Torque Wrenches - my answer...

[Huw Powell]

> im afraid im having great difficulty with that concept. After all,
> torque is a force about an axis at a given distance. As long as you
> are rotating that wrench about the axis of the square drive the
> mechanism will always react at the same point.

> I think the point here is that torque wrenches don't actually measure
> torque.  Somehow (and after all this thread I have a killer urge to
> take a clicker torque wrench apart) what it actually measures is the
> applied force.  The torque scale etched on the wrench is just the
> measured force multiplied by the assumed constant length of the 
> wrench.

I think there are some glimmers of useful truth contained in these two 
paragraphs.

I also think we are wrestling with two or three separate concepts that 
make getting lost very easy.

One applies to the internal workings of the typical "click" type torque 
wrench.  I do not think that they measure torque at the drive square - I 
think, as it says above, they are more complicated, which is why you 
have to hold them correctly, etc.

That said, we move along to our original question - what is the 
resultant torque on the crank bolt when using a torque wrench set at 258 
ft-lb with the extension tool?

I say... 258 ft-lb.

Keep in mind two things, the first of which Phil mentioned.  The reason 
for the tool is not to multiply torque, it is to make the car "easier" 
to work on, ie it moves the point of applied force out from behind the 
bumper of the car.

The second is more abstract.  I will ignore all internal mechanisms of 
torque wrenches, since the wrench can be considered to be a "black box" 
that simply creates a *rotational force,* or torque, in a circular 
fashion about a small square drive nubbin thing.

As far as the 2079 "knows," all that is happening is that one end is 
stuck on the crank bolt, and the other is being *twisted* at up to 258 
ft-lbs.

I don't think it even matters what angle is made between the torque 
wrench and the 2079, the effect is the same.

This twisting force is present at one end of the 2079, and since the 
other end is fixed in space (it can only rotate about the crank axis), 
the torque is transferred from one end to the other.

For all the lever arm math to apply, the initial torque has to be turned 
back into a lever arm with force on it, and then connected to another 
lever arm, like the way gears are used to trade off rpms for more or 
less torque output.  The first gear turns the input torque back into a 
lever arm (the radius of the gear) and the second gear is the second 
(usually different length) lever arm.

It is important to rememeber that torque, expressed as foot-pounds (or 
any other "distance-force" configuration) cannot be turned back into a 
lever arm and force from raw, twisting torque for the purpose of 
calculation.

It would make no difference whether the 258 ft-lb were applied to the 
2079 by a typical 2 foot torque wrench and a fair amount of grunt, or a 
four foot torque wrench and less operator effort.  It would still be 
twisted at one end with 258 ft-lb of torque.  And thence, transfer that 
same torque to its other end, less any tool bending.  And I bet the 2079 
is made not to bend at 258 ft-lbs.

We need two things to "verify" what I am saying - one would be someone 
who is comfortable with the simple physics involved (where the heck are 
you, Larry?), and the other would be a few simple experiments with dry 
fasteners, torque wrenches, and 2079-like extenders.

Well, that's what my brain churned out after three days of thinking 
about this problem.

-- 
Huw Powell

http://www.humanspeakers.com/audi

http://www.humanthoughts.org/