Skye:
<<<snip>>>
I have a cheapo Craftsmen deflection type torque wrench, you
know with
the shaft that bends and the needle measures the
deflection...
How accurate are these?
<<<snip>>>
More accurate mechanically than the click type. The manuals
on click type often say something like, "When used in
exacting situations or extensively, the wrench should be
periodically calibrated" The click type uses a spring
loaded ball pressing into a detent. What happens as the
ball, or detent wears? The release point changes. The more
you use this type, the less the actual torque becomes. Then
it needs to be recalibrated.
The error of my 1/2" drive beam Craftsman torque wrench is
list as 2%. The error of the click Craftsman torque wrench
is 4% over 30 lbs, and 1 ft pound under 30.
My ex Navy relative had a MOS such as "gauges and
calibration." All click type wrenches had to be subject to
calibration. He tested one of my Craftsman click 1/2 drive
wrenches and found it to be right on. It surprised the hell
out of him as it was very exceptional.
With the beam type if the "zero" is off you just bend the
needle. It will never change calibration unless the needle
gets bent. The calibration of the beam is in the molecular
alloy structure of the beam itself.
So, don't call every beam type torque wrench a "cheapo." If
fact one of the best torque wrenches we have is a dial
indicating with telltale Blackhawk wrench. The innards are
a beam with the pointer linked to the dial. If the beam
wrench comes from a reputable company it should be very
good. Foreign built beams are another matter as you don't
know the quality control of the material used to make the
beam.
The only errors with the beam type is the operator not
reading the pointer right and the difficulty in reading a
pointer in some locations.
So, I use both depending on the use, torque needed and
visibility.
--
Blake
NAMGAR Technical Reprint Provider
http://www.mgcars.org.uk/namgar/techpage.htm
MG TD
MGA twin cam
http://members.nvc.net/yd3/
/// or try http://www.team.net/cgi-bin/majorcool
/// Archives at http://www.team.net/archive
|