Whatever current Proto torque wrenches claim, I'm not sure. But at the time,
I had access to a torque wrench tester, and had drawn a curve for my wrench
vs. the actual reading. IIRC it was fairly close in the middle of the range,
but quite a bit off (like almost 20%) at the high and low ends. This was a
clicker-style wrench from the 1970's, 1/2" drive, 250 lb-ft capacity.

The point is, torque wrenches are not laboratory instruments - - you should
expect them to have a tolerance, just as one should expect EVERY measuring
instrument (even laboratory instruments) to have a tolerance, which is
another way of saying, they may be off by a certain amount. Just a part of
basic awareness, one should know what sort of tolerance to expect from
whatever you are using to measure anything that is measurable. That being
said, at the time, a beam-type torque wrench was considered to be more
accurate than a clicker-style torque wrench, but as I mentioned in my
previous message, just how many people took care to get a proper reading,
i.e., without a parallax error? I remember that the Snap-On torque wrenches,
the beam type, had optional electrical contacts that would turn on a light,
or sound a beep, at the selected torque value, and if I remember correctly,
Snap-On claimed 5% tolerance for their wrenches. Not everybody could afford
Snap-On, however!

I believe the engineers know this, and set torque values to a number that
they believe will do the job, without over-stressing the part. And as has
been pointed out by others, it's the uniformity of torquing fasteners (at
least on parts having more than one) that is probably more important than
the actual torque value. So if you are using a reasonably good torque
wrench, even if it may be off by quite a few percent, it is probably
repeatable, which is a different thing than being accurate. In other words,
if all five lug nuts, for instance, are torqued to the same value, and are
in the proper range of torque, that is probably more important than whether
or not they are all torqued to 118 lb-ft, or 133 lb-ft. But as I am not an
engineer, it is just IMHO!

And just for grins, I think I may still have one of the old Proto catalogs,
so I'll look and refresh my memory of what they claimed for that era. Also,
I should point out that the statement, "the technology in clicker-type
wrenches hasn't changed", may be true, but the technology MAY have been
refined. :^)

Gary

> At 12:02:24 27 Sep 2011, Mike S <mikes@FLATSURFACE.COM> scribed,
>
> I'm not sure Gary's remembering accurately. Current Proto torque
> wrenches are "Calibrated to +/-3% in clockwise direction and +/-6% in
> counter clockwise direction," so much better that 20%. The technology
> in "clicker" type wrenches hasn't changed. If you don't get them
> calibrated occasionally, or don't reset them to the minimum setting
> after every use, then the calibration could be off, but probably not by
> 20%.
>

At 07:35 PM 9/26/2011, John Rodgers wrote...
>If Gary Bawden's plus or minus 20% rule applies - the would be low
>-98.4
>ft-lbs and high - 147.6 ft-lbs and that figure is a whole bunch!