Re: Regarding friction factor

Hi Christopher

If you want to see metal surfaces stick together, ask your nearest precision machine shop to show you their gauge blocks. These are precision finished blocks, normally made of steel or tungsten carbide, which are stacked together to give an accurate thickness, sort of like super feeler gauges. The blocks are not actually stacked, they are twisted together - called "wringing". When two blocks are wrung together, they need to be slipped apart by shearing motion to separate; trying to part by a tensile pull results in a blue face. This attraction gives frictional resistance even with a negative reaction.

I feel that the biggest problem with frictional coefficients is their unpredictability and the difficulty in arriving at a suitable safety factor to apply to an experimentally determined value. For example, shrink fits, steel on steel, a coefficient of friction of 0.1 to 0.15 is commonly allowed, even though the actual coefficient will be higher.
With things like pipe shoes the danger is that the actual frictional coefficient will be higher that allowed for in calculations, and will result in distress to pipework and structures. I do not think that single experiments are the right way to estimate design values for frictional coefficients. In many cases frictional properties will worsen with time; e.g. corrosion, creep,dirt, misalignment, lubricant loss. Using experimental data could paint a far rosier picture than will exist 5 to 50 years down the track. Reliance on accuracy is, in my opinion, dangerous in some circumstances.

My self-developed rule of thumb is measure it and then multiply by three and a bit. It works sometimes.

Cheers

Steve McKenzie

• In PipingDesign@y..., Christopher Wright <chrisw@s...> wrote:
  > >why the value of friction factor varies from
  > >project to project and what are the main factors
  > >governs the value of fiction factor.
  > Friction varies with surface finish, surface cleanliness the
  presence or
  > absence of lubrication or oxidation and to a degree with load
  (Unlike
  > what we were all taught...) The actual mechanism (or so I was
  taught) is
  > that friction results from microscopic pressure welds between
  surface
  > asperities. The frictional force reflects the load required to
  break the
  > welds. The more pressure the more asperities are brough into
  contact so
  > the higher the friction force, so the friction coefficient is taken
  as a
  > constant although it isn't. A tale I heard once (I never verified
  it) is
  > that highly polished surfaces which are absolutely clean and flat
  will
  > weld themselves in a vacuum. Galling in stainless steel is a form
  of
  > pressure welding of this sort, so I'm inclined to believe the tale.
  >
  > I've seen a study of friction coefficients for austenitic stainless
  > against itself where the coefficients vary from 0.2 to 0.8.
  Actually when
  > it galls it's a helluva lot higher than that. The high values
  reported
  > for steel in some handbooks (around 0.7 I think) are for very clean
  and
  > smooth surfaces, probably in a lab somehwere. The usual static
  values are
  > around 0.2-0.3 for machined surfaces and 0.1-0.2 for kinetic
  friction.
  > Your values of 4 or 5 aren't true friction coeficients--the highest
  I've
  > ever seen reported are for rubber in concrete at around 1.3 under
  very
  > special conditions.
  >
  > A good way to estimate static friction is with a simple test. Find
  > yourself samples of the materials in question mount one to a
  straight
  > stiff plank and the other to a weight that will produce a contact
  > pressure about what you expect in service Put the two sample
  surfaces in
  > contact and raise one end of the plank until the weighted sample
  slips.
  > The tangent of the angle of inclination will be the friction
  coefficient.
  > You can try the same test with lubricants or water or finish the
  two
  > surfaces and see the effect of surface condition. The results will
  > probably be as good as you'll find in all but the most
  comprehensive
  > handbooks.
  >
  >
  >
  > Christopher Wright P.E. |"They couldn't hit an elephant at
  > chrisw@s... | this distance" (last words of Gen.
  > ___________________________| John Sedgwick, Spotsylvania 1864)
  > http://www.skypoint.com/~chrisw