Proposal: Clutches Should Be Rated for Torque, not Horsepower

[jlccc]

Everyone here on the TDR is probably familiar with the fact that most clutch vendors rate their clutches based on the amount of horsepower they can hold without slipping. This rating system works OK when applied within a specific service class or type of application (e.g. a diesel-powered Dodge pickup). But when applied broadly, the terminology is misapplied and the rating system breaks down to the point that meaningful comparisons cannot be made, and can even be grossly misleading. For example, does anyone believe that the same clutch which transmits 400hp from a 454 Chevy big block will stand up to 400hp from a Cummins diesel?

The solution to this rating problem is for clutch manufacturers to specify TORQUE ratings rather than horsepower ratings. Let’s lay a little mechanical groundwork and then consider several examples.

Most of you have seen the following formula for calculating transmitted power using a conversion factor for the use English units.

hp=torque x rpm
------------
5252

which can be rearranged as

T=hp x 5252
---------
rpm

With formula in hand we are ready to analyze the Chevy big block we referenced in the opening paragraph. Let’s assume that this motor makes 400hp at an even 5000 rpm. Plugging these values into our formula yields a torque value (at peak horsepower) of 420 ft-lb (400*5252/5000).

Compare this to a stock 24v ETC Cummins that makes peak power of “only” 235hp at 2700rpm. Plugging these values into our formula yields a torque value (again, at peak horsepower) of 457 ft-lb (235*5252/2700).

Can you see why the clutches on our diesel trucks are so large? Obviously it’s because they have to transmit a lot of torque and not necessarily because they have to transmit a lot of horsepower.

The strain on a diesel truck clutch is often even greater than the example above because peak torque is generally higher than the torque at peak horsepower. My truck for example reaches peak horsepower at 2350 rpm but peak torque occurs 450 rpm earlier and is 95 ft-lb higher than the torque at peak horsepower.

From the two scenarios above we can see that the system of rating a clutch by horsepower has already broken down beyond the point of being useful. But, let’s push things a little bit more and see if we can come up with two even more extreme examples.

Consider a standard 5hp 1750 rpm AC motor (yes, only 5hp) attached to a large, multi-stage gearbox with reduction ratio of 100:1, meaning that the output speed of the shaft is about 17.5 rpm. Neglecting any losses of efficiency through the gearbox, the output torque of the gearbox is 1500 ft-lb! What automotive clutch is going to hold 1500 ft-lb of torque? Perhaps a sled pulling clutch, but few others.

Now consider the same 5hp AC motor, this time transmitting its power at 1750 rpm instead of 17.5 rpm. Running the numbers once more yields a torque of just 15 ft-lbs. Heck, a worn out clutch from a Yugo will hold that much torque without slipping.

I hope you’ve found each of the examples interesting and trust you have noted that in each of the four the essential characteristic has not been power, but torque. Owing to this fact I think it behooves clutch vendors to rate their clutches by torque rather than horsepower.