Metallurgical Question

[pugwash]

Are all halfshafts destined to snap?

For example i have an FWD car that has done ~200k miles. Considering the lack of impact and shock loading (as the car is an auto FWD car) will thes halfshafts eventually snap? Does metal get weaker due to use or does it "work harden" with time?

Will the halfshafts in my car be stronger or weaker than when they were first installed?

if metal isn't "worn" down, or destroyed by rust, does it have a useful life? i would guess it would be in the tens or hundreds of years?

This got me thinking of LR halfshafts, do they break with such frequency purely because they aren't specified for our needs? or do older ones break with due to "age"?

(this question discounts vagaries of manufacturing and wear patterns- ie ignore fractures due to bad manufacturing)

[rtbarton]

Work hardening will increase the chance of snapping. Ideally halfshafts should have some give.

I'm sure all shafts will eventually snap, perhaps after 100s of years use, perhaps after one year, depends on useage and design.

In the Riley RM series of cars the 2-1/2 litre cars are notorious for snapping shafts, the 1-1/2 not so, but they do snap. The reasons being the more powerful cars had a poorly designed shaft.

Shocks will increase the chance of snapping, so as the diff and drive members wear risk of snapping will increase.

Shafts don't always snap under heavy load, I snapped one (Riley 1-1/2) pulling away gently from traffic lights.

[bishbosh]

Fatigue life is a function of numerous factors including geometry, material spec and not least cyclic loading.

So a metal element subjected to repeated cyclical loading will eventually suffer from fatigue and cracks will start to form and propagate leading to the eventual failure of the element.

As you suggest, fatigue life (number of cycles) can be very short for poorly detailed highly stressed elements to hundreds of years for well detailled elements under low cyclical stress variations.

I think work hardening would only affect the surface of the element in contact with something else (like the splines for example) but you'd need to ask a grown up to be sure...

[Superpants]

I will have to dig out my books at home for a fuller answer, but most metals have a fatigue limit, a nominal stress under which you can cyclically load the material pretty much indefinately, however this figure is very much lower than the yield stress of the material (I forget how much by), and I suspect we regularly exceed it, therefore driving fatigue of our parts.

Above the limit the life will be reduced by cyclic loading in proprotion to the load (although not directly proportional). Therfore if a part is subject to a cylic load of a constant stress, then the fatigue life of the part is reasonably predictable, if however the stress applied varies (as in many real life cases, including halfshafts) then predicting the failure will prove extremly difficult.

The effects are similar to the workhardening you have mentioned, and are driven by the microsture of the material (grain size, inclusions, dislocations, crystal structure etc), but I forget the details now as I haven't picked up my fracture mechanics books since I left uni!

"This got me thinking of LR halfshafts, do they break with such frequency purely because they aren't specified for our needs? or do older ones break with due to "age"?"

This will be a combination of factors- there will be an effect due to "age"- the fatigue life above, but I suspect (and can't be sure because I haven't been inspecting broken half shafts) that the failures we see are probably more due to a higher load than the tensile strength of the material- however I stand to be corrected by someone with more experience!

[edwardatherton]

Shafts don't always snap under heavy load, I snapped one (Riley 1-1/2) pulling away gently from traffic lights.

Dad snapped the one in the 2A setting off downhill from the end of the drive - this was after going across the car park, along the drive, and over a speed bump...

[will_warne]

I will have to dig out my books at home for a fuller answer, but most metals have a fatigue limit, a nominal stress under which you can cyclically load the material pretty much indefinately, however this figure is very much lower than the yield stress of the material (I forget how much by), and I suspect we regularly exceed it, therefore driving fatigue of our parts.

Above the limit the life will be reduced by cyclic loading in proprotion to the load (although not directly proportional). Therfore if a part is subject to a cylic load of a constant stress, then the fatigue life of the part is reasonably predictable, if however the stress applied varies (as in many real life cases, including halfshafts) then predicting the failure will prove extremly difficult.

The effects are similar to the workhardening you have mentioned, and are driven by the microsture of the material (grain size, inclusions, dislocations, crystal structure etc), but I forget the details now as I haven't picked up my fracture mechanics books since I left uni!

"This got me thinking of LR halfshafts, do they break with such frequency purely because they aren't specified for our needs? or do older ones break with due to "age"?"

This will be a combination of factors- there will be an effect due to "age"- the fatigue life above, but I suspect (and can't be sure because I haven't been inspecting broken half shafts) that the failures we see are probably more due to a higher load than the tensile strength of the material- however I stand to be corrected by someone with more experience!

Superpants is totally right about this - there is a lower limit for fatigue. When material is fatigued it is changing, these changes can be very slight but will eventually lead to fractures and eventual failure. The reason work hardening occurs is due to the presence of small defects in the material – all materials have them and one of the measures of its quality is what size and how numerous the defects are. Around the defect the structure of the material has, by definition, lost its uniformity meaning some of the atoms won’t be in ideal positions relative to one another. This means some of the intermolecular bonds are longer (and so weaker) than they would be in the uniformly structured material. When a load is applied it is ‘concentrated’ around the defect as the weak intermolecular bonds are the ones which’ll break first. This allows the defect to propagate and the beginnings of a fracture to form. These failures are very small scale but over time add up. However, if the load applied is not enough to break any bonds (or more accurately in metals move atoms given the type of bonding they exhibit) then no fatigue occurs.

I'm not sure about the effects of just time on the material - I don't know enough about alloys. Clearly oxidation will have an effect but only at the surface. The operating temperatures shouldn't be high enough to allow the crystal structure of the shaft to change or to allow the elements in the alloy to come in or out of solution so I doubt the materials have a shelf life. However, I don't know enough to say for certain.

[Mark]

I have snapped shafts and blown diffs doing very mundane road driving in the IIa.. I have always had to work harder in the disco, but the reult is still the same...

[Turbocharger]

I seem to remember it's as mentioned above - the fatigue stress limit is that at which the material will withstand 10 million cycles (I think). The general theory is that, if it'll do that many, it'll go on forever. Otherwise you're playing a waiting game based on the magnitude of the loading and the different between the max and min loadings.

Unusually, Wikipedia is quite helpful on the subject.

[Roverdrive]

Superpants

Sounds about right to me. That's why when you build things like a brick sh*t house they don't break as you are not exceeding the lower elastic limit ( or whatever it is called)

As for snapping half shafts under low load, it is probably after over stressing it previously, and starting a crack in the material. The crack propogates, but there is still enough metal to transmit the applied power until it eventually breaks.

When I used to comp a lightweight, if I had a really heavy landing with power on I would change the half shafts - assuming I got to the end of the event without incident.

[Tobias]

Are all halfshafts destined to snap?

For example i have an FWD car that has done ~200k miles. Considering the lack of impact and shock loading (as the car is an auto FWD car) will thes halfshafts eventually snap? Does metal get weaker due to use or does it "work harden" with time?

Will the halfshafts in my car be stronger or weaker than when they were first installed?

if metal isn't "worn" down, or destroyed by rust, does it have a useful life? i would guess it would be in the tens or hundreds of years?

This got me thinking of LR halfshafts, do they break with such frequency purely because they aren't specified for our needs? or do older ones break with due to "age"?

(this question discounts vagaries of manufacturing and wear patterns- ie ignore fractures due to bad manufacturing)

Basics of fatigue has nbeen farily well covered already.

The material itself does not age, so the material itself is as strong as when the car was new, but it may well contain cracks, which weakens the component. As the crack (s) propagate there is less and less area left which is accying the load and eventually it is small enough that a very small load can snap the last part.

Aluminium notably does not have an endurance limit, as steel, hence crack tests in airfraft components. When the crack is long enough the part is changed out.

For engine components, which see an enourmous amount of loadings and unloadings it is often the endurance limit that is the designing factor, while Axle/chassis components see fewer rpms and hence fewer loadings and can be designed to a higher level of stress and a set fatigue life.

The simple answer is that LR half shafts are not designed for the loads. The reason the RR was full time 4x4 I think was just to spread the load on 4 halfshafts because 2 were deemed not strong enough.

Tobias (works with fatigue related problems ;-) )

[rtbarton]

The reason the RR was full time 4x4 I think was just to spread the load on 4 halfshafts because 2 were deemed not strong enough.

Tobias (works with fatigue related problems ;-) )

That's my understanding too, also the reason why Series Landies select 4WD automatically in low ratio.