OT: torsion spring strength equivalence: wire diameter vs length
Close
Login to Your Account
Results 1 to 16 of 16
  1. #1
    Join Date
    Aug 2007
    Location
    GA
    Posts
    927
    Post Thanks / Like
    Likes (Given)
    1
    Likes (Received)
    182

    Default OT: torsion spring strength equivalence: wire diameter vs length

    Just broke a garage torsion spring (one of 2). Existing springs are .218 x 2" ID x 25", which is out of stock around here. I can get .225 x 2 x 27; is the additional 2" of coil length enough to compensate for the .007" thickness difference? TOO much? Also, the .225's are 1/3 the price....I've replaced these before with the winding bars, its not bad as long as you have bars that fit the cone holes properly.

  2. #2
    Join Date
    Jan 2005
    Location
    Victoria, Texas, USA
    Posts
    4,832
    Post Thanks / Like
    Likes (Given)
    3515
    Likes (Received)
    1373

    Default

    I say go for it!

    The difference in forces will never be notice when fighting with the drawing up of the springs. Without doing calculations, I doubt stress levels will be any different between the two springs. Would love to see a set of garage door springs made from Inconel X750 wire!

  3. #3
    Join Date
    Mar 2011
    Location
    Geneva Illinois USA
    Posts
    6,631
    Post Thanks / Like
    Likes (Given)
    2974
    Likes (Received)
    2664

    Default

    For helical torsion springs the rate is proportional to the wire diameter to the forth power divided by the number of turns,(d^4)/N. The heavier spring with be about 12-13% stiffer than the old one. This may cause problems with door balance.

    Tom

  4. Likes 4GSR liked this post
  5. #4
    Join Date
    Apr 2018
    Country
    UNITED STATES MINOR OUTLYING ISLANDS
    Posts
    9,201
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    4468

    Default

    Quote Originally Posted by TDegenhart View Post
    This may cause problems with door balance.
    Then bolt two bricks to the bottom ...

  6. Likes johansen liked this post
  7. #5
    Join Date
    Aug 2007
    Location
    GA
    Posts
    927
    Post Thanks / Like
    Likes (Given)
    1
    Likes (Received)
    182

    Default

    Oooh, thanks for that formula. Looks like they have .225x2x29" springs also, only $10 more, which should be an even closer match at +7% over the original .218x2x25's if my math is correct. (is it?) I'll get those...I'm pretty sure I can easily de-tension by that much if I need to. Or bricks, yes, lol.

  8. #6
    Join Date
    Nov 2018
    Country
    UNITED STATES
    State/Province
    Michigan
    Posts
    826
    Post Thanks / Like
    Likes (Given)
    456
    Likes (Received)
    565

    Default

    Not so fast, don't forget to divide by the number of turns.

    .225/.218 = 1.032

    1.032^4 = 1.134 (there's your +13% stiffness based on diameter)

    The 27" long spring is 120/114 turns compared to the original, so +4.6% length. 1.134/1.046 = +8.3% stiffness
    The 29" long spring is 129/114 turns compared to the original, so +12.4% length. 1.134/1.124 = +0.9% stiffness

    The 29" .225" diameter spring is a near exact match if you have the space for it.

    Edit, fixed my numbers to account for diameter difference influencing number of turns along its overall length.

  9. #7
    Join Date
    Aug 2007
    Location
    GA
    Posts
    927
    Post Thanks / Like
    Likes (Given)
    1
    Likes (Received)
    182

    Default

    The 29" .225's are on the way, in the meantime while waiting I cut the old stub off the broken one with a zip wheel, screwed the remaining spring on the cone, tensioned it and it is working fine. I know it won't last long because the spring is so short but it only has to make it a few days

  10. #8
    Join Date
    Sep 2004
    Country
    UNITED STATES
    State/Province
    Wisconsin
    Posts
    276
    Post Thanks / Like
    Likes (Given)
    1
    Likes (Received)
    75

    Default

    Many of those springs are sold by expected number of cycles to failure with 10,000 being about the lowest. Larger diam wire and longer spring should give more cycles. You should weigh the door and get a spring rated for the weight. When tensioning you can count the bars(1/4 turns) to get the balance point close. Make sure to use good quality alloy bars.

  11. #9
    Join Date
    Nov 2003
    Country
    UNITED STATES
    State/Province
    Texas
    Posts
    6,064
    Post Thanks / Like
    Likes (Given)
    219
    Likes (Received)
    2094

    Default

    Any decent garage door company should be able to both recommend the proper replacement and provide it from their stock.

    Whatever you do, be sure to run a safety wire down the center of the springs to prevent pieces from flying at you the next time you have a failure. I used 1/8" stranded steel cable.

  12. #10
    Join Date
    Nov 2019
    Country
    CANADA
    State/Province
    Ontario
    Posts
    687
    Post Thanks / Like
    Likes (Given)
    1104
    Likes (Received)
    167

    Default

    dont use 2 different spings. it will cause nothing but problems opening and closing uneven and getting jammed

  13. #11
    Join Date
    Feb 2010
    Location
    ch
    Posts
    3,381
    Post Thanks / Like
    Likes (Given)
    279
    Likes (Received)
    471

    Default

    are you guys sure about the formula? what about the diameter of the coils?

  14. #12
    Join Date
    Nov 2018
    Country
    UNITED STATES
    State/Province
    Michigan
    Posts
    826
    Post Thanks / Like
    Likes (Given)
    456
    Likes (Received)
    565

    Default

    Quote Originally Posted by dian View Post
    are you guys sure about the formula? what about the diameter of the coils?
    The full equation is E*d^4 / 64*D*N

    Where:

    E = Young's Modulus
    d = Material Diameter
    D = Coil Diameter
    N = Number of Turns

    Young's modulus should cancel out if they are both steel, coil diameter is 2 in both cases, so that'll be fine (no need to account for the .007" difference even if it wasn't made tighter to be a true 2" anyway), 64 is a constant so doesn't matter, and d and N were both accounted with their proper respective powers in the relative stiffness comparison calculation.

    Math checks out.

  15. #13
    Join Date
    Apr 2004
    Location
    Modesto, CA USA
    Posts
    8,801
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    1731

    Default

    Too much math for me to figure out today. Is it really just the length and diameter of the straight rod before it is bent into a coil? Does the fancy math just account for the circumference of each coil and add them up to a total straight length or is there more to it? Car torsion rods are just straight bars with a lever on one end that rotates up and down while the other end is fixed.

  16. #14
    Join Date
    Nov 2018
    Country
    UNITED STATES
    State/Province
    Michigan
    Posts
    826
    Post Thanks / Like
    Likes (Given)
    456
    Likes (Received)
    565

    Default

    Quote Originally Posted by Bill D View Post
    Too much math for me to figure out today. Is it really just the length and diameter of the straight rod before it is bent into a coil? Does the fancy math just account for the circumference of each coil and add them up to a total straight length or is there more to it? Car torsion rods are just straight bars with a lever on one end that rotates up and down while the other end is fixed.
    Yes, all springs are just elastic bending of a rod or bar. That rod or bar can take many forms, but in the case of twisting a torsion spring, it is bending each differential element of the rod (i.e. zoom in until the coil radius is near-straight, and the bending is the same as a straight bar).

    Because of that, it is not actually getting any longer (neither is an extension spring for that matter), they are bending, making the angle between every set of those differential units slightly different. For a torsion spring, this means the radius of the coil will contract slightly to account for the new coils being added to the end of it, preserving total bar length.

    Hope that's clearer than mud for ya. Don't forget, everything's a spring!

  17. #15
    Join Date
    Jan 2011
    Country
    UNITED STATES
    State/Province
    Michigan
    Posts
    236
    Post Thanks / Like
    Likes (Given)
    3
    Likes (Received)
    114

    Default

    Unintended consequences. Your post reminded me to oil the springs and hinges on my door. A little bit of rust on a spring can cause an early failure. Thanks for the post.

  18. #16
    Join Date
    Feb 2010
    Location
    ch
    Posts
    3,381
    Post Thanks / Like
    Likes (Given)
    279
    Likes (Received)
    471

    Default

    Quote Originally Posted by BoxcarPete View Post
    The full equation is E*d^4 / 64*D*N

    Where:

    E = Young's Modulus
    d = Material Diameter
    D = Coil Diameter
    N = Number of Turns

    Young's modulus should cancel out if they are both steel, coil diameter is 2 in both cases, so that'll be fine (no need to account for the .007" difference even if it wasn't made tighter to be a true 2" anyway), 64 is a constant so doesn't matter, and d and N were both accounted with their proper respective powers in the relative stiffness comparison calculation.

    Math checks out.
    i guess i missed the diameter was the same. what would be the formula for a coil spring in tension/compression? same thing?


Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •