Fixture / Component Thermal Expansion advice needed. 1

[adamski]

Hello fellow engineers.

Tried searching for about a hour on the site, but can find nothing on this.

I am looking to design/manufacture an assembly fixture out of steel.
This fixture will be for aluminium components.
The components could measure around 20 Meters in length, 4 metres wide and a couple of mm thick.

Could someone post advice on how we should cater for the potential thermal expansion differences between the fixture and the component material?

Please don't ask me the specifics of material or sizes - I am just trying to cost the job at them moment, but ned to know if there are any special considerations I need to account for with regard to the fixture design.

Trust some one can help!

 

[gwolf]

I estimate that for every 10 centigrade above ambient the steel and aluminium structures will show a differential thermal expansion of 2.2mm assuming a 20m initial length. Note that this is the difference between the two structures were they not connected, not absolute length.

You will have to account for this somehow if the structure is likely to get hot, otherwise the least stiff of the two structures will start to buckle and the weakest of the two structures may break.

 

[adamski]

Many thanks Gwolf.

We are looking to locate the aluminium part at either end on some 20mm dia location pins, so it is a worry that after a period of time and with a variance of temperature, the part would 'lock' onto the pins.

Anyone else out there got suggestions?

 

[Kapitan]

This sounds like an aicraft wing or a rail car, by the dimensions. There are two constants to deal with, the coefficients of thermal expansion of steel and aluminium, the former being about half the latter.There are also some undeclared variables, including,

[ul]
[li]Tolerances.[/li]
[li]Temperature Range.[/li]
[li]Operation details, what's going on ?[/li]
[ol circle]
[li]Clamping.[/li]
[li]Bonding.[/li]
[li]Drilling for fasteners.[/li]
[li]Welding.[/li]
[/ol]
[li]Number of Assemblies.[/li]
[li]Budget (probably cheap as possible).[/li]
[/ul]

Some simple questions,

[ul]
[li]Is there any temperature control in the building ?.[/li]
[li]Could the assembly jig structure be made of aluminium ?[/li]
[/ul]


Without knowing the answers to this lot it's difficult to say much.

 

[gwolf]

The general solution to this sort of problem is the "minimal restraint" approach. In this context you would only fix the structure rigidly in three dimensions at one point, two dimensions at another point, and one dimension at a third point.

Alternatively you can place a rubber-like pad between the two structures at all contact points, this way the rubber takes up the differential thermal expansion whilst still holding the component tightly.

 

[adamski]

Many thanks Kapitan / gwolf.

You are on the right lines when you mention Wings.

In the interests of confidentiality, I wish to say no more.

gwolf, you are correct with the "minimal restraint" approach. Not too sure of a rubber interface though. Have discussed encoding one end on a slide, with laser tracking linked to temperature measurement to compensate, but there is a lot of tooling over 20 meters!!

I calculated 1.28 mm difference in expansion / 10deg.
(without specifying material make-ups)

I imagined a facility with sky lights in the roof, beating down sunshine in the summer, getting the tooling/component nice and warm as a worst case scenario.

I just wondered if there was anyone out there who does this type of large scale tooling on a daily basis that could offer some advice.

As for your questions Kapitan:
Tollerance would be +/-0.1mm between location pins.
Temp range would be +/- 10 Deg (estimate)
Operation: Clamping, drilling & fastener insert.
No. of assy's: 15/month.
Budget: correct - economical (I don't like to say the word "cheap", as you get into the habit and then say it in front of the customer)
Temp control: see note above about sun rays.
Aluminium tooling...mmmm...expensive.

Still open to any advice.

 

[GregLocock]

I wonder if anyone makes temperature compensating pins, that would move with temperature? Why not use a temperature controlled room? Is a spring out of the question?

To be honest 1mm in 20000 is well inside my comfort zone for strain, just make the steel stiff enough.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[BillPSU]

On occasions where locating on two round holes the standard process is to use a round and a diamond shaped pin. The round pin becomes your primary locator and the diamond pin holds orientation allowing the part to grow and shrink in the longer dimension. If the tolerance needs even better location an expanding mandrel or a expanding split pin can be used for the primary locating point and a solid diamond pin for the other.

Is there a reason why a steel fixture is required? I would recommend to the customer to use an aluminum plate/tube fixture and make the differential expansion problem go away.

 

[adamski]

Thanks for joining in GregLocock & BillPSU.

I think we will push for a primary hole and a secondary slot in a direction that will accomodate the expansion. We have also discussed an aluminium frame for the component tooling mounted to a Steel Sub Frame - but I guess we just end up moving the problem to the connection between these two.

Off to see some big tooling manufacture people later today - I'm sure they will have the correct solution. Rest assured I will let you know how it goes.

Otherwise I wil cost for a couple of lump hammers to knock the panel off the pins when the assembly is complete!

Cheers all.

 

[tlee123]

The best answer (as mentioned above) is to use a temperature controlled room.

If that's just not possible, then mount to the fixture just constraining the 6 degrees of freedom of the part. This will ensure that the part can't get stuck in the fixture.

Also, there is a clever way to make one feature on the part insensitive to thermal expansion. It's hard to explain in words, but look into Exact Constraint: Machine Design Using Kinematic Principles (author is Blanding by ASME Press), p.21.