Application Center - Maplesoft

App Preview:

Optimizing the Design of a Coil Spring

You can switch back to the summary page by clicking here.

Learn about Maple
Download Application




Optimizing the Design of a Helical Spring

Introduction

The design optimization of helical springs is of considerable engineering interest, and demands strong solvers. While the number of constraints is small, the coil and wire diameters are raised to higher powers; this makes the optimization difficult for gradient-based solvers working in standard floating-point precision; a larger number of working digits is needed.

 

Maple lets you increase the number of digits used in calculations; hence numerically difficult problems, like this, can be solved.

 

This application minimizes the mass of a helical spring. The constraints include the minimum deflection, the minimum surge wave frequency, the maximum stress, and a loading condition.

The design variables are the diameter of the wire d, the outside diameter of the spring D, and the number of coils N.

 

Reference: "Introduction to Optimum Design", Jasbir S. Arora, 3 Edition 2012.

 

restart; _local(gamma)

Parameters

Gravitational constant (in s)

g := 386

Weight Density of spring material (lb in)

gamma := .285

Shear Modulus (lb in)

G := 1.15*10^7

Mass density of material (lb s in)

rho := gamma/g

Allowable shear stress (lb in)

`τ__a` := 80000

Number of inactive coils

Q := 2

Applied Load (lb)

P := 10

Minimum spring deflection (in)

Delta := .5

lower limit of surge wave frequency (Hz)

`ω__0` := 100

Limit on outer diameter of Coil (in)

D__0 := 1.5

Engineering Relationships

Spring Constant

K := d^4*G/(8*D^3*N)

Shear stress

tau := 8*k*P*D/(Pi*d^3)

Wahl stress concentration factor

k := (4*D-d)/(4*(D-d))+.615*d/D

Frequency of surge waves

omega := d*sqrt(G/(2*rho))/(2*Pi*N*D^2)

 

Constraints

Minimum deflection.

cons1 := P/K >= Delta

.5 <= 0.6956521739e-5*D^3*N/d^4

(4.1)

The outer diameter of the spring should smaller than or equal to D0

cons2 := D+d <= D__0

D+d <= 1.5

(4.2)

Avoid resonance by making the requency of surge waves along as spring as greater than a minimum defined value.

cons3 := omega >= `&omega;__0`

100 <= 44124.02775*d/(Pi*N*D^2)

(4.3)

The shear stress cannot exceed the allowable shear stress.

cons4 := tau <= `&tau;__a`

80*((4*D-d)/(4*D-4*d)+.615*d/D)*D/(Pi*d^3) <= 80000

(4.4)

Collect all  the constraints

cons := {cons1, cons2, cons3, cons4}

{100 <= 44124.02775*d/(Pi*N*D^2), .5 <= 0.6956521739e-5*D^3*N/d^4, 80*((4*D-d)/(4*D-4*d)+.615*d/D)*D/(Pi*d^3) <= 80000, D+d <= 1.5}

(4.5)

Objective function

Mass of spring

mass := proc (d, D, N) options operator, arrow; ((1/4)*N+(1/4)*Q)*Pi^2*D*d^2*rho end proc

 

Optimization

bounds := N = 2 .. 15, d = 0.5e-1 .. 2, D = .25 .. D__0

Digits := 20

Hence the optimized design variables are

Optimization:-Minimize(mass(d, D, N), cons, bounds, iterationlimit = 10^5)

[0.23096520750490060772e-4, [D = .35700930780104652629, N = 11.285519806287015866, d = 0.51700599812014672263e-1]]

(6.1)