Rivets versus welds?
 

At one time steamship hulls were formed from riveted plates, as were automobile bodies. Both have evolved to welded joins. In contrast, aircraft -- from the Ford Trimotor to modern jets -- seem to continue to use rivets to join fuselage members. Given that welds have become so sophisticated -- and corrosion or weak spots at rivet holes remain a problem (see JAL loss of aft pressure bulkhead), why haven't welds replaced rivets on planes? Is it just because aluminum is so much harder to weld than ferrous metals? Or another reason?

I don't think this is the definitive answer, but my view is that it makes panel repairs easier insofar as it's easier to remove smaller aircraft sections than a large welded assembly.

Aluminum is a bit of a bastard after its welded or is exposed to any high tempratures.

It can warp and creep over a period of time sometimes years after you have welded it.

Also as well controlling any welding process is critcal to the quality of the weld and the resulting structure.

Its relatively easy to heat treat large structural steel to normalise the stress levels. From memory ALMG alloy you need to get it up to about 350 degs then cool it in a controlled manner 15deg per hour or some such. The bigger the bit being weld the longer its needs to soak at 350 to get everything to the same temp. Also alot of riveted structures have wires and other non metalic substances under them that wouldn't survive getting heated to 350 degrees for 6-10 hours.

Also thin late is hard to get a smooth surface after welding. Just look at boats and what the call hungry horse between the support ribs. The profile of the surfaces needs to be exact on a aircraft.

Aren't aircraft designed also to have something like 'tear sections' which are small enough to fail if there is fatigue - but not so much as to rip the 'plane to bits like the early Comets...?

Most safety related structures will have crack analysis done to create ductile tearing instead of fast fracture on failure. Its not just aviation.

Its more because of the material properties of Al-Mg alloys after welding that they don't do it.

I thought that both rivets and welds were once (and still?) regarded as outmoded, and that as long ago as the VC10 large sections were effectively glued together.

Gluing is better at carrying certain types of load and rivets other types so they are complimentary.

Also to get the required amount of loadbearing it may be that rivets are lighter than glue.

the h-p herald used the spot welding system, fwd fuse skin to stringer not sure about rear fuse, thats handly-page not the comput......

gs

The standard for aluminum alloys that aircraft are made from would change if welded. A pressurized and stressed airframe would crack at the welds.

Spot Welds
 

Sppot welding aircraft structures (skins to stringers) was a process adopted in the 60's - 70's. The problem with it is that the joint has to be clean on assembly with no interface sealants etc. They are therefore more likely to corrode.
I work on Dassault F20's that have spot welded skins and believe me when I say the process isn't good. Not all the F20's were welded as Dassault even realised it was S**T after a while.
We have to NDT the welds on a regular basis etc. very time consuming and you normally finsd one or two corroded. Welds induce impurities.
A properly sealed and formed rivet is much better and easier to assemble in certain places.
With more and more composite structure these days even the rivets will be disapearing.

The yield strength of most alloys will reduce about 40% in the immediate vicinity of a weld (up to about 2-3 times the material thickness).
It's also harder to verify the integrity of a weld compared to a rivet.

In the automotive industry self-piercing rivets and epoxy seam adhesive is common practice in aluminium bodies, but the SPR technology is better suited to high volume production.

Rivets aren't as strong as the surrounding material, but their performance is consistent and they don't create a heat affected zone.

Pressure and fuel sealing is achieved by 'wet rivetting' with a poly-sulphide or epoxy application to the mating surfaces, which has the added bonus of creating a stiffer structure.

This accident was caused by failure to effect a proper repair according to the manufacturer's instructions.