4 - h^4)/12
    J = (H^4 - h^4)/6
    S = (H^4 - h^4)/(6*H)
    k = sqrt(I/A)
    y (diagonal) = H/sqrt(2)

    Tension = Sy * A
    Bending (square) = Sy * I/(H/2)
    Bending (diagonal) = Sy * I / y
    Torsion = Shear * J / y

    I'm looking at buckling now... [Edit] Looks like the buckling formulas are the same as round tube :)



    ps:

    I need some definitive material specs :
    Yield
    Shear
    Density
    E

    The numbers I've found for 4130 and 2024 online vary from site to site.
     
    Last edited: Nov 9, 2019
  8. Nov 9, 2019 #28

    wsimpso1

    wsimpso1

    wsimpso1

    Super Moderator Staff Member

    Joined:
    Oct 19, 2003
    Messages:
    5,893
    Likes Received:
    3,119
    Location:
    Saline Michigan
    Math looks right for square tubes. AdrianS is using the common assumption that we neglect the corner radii. This only very slightly overstates our stiffnesses and strengths, but is commonly done. One issue with square tubes is the straight walls will get into crippling (also called elastic instability) more easily than circular tubes and so an additional check should be performed using the criteria for this published in the classic texts. I will run that in mine too.

    Yield and shear strengths published do vary because alloy content, impurity content, and heat treat do vary. If we have statistically valid test data from our supplier and the process that supplier uses for our material, we can use some statistical minumum off that data. One source for steel or aluminum? Ha. I never did that, and we bought steel for many of our part numbers in the millions of pounds per year. Just use the lowest value you come across, ensure your design is secure at that level, and know that most of the time, your material is better than that. The difference in design when selecting stock sizes is usually insignificant to your decision. Even when specifying material thicknesses, any weight reductions you can get are generally tiny and not worth the fuss. Use min strength values for any material and be happy.

    Material Characteristics?
    Steel density is 0.282-0.283 lb/in^3, E is 30 E6 psi. In the deep dark past, we used to use 29E6, but somewhere in there, the world settled on 30E6.
    Aluminum density is 0.100, E is 10.3 E6.

    I have included the latest version of the steel and aluminum tables. Thanks to Adrian for noticing my typos. Note that I omitted 1-1/4 x 0.065 aluminum. 1-1/4 x 0.058 does the job on everything and is lighter.

    On to square tubes.

    Billski
     

    Attached Files:

    dino and AdrianS like this.
  9. Nov 9, 2019 #29

    AdrianS

    AdrianS

    AdrianS

    Well-Known Member

    Joined:
    Jul 5, 2014
    Messages:
    378
    Likes Received:
    228
    Location:
    Australia
    It's been freezing and raining here, so I stayed indoors all day.
    Big mistake

    Did you know that using data tables and local names, you can end up with cell formulas like =Pi/4*(Osize^2-Isize^2)

    or even
    =IF(Shape="Round",Pi/4*(Osize^2-Isize^2),Osize^2-Isize^2)


    My round tube calcs agree with Billski to a % or so.

    RoundTubes.PNG

    So I did my version of square with the same sizes :)
    SquareTubes.PNG

    That's more than enough for now.
    note : THIS SPREADSHEET MAY CONTAIN ERRORS!

    [Edit] removed the spreadsheet, because I just thought of a way to massively simplify the data table.
     
    Last edited: Nov 9, 2019
    dino and wsimpso1 like this.
  10. Nov 9, 2019 #30

    Pops

    Pops

    Pops

    Well-Known Member

    Joined:
    Jan 1, 2013
    Messages:
    7,131
    Likes Received:
    6,022
    Location:
    USA.
    Look at the bad things when you take a round 4130 tube and make a streamline tube from it.
     
  11. Nov 9, 2019 #31

    AdrianS

    AdrianS

    AdrianS

    Well-Known Member

    Joined:
    Jul 5, 2014
    Messages:
    378
    Likes Received:
    228
    Location:
    Australia
    You're not going to sucker me into adding asymmetrical shapes...yet.
     
    Pops likes this.
  12. Nov 9, 2019 #32

    Pops

    Pops

    Pops

    Well-Known Member

    Joined:
    Jan 1, 2013
    Messages:
    7,131
    Likes Received:
    6,022
    Location:
    USA.
    Inquiring minds want to know. I can wait. I'm still a half of a mile (804.672 meters) down the road.
     
  13. Nov 10, 2019 #33

    AdrianS

    AdrianS

    AdrianS

    Well-Known Member

    Joined:
    Jul 5, 2014
    Messages:
    378
    Likes Received:
    228
    Location:
    Australia
    This has got to be the first spreadsheet I've some in a looong time that's purely in kings' units.



    So I've got to the stage of having drop down selections for material, shape and section, and a colour-coded results chart like Billski's.
    I've added filtering for the sections, so you can limit the list to what's available.

    What next?

    I can't do carbon without someone providing more information : I'm a computer guy - manipulating formulas is my thing, not deriving them.

    I could do rectangles and ellipses, but I don't really see the point.
     
  14. Nov 10, 2019 #34

    wanttobuild

    wanttobuild

    wanttobuild

    Well-Known Member

    Joined:
    Jun 13, 2015
    Messages:
    597
    Likes Received:
    57
    Location:
    kuttawa, ky
    AdrianS

    Thank you for this information. Very nice of you sir!
    I believe the plan is to work out the alum to 4130 then move on to carbon. As others see this there is gonna be a lot of conversation. Maybe too much.
    Looking at the results, might sway me toward aluminum. Cherrymax makes a reliable product for attachment.
    People are gonna ask about 6061T6.
     
  15. Nov 10, 2019 #35

    AdrianS

    AdrianS

    AdrianS

    Well-Known Member

    Joined:
    Jul 5, 2014
    Messages:
    378
    Likes Received:
    228
    Location:
    Australia
    I have a "drop down" select from a list for materials, so adding another metal is easy - I could probably add titanium too : why not?

    I'll see if I can get specs for 6061-T6 (and titanium ;)) and add them tomorrow.

    This has been a bit of an Excel learning exercise, but I've enjoyed it. Some things would be easier in later Excel versions, but I'm keeping it as backwards compatible as possible.

    ps[-]The stress/strain theory isn't new to me, but it's the first time I've actually used it.[/-]
    Edit : I clean forgot I used this stuff for checking some flywheel designs.
     
    Last edited: Nov 10, 2019
    wanttobuild likes this.
  16. Nov 10, 2019 #36

    Geraldc

    Geraldc

    Geraldc

    Well-Known Member

    Joined:
    Nov 12, 2011
    Messages:
    134
    Likes Received:
    60
    Location:
    nz
  17. Nov 11, 2019 at 7:41 AM #37

    AdrianS

    AdrianS

    AdrianS

    Well-Known Member

    Joined:
    Jul 5, 2014
    Messages:
    378
    Likes Received:
    228
    Location:
    Australia
    Composites are HARD, because both strength and stiffness are different between tension and compression.

    I've found plausible data on cf including those values, but the math will be a while yet (or I'll just plagiarize Billski)

    I'm off to finish reading the internet.
     
  18. Nov 11, 2019 at 10:38 AM #38

    AdrianS

    AdrianS

    AdrianS

    Well-Known Member

    Joined:
    Jul 5, 2014
    Messages:
    378
    Likes Received:
    228
    Location:
    Australia
    Ok, Titanium is interesting.

    Data from sandvic metals, grade 9 drawn tube :
    Yield = 105,000, Density = 0.160, E = 16,500

    It's stronger and lighter than steel, but much less stiff. So for long tube compression, we need to go up in diameter.


    Round Ti.PNG
    note: I have just assumed 1/8" sizes are available
     
  19. Nov 11, 2019 at 1:13 PM #39

    AdrianS

    AdrianS

    AdrianS

    Well-Known Member

    Joined:
    Jul 5, 2014
    Messages:
    378
    Likes Received:
    228
    Location:
    Australia
    The titanium figures got me thinking : we're comparing tubes for equivalent strength.
    That's not the same as equivalent stiffness.

    I can work out twist at Tlimit for torque : what loading would the bending deflection be?
    I was thinking a cantilever of a given length, with an end weight stressed to load limit, then look at end deflection.
    Is that a reasonable measure?

    ps Billski - I hope you don't mind me hijacking your thread. If you'd rather I pipe down, let me know.
     
  20. Nov 11, 2019 at 5:00 PM #40

    wanttobuild

    wanttobuild

    wanttobuild

    Well-Known Member

    Joined:
    Jun 13, 2015
    Messages:
    597
    Likes Received:
    57
    Location:
    kuttawa, ky
    He is working in the background, or raking leaves, carry on!
     

Share This Page

arrow_white