Metal vs. Metal - Friction Coefficient Optimization

First off, what material are you using as the stop?

Not just the material, what components? Electro-mechanical stop? Limit switch?

Sensor, proximity? NPN/PNP? Is this a rod you are trying to stop? Dia? What force is exerted if X-Y? Your proposed stopping is Z? Are you using air pressure? Is it a clamp type device? What is the control scheme?

So many questions need to be answered to give a relatively decent answer.

Please advise

My question is relating to metalurgy, not instrumentation.

Just in case it may interest you, the coefficient of friction for various material combinations may be found on these pages

http://www.engineeringtoolbox.com/friction-coefficients-d_778.html

http://www.carbidedepot.com/formulas-frictioncoefficient.htm

http://www.physlink.com/reference/FrictionCoefficients.cfm

http://www.roymech.co.uk/Useful_Tables/Tribology/co_of_frict.htm

etc.

However the braking force to be calculated and based on that the pressure. However ensure that the 3mm rod don't bend or break under brake .

Thank you SB,

These are the first steps taken in order to choose our geometry and materials.

As such, the first design iteration does brake 30N.

The tables have a limited usefulness as to help optimize the design.

It is stated in literature that the friction force will rise proportionally with the normal force up until a certain pressure level where the surfaces will tend to seize together.

At this pressure point the friction coefficient is quite higher than at lower pressures.

I did not find exhaustive information on this yet.

sb -- GA!! Thanks for the great links. Especially the roymech site. I'd heartedly recommend anyone with an interest in lubrication at the component level check out the various pages and links there.

Ed Weldon

This roymech site is great. Infact go to the home page and you will find links to a lot of mechanical engg topics- starting from metallurgy onwards.
This site I found just by coincidence about 3-4 years ago (in fact through metallurgy links - had some problem in - of all things greay cast iron metallurgy )

And the links on bottom are really great, he has collected these well.

Generally, bronze interacts with stainless more benignly that does aluminum, especially in the presence of moisture. You'd lose some friction coefficient, but gain some reliability. If getting sufficient braking force is an issue, perhaps a cam (as used on some pneumatic cylinder rods) would work? A lever for force multiplication? I'd think that bronze could be used to apply several hundred N to an area of several square millimeters without the shaft or brake showing any wear or deformation.

If this is a particularly difficult application, perhaps the traction fluid used in CVT transmissions would help? Sprag clutch principal? Microteeth? Thread the rod and use half nuts?

Hi,

friction is a nightmare if you want to exceed known benign levels of stress and sliding velocity.

The basic physical process that triggers the material response that we see as a "constant" friction coefficient is the contacting points where roughness and form lets to surfaces to contact at very small "points".

These contacting points are deformed, first elastic then plastic deformation.

Ther intermediate layer between two contacting points then subject to shear load: good if lubricated, bad if pure metallic.

Pure metallic is only existing in high vacuum: there we get friction coefficients of 1 to 100 by immediate metallic friction welding.

In atmosphere there is absorbed water at any surface so you get a shear strain that is dependent on a lot of different factors: that#s why we have to lubricate, to establish a stable - electrostatically attached - film of low shear strength.

Back to the problem:

Neither aluminum (any alloy) nor SS is a good choice.

Anodised aluminum (6061 will have a much better coating than 7075), preferably hard anodised will be a good choice.

With hard anodised surfaces you are allowed to do both parts of the same material and coating. (Ordinarily in most materials this would not work, but in ceramics that are a little bit lubricated by absorbed water this will work, and in anodised with the nano-pores this is good.)

The above remarks are for survival of sliding surfaces, not for maximum friction coefficient.

If you need maximum and constant friction you should switch your material selection towards steel (ordinary) against phenolic reinforced ceramic fibers. This is used in advanced car brakes, get your material from a new or used brake, easy, cheap, fast, proven, reliable.

RHABE

Thank you all for your replies,

This being part of a portable field instrument, stainless steel is preferred.

The rod is used as a guide for operating the instrument, sliding against plastic bushings.

At some point we need to block it without creating axial forces or displacement.

The idea is to use an Al bushing witch is rotated to interfere with the rod.

The available force is limited to 10N, and space is within a handheld instrument so lever length is also limited.

The available torque is 0.625Nm.

I can play with the braking material length, part of the lever equation to increase the contact force, within the braking material's elastic regime. There is also some limitation on the rotation angle to approx. 1°.

This is simple and it works, now I want it failsafe.

Why not cut a slot axially in the bushing, then apply a clamping force on the slotted area, allowing the inner diameter to grab the shaft. This is commonly used method to grasp a shaft. Just adapt the clamping mechanism to suit the braking application.

This may not help with surface friction but the others have already covered that.

The bushing was for simplicity, actually the brake is a single piece acting as a lever, pivot and brake.

The braking surfaces are machined-in.

The split bushing or flexure arrangement was dismissed due to the very limited forces available for the brake actuation and added machining.

The inspiration came from a caulking gun.

Al is chosen not to mark the steel rods, micron position is a requisite.

The observed friction coefficient varies between 0.54 and 0.8, witch is not that bad.

I wish it was more predictable.

Hi,

if you look to the machine tool clampings:

often amplification of clamping force by conical surfaces,

never aluminum nor SS-alloys - with the exception of hardenable (non austenitic) alloys as 440C and similars (used in stainless ball bearings).

Bronze (CuSn8) to hard (HRc 56 to 63) chromium steel, slightly polished - do not remove totally the grooves and scratches of prior machining, these act as oil or grease reservoirs and ducts. Chromium steel as in ordinary ball-bearings or any steel that can be treated to similar high hardness. High hardness equivalent to high yield strength equivalent to good sliding surfaces.

If this is not practical then switch to 6061 hard anodised (20 to 50µm) and slightly polished. This is very good with any lubricant, also water. Ask for not sealing after anodising and impregnate with oil or any other lubricating stuff for absorption and protection.

I made 3mm thermal clamping chucks from 6061 anodising these inside to very precise tolerances by pumping the electrolyte through the reamed bore and determined by current x time the thickness of anodising.

Also this dynamic airbearing from 6061 - is surviving start and stop cycling without lubrication. (Second axial surface missing on the photo).

RHABE

Thank you RHABE, your experiences and observations are what makes CR4 great.

I looked at taper clamping systems such as Shaftloc from http://www.sdp-si.com/

I had doubts on how to implement this in a small and low cost instrument where actuation energy and movement amplitude are quite limited.

Many locking mechanisms require some micro-movements in order to exert their full force. It may be the case for tapers. This is why I also dismissed the ball locking scheme used in one way pulleys. (one way braking could be acceptable) This is why I liked the bushing rotation as the forces and movement are quasi radial and are balanced.

The rods we are using have a Ra 0.25 finish and I have been contemplating other materials and finishes.

The Al contact surfaces have a reamed finish specification for Ra 0.8

My understanding is the finer may be better.

Other materials tested on SS 303 were brass 936 and 944 alloys and phenolic grade XX and G11 Garolite, witch provided a lower coefficient of friction. I may revisit brass for stability against humidity.

Hi,

think about a roller from a tapered roller bearing as rod?

Phenolic is really good as acting against this, if relative movement is existing then I would prefer non-glas types.

Likely that clamping is easy but unclamping may require too high axail force.

RHABE

No, the 3mm rod is 145mm long.

Misumi has a great choice of 3mm shafts in different materials. As some are hard to machine I just removed a 2-56 threaded hole requirement on one end.

gigaconcept --

After giving some thought to your problem and noting your observation of variations in apparent coefficient of friction I would offer that the root of the problem is the in the hole in the aluminum brake lever. The sharpness of the outer corners of the hole and its angularity to the rod (centerline to centerline) are important.

When the brake is applied the extent to which the corner digs into the rod needs to be understood. When the contact stresses reach the point of general elastic deflection of the surface of the rod the constants of traditional coefficient of friction begin to go away. The sharper the edges and the greater the clearance between rod and hole the worse the problem. I suspect your drawing tolerances are wide enough for you to get unwanted variation in the product performance. Often so called "block tolerances" on the drawing are the culprit.

Also worthy of note here is the tendency of thin hard oxide layer to develop on all aluminum surfaces exposed to air within minutes or hours after exposure. Whether this is a factor in this situation is something I cannot intelligently comment on.

As a first cut at the problem I would seek a simple way of "bellmouthing" or radiusing each end of the aluminum hole and retesting.

To better understand the problem in detail you may need to be able to actually "look" at it. A low power binocular microscope of the type commonly used for small device assembly will be helpful. Depending on how you must communicate the results of your work in your company the use of a digital camera to record microscope images may be helpful. If such a purchase cannot be made in these times of tight budgets at least equip yourself with a 10x Coddington hand magnifier for observations.

Accurate measurement of the 3mm hole diameter especially at the ends is best done with Deltronic Tenth Pins (also available in comparable metric increments) A set of 25 of these costs around US$150 and delivery is something like a couple of weeks. I strongly urge you to move to equip your inspection department with a set that they will let you use in your engineering work.

Should your work show this to be a good solution then the question will be how to implement at minimum cost. Two approaches are suggested:

1. Modify the tooling that creates the hole to produce a desired configuration at each end. This may be special deburring cuts after conventional drilling and reaming or a change in the shape number of core pins (from one to 2) in a die casting die and implementation of a secondary operation to size the hole to required tolerances and remove the internal flash line.

2. Implement a special "precision" controlled edge forming operation to rework parts at the final assembly stage or someplace close to that stage. I would favor some of this in a pilot production environment should it be desirable to go through a learning curve with you present for oversight. Very soon you will want to eliminate this operation for cost reasons. This approach usually leads to well developed process controls and instructions to support future manufacturing quality while arming you with the knowledge necessary to confront future challenges and product design changes should they become necessary.

I could make other recommendations that involve materials or major design changes; but it seems to me that you are beyond the stage in development of this product where sch changes are practical.

Ed Weldon