The BSA A7-A10 Forum

Technical => A7 & A10 Engine => Topic started by: MikeN on 20.05. 2009 12:02

Title: metalurgy.
Post by: MikeN on 20.05. 2009 12:02
Any metalurgists out there got any knowledge of the life of highly stressed alloy or steel engine components?
 Do they become "time expired"?
 Anyone ever x-rayed or ultra sounded a con rod or crank?
Im just curious.
Title: Re: metalurgy.
Post by: fido on 20.05. 2009 13:19
I used to work as a metallurgist back in the mid '80s but have never had chance to do the tests you mention on bike components. Manufacturers tend to work on a minimum fatigue life for components, only in aerospace do they test for maximum life. For example, back in the '70s I once read that Honda considered an engine to be reliable if it could be run at maximum power for 50 hours.
For the DIY bike restorer it would not be worthwhile buying expensive test equipment and sending parts to a test house would also be quite costly. An alternative is the dye penetrant test, which might be worth a try:
Title: Re: metalurgy.
Post by: BSA_54A10 on 20.05. 2009 13:53
Gees why do I feel like I am being baited.
The answer to that question would be the result of several thousand doctorial thessis ( thesi ?).
Con rod makers got their act together very early re quality control.
It is very rare for one to fail from the inside out they usually fail from the outside in.
When I was actively doing ultra sound we could not do castings as the coarsest wave we could generate either got reflected or refracted by the rather large grain boundries. Then again these were the days of a trace on a screen or if you had the research grade gear ( and I did ) you could get 3 traces on the screen.
Metallurgy would have to be one of the most conservative fields that I have ever come across.
I was banned from tendering to one government department because my CRT had a "revolutionary" square screen.
As for x-raying yes done thousands of them, tiny little buggers from refrigeration compressors a right royal PIA and very expensive ( 4 exposures each one), Then again so is a container full of rancid meat or cooked vaccines.
Never done a motorcycle one.
The grade of aluminium used for rods ( it is an alloy specifically designed for this application) is very fracture tough and extreamly notch insensitive so they will take a lot of abuse before they give up the ghost. Design wise the stuff we play with is also about twice as heavy as it needs be .( stretch excluded).

Steels otoh tend to fracture from the inside out.
The actual mechanism for this is migration of dissolved hydrogen atoms to the points of greatest internal stress . When you get a few of the buggers there they form a hydrogen atom and this causes the crack to start and travel through the body of the metal.
My old prof (Hugh Muir ) got an Acta Metalurgica ( metallurgy's Nobel prize) for his work on this and as an undergraduate I did about 0.0000000000000000001 % of the work on the project.
This is why double vacuum melted steels are so strong ( no dissolved hydrogen) they are also just a tad dear for fitting onto the average A10.

There are tables of the fatigue limits of just about every material known to man and without rummaging around to find my ASTM handbooks & data sheets Aluminium casting alloys go around 10,000,000 to 100,000,000 cycles to failure. The best alloy for fatigue resistance is the one used for wheels AP601 or BP601 and this one is good for 1,000,000,000 cycles.
Note that is not how many stress cycles that your rods can go through, that is the number of cycles a prepared bar can go through on a test rig and is purely for comparing one alloy to another.
Heat treating can vary these results and elevated operating temperatures tend also to increase the cycles to failure. OTOH creep ( or stretch) is increased with temperature so as always it is a balancing act.

Working life of  component is used to be determined by running one in a rig till it fails, Then doing the same till you have enough data to fulfill the percentage certency required.
If you want to be 100% sure, tough luck.
99,99% is around 1473 determinations
90% is around 1200.
If I was making bits for an aeroplane or rocket ship then I could justify destroying 1500 of them first but for a BSA it might be just a tad dear.

However now days you an do some really clever computer modeling because so much more data is available and we know so much more than we did in the days when Hele & Hopwwod were designing motorcycles.

So what exactly did you want to know ?
Title: Re: metalurgy.
Post by: A10Boy on 20.05. 2009 18:09
You know, that's just what i was going to say!!
Title: Re: metalurgy.
Post by: MikeN on 20.05. 2009 20:59
BSA 54-A10,
thanks for the reply.
  I was thinking about 55 year old alloy con-rods. Are they likely to go "snap".
   Or does age not bother them?
Title: Re: metalurgy.
Post by: beezalex on 20.05. 2009 21:56
Mike, age has little to do with it, but the number of times the load has changed on them and how much load was applied in each of those changes.  Spin it up to 10Krpm and it won't last long.  If most of it's life has been spent bimbling around the countryside, it'll likely last a very long time.  IMHO, best thing to do is to inspect the rods for cracks using magnaflux.  As trev says (and I'm roughly paraphrasing into english here) they cracks tend to start from the surface, so they are typically not hard to find.  That said, I've done this on several rods, particularly the ones on my A10, which appears to have had a very hard life (if sleeved AND overbored to .040" is any indication) and have yet to find any cracked rods.  I've seen a few broken and bent ones, but no cracked ones.
Title: Re: metalurgy.
Post by: MikeN on 20.05. 2009 22:05
Thats interesting,Thank you.
Title: Re: metalurgy.
Post by: BSA_54A10 on 21.05. 2009 01:45
Calendar age has little to do with most alloys unless they have been heat treated prior to use. Any aging on a rod would tend to be countermanded by the constant heat cycles.
Operating hours is what you should be concerned about with regards to "age" and in particular operating hours at various levels of operation. Engines that have been raced and run up to 8000 regularly with gay abandon will not last long as they are operating right at the top end of their design parameters as would rods from a sidecar hack geared up to "save fuel" and dragging around mum on the back and 3 kids + gear in the double adult car.

Next worry is storeage conditions. if a piston has been stored in a damp condition then it could fail due to intergranular corrosion that you can not see with the naked eye. Even then the piston needs to be polished then etched then examined with very powerful magnification or tested with dye penetrants.

So as Fido said, I would trust an old factory rod or a modern brand name rod ( Corrillo etc) over anything that I see on evilbay and in particular any thing that is "machined from billet".
Items like rods & pistons derive their strength and stiffness from very carefully controlled grain structure and size and just like you can not make a wax impression of a forging then expect the casting to have the same strength you can not expect a machined perfect surface reproduction of an item to have the same properties of the cast or forged items.