Latest / Mechanical Engineering Made Simple / (#36) From Parallel Axes to Perfect Gears: The Precision Blueprint for Designing High-Performance Power Transmission

Transcript

(#36) From Parallel Axes to Perfect Gears: The Precision Blueprint for Designing High-Performance Power Transmission · Mechanical Engineering Made Simple

• 0:00
  Welcome back, design junkies, to Mechanical Engineering Made
• 0:02
  Simple. We are cracking open a pretty
• 0:05
  hefty stack of technical sources.
• 0:07
  Today. We're really going to dive deep
• 0:09
  into the architecture of power transmission.
• 0:11
  We're moving past just, you know, the idea of a spinning
• 0:14
  wheel and getting right into precision gear design, covering
• 0:18
  everything from like basic tooth definitions all the way to
• 0:21
  complex material stress factors. Exactly.
• 0:24
  Our mission really is to boil down this mountain of data,
• 0:28
  extract that foundational knowledge you absolutely need to
• 0:31
  understand, and crucially, to design truly functional, high
• 0:35
  precision gears. We're aiming to build a complete
• 0:38
  engineering blueprint, you know, moving step by step from basic
• 0:41
  categories right through to what makes a gear last.
• 0:44
  And before we get totally lost in the weeds, there's something
• 0:46
  really critical that jumped out from these manuals.
• 0:48
  It kind of sets the tone for everything else.
• 0:50
  The goal isn't just making gears that work.
• 0:52
  The sources really hammer this home.
• 0:54
  Modern engineers have to aim for perfect gears right from the
• 0:57
  start. That's absolutely the key take
• 0:59
  away here. The material is really insistent
• 1:01
  on this. If you ignore the advances in,
• 1:04
  well, scientific gearing theory, if you just simplify things too
• 1:07
  much at the beginning, you end up with what they call
• 1:09
  approximate gears. And these approximate gears,
• 1:13
  they just inherently can't meet today's needs for high power
• 1:17
  density, low noise, low vibration.
• 1:19
  That perfection has to be designed in from the get go.
• 1:22
  OK, so let's unpack this blueprint, starting at the
• 1:25
  highest level, the basic classification.
• 1:28
  How is the whole world of Gears first divided up?
• 1:31
  Based on their axis, right? That's right, it all comes down
• 1:33
  to how the axis of rotation relate to each other in space.
• 1:37
  Three main families. We start with the most
• 1:39
  straightforward parallel axis gears.
• 1:42
  The driving gears axis and the driven gears axis are perfectly
• 1:45
  parallel. And the big one here, the most
• 1:47
  common, that's the spur gear, isn't it?
• 1:48
  Correct. It's cylindrical, teeth are
• 1:50
  straight run parallel to the axis.
• 1:52
  It's the easiest to manufacture, most cost effective, definitely
• 1:55
  the workhorse. But right alongside it you have
• 1:57
  the helical gear, still cylindrical, but the teeth are
• 2:00
  cut in a Helix like a screw thread.
• 2:02
  OK and the advantage there? Well, it.
• 2:04
  Gives you smoother, quieter power transfer, and it can
• 2:07
  generally handle higher load than a spur gear of the same
• 2:11
  size. Sounds like a direct upgrade
• 2:12
  then. Why not use helical gears for
• 2:14
  everything? What's the cat?
• 2:16
  The catch is axial thrust, that Helix angle.
• 2:19
  It creates a force that pushes the gear sideways along its
• 2:23
  axis, so you need bearings designed to handle that thrust.
• 2:27
  Engineers found a way around that too with the double helical
• 2:29
  gear I mentioned. 2 helical gears, one with a right hand
• 2:32
  Helix, one with a left hand machine back-to-back on the same
• 2:36
  shaft. The opposing Helix angles cancel
• 2:39
  out the thrust forces perfectly. Think herring bone pattern.
• 2:42
  Clever. And there is one more in this
• 2:44
  family. Yes, the internal gear here.
• 2:47
  The teeth are cut on the inside surface of a ring.
• 2:50
  It often meshes with the standard spur gear.
• 2:52
  You see these a lot in planetary gear systems for compact speed
• 2:55
  reduction. Got it.
• 2:56
  Parallel axis family spur helical, double helical
• 2:59
  internal. Makes sense.
• 3:01
  OK, Category 2 intersecting axis gears.
• 3:04
  Right now the axis cross typically at a 90° angle, though
• 3:08
  other angles are possible. This is mainly the double gear
• 3:11
  family. The simplest form is the
• 3:12
  straight bevel gear. The teeth are tapered like
• 3:16
  sections of a cone. It's the easiest bevel gear to
• 3:19
  produce. So like the spur gear is to
• 3:22
  parallel axes, the straight bevel is the basic for
• 3:25
  intersecting axes. Pretty much, yeah.
• 3:27
  But if you need higher performance.
• 3:29
  There's always something better. Usually in this case it's the
• 3:32
  spiral bevel gear. The teeth are cut along the
• 3:34
  spiral curve on the cone face. This gives you significantly
• 3:37
  higher strength and much, much quieter operation compared to
• 3:40
  the straight bevel. But as you'd expect,
• 3:43
  manufacturing is way more complex.
• 3:46
  Spiral for strength and quietness.
• 3:47
  Is there anything else here? There's a sort of hybrid called
• 3:50
  the 0 bevel gear. It's technically a type of
• 3:52
  spiral bevel gear, but it's designed with a 0 spiral angle
• 3:56
  so it shares the force characteristics of a straight
• 3:58
  bevel. No significant thrust loads, but
• 4:01
  keep some of the smooth meshing benefits of the spiral tooth
• 4:03
  form. Kind of an interesting middle
• 4:05
  ground. And the final family, the ones
• 4:07
  that break the rules, crossed axes, gears neither parallel nor
• 4:13
  intersecting. Exactly.
• 4:15
  These axes exist in different planes and would pass each other
• 4:17
  if extended. This category is often about
• 4:20
  achieving really high gear ratios.
• 4:22
  The classic example is the cylindrical worm gear pair.
• 4:26
  You have a screw like worm driving a helical like worm
• 4:29
  wheel. Its main advantage, its standout
• 4:31
  feature, is getting a massive gear ratio in just one single
• 4:35
  mesh, often 50.1, a 100.1 or even more.
• 4:38
  Plus it's usually very smooth and quiet.
• 4:40
  Sounds amazing. High ratio, quiet.
• 4:43
  What's the downside? There's always a downside.
• 4:44
  Efficiency. The transmission efficiency is
• 4:47
  generally poor compared to other gear types.
• 4:49
  There's a lot of sliding friction between the worm thread
• 4:51
  and the wheel teeth, which generates heat and wastes
• 4:54
  energy. OK, efficiency is the trade off
• 4:56
  for that huge reduction. Got it.
• 4:57
  So that covers the main classifications.
• 4:59
  Now let's move from the big picture down to the nitty
• 5:01
  gritty. Designing that perfect gear.
• 5:03
  We need the language of precision, right?
• 5:06
  The sources call out three core concepts for tooth dimensions.
• 5:09
  Absolutely crucial. These three define the tooth
• 5:12
  basic geometry. The first up is the module M
• 5:16
  Think of the module as the fundamental measure of tooth
• 5:18
  size. It's a metric unit.
• 5:20
  In the English system it's related to something called
• 5:22
  diametral pitch, which is teeth per inch of diameter.
• 5:25
  But module is simpler conceptually.
• 5:27
  Bigger module, bigger teeth. OK, module equals tooth size.
• 5:30
  What's next? Next is pitch.
• 5:32
  If modulus size pitches the spacing.
• 5:35
  Specifically, it's the distance along the pitch circle between
• 5:38
  the same point on adjacent teeth.
• 5:40
  And the key thing is reference pitch P is directly calculated
• 5:43
  from the module P = π * m So the absolute fundamental rule for
• 5:47
  any 2 gears to mesh correctly is they must have the same pitch,
• 5:50
  which means they must have the same module.
• 5:51
  Makes sense. Same size teeth, same spacing
• 5:54
  and the third concept. The pressure angle alpha.
• 5:58
  This is the angle between the line of action, the direction
• 6:00
  the force pushes between teeth and a line tangent to the pitch
• 6:04
  circles. It defines the shape of the
• 6:06
  tooth profile. Standard pressure angles like
• 6:09
  20° are common to ensure gears from different manufacturers can
• 6:12
  potentially work together. It affects strength, contact
• 6:16
  ratio, all sorts of things. Module pitch, pressure angle,
• 6:20
  the building blocks. But making these things isn't
• 6:23
  perfect. The sources really flag there.
• 6:25
  Undercutting is a major issue, especially with smaller gears,
• 6:29
  Pinions specifically. What exactly is that?
• 6:31
  Undercutting happens during the manufacturing process, typically
• 6:34
  when using a hob cutter to generate the teeth on a gear
• 6:36
  with a small number of teeth, like a small pinion, The cutting
• 6:39
  tool essentially scoops out too much material near the base, the
• 6:42
  root of the tooth. It gets deeper than where the
• 6:44
  mating gear tooth profile would actually contact.
• 6:47
  And that weakens the tooth. Severely.
• 6:49
  It removes critical material from the Dendum area, right
• 6:52
  where the bending stresses are highest.
• 6:54
  It creates a stress concentration point and makes
• 6:57
  the tooth much more likely to just snap off under load.
• 7:00
  This is exactly the kind of flaw you get with approximate
• 7:03
  geometry. OK, so how do engineers prevent
• 7:07
  this? How do you make a strong small
• 7:09
  gear without this undercut? They use a really elegant
• 7:12
  technique called profile shifting, or sometimes just
• 7:15
  correction. The analysis shows that for that
• 7:18
  common 20° pressure angle, if you have 17 teeth or more,
• 7:22
  you're generally safe from undercut.
• 7:24
  But if you have 16 teeth or fewer, oh, you're in the
• 7:26
  undercut risk zone. The fewer teeth, the worse it
• 7:29
  gets. So the correction fixes this.
• 7:31
  Yes, you apply what's called a positive correction.
• 7:34
  This involves intentionally feeding the gear cutter slightly
• 7:36
  further out from the gear center than the standard death.
• 7:39
  It effectively shifts the whole tooth profile radially outwards,
• 7:43
  moving the critical root section away from the damaging part of
• 7:46
  the cutter's path. It prevents the scoop out the
• 7:49
  undercut. That's clever.
• 7:50
  You're basically adjusting the manufacturing process to enforce
• 7:53
  a stronger geometry at the toothpaste.
• 7:55
  Does this profile shifting do anything else?
• 7:57
  Yes, and this is where it becomes a really powerful design
• 8:00
  tool. It's got a dual purpose.
• 8:03
  Besides preventing undercut, profile shifting is also used
• 8:06
  deliberately to fine tune the center distance between two
• 8:09
  mating gears. Let's say your design needs a
• 8:12
  very specific distance between shafts and the standard module,
• 8:16
  and tooth counts don't quite hit it.
• 8:18
  You can apply specific amounts of positive or even negative
• 8:21
  correction to one or both gears to precisely adjust their
• 8:25
  operating center distance. So it's not just fixing a flaw,
• 8:29
  it's also a precision adjustment tool.
• 8:31
  Very neat. It's fundamental to high
• 8:33
  precision gear design. OK, let's zoom out again from a
• 8:36
  single pair of gears to whole systems gear trains.
• 8:39
  How do we figure out the speed changes in direction when we
• 8:42
  link multiple gear pairs together?
• 8:44
  Well a simple meshing pair is what we call a single stage gear
• 8:46
  train. When you connect these in
• 8:48
  series, say the output of the first pair drives the input of a
• 8:51
  second pair, you have a two stage gear train or multi stage
• 8:54
  if there are more. Calculating the overall speed
• 8:56
  ratio, often denoted by I, is actually pretty simple.
• 8:59
  It's just the product of the individual stage ratios.
• 9:02
  So I total equals I stage one times stage 2 and so on.
• 9:06
  And what does the value of I tell us?
• 9:09
  If I is greater than one, it's a speed reduction.
• 9:11
  The output is slower than the input.
• 9:13
  If I has less than one it's a speed increase and if I equals
• 9:16
  one, well, speed is unchanged. Maybe just changing direction or
• 9:19
  transmitting power? Speaking of direction, what are
• 9:21
  the rules there? For standard external gear
• 9:24
  meshes like 2 spur gears or two external bevel gears meshing,
• 9:28
  they always reverse the direction of rotation.
• 9:30
  Input clockwise means output counterclockwise.
• 9:32
  However, if you use an internal gear mesh where one gear's teeth
• 9:36
  are on the inside, then the input and output shafts rotate
• 9:39
  in the same direction. That can be useful sometimes.
• 9:42
  OK, simple multiplication for ratios, reversal for external
• 9:45
  gears, same direction for internal.
• 9:47
  Got it. Now here's something that always
• 9:50
  seems counterintuitive, especially after talking about
• 9:53
  perfection backlash, The sources say we need it.
• 9:56
  Why design looseness into a precision system?
• 9:59
  Right, it does seem odd at first.
• 10:01
  Backlash is basically the small amount of clearance or play
• 10:04
  between maybe teeth. It's defined as the amount by
• 10:07
  which the width of a toothpaste exceeds the thickness of the
• 10:11
  engaging tooth on the pitch circle.
• 10:13
  And it's absolutely essential. Think about it.
• 10:16
  Gears aren't perfectly made, they heat up and expand when
• 10:18
  running. Lubricant needs space to get in
• 10:20
  between the teeth. Without that tiny bit of
• 10:22
  backlash, the teeth would bind up instantly.
• 10:25
  You'd get interference, massive friction, overheating, noise,
• 10:28
  and likely very quick failure jamming.
• 10:30
  OK, so it's a necessary evil maybe?
• 10:33
  How is it normally created? You just mount the gear slightly
• 10:35
  further apart. You can increase the center
• 10:37
  distance slightly to create backlash, but that's often less
• 10:40
  precise and can affect the contact pattern.
• 10:43
  More common and generally preferred method is to
• 10:46
  intentionally make the teeth slightly thinner than their
• 10:48
  theoretical size. During manufacturing, you reduce
• 10:51
  the thickness of each mating tooth by a small calculated
• 10:54
  amount. The total reduction across both
• 10:57
  teeth creates the desired circumferential backlash,
• 11:01
  usually denoted JT, right at the pitch line.
• 11:05
  So you control it by removing material from the teeth
• 11:07
  themselves. Exactly, and controlling it is
• 11:09
  critical. Too little backlash and you risk
• 11:11
  jamming, pulsating transmission, overheating.
• 11:15
  Too much backlash and you get excessive noise, impact loads
• 11:18
  when direction reverses and potentially faster wear.
• 11:21
  It's another one of those critical details for perfect
• 11:24
  operation. OK, backlash is essential
• 11:26
  clearance controlled by tooth thinning.
• 11:28
  Now let's shift focus completely.
• 11:29
  Materials and longevity. What are gears actually made of,
• 11:33
  especially for demanding applications?
• 11:35
  For high performance, high strength gears, we're almost
• 11:37
  always looking at alloy steels. Things like SCM 440 which is a
• 11:42
  chromium molybdenum steel or S 45 CA medium carbon steel.
• 11:47
  But the key thing is the steel itself is only part of the
• 11:50
  story. The real performance comes from
• 11:52
  heat treatment. That's what unlocks the required
• 11:54
  strength and durability. What kinds of heat treatment are
• 11:57
  common? The sources highlight two main
• 11:59
  approaches. First is thermal refining.
• 12:01
  This typically involves quenching, rapid cooling
• 12:03
  followed by tempering at a relatively high temperature.
• 12:06
  This gives a good balance of through hardness and toughness
• 12:09
  for the entire gear body. OK, balance properties, but what
• 12:13
  if you need extreme surface durability, like where the teeth
• 12:16
  are constantly meshing under heavy load?
• 12:18
  Then you need a surface hardening treatment.
• 12:21
  The gold standard is carburizing and crunching.
• 12:24
  You introduce extra carbon into just the surface layer of the
• 12:27
  steel, then you quench it. This creates an extremely hard
• 12:31
  outer case. We're talking Vickers hardness
• 12:33
  of maybe 600 to 640 HPV while keeping the core of the tooth
• 12:39
  tougher and less brittle. That super hard surface is
• 12:42
  fantastic at resisting wear and pitting fatigue, which are
• 12:45
  common failure modes right at the tooth legs.
• 12:48
  Makes sense. Hard surface, tough core.
• 12:50
  Now steel obviously dominates, but we do see plastic gears too.
• 12:53
  What's the story there? Materials like nylon or acetyl.
• 12:57
  Right plastics like MC nylon or acetyl copolymer, often called
• 13:00
  Duracon, have their place. Their big advantages are being
• 13:02
  lightweight, naturally self lubricating to some extent, and
• 13:06
  operating very quietly. Great for noise reduction.
• 13:08
  But they must have limitations, right?
• 13:09
  Oh definitely big limitations. The main one is dimensional
• 13:13
  stability. Plastics expand and contract a
• 13:16
  lot more with temperature changes than steel does.
• 13:19
  They also tend to absorb moisture from the air, which can
• 13:22
  cause them to swell. This makes holding tight
• 13:24
  tolerances like for backlash much trickier over different
• 13:28
  operating conditions. And I bet they can't handle the
• 13:30
  same speeds or lows as steel. Absolutely not.
• 13:34
  Their strength is much lower and they're very sensitive to heat
• 13:37
  generated by friction. The sources give clear limits,
• 13:40
  for instance, even with continuous lubrication.
• 13:43
  A plastic worm gear pair made of MC nylon has a maximum
• 13:47
  recommended sliding speed of only about 2.5 meters per
• 13:50
  second. Push it faster and it'll likely
• 13:52
  overheat and fail quickly. You get to swap plastic in for
• 13:55
  steel without careful analysis. OK, so plastics are for lighter
• 13:58
  loads, lower speeds where noise or weight is key.
• 14:01
  Finally, let's talk about keeping gears alive.
• 14:03
  Lubrication. It seems critical.
• 14:05
  What are the main ways gears get lubricated?
• 14:07
  It's absolutely vital. Lubrication reduces friction,
• 14:10
  carries away heat and flushes out wear debris.
• 14:14
  There are basically 3 common methods, scaling up with speed
• 14:17
  and load. The simplest is grease
• 14:19
  lubrication. Good for open gears or enclosed
• 14:23
  gearboxes running at low speeds. It sticks in place well, but
• 14:26
  it's not great at cooling. OK, grease for low speed.
• 14:28
  What's next? For medium to reasonably high
• 14:30
  speeds, you typically use a splash or oil bath system.
• 14:34
  The gearbox housing acts as a reservoir and the rotating gears
• 14:38
  dip into the oil, splashing it around to lubricate all the
• 14:40
  moving parts. The key here is managing the oil
• 14:43
  level carefully. Too high and the gears churn the
• 14:46
  oil excessively, causing energy loss, agitation loss and
• 14:49
  overheating. Too low and you don't get enough
• 14:51
  lubrication or cooling. Right, the Goldilocks zone for
• 14:54
  oil level and for the really demanding stuff, high speed,
• 14:57
  high load. That requires forced oil
• 14:59
  circulation. Here you have an oil pump that
• 15:02
  actively takes oil from a sump, often filters and cools it, and
• 15:05
  then sprays it directly onto the gear mesh and bearings.
• 15:08
  This provides the best lubrication and crucially, the
• 15:11
  best cooling for heavy duty applications.
• 15:13
  And regardless of the method, there's a critical temperature
• 15:16
  limit we need to respect. Yes, a hard rule from the
• 15:19
  sources keep the oil temperature below about 80 to 90°C.
• 15:24
  That's roughly 175 to 195 Fahrenheit.
• 15:28
  Go above that and you risk serious trouble.
• 15:31
  The lubricant itself starts to degrade rapidly, losing its
• 15:34
  protective properties. The gear housing might distort
• 15:36
  from the heat and critically, thermal expansion can eat up
• 15:39
  your designed in backlash. The backlash decreases.
• 15:43
  Exactly. And as we discussed, lose too
• 15:46
  much backlash and the gear start to bind, leading to rapid wear,
• 15:49
  potential jamming and catastrophic failure.
• 15:52
  That 8090 C window is really the upper limit for safe continuous
• 15:56
  operation. Wow.
• 15:56
  OK, so that really covers the whole life cycle.
• 15:58
  That was an incredibly comprehensive look really at the
• 16:01
  whole mechanical engineering journey of a high performance
• 16:04
  gear. We went from the basic
• 16:05
  classification through the nitty gritty of tooth geometry and
• 16:08
  material science, all the way to lubrication and temperature
• 16:11
  control affecting its lifespan. And the clear thread running
• 16:14
  through all of it is it achieving genuine high
• 16:17
  performance, meaning low noise, high power density, long life,
• 16:21
  heart? It really is a master class in
• 16:24
  controlling errors. It boils down to meticulously
• 16:27
  managing those tiny geometric deviations, things like single
• 16:30
  pitch errors, run out errors, while also nailing the big
• 16:34
  operational factors like having the correct amount of backlash
• 16:37
  designed in and ensuring proper cool lubrication throughout its
• 16:41
  life. Which brings us right back to
• 16:43
  that powerful idea we started with that quote from the
• 16:45
  sources. No perfect gears can be designed
• 16:48
  and manufactured if advances in the theory of gearing are
• 16:51
  ignored. And that leaves a really
• 16:53
  interesting question for you, the listener, whatever your own
• 16:55
  field might be. How often are you maybe relying
• 16:58
  on approximate solutions? You know, the way it's always
• 17:00
  been done, the easier path? Especially when there's likely a
• 17:03
  more detailed, established theoretical framework out there
• 17:06
  that could help you achieve true perfection, or at least much
• 17:08
  higher performance. What detailed technical manual?
• 17:11
  What foundational theory in your area might you be overlooking
• 17:14
  right now? Maybe it's time to go find it.
• 17:17
  We'll see you next time.