Estimating break-away force for an interference fit?

[mrpi]

I have a 45mm ball bearing shrink fitting onto a 303 shaft.

The ball bearing has an ID of 45mm at P6 tolerance.
The shaft has an OD tolerance of k5 as recommended by the bearing manufacturer (SKF).

This makes for an interference of:
maximum = 0.013mm
minimum = 0.002mm

How can I estimate the torque to break away this interface?

What should I use for friction coefficient?

steel-steel = 0.80?

Are there tables for these kinds of fitments?

Thanks.

Beat to fit, paint to match.

 

[KENAT]

faq404-1230 might be of some help although I think it only covers dowel pins. I thought a similar question had been asked before, you might be able to find it with the search above.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[electricpete]

Torque = (f*Pc*A) * radius
where
f = static coefficient of friciton
A = 2*pi*r*L
Pc = given in the FAQ above


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[GregLocock]

You would be most unwise to use mu=0.8, that implies great cleanliness.

Typical mu for steel on steel is around 0.2-0.3. This is what I use when calculating press fits for bearings and it agrees pretty well with the measured insertion force.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[corus]

Coefficient of friction for steel is typically 0.3 but for design purposes we used to use a value of 0.1 given that in the workshop you might have oily dirty surfaces that reduce the coefficient on a subsequent refit or whatever.

ex-corus (semi-detached)

 

[TheGopher]

http://www.tribology-abc.com/calculators/e6_2.htm

or

http://www.engineersedge.com/calculators/machine-design/press-fit/press-fit-calculator.htm

These are usually simplifications of Lame's equation considering plastic deformation.

You must also consider the burst stresses and deformation consequences (ie if an internal (female) thread for a flange bolt is going to exist on an internal diameter of the shaft).

A lot of this is best done in FEA to confirm the calculator results.

Poisson's for steel: 0.3
coeff. friction steel: 0.2-0.3
(depending on surface finish, search online for data)
modulus of steel: 200Gpa
Yield: 50Mpa
(depending on carbon content/quality)

 

[mrpi]

Thanks for the quick replies, I had searched the forum but didn't find anything.

I didn't know about the FAQ, that's pretty handy.

Greg & corus, the friction coefficient information is very helpful.

Gopher, I was actually just looking through the equations that calculator uses. This led me to the NTN bearing website, and this handy "Rolling Bearing Handbook"

http://www.ntn.co.jp/english/products/pdf/bearing-hb/index.html

It looks like that has most of the information I need.

Thanks again.

p.s. I also just read the "rules" that mention performing internet searches and FAQ searches before asking questions. Oops! But then again, what else would we be talking about around here!

Beat to fit, paint to match.

 

[mrpi]

Ok, I ran the calcs using us = 0.1 to be conservative and got:

Tmax = 300Nm @.013mm interference
Tmin = 50Nm @.002mm interference
Tavg = 180Nm @.0075mm interference

Does that sound reasonable?

Beat to fit, paint to match.

 

[mrpi]

I think that FAQ has a typo, there aren't enough "]" brackets and no "Uo".

x/[Dc*[((Dc^2+Di^2)/(Ei(Dc^2-Di^2))+((Do^2+Dc^2)/(Eo*(Do^2-Dc^2))-((Ui/Ei)+Ui/Eo))]

Should it read?:

x/[Dc*[((Dc^2+Di^2)/(Ei(Dc^2-Di^2))+((Do^2+Dc^2)/(Eo*(Do^2-Dc^2))-((Ui/Ei)+Uo/Eo))]]

???

I attached a .JPG of the equation in question.

Beat to fit, paint to match.

 

[mrpi]

Ok, I re-calculated using the NTN bearing handbook equation for surface pressure and get:

NTN = 29.2 Nm

Then I used the tribology calculator here:

http://www.tribology-abc.com/calculators/default.htm

And I got:

Trib calculator = 29.2 Nm

Which makes sense since NTN equation assumes a steel shaft and a steel collar, so the Poisson's ratio's etc cancel out.

Beat to fit, paint to match.