Shrink fit steel pin into an aluminum piece

[WobblyHand]

You're not shrink fitting, you're force fitting. It should be no problem pressing your pins into a 0.002" undersized hole - even 0.003" should be doable. Heat shouldn't be required, but it can't hurt to heat up the Al bar before you press. Any press will do this with ease. I've even used a mill vice to do the pressing.

Umm, no, I want to shrink fit, by heating the aluminum and hopefully dropping the cold pin in place. Don't know if this will really work in practice without an arbor press, so I will make a couple practice tries before attempting the real pieces.

Other posters mentioned using arbor presses, which I do not have. As for using a vise, that will be a hard pass, as the last time I attempted using a vise as an arbor press, the cast iron nut in the vise broke into half a dozen pieces. And that nut was unobtanium. Nope, not using a vise as an arbor press again, learned that expensive lesson.

 

[SLK001]

Okay, be careful with the liquid nitrogen - it can freeze you solid.

 

[JPMacG]

The coefficient of thermal expansion for aluminum is 13 parts per million per degree F. For example a one inch long bar of aluminum will expand in length by 13 micro-inches for every degree F it is heated. Or, a 10 inch bar of aluminum would expand by 130 micro-inches for every degree heated. Or, a 0.1 inch hole would expand by 1.3 micro-inch for every degree heated.

The second paragraph of your first post was confusing to me. I'm just trying to clarify how it works in case you had any misunderstanding.

 

[Deleted member 69715]

Okay, be careful with the liquid nitrogen - it can freeze you solid.

Nitrogen can do that. The method I've used to "shrink" metal is to use a container of acetone set into and surrounded by dry ice. Leave for a short time until I see the acetone evaporating and about to expose the part. Use a suitable set of tongs or such (NOT your fingers!) to grab it and slip it in where you want it.

 

[WobblyHand]

The coefficient of thermal expansion for aluminum is 13 parts per million per degree F. For example a one inch long bar of aluminum will expand in length by 13 micro-inches for every degree F it is heated. Or, a 10 inch bar of aluminum would expand by 130 micro-inches for every degree heated. Or, a 0.1 inch hole would expand by 1.3 micro-inch for every degree heated.

The second paragraph of your first post was confusing to me. I'm just trying to clarify how it works in case you had any misunderstanding.

Thank you for the clarification. This is why I originally posted, to see if I misunderstood. So If I am understanding this correctly, if I heat up aluminum from ambient, the hole size would change by 0.2490 * ( 1 + 13e-6 * deltaT1), where deltaT1 is positive. Likewise if one were to cool the steel pin, the change would be 0.2495 * (1 + 6.5e-6 * deltaT2), where deltaT2 is negative. It seems the expansion and contraction are as they say, not much. Going to take relatively high temperature differences to make this work. Possibility of screw ups is increasing...

Did I get the math right? How much clearance between the pin and hole do I need for it to drop in easily? 0.001? 0.005?

 

[rabler]

If you are using this on your cabinet brackets, where the exact diameter of the steel pin is not critical, one way to cheat is to buy steel taper pins from McMaster, and a taper pin reamer. That would allow you to easily and quickly set the pin, and a little temperature difference would lock it in. Admittedly this lacks the elegance and learning experience of working the clearances.

 

[WobblyHand]

I need to rethink this. With these small diameters, if I heat the Aluminum to 450 F, and cool the steel pin to -320 F, (liquid nitrogen) the total clearance of the pin to the hole will be 1.36 mils. I don't see how one could get a drop through with only 0.68 mil gap around the pin before it seizes up. Probably will have to do something more mundane. Was interesting to learn about this, though.

[FONT=courier new]#!/usr/bin/env python3

from numpy import around

if __name__ == '__main__':
  roomTemp = 70
  ovenTemp = 450         #
  dHot  =  ovenTemp - roomTemp   # set oven to ovenTemp
  dryicenacetone = -90    # -90F
  liquidnitrogen = -320   # boiling point
  dCold = dryicenacetone - roomTemp
  dCold = liquidnitrogen - roomTemp
 
  TCEAl = 13e-6           # 13ppm/inch/deg F
  TCEFe = 6.5e-6
  pin_dia = 0.2495        # measured
  hole_dia = 0.2490       # reamer size
 
  coldPinDia = pin_dia * ( 1.0 + TCEFe * dCold )
  hotHoleDia = hole_dia * ( 1.0 + TCEAl * dHot )
 
  clearance = hotHoleDia - coldPinDia
 
  print("Hot Temp  = ", ovenTemp, "F")
  print("Cold Temp = ", liquidnitrogen, "F")
  print("Room Temp hole diameter = ", hole_dia)
  print("Room Temp pin diameter  = ", pin_dia)
  print("Hot Hole Dia = ", around(hotHoleDia,5))
  print("Cold Pin Dia = ", around(coldPinDia,5))
  print("Cold/Hot Total Clearance = ", around(clearance*1000,2), " mils")
 
  # what about if both are heated?  Both expand but at a different rate
  hotPinDia = pin_dia * ( 1.0 + TCEFe * dHot )
 
  print("Hot hole diameter = ", around(hotHoleDia, 5))
  print("Hot pin diameter  = ", around(hotPinDia,5))
 
  print("")
  print("Hot clearance = ", around((hotHoleDia-hotPinDia)*1000, 2), " mils")[/FONT]

Output:

In [21]: run shrinkfit.py
Hot Temp  =  450 F
Cold Temp =  -320 F
Room Temp hole diameter =  0.249
Room Temp pin diameter  =  0.2495
Hot Hole Dia =  0.25023
Cold Pin Dia =  0.24887
Cold/Hot Total Clearance =  1.36  mils
Hot hole diameter =  0.25023
Hot pin diameter  =  0.25012

Hot clearance =  0.11  mils

 

[jwmelvin]

I agree this isn’t a good application for heat-shrink fit. Just hammer the pin in.

Or use a screw and thread the aluminum?

Edit: hammer isn’t really a joke. You can drill a sliding-fit hole for half the depth and that would hold the pin straight to get hammered through the interference-fit portion.

 

[SLK001]

Edit: hammer isn’t really a joke. You can drill a sliding-fit hole for half the depth and that would hold the pin straight to get hammered through the interference-fit portion.

If I didn't have a press, or didn't want to use my vise to press them in, I would also resort to a hammer. Just make a jig of some metal that the pin will slide into easily, place it over the hole, put your pin into the jig and hammer it home.

 

[WobblyHand]

So far, a screw with locktite sounds like the best idea for me. Was hoping to avoid it, but with no arbor press, my options are limited. Not anxious to hammer on the brackets I made. Still looking for an arbor press, though.