Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Does the load matter to a clutch?

  1. Apr 7, 2017 #1
    Does the weight of a car or length of gears it has change the requirements for clutch strength or is the requirement solely dependent on the torque the engine applies to it? I see a lot of consensus that says the former, but I think it is the latter.

    An engine making 300 lb-ft should need a clutch that holds 300 lb-ft regardless if its hauling a 150kg motorcycle or a 5000kg truck. The gear or mass that applies resistance shouldn't change anything because it isn't a counter-acting torque. I believe a stronger clutch is required than the engine when there is a counter-acting torque applied such as if a vehicle is going on an upward incline (a hill).

    Now, the question from this of course is why do cars tend to have slipping clutches in higher gears first. I believe the answer to that is acceleration. A higher gear means less acceleration so less power diverted to spooling up the rotating masses of the engine internals and flywheel and more of it moving through the clutch, right? In a low gear with a lot of acceleration, a bunch of power is sucked up spooling up the crankshaft and flywheel which reduces torque seen by the clutch. That is why it appears the load is what is causing the slip as opposed to less power (and therefore torque) left over to spin the clutch.

    Am I entirely wrong? Would an equivalent engine really require a stronger clutch in a higher load situation?
     
  2. jcsd
  3. Apr 7, 2017 #2

    JBA

    User Avatar

    The amount of torque on the clutch is a function of the resistance of the vehicle to accelerate, climb a grade, or of the tires to lose their grip, relative to the amount of torque an engine can provide.

    As for your "higher gear" question. all torque multiplication in a drive train occurs beyond the clutch in the transmission; the higher the selected gear, the less torque multiplication and therefore the higher amount of the vehicle resistance force that is delivered back to the clutch and the engine.
     
  4. Apr 7, 2017 #3
    Let me add an analogy.

    Suppose the muscle of my arm can apply 10lbs of force and my bone can handle 10lbs of force before breaking. Suppose I want to lift a stationary 100lbs weight from the ground.

    My muscle is the engine, my bone is the clutch, and the weight is a car.

    Will my bone break from trying to lift the weight? Of course not. My arm applies 10lbs of force, I don't move anything because its too heavy and my bone is fine because no more than 10lbs of force is applied to it. Saying that a higher gear puts more load on the clutch and increases slip probability from "resistance" is equivalent to saying that me trying to lift a 200lbs weight increases the chances that my bone breaks. It doesn't matter if I try to lift 100lbs or 10^100th lbs, no more than 10lbs of force is ever applied to my bone.

    Is this analogy wrong?
     
  5. Apr 7, 2017 #4

    JBA

    User Avatar

    The anaolgy is backwards, if I have an arm muscle (engine) that is capable of lifting 20 lbs but my arm bone (clutch) is only strong enough to lift 10 lbs then I can lift any weight up to 10 lbs without breaking my arm (slipping the clutch) but if I use a rope with two pulleys (transmission) then I can lift up to 20 lbs without breaking my arm (slipping the clutch).
     
  6. Apr 7, 2017 #5
    Gearing comes after the clutch in our case so it shouldn't matter. Every conclusion if my analogy is wrong leads to the inevitable scenario that any arbitrarily weak engine can over power any arbitrarily strong clutch so long as there is some arbitrarily large mass or resistance behind that clutch. So if my 300 lb-ft engine attached to a clutch capable of handling 1000 lb-ft torque can over-torque this clutch and make it slip so long as I have some infinitely large mass or resistance behind that clutch?

    Also keep in mind, potential energy changes everything as I stated earlier. Since that applies a force back on me. I'm talking about a mass with zero potential energy and no desire to accelerate anywhere on its own.
     
  7. Apr 7, 2017 #6

    JBA

    User Avatar

    If the engine torque is less than the clutch rating and the rear drive is locked then your engine will stall.
     
  8. Apr 7, 2017 #7

    JBA

    User Avatar

    I have driven standard transmission vehicles for over 40 years in competition and on the road and still do. Clearly you are lacking any similar experience with this type of drive.
     
  9. Apr 7, 2017 #8
    In my car, the multiplication difference between my first gear and last gear is about 5x (4.23 vs 0.83). I never had any clutch slip in any gears despite my 6th gear putting 5x the load on the engine. As soon as I bumped engine torque from a modification by about ~15%, it started to slip. I changed load by 400% with no problems or slip but changed engine torque by ~15% and got clutch slip.

    If load matters, can you explain how this occurred?
     
  10. Apr 7, 2017 #9

    Randy Beikmann

    User Avatar
    Gold Member

    Under steady acceleration, it doesn't matter what gear you are in - the torque in the clutch is essentially the same as the output from the engine. So by shifting from first to top gear you are changing the torque (load) on the clutch by 0%, not 400%.
     
  11. Apr 7, 2017 #10
    Thank you for confirming. Thats what I originally thought.
     
  12. Apr 7, 2017 #11

    Randy Beikmann

    User Avatar
    Gold Member

    To your original question, there are other factors besides the mean engine torque. Some are much bigger, in fact:

    1) Engine torque isn't steady, but delivered in pulses from each combustion event. So the peak torque is higher than the average torque we normally speak about.
    2) Bigger yet is impulsive torque from sudden engagement of the clutch. Imagine your engine and flywheel are rotating at 6000 RPM (clutch disengaged), and the car is stationary in first gear. You let the clutch engage. If the clutch engaged instantly, it would cause a torsional "collision" between the engine and transmission/driveline, creating a very high transient torque (many times higher than the rated engine torque). If the tires can slip easily, this limits the torque in the driveline, and therefore the clutch - if not, the peak torque can be very high.
    3) Slipping the clutch produces heat, and a heavier vehicle requires more slipping during engagement. Too small a clutch for the vehicle will overheat.

    I'm sure I've forgotten some factors, but already you can see, for instance, that a heavier car with more traction will need a stronger clutch (not necessarily more torque capacity) than a lighter one with less traction.
     
  13. Apr 7, 2017 #12

    JBA

    User Avatar

    Actually for the same acceleration rate/load, the torque load on the engine and clutch will be higher as the gear ratios reduce that is the reason you can start a car moving in 1st gear with much less clutch slippage and risk of an engine stalling than if you are in 4th gear.

    That said, as Berkeman stated, a 15% increase in the maximum available torque of an engine will result in a potential 15% increase in the torque that the clutch can experience.
     
  14. Apr 8, 2017 #13

    Randy Beikmann

    User Avatar
    Gold Member

    It's true that at a given acceleration rate, the torque from the engine and in the clutch will be higher in a higher gear. But to the original question on what determines clutch design, the engineer has to design the clutch for worst case conditions.

    For preventing steady state slippage, the worst case is at WOT, regardless of gear - not really dependent on vehicle mass either. For strength, it's probably when dumping the clutch at WOT, to launch the car. For heat capability, it could be launching the car with a heavy trailer. The design engineer must come up with a clutch design that can stand up to all of these.
     
  15. Apr 8, 2017 #14

    JBA

    User Avatar

    Wait a moment, I believe that a car and a trailer for that matter are masses and the inertia of those masses is what creates the problem at launch. (This is not really a serious response, but someone has to be the science guardian form time to time)
     
  16. Apr 8, 2017 #15
    Acceleration and load are different things. An engine at full throttle is under full load regardless of gear or weight being pulled.

    F=MA where force applied (in this case torque) is the same regardless of mass because acceleration changes to equalize.

    I agree with your second statement.

    Now as for why you cannot start moving from a higher gear from a stop, its because a gas engine needs a minimum idle RPM and if that gear ratio brings the RPM too low, it stalls. For simplicity, assume we have electric motor that makes a flat torque curve at any RPM.

    Clutch slippage time from a start depends on gearing, not mass or inertia for the reasons I stated above I think.
    ========

    Either way, as Randy said:
    That confirms my question then.
    ==========

    Randy, when you said:
    This for the reasons I stated in my opening post, right? That high acceleration means more energy is diverted to spool up the engine's own internal masses plus its flywheel?
     
  17. Apr 8, 2017 #16

    Ranger Mike

    User Avatar
    Science Advisor
    Gold Member

    Clutch's slip because the engine torque is greater than the mechanical advantage of the drive train to translate rotary motion into traction. The tires have too much adhesion for the torque so the clutch must slip. You are forgetting one huge variable. The differential or final drive.
    In read end drive vehicle this is the rear differential. This is also present in front wheel drive vehicle but not so readily apparent.

    Engine torque varies with RPM. Tires adhesion varies with the weight of the vehicle. What does not vary is the mechanical advantage of the first gear in the transmission and the final drive ratio of the “rear end”. More on this late.

    To simplify and explain – a three speed manual transmission has a very numerically high, mechanically ” low “ gear of 2.54 to 1 ratio. This means the engine side of the transmission ( input shaft) turns one time and the out put shaft going to the differential turns 2.54 times. Second gear was around 1:48. High gear or 3rd gear of this transmission has a 1:1 gear ratio meaning the input and output shaft turn the same number of revs. It is apparent that we have significant torque multiplication with the first gear.
    Back to the rear end. Typical passenger car rear end rations were 3.23 :1 to 4:10. Every 3.23 turns of the drive shaft the rear wheel would make one revolution.
    To calculate Final drive ration simply multiply the transmission gear x rear end ratio.

    2.54 x 3.23 = 8.128 to 1 final drive.
    carrying this drill on to your question about the effect of a numerically low first gear ??

    2 x 3.23 = 6.46 to 1 and we may not have enough mechanical advantage to move the automobile. So what happens? The clutch slips.

    Going the other way, what happens when we have too much mechanical advantage?

    first gear of 3.54 and rear end gear ratio of 6.13 to 1

    3.54 x 6.13= 21.7 to 1 and we will slip the tires almost on release of the clutch. This is why I have to start out in 2nd gear in my Dodge Cummins Diesel pick up when not loaded!

    So why have this set of gears if you can not use it?
    When we have very heavy load on the pick up truck, like adding a 5000 pound trailer and race car, we put additional load on the rear tires and now we better traction and thus no slippage.
    Care must be taken not to over rev the engine and “ slip the clutch”

    hope this helps

    rm
     
  18. Apr 8, 2017 #17

    Randy Beikmann

    User Avatar
    Gold Member

    That's true. A higher percentage of engine torque goes to accelerate itself, rather than the vehicle. Sometimes you'll see the influence of the inertia from the engine, added to the mass of the car as an effective mass: m_Eff = J_Engine x (R_OA/r_Tire)^2, where R_OA is the overall gear ratio. Since this grows as the square of the overall ratio, it is a very small factor in high gears.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: Does the load matter to a clutch?
  1. Clutch Question (Replies: 2)

  2. Gears and Clutch (Replies: 0)

Loading...