Fitting a pressfit in this heat

Short answer - Yes it would cause the frame to expand and the bb to contract :o)

Yeah, that'll be like chucking a sausage down an alleyway, that will.

Ok, so it's in the freezer next to the half empty tube of wetsuit glue.

If you do it right it will.

When you go to put the bb in the freezer, get yourself an ice cream.

When you go to put the frame out in the sun, stand next to it for a while eating your ice cream.

Putting the bb in will still be a PITA but you'll be an ice cream up whatever.

Freezing bearings and heating the housing is a tried and tested engineering technique. I would always suggest doing it whatever the weather. Some tolerances are made so fine it is often the ONLY way to fit bearings.

It would but you don't want it too loose, the point of pressfit is that it's an interference fit and if it's rattling around in there as they return to the same ambient temperature will they be properly aligned at that point.

Short answer – Yes it would cause the frame to expand

Does the hole get bigger or smaller though as [material around] it expands?

Straight answer John is it isn't that hot. They manage to do these things just fine in places where 30 degrees is mild weather.

PF should be snug to fit and take back out. Go slowly and make sure you go square and you'll be fine. Freezing the BB etc will likely help it go in easier but beware that can make the job harder - if it is *easy* chances are it will move out of square as it all reaches ambient.

Well there's a little lip at the ends so when it's all at the same temperature I can see if everything is square. If I can get the chock ice wrapper in I'll give it a tap.

Does anyone else use a mallet to fit press fit bb? Going gentle obviously. Works fine for me and takes minutes

Does anyone else use a mallet to fit press fit bb?

Yes. That being said people use butter knives as screwdrivers.

Have you never made a sandwich using a screwdriver?

Have you never made a sandwich using a screwdriver?

All the orange juice makes the bread go soggy.

Very good.

Plastic pressfit bb might actually harden and be more brittle if you put it in the freezer. Possibly make it more difficult to press in as it won't deform to the bb diameter.

Straight answer John is it isn’t that hot. They manage to do these things just fine in places where 30 degrees is mild weather.

Pressfit is banned in Australia.
And BSA isn’t much better because the threads go the other way down under.

Pressfit is banned in Australia

Are they any good or the antipodean Ed sheeran?

I use that technique to change the bearings in my Hope BB without tools! Put the old cup and bearing in a mug of hot water with the outside of the cup facing down straight out of the kettle, leave for a minute and you'll hear a 'ting' as the bearing falls out. Have your pre-chilled new bearing ready, fish the hot cup out of the mug and place it on a suitable surface facing up. Drop the new bearing in and leave it for a minute or two for the temperatures to equalise and voila, one perfectly seated bearing with no tools and just some water to clean up!

Just be aware if it’s a carbon frame that some (I don’t know if it’s all, I suspect not) carbons have negative expansion coefficients.

They use them as the core of high capacity overhead lines so the heat generated at maximum load doesn’t cause the line to sag.

That's a good point. The bike is aluminium.
It might explain why a carbon windsurf mast snapped in the heat a few years ago. If it was a mix of carbon, glass and epoxy it might have been trying to expand and contract at the same time

That went well!

Apart from the bit where the sun went behind a wall a couple of hours before I planned to do it so I had to move the bike somewhere else to warm up again. But the cups slid in just right - I even remembered that right and left are the other way round when the bike's upside down.

Test ride - nice and quiet. As quiet as your car when it's just been washed.

I always freeze the bearings overnight before fitting. No need to hear the frame unless it's been in a very cold garage/shed

How will you get them out? Happen you won't fit a whole bike in a freezer.

Because I work with 800 physicists and engineers, some of whom ride bikes, I asked a couple at lunch and we don't know exactly why this works.

What we're doing here is using the coefficient of thermal expansion of aluminium, which happens to be one of the highest at 24 x 10^-6 m/(m x K)

https://www.engineeringtoolbox.com/linear-expansion-coefficients-d_95.html

(actually says 21-24, but I'm going biggest case)

So length at T2 = Length at T1 + (Length at T1 x rise in temp x 24x10^-6)

ie a 1 metre length bar raised by 100C (or K) will now be 2.4mm longer. And depending on the other dimensions, will also be bigger in Y and Z co-ordinates too.

BUT

If we take a BB shell to be a essentially a cylinder, and its thickness is maybe a couple of mm, ie 1/500 of a metre. So even if you really heated it by 100C that would mean it would expand by 2.4/500 = about 5 microns. 1/10th of the thickness of the proverbial bawhair? Does that really make a difference, even if the steel's also contracting.

And then as someone pointed out.....if the cylinder is expanding does it not do it equally so is the hole though it, the ID not actually getting smaller as a result of the thickness of the shell expanding slightly?

Should we not freeze them both and make use of the Al contracting?

It's called the torus effect I think. Like when you cook a ring doughnut or bagel and due to the expansion the hole gets smaller. Anyway it worked. And I shouldn't have to replace it ever again because pressfit DUB bearings last forever. Don't they?

but you don't want the hole to get smaller. So you should freeze both so the ID of the shell gets bigger and the OD of the bearing gets smaller

Because I work with 800 physicists

I then fell asleep

Not sure that’s safe with 800 physicists around. They might try and expand your holes.

If we take a BB shell to be a essentially a cylinder, and its thickness is maybe a couple of mm, ie 1/500 of a metre. So even if you really heated it by 100C that would mean it would expand by 2.4/500 = about 5 microns.

But what if we take the BB shell to be essentialy a flat bar coiled in a circle. How much expansion would we get along the length of said bar?

I'll let the physicist do the maths cos I can't be arsed.

If we take a BB shell to be a essentially a cylinder, and its thickness is maybe a couple of mm

OK, I'll have a shot;

It's like inflating an inner tube out of a tyre, innit?

The tube wall is getting a bit thicker, but the dimension I would calculate is expansion of the material that makes up the circumference if you were to unroll it. Since it's constrained as a circle, the growth of the circumference drives the diameter of the cylinder to increase and ease the fit.

A 46mm shell therefore expands 70 times (Pi x D = 144.5ish) more than a 2mm wall thickness.

70 times 5 microns = 350 microns or about 5 hairs, so you now have a sausage in an alley situation...

OK, I did it for a 46mm PF BB shell not headset but the idea is the same...

Hmmm,,,,I see your theory but I don't know if it expands like an inner tube.

I was considering the shell to approximate to a series of bars of say 2mm x 2mm x 68 (or 73) mm.

As per your maths, there would be ca 70 of them that when laid all together long edge to long edge could make a cylinder. OK, they'd all have to actually be slightly wedge shaped but it's an approximation.

I calculated the effect of expansion radially, but as you say if they are 2mm square rods then they'd also expand circumferentially. So yes, the circumference of the shell would expand by 70x while the inner radius would shrink by 1x. Actually I think the radius would shrink by 1/2x because the outer diam would also expand, the expansion is not only inward. So 140x = 10 bawhairs. A 70's German plumber level of hairyness.

As quiet as your car when it’s just been washed.

My car doesn't get washed but if I hoover out all the detritus from kids and dump runs it has this effect.

Deliberate mistake, that 70x is the circumference so the diameter grows by 70/Pi so only 22 times more than the change in wall thickness.

Add back in the impact of the wall, as you say, 1/2 a radial expansion for the ID, but double it for diameter impact brings us down to 21 times or 100 microns, or more than the tolerance of an interference fit.