Latest / Mechanical Engineering Made Simple / (#87) Combined Stress Core and Beam Deflection

Transcript

(#87) Combined Stress Core and Beam Deflection · Mechanical Engineering Made Simple

• 0:00
  Welcome back design junkies, to mechanical Engineering Made
• 0:02
  Simple. Today we are moving beyond the
• 0:05
  the simple one-dimensional analysis.
• 0:07
  We're sort of taking the safety goggles off and diving into the
• 0:09
  real world. A world where structural members
• 0:12
  rarely fail because of just one single mode of stress acting
• 0:15
  alone. Exactly.
• 0:17
  For too long, you know, foundational mechanics teaches
• 0:19
  us to solve for peer tension, pure bending, or pure
• 0:22
  compression. But in reality, members often
• 0:24
  face all of these at the same time.
• 0:26
  Our mission today is to look at structural integrity through two
• 0:29
  critical integrated lenses. First, the concept of combined
• 0:32
  stresses, which really determines a member's ultimate
• 0:35
  strength. And 2nd, a deep analysis of beam
• 0:38
  deflection, which governs its critical performance.
• 0:40
  Or you could say its stiffness. We've pulled from foundational
• 0:44
  mechanics texts focusing on how we calculate the resultant
• 0:47
  behavior when members are subjected to both axial forces.
• 0:51
  That's thrust or pull and bending moments simultaneously.
• 0:54
  We're giving you the shortcut. OK, let's unpack this combined
• 0:56
  stress situation first, because it's, I mean, it's incredibly
• 0:59
  common. Think of a short stocky column
• 1:02
  or strut in a building. It's primarily holding up
• 1:05
  weight, which is a big axial compressive load P But maybe
• 1:09
  it's not loaded perfectly in the center.
• 1:11
  Or, you know, perhaps there's persistent wind pushing on it,
• 1:14
  inducing A bending moment M. You just can't analyze the
• 1:17
  effect of P&M in isolation, right?
• 1:20
  If you try to treat them separately, the math is simple,
• 1:23
  but it's wrong. P acts down the center, causing
• 1:26
  uniform compression. M causes that gradient of stress
• 1:29
  tension on one side, compression on the other.
• 1:31
  But when they combine, the resultant longitudinal stress,
• 1:35
  which we call Sigma, is the algebraic sum of the two
• 1:38
  stresses acting at any given point across the cross section.
• 1:41
  So if we look at the direct stress from P alone, that's
• 1:44
  purely compressive, uniformly distributed, and it's equal to
• 1:47
  the load P divided by the area A.
• 1:49
  That's the baseline. Correct the safety net as you
• 1:51
  said, and if MX alone you get the familiar bending stress
• 1:55
  formula M * y which is the distance from the centroid all
• 2:00
  divided by I sub X the second moment of area.
• 2:03
  And that stress is what creates the tension and compression
• 2:06
  symmetrically around the neutral axis.
• 2:08
  OK, so here is where it gets interesting and a bit algebraic.
• 2:12
  When you add them together, the resultant stress Sigma is
• 2:16
  negative P / a ± m Y over I sub X.
• 2:21
  We use the negative sign on that first term because by convention
• 2:24
  compression is negative, so that P / A is constantly trying to
• 2:27
  keep the whole section pushed together.
• 2:29
  And this combination leads to really two major outcomes for
• 2:32
  the internal stress distribution.
• 2:34
  In the 1st and most desirable case, the direct compressive
• 2:37
  stress, that P / A component is so large that it just overpowers
• 2:41
  the tensile part of the bending stress.
• 2:43
  This means the entire craft section remains compressive,
• 2:45
  which is ideal for materials like.
• 2:47
  Like concrete? Exactly like concrete.
• 2:48
  But what if that bending moment M starts the Dom?
• 2:51
  The equation maybe because P is smaller or the beam is more
• 2:54
  slender. Then you get the second, more
• 2:56
  dangerous outcome. The stress on one side of the
• 2:59
  member can become zero, or even worse, it can become tensile.
• 3:03
  The engineering consequence here is huge, Means the neutral axis,
• 3:07
  that line of 0 strain is no longer at the physical center of
• 3:10
  the beam. It's been pushed, maybe even
• 3:13
  outside the section entirely, or it's migrated way closer to the
• 3:16
  side with the highest compression.
• 3:18
  It just shows how much the bending moment has distorted
• 3:21
  everything. That immediately makes me think
• 3:23
  of eccentric thrust, because sometimes the bending moment
• 3:26
  isn't from a separate force like wind, but from the load P itself
• 3:30
  not being applied right down the middle.
• 3:31
  So what if P is applied at a distance, let's say E sub Y from
• 3:35
  the centroid C? That is the beautiful
• 3:38
  mathematical simplification that eccentric load P is statically
• 3:42
  equivalent to two things happening at once.
• 3:44
  It's an axial for us P plus an induced bending moment M that
• 3:47
  equals P * e sub Y. So we don't need a new set of
• 3:50
  equations. Not at all.
• 3:52
  We just use the combined stress equation we just talked about
• 3:54
  and we substitute PE sub Y in for M.
• 3:57
  It's elegant. This leads to one of the most
• 3:59
  critical insights for civil and Mechanical Engineers we know.
• 4:03
  Materials like masonry, stone, or unreinforced concrete are,
• 4:08
  well, they're terrible intentions.
• 4:09
  So if we need to prevent tensile stress from developing at all,
• 4:13
  there must be some kind of spatial limit on this
• 4:15
  eccentricity on east sub Y. Exactly.
• 4:17
  We need to prevent the bending stress component from ever
• 4:20
  overcoming the direct compressive stress.
• 4:23
  For a simple rectangular cross section with depth H you can
• 4:27
  completely avoid tensile stresses only if that
• 4:30
  eccentricity E sub Y is less than or equal to H / 6.
• 4:35
  H / 6 That sounds like a tiny sliver right in the center.
• 4:38
  Is that really where the load has to stay?
• 4:40
  That region is formally defined as the core of the section.
• 4:43
  If the applied load P stays strictly within this core, the
• 4:46
  resulting combined stress guarantees that the entire
• 4:48
  section stays under compression. But if you let that load step
• 4:51
  outside that boundary even slightly, you will inevitably
• 4:55
  induce tensile stresses somewhere in the member, and
• 4:58
  that leads to micro cracking and eventual failure.
• 5:00
  And the geometry of this core, it changes dramatically
• 5:04
  depending on the member shape, right?
• 5:06
  It does, and it really highlights the challenge of
• 5:07
  geometric design. For that rectangular section,
• 5:11
  the core is a diamond shape or a parallelogram with diagonals of
• 5:15
  H / 3 and B / 3. But if you have a circular cross
• 5:19
  section, common in piles or chimneys with radius R, the core
• 5:23
  shrinks even further. It's a small circle with a
• 5:26
  radius of just r / 4. R / 4 I mean, that means in a
• 5:30
  column that's a meter wide, your load has to stay within a 12 1/2
• 5:35
  centimeter radius of the center. That is a tight constraint,
• 5:38
  especially when you factor in construction tolerances or
• 5:40
  dynamic loads. It's the non negotiable safe
• 5:43
  zone, and to give you a feel for the magnitude, one of our
• 5:45
  sources ran a problem on a massive masonry pier, 3 meters
• 5:49
  by two meters, subjected to 1000 kilonewtons.
• 5:52
  Even with what seemed like minor eccentricities, the resulting
• 5:55
  calculations showed a maximum tensile stress of 0.008 Mega
• 6:00
  Newtons per square meter. That sounds small, but in
• 6:02
  concrete or masonry 0.008 mega newtons per square meter of
• 6:06
  tension is basically a crack forming.
• 6:08
  Precisely and at the same time, the maximum compressive stress
• 6:13
  was only -0.342 mega newtons per square meter, so the data
• 6:17
  confirms it. In materials with basically no
• 6:20
  tensile capacity, even a tiny deviation from true axial
• 6:24
  loading can start a failure, because that slight eccentricity
• 6:27
  will pull the load outside the core and induce tension.
• 6:30
  Which brings us to a really revolutionary application of
• 6:33
  this pre stressed concrete. Since concrete is so notoriously
• 6:38
  weak in tension, how does prostressing use this analysis
• 6:41
  to essentially, I guess, trick the material into behaving
• 6:45
  better? Pris Crest concrete is
• 6:47
  engineering genius and it's based directly on that combined
• 6:50
  stress equation. You apply a massive permanent
• 6:53
  axial pre compression force P to the beam.
• 6:56
  Just acts as a constant uniform compressive floor.
• 6:59
  This pre compression counters the tensile stresses that the
• 7:02
  eventual service bending moment M would naturally induce your
• 7:05
  building in a permanent compressive buffer.
• 7:07
  Wait, if we're trying to avoid tension entirely, why not just
• 7:10
  use that simple axial pre compression every time?
• 7:12
  Why bother with the incredible complexity of eccentric pre
• 7:15
  compression? That's a great question, and
• 7:17
  it's all about optimization and efficiency.
• 7:19
  Simple axial compression is inefficient.
• 7:22
  It wastes compressive force on the top fibers which will get
• 7:25
  compressed by the service load M anyway.
• 7:28
  Eccentric pre compression is the superior solution because it
• 7:31
  custom tailors the internal stress state before the service
• 7:34
  load even arrives. How does that tailoring work
• 7:36
  exactly? Well, if you apply the pre
• 7:38
  compression thrust P on the edge of the core, specifically at
• 7:42
  that H / 6 location below the center line, you create a very
• 7:46
  specific internal stress state before any service load is
• 7:49
  applied. Because P is acting right at the
• 7:51
  boundary of the core, the upper fibers are at near 0 stress
• 7:55
  while the lower fibers are at a maximum high compression,
• 7:58
  specifically two P / A. So you're preemptively
• 8:01
  strengthening the exact region where the future tensile stress
• 8:04
  will hit the hardest, and you're leaving the top fibers neutral.
• 8:07
  Precisely so when the structures own weight and external service
• 8:11
  loads induce a typical sagging moment M that moment tries to
• 8:15
  pull the bottom fibers into tension and push the top into
• 8:18
  compression. But because we preloaded the
• 8:21
  bottom fibers with all that compression, we can perfectly
• 8:24
  balance the resultant stress. We keep all stresses safely
• 8:27
  within the concrete's allowed compressive limit and ensure 0
• 8:31
  tensile stress. It's incredible.
• 8:33
  So what's the, you know, the bottom line difference in
• 8:35
  performance between the two methods?
• 8:37
  The. Key insight is that pre
• 8:39
  compression by an eccentric load gives a considerably higher
• 8:42
  bending strength than simple axial pre compression.
• 8:45
  Considerably higher. In one comparison, the eccentric
• 8:48
  method allowed for nearly double the maximum allowable bending
• 8:50
  moment. It moves pre stressing from just
• 8:53
  being a safe solution to an incredibly efficient 1.
• 8:56
  That really wraps up the strength side of the equation.
• 8:58
  The internal stresses, the material limits, but as
• 9:00
  engineers we know strength isn't enough.
• 9:03
  We also have to ensure stiffness so the beam doesn't deflect too
• 9:06
  much and, you know, fail functionally.
• 9:08
  So now we need to pivot and study deformation.
• 9:11
  That's the critical link, the link between internal stress
• 9:14
  capacity and external performance.
• 9:16
  To analyze stiffness, we have to rely on the fundamental
• 9:19
  governing equation for the elastic bending of straight
• 9:22
  beams. It's EI multiplied by the 2nd
• 9:25
  derivative of V with respect to Z ^2 = -m.
• 9:30
  OK, let's clarify those variables for a second.
• 9:32
  M is the bending moment, Z is the distance along the beam, and
• 9:35
  VV is the vertical displacement or deflection.
• 9:38
  That's what we're trying to minimize.
• 9:39
  In that product, E times I is the most important measure.
• 9:42
  Here it's the product of Young's modulus E, the material
• 9:46
  stiffness and the second moment of area I, the geometry's
• 9:49
  contribution. This EI product is what we call
• 9:52
  the beams flexural stiffness. Think of it as the beams
• 9:55
  inherent resistance to bending. If you double I or E, you double
• 9:59
  the stiffness. And by integrating this
• 10:00
  governing equation twice, we get those quick reference formulas
• 10:03
  that every structural engineer keeps handy.
• 10:06
  The fast checks, for example, a simply supported beam with a
• 10:09
  uniform load W spread across it. The Max deflection V is five WL
• 10:13
  to the 4th divided by 384 EI. That exponent on L is what's
• 10:17
  terrifying for designers. Since the length is raised to
• 10:20
  the 4th power, even a small increase in the span lengths
• 10:23
  just crushes the stiffness. The source material showed a 6m
• 10:27
  beam with A10 kilonewtons per meter load.
• 10:31
  It's Max deflection was calculated to be 0.00844 meters.
• 10:37
  Sounds small, but if your design limit is, say, 150th of the
• 10:40
  span, that number is critical. And for cantilever beam the
• 10:45
  classic balcony or shelf with a concentrated load W on the end.
• 10:49
  The Max deflection is WL cubed divided by three EI.
• 10:53
  Again, that length cubed means cantilevers are highly sensitive
• 10:56
  to length changes. OK, so direct integration works
• 10:59
  beautifully for those simple cases, but the real world often
• 11:02
  involves complex loading, a mix of distributed loads, multiple
• 11:05
  point loads, moments apply to odd points.
• 11:08
  Right, and analyzing those individually means breaking the
• 11:10
  beam into a dozen separate segments.
• 11:12
  It leads to endless simultaneous equations to solve for all the
• 11:15
  constants. It's a mess.
• 11:17
  That's where the elegant mathematical solution comes in.
• 11:20
  Macaulay's method. Anyone who's tried to solve a
• 11:23
  complex deflection problem late on a Friday knows just how
• 11:27
  powerful this technique is. It really is a brilliant bit of
• 11:30
  algebra. Macaulay's method lets you write
• 11:32
  a single continuous bending moment equation that's valid for
• 11:36
  the entire beam length, no matter how many point loads or
• 11:38
  moment changes you have. It just eliminates the need for
• 11:41
  all that separate segmented analysis.
• 11:44
  The trick is that famous Macaulay bracket, right?
• 11:47
  It's written like a open bracket Z -, a close bracket to the
• 11:50
  power N When you integrate, you treat the expression inside
• 11:53
  those brackets just like any other term.
• 11:55
  But here's the clever bit. If the term inside the bracket
• 11:58
  becomes negative, which means the section you're analyzing is
• 12:01
  before the load at distance a, you just algebraically ignore
• 12:04
  that entire term. You set it to 0.
• 12:06
  Think of that Macaulay bracket as a gate or a software switch.
• 12:10
  The equation is running down the beam, but the switch stays off,
• 12:13
  contributing nothing until you physically get to the point
• 12:16
  where that specific load W at distance A starts to have an
• 12:21
  effect. Once you pass A, the switch
• 12:24
  flips on and that force gets integrated as normal.
• 12:27
  It's a systematic approach that just simplifies everything.
• 12:29
  It minimizes the number of arbitrary constants and gives
• 12:32
  you 1 continuous equation for both slope and deflection across
• 12:36
  the whole structure. But we have to move beyond the
• 12:38
  ideal. The real world doesn't always
• 12:40
  give us uniform cross sections. What happens when I second
• 12:44
  moment of area changes along the length?
• 12:47
  It happens a lot in weight optimizer taper designs.
• 12:50
  So if I isn't constant, that EI term becomes a function of Z.
• 12:53
  Exactly. So in that case, EI is no longer
• 12:55
  a constant you can pull out of the integral.
• 12:58
  The differential equation now requires you to integrate the
• 13:00
  ratio of M over I along the length instead of just
• 13:03
  integrating M. This is vital for, say, columns
• 13:06
  or beams designed for specific nonlinear performance.
• 13:09
  One design application in the sources showed just how precise
• 13:12
  this can get. It was a cantilever strip with
• 13:15
  constant breadth but a variable thickness T designed to bend
• 13:20
  into a perfect circular arc, which implies constant curvature
• 13:23
  along its length. And to get that constant
• 13:26
  curvature, the second moment of area I had to be proportional to
• 13:30
  zed, the distance from the free end.
• 13:32
  Since I is proportional to the thickness cubed t ^3, this meant
• 13:35
  the thickness T itself had to vary so that t ^3 was
• 13:38
  proportional to zed. It's a stunning example of
• 13:40
  tailoring material based on a deflection requirement.
• 13:43
  And finally, we've focused almost entirely on bending
• 13:47
  deflection. We've assumed that almost all
• 13:49
  the vertical displacement comes from the moment M.
• 13:51
  But we can't forget about shearing deflection.
• 13:54
  Does that truly matter in the big picture?
• 13:56
  It depends entirely on the geometry, but yes, it can matter
• 13:59
  significantly. All those previous calculations
• 14:02
  assumed the shear force had a negligible impact on
• 14:05
  deformation, and that holds true for long slender beams.
• 14:09
  But shear stress also causes its own lateral deformation.
• 14:13
  The shear strain gamma is proportional to the shear stress
• 14:16
  at the neutral axis divided by the shearing modulus G.
• 14:20
  This causes additional vertical deflection.
• 14:22
  So when does this shear deflection stop being
• 14:25
  negligible? When we have to worry about it.
• 14:27
  The ratio becomes significant for deep beams where the ratio
• 14:30
  of depth H to length L is large. If a beam is short and squat,
• 14:34
  the shear forces have a much stronger influence on the
• 14:37
  overall shape change than the bending moment does.
• 14:39
  There was a specific data point from one of the design problems
• 14:43
  that should, I think, shock every engineer listening.
• 14:45
  It was a timber beam carrying a concentrated load.
• 14:48
  When they analyzed it, the total shear deflection was calculated
• 14:51
  to be nearly 30% of the total bending deflection. 30% That is
• 14:56
  an enormous amount of unaccounted displacement.
• 14:59
  It proves that depending on the material, especially softer
• 15:02
  materials like timber and the geometry, neglecting shear can
• 15:06
  lead to seriously underestimating the total
• 15:08
  displacement. If you ignore that 30%, you have
• 15:12
  fundamentally jeopardized the required stiffness and
• 15:14
  performance of the structure. That was quite a journey.
• 15:17
  Today we started by tackling internal strength, understanding
• 15:20
  how compression and bending combine, defining that non
• 15:23
  negotiable region, the core of the section to avoid tension.
• 15:27
  Then we moved into calculating structural performance using
• 15:30
  flexural stiffness EI and using powerful tools like Macaulay's
• 15:34
  method to manage complex deflection equations.
• 15:37
  And we learned that strength isn't just about the macro and
• 15:40
  stress, it's also about setting up a favorable stress state,
• 15:43
  like we saw with the sheer optimization of eccentric pre
• 15:45
  stressing. It's an elegant application of
• 15:48
  combined stress theory that nearly double S the bending
• 15:51
  capacity for the same material. So as you move forward and apply
• 15:54
  this knowledge, here's a final provocative thought for you to
• 15:57
  explore. We found that the core of a
• 15:59
  circular section is r / 4. That limit is critical for
• 16:03
  static structures, but it becomes exponentially more
• 16:06
  important for dynamic ones. So how does understanding this
• 16:09
  limiting geometry influence modern design standards for
• 16:12
  things like concrete piles or, say, tall, slender structures
• 16:16
  like chimney stacks that are constantly subjected to swaying
• 16:19
  forces from acceleration or highwind that are divided by 4?
• 16:23
  Limit is the key boundary that dictates whether the entire
• 16:25
  structure stays in safe compression or begins to tear
• 16:28
  itself apart with tension. Always more to learn.
• 16:30
  Thanks for diving deep with us.