0% found this document useful (0 votes)
112 views27 pages

Basic Structural Design For Formula Sae: Dave Dejohn

Formula SAE covers design and analysis fundamentals with FEA or hand calculations. Material selection based on physical and mechanical properties (no chemistry) design metrics if your goal is to finish well, the measure is competition points.

Uploaded by

Nathan MacIntire
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
112 views27 pages

Basic Structural Design For Formula Sae: Dave Dejohn

Formula SAE covers design and analysis fundamentals with FEA or hand calculations. Material selection based on physical and mechanical properties (no chemistry) design metrics if your goal is to finish well, the measure is competition points.

Uploaded by

Nathan MacIntire
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 27

Basic Structural Design for

Formula SAE
Dave DeJohn
[email protected]
Purpose
• Cover design and analysis fundamentals
with FEA or hand calculations
– Take a crash course, idealistic approach
– Give a systematic method
– Stay general – we can discuss specific parts
of the car in another session
• Discuss material selection based on
physical and mechanical properties (no
chemistry)
Design metrics
• If your goal is to finish well, the measure is competition points
• Break it down:

Criteria Measured by:


Performance Acceleration time
Skidpad time
Autocross time
Endurance time
Fuel mileage
Cost Design report formulas
Actual budget
Safety Rules requirements

Durability Time between service


Time to complete service

• Design decisions often improve one metric at the expense of another. For a
cohesive design the leads must have the same priorities.
– It’s easy to lose focus!
• Engineers should think quantitatively – how much faster? How much more
expensive? How many points does that work out to?
• Structural design influences all of these categories.
Structural design and analysis
happens on a sliding scale…
Simpler More Complex

Design Analysis Prediction


•Simplified methods •More advanced •Most advanced
methods methods
•Bounding material
properties •Bounding •Account for
material properties material scatter
•Parametric
•More accurate •Most parameters
•Focus on comparing
answer, but more are fixed
options, rather than the
parameters are
“correct answer” •Probabilistic –
fixed
likelihood of failure

For Formula SAE, stay to the left side of the scale


Refresher - basic material behavior
• The point of most structural analysis is to use properties from a tension
specimen to design… everything else!

Ultimate Strength

Yield Strength Elongation


Stress:
at fracture
Force
Area

Elastic Modulus

Strain: Change in Length


Length
Trends in “low strength” vs “high
strength” material (same family)

Stress:
Force
Area

Strain: Change in Length


Length
Brittle Fracture
• The materials you use will usually only experience “nonductile fracture” –
failure without permanent deformation – if they are cracked or have
manufacturing defects

Stress:
Force
Area

Strain: Change in Length


Length
FEA Considerations
Default is linear elastic behavior;
For stress > yield strength,
•Overpredicts stress (& strength)
•Underpredicts strain (& deflection)

Stress:
Force
Area

Strain: Change in Length


Length
FEA Considerations
Default is linear elastic behavior;
For stress > yield strength,
•Overpredicts stress (& strength)
•Underpredicts strain (& deflection)

Stress:
Force Elastic-Perfectly Plastic is a very practical simplification
Area

Strain: Change in Length


Length
FEA Considerations
Default is linear elastic behavior;
For stress > yield strength,
•Overpredicts stress (& strength)
•Underpredicts strain (& deflection)

Stress:
Force Elastic-Perfectly Plastic is a very practical
Area simplification

Elastic-Plastic analysis is most advanced; is


in terms of true stress-strain, rather than
engineering stress-strain
• ultimate strength “disappears”

Strain: Change in Length


Length
Nominal stress and peak stress
F

M T Nominal Stress Peak Stress


V
Tension ST = P/A σT = kTST (or FEA)

Bending SB = Mc/I σB = kBSB (or FEA)


“Gross Area”

Stress Transverse Tv = VQ/(It) τV (use FEA)


Concentration Shear

Torsional Shear Tt = Tc/J τt (use FEA)


“Nominal Area”
or “Net Section”
(A, I, J, t)
• FEA gives peak stress, usually as the
combined effect of all these load types.

• BUT, peak stress is often not a good


indicator of when or how the part will fail
V
M • When is nominal stress more important, &
how do you “extract” it from FEA?
F T
We will focus on these two
challenges in structural design:
1. Peak stress is often not a good predictor
of when or how the part will fail
2. Sometimes stronger materials make
weaker parts

To overcome these challenges you need to


understand the basic failure modes
Design checks and failure modes
1. Load Carrying Capacity
2. Elastic Stiffness
3. Buckling Capacity
4. Fatigue Life
5. Brittle Fracture
• Due to manufacturing defects
• Due to fatigue cracks
Load Carrying Capacity
What it is:
• When the combined nominal stress due to the load conditions reaches the yield strength, the part
will begin to permanently deform
• If the loads continue to increase, the part will collapse or fracture
– Note: the ultimate strength (Su or tensile strength) is only a good predictor of fracture load for
pure tension; but we still use it as a rule of thumb for material choice

What is the critical material property?


• Yield strength, Sy, used with appropriate safety factors
• Or, “design strength:” lesser of 2/3(Sy) or 1/3(Su)
• Thermal stress doesn’t affect load carrying capacity, but higher temperature will decrease strength

Hand calculation methods:


• “Effective stress” < 2/3(Sy) or 1/3(Su)
For most of your cases, this simplification will work:
2 2
1  P Mc   V Tc 
S EFFECTIVE      3  
2 A I  A J 

• Partial allowables (allows different safety factors for the loading conditions)
P  Mc   V Tc 
      
 A  I  A J 
   WEB 1
min(0.60Sy,0.50Sy) 0.66Sy (tube) or 0.75Sy (solid) 0.40Sy

FEA methods:
• Stress “linearization” or “planarization” (old school)
• Limit load analysis – use elastic-perfectly plastic properties, ramp load. plot deflection vs load
Load Carrying Capacity t
t

d d h h

Tension / Fty 

4
d2y Fty 

4
d 2
 d  2t   y
2
 Fty  h 2  y Fty  h 2  h  2t 
2

Compression
Fty*   d t  y Fty*  4 h t  y

Bending    d  2t 4   1 3 1 h  2t 4  
My  d3 y My  d3  My  h y M y   h 3 
32 32  d  y 6 6 h  y
(Elastic Limit)  
 4
M *y  d 2t  y M *y  h 2 t  y
4 3

(Plastic
M p  1.70 M y M p  1.27 M y M p  1.50 M y M p  1.12 M y
Reserve)


Transverse
Shear
Fsy  d 2  0.57 y  Fsy 
4
d t  0.57 y  Fsy  h 2 0.57 y   
Fsy  2ht  4t 2 0.57 y 
4 3


d 0.57 y 
  d  2t  
4
 0.57 y  Ty  0.208h 3 0.57 y 
Ty  Ty   Ty  2t h  t  0.57 y 
3 3 2
d
16 16  d

Torsion

Ty  d 2t 0.57 y 
2
Elastic Stiffness
What it is:
• Even if nominal stress is kept in the elastic range, the part is not stiff enough to:
– Function properly
– Keep loading “quasi-static,” or prevent load amplification due to dynamic conditions

What is the critical material property?


• Elastic modulus, E

Hand calculation methods:


• Deflection Stiffness
Tension δ = PL/(AE) k = AE/L
Bending δ = ML2/(CEI) k = CEI/Lx
x = 2 for moment, 3 for point load,
4 for distributed load)
Torsion δ = TL/(GJ) k = GJ/L (radians)
G = E/2.6

• Keep period of natural frequency, T = 2π/sqrt(k/m) where m = supported weight/(gravity


constant), less than load application time
• Try to keep similar stiffness for connected parts

FEA methods:
• For linear elastic analysis, k = Load / Deflection
• Natural frequency (eigenvalue) extraction
Elastic Stiffness
Load amplification due to dynamic conditions

2
Max Dynamic Load
Static Load

~120%
~115% ~110%
1
1 2 3 4
Load Ramp Time
Period of Natural Frequency
Buckling Capacity
What it is:
• Characteristics of load carrying capacity and elastic stiffness – as the nominal compressive stress increases,
the deflection response suddenly becomes nonlinear… but determined by elastic modulus, not strength
• Structural equivalent of standing on one foot

What is the critical material property?


• Elastic modulus, E

Hand calculation methods:


• Euler column formulas
• AISC/Column safety modifications; L is effective length, r is radius of gyration, safety factor of roughly 2:

PC Sy Sy 2  L  L 2 2 E
    if 
A 1.92 7.68 2E  r  r Sy
FEA methods: PC  2E L 2 2 E
 if 
• Bifurcation analysis – elastic only A L
2
r Sy
1.92 
•  r  geometry (large strain, large displacement)
Elastic-plastic analysis accounting for nonlinear
• Be careful – FEA might not pick-up real life concerns like eccentricity, friction, etc.
Fatigue Life
What it is:
• Repeated peak stress (or peak strain) reversals cause small cracks to initiate and grow,
especially in areas of stress concentrations
• Eventually a crack grows large enough that a part loses its load carrying capacity (collapses) or
experiences brittle fracture or ductile tearing

What is the critical material property?


• For rotating parts: the endurance limit, S E (probably with a safety factor)
• For nonrotating parts: fatigue curves - stress vs cycles or strain vs cycles
– Probabilistic; recognize that data scatter, surface finish, part size, environment, and other
factors can reduce predicted life by a factor of 20!

Hand calculation methods:


• Compare kt and ΔS to stress-life curves
• Compare ktΔS/E to strain life curve, if ktΔS < Sy (peak stress less than yield)
• Compare (ktΔS)2/(SyE) to strain life curve, if ktΔS > Sy (peak stress greater than yield)

FEA methods:
• Linear elastic analysis:
– Compare Δσ to unnotched and notched stress-life curves
– Compare Δσ/E to strain life curve, if Δσ < Sy (peak stress less than yield)
– Compare (Δσ)2/(SyE) to strain life curve, if Δσ > Sy (peak stress greater than yield)
• Elastic-plastic analysis: compare Δε to strain-life curve (tricky)
Fatigue Life
Even though fatigue cracking is driven by peak stress, the difference between the peak stress and the
nominal stress is still a factor due to notch sensitivity
ΔF ΔF 2ΔF

kt = 2

ΔF ΔF 2ΔF
Multiply “Maximum Stress” by
Shorter average life 2 before comparing to FEA!

4130 Steel Sheet


Bar 3
ΔF/A = 60 ksi

← unnotched
Bar 1
Bar 2
notched, kt = 2.0 →
Alternating Peak Strain (log scale) Fatigue Life

“Weaker” Material

“Stronger” Material

0.01

1000 Cycles to Failure (log scale)

Approximate conversion from stress-life to strain-life: Δε = (ktΔS)2/(SyE)


Brittle Fracture
What it is:
• As the nominal or peak tensile stress increases, a part made from a normally ductile material
suddenly fractures or tears due to a manufacturing defect or fatigue crack
• Not likely to be an issue for steels, but could be for aluminum

What is the critical material property?


• Critical plane-strain fracture toughness, K IC

Hand calculation methods:


• Calculate stress intensity factor, KI, based on crack size, a, and geometry parameters, C and Q
a
K I  C  K Ic
Q

• Prediction can still be difficult, but a useful simplification to compare materials is:
2
1K 
acritical   Ic 
  Sy 

• Fracture is less likely if the material is less than a critical thickness


2
K 
tcritical  1.0 to 2.5 Ic 
 Sy 

FEA methods:
• Complicated
Material Comparison
Fully-
70ºF 70ºF 70ºF 70ºF Max
Reversed Critical
Design Fracture Thickness to
Density Yield Ultimate % Elastic Peak Stress, Defect
Nominal Toughness Avoid
(lbs/in) Strength Strength Elongation Modulus Nf = 30,000 Size
Strength (psi√inch) Fracture
(psi) (psi) (psi) cycles (inch)
(psi) (inch)
(psi)

65,000 psi
4130 N/A if
(kt = 1) N/A if T> N/A if T>
Steel 0.283 75,000 90,000 30,000 8-12% 29,000,000 T>
Sheet 160,000 psi -100ºF -100ºF
-100ºF
(kt = 5)

7075-T6/ 46,000 psi


T651 (kt = 1) 0.117 –
Aluminum 0.101 70,000 80,000 26,600 5-7% 10,300,000 24,000 0.034
60,000 psi 0.294
Plate (kt = 5)
Material Comparison
Fully-
70ºF 70ºF 70ºF 70ºF Max
Reversed Critical
Design Fracture Thickness to
Density Yield Ultimate % Elastic Peak Stress, Defect
Nominal Toughness Avoid
(lbs/in) Strength Strength Elongation Modulus Nf = 30,000 Size
Strength (psi√inch) Fracture
(psi) (psi) (psi) cycles (inch)
(psi) (inch)
(psi)

65,000 psi
4130 N/A if
(kt = 1) N/A if T> N/A if T>
Steel 0.283 75,000 90,000 30,000 8-12% 29,000,000 T>
Sheet 160,000 psi -100ºF -100ºF
-100ºF
(kt = 5)

7075-T6/ 46,000 psi


T651 (kt = 1) 0.117 –
Aluminum 0.101 70,000 80,000 26,600 5-7% 10,300,000 24,000 0.034
60,000 psi 0.294
Plate (kt = 5)

Weight Load Carrying Capacity Overload Elastic Fatigue Life Likelihood


Capacity Stiffness, of Brittle
Buckling Fracture
Capacity
Material Comparison
Conclusions
In terms of the two challenges:
1. Peak stress is often not a good predictor of when or how the part will fail
– Load carrying capacity, elastic stiffness, and buckling capacity are driven by
nominal stress
– Fatigue life is driven by peak stress, but notch sensitivity and material scatter
make life prediction difficult

2. Sometimes stronger materials make weaker parts


– If choosing higher strength material, be careful not to make the part more likely
to fail by making it more notch sensitive or likely to fracture
– Make sure you need the strength (load carrying capacity) – otherwise higher
strength material will not make the part stiffer or less likely to buckle, and may
not improve fatigue life
Books
• “Roark’s Formulas for Stress and Strain” (buy
old and cheap)
• “Fracture and Fatigue Control in Structures” by
Rolfe/Barsom (buy old and cheap)
• “Deformation and Fracture Mechanics of
Engineering Materials” by Hertzberg (buy old
and cheap)
• “Design of Welded Structures” or “Design of
Weldments” by Blodgett (buy new and cheap,
direct from Lincoln Electric)

You might also like