Aluminum Work Hardening 5

[AlanD]

I've been involved in a long-term debate with people who in my opinion do not have the technical background, over whether Aluminum mast (for sailing) will work harden without the masts becoming permanently bent. The masts are a tubular section of either 6061 or 6063 grade and heat-treated to T6

From my knowledge of dislocation movement work hardening will only occur when the material yields. But below the yield point, the dislocation is unimpeded.

The masts do flex, but it seems wrong to me to think that they work harden while they flex. To make things more difficult, the mast suppliers are now labelling the masts to indicate that new masts should not be used in strong breezes, until the masts have been "broken in".

Does anyone have any comments? I'm particularly interested in any papers written on the subject.

 

[kenvlach]

Please drop the notion of fatigue design stress or failures.
We are dealing with failures of ‘fresh, unbroken-in’ masts, and (to me at least) the amount of mast bowing in some of the photos demonstrates enormous strains impossible for 6061-T6. I agree with bradh’s presumptions although not about work hardening below YS; rather think that there’s variation in the material or an unnoticed amount of plastic yielding.

TVP & CoryPad, I think the mast mfrs. have investigated and aren’t talking. Suppose there are some lower strength areas in the mast due to unevenness of heat treat or wall thickness. Wouldn’t these areas plastically work-harden to bring them up to strength of the elastic regions? Although I would expect some permanent set because I feel the loading is fairly unidirectional (unlike BespinSunset), it may be ~0.5% over a limited span and unnoticed.

Also, with stress concentrations at mast fittings (some of the masts look like rounded vees rather than smooth curves [due to splice?]), so there may be a little bit of plastic deformation of the order 0.05% for stress redistribution. And, a 12+% increase in strength is significant with mast stresses at the point of breaking.

I have persuaded myself that there are some valid reasons for a ‘break-in.’
Alan, any more info on straightness & thickness would be helpful.
I hope someone with more structural expertise than myself can demonstrate a convincing answer.

 

[Metalguy]

One more time, IF the assumption is correct that these masts see varying stresses during operation, can you understand why avoidance of fatigue cracking would be important? Can you also see why the mast would/should be designed such that the bending-imposed stresses would not be anywhere near the materials yield strength?

Therefore a sudden "overload" fracture without a higher-than-normal stress just screams "defective mast".

 

[kenvlach]

Metalguy,
You are pursuing a non-issue. Obviously, the masts see varying loads (see photos). Yes, the masts should be designed to avoid fatigue cracks.
Fatigue design implies S < YS which the bending masts in photos show is untrue (since elongation is only 0.4% at YS). IF the masts were originally designed for a high fatigue life, then either the design was faulty or else inferior material.

“can you understand why avoidance of fatigue cracking would be important? “
-- If masts fail catastrophically when new, fatigue cracking is irrelevant. What does a cure for arthritis matter if the baby is murdered?

“Can you also see why the mast would/should be designed” –- you are proposing a redesign to lower the stress, this thread concerns the actual existing design.

What possible relevance is fatigue design to the manufacturers' recommended ‘break-in’ for new masts?
--maybe S-N curve is upside down????

Alan, wouldn’t the development of stronger sail material over the past 25 years allow letting out more curvature of the sail to catch more air (sorry for non-nautical phrasing)? And, also allow use in stronger winds?

 

[Metalguy]

I'm not proposing any &quot;redesign to lower the stress&quot;. I simply pointed out that the ORIGINAL design must/should have taken fatigue limits into account, and therefore the levels of stress should be far below the YS. So why do some new masts bend permanently in winds much slower than what other masts (apparently same design/size)? Because the stresses exceeded the YS in the area of bending, or there is a buckling collapse on the compression side. So either something other than the moderate winds caused these high stresses, or the mast had lower than normal YS in that area, or there was a buckling problem with the bent masts.

In re-reading most of the earlier messages, it seems that there haven't been any fractures, only bending. So I will change the words &quot;sudden overload fracture&quot; in my earlier post with &quot;bent mast&quot;--the meaning remains the same.

 

[kenvlach]

Metalguy,
Glad to have you aboard the main thread & in agreement that the design is inadequate, etc.
Ken

 

[Metalguy]

Thanks, Ken

Glad I found this website-some excellent answers in every area I've visited.

Metalguy

 

[kenvlach]

Alan,
I read through the Laser rules (earlier, had stopped due to nautical terms I don't understand) and have some questions for you, plus some points for everyone to consider.

&quot;FUNDAMENTAL RULE
The Laser shall be raced in accordance with these rules, with only the hull, equipment, fittings, spars, sail and battens manufactured by a licensed builder in accordance with the Laser design specification (known as the Construction Manual) which is registered with ISAF.&quot;

Any chance to obtain one of these Construction Manuals? We especially need the wall thickness for the mast and verification that the material is Al 6061-T6.

&quot;5. MAST
No mast which has a permanent bend shall be used at any time.&quot;

Below, I found a rule referring to a mfr.-provided 5 degree aft pre-bend. Are there 2 classes of masts (if so, do the mfrs. recommend 'break-in' for both) or do they all have 5 degree prebend?

&quot;PART FOUR
LASER RADIAL RIG AND LASER 4.7 RIG OPTIONS
Part 4 of the Laser Class Rules shall be read in conjunction with the remainder of the Laser Class Rules.
28. LASER 4.7
(f) MAST
Rule 5 shall be amended to read as follows:
5 The Laser 4.7 bottom mast is supplied with a pre-bend aft of approximately 5 degrees. The pre-bend shall not be increased or decreased. No top mast that has permanent bend in it shall be used at any time.&quot;

Alan, for both rules any permanent bend is prohibited (other than by mfr.).
a) Do people try to gain some advantage by cheating a little on this rule?
Or maybe bend & then straighten???
b) There must be some enforcement of this rule, and there must be some tolerance due to original variation as manufactured? ½ degree? What happens to out-of-tolerance masts?

To all: This approximately 5 degree prebend was probably conducted using some type of giant tubing bender to prevent any buckling of the aft surface. This amount of bend is about 1/2 way to the breaking point for Al 6061-T6 and indicates that work hardening to the UTS did occur, and furthermore, there must be some material thinning of the forward mast surface. My interpretation of &quot;approximately 5 degrees&quot; is greater than 4.5 and less that 5.5 degrees, and this variability is a significant matter. Comments?

To all: I measured a total mast curvature of 9.0 degrees from photo 11agstrt. Has anyone else tried to measure mast bend from the photos?

Alan, can you tell us the radius of curvature of this bend or more simply, the length of the non-straight portion of the mast?

Alan, I noticed earlier that non-racing masts were either anodized or painted. Do the Laser masts have anodized finish? (If so, I sure hope it was done after mfr. did bending). Any idea how thick? If no data, I’d guess 0.0005 inch. This would reduce the Al 6061 wall thickness by about 0.00025 inch (presuming only the outer surface was anodized).

To all: (If anodized masts) The anodize could crack and offer lots of crack initiation points if the mast flexed significantly during racing, but how relevant that is for overloading-type failure? My guess is less important than wall thinning.
[it would certainly decrease fatigue properties if those were being considered].

Please, will some mechanical or structural engineer do some computer modelling (maybe backcalculate the wall thickness from the degree of mast bending and Al 6061-T6 props.); I don’t have the means.

Ken V.

 

[mikemoore]

Alan,

99% of all aircraft structure is 2024 T3. There is some 2011 or 2017. The loads that a plane fuselage sees are obviously a function of the load induced, by either flight profiles, landing loads or manufacturing/assembly loads. 6061 is not cosidered to be &quot;structural&quot; or aircraft grade aluminum at all, though some people refer to it as such. The idea of breaking in a mast is proposterous, as someone mentioned earlier there is no break in for a plane. I can however reccomend 2024 T3 or any 2000 series aluminum for a mast. 2000 series aluminum's major alloy is copper, which does work or strain harden over time, but only if flexed or worked past its yeild point. If that point is not ever reached, it could last forever. I can recommend a good friend for this, world famous One design boat racer/builder Mike Mahar of Mahar Spar here in Cleveland OH. His number is 216-249-7143. Tell him Mike Moore said to call him. He loves cheap wine and good cheese.

 

[kenvlach]

mikemoore,
Al 6061- or 6063-T6 is commonly used for masts due to superior corrosion resistance (2024 is perhaps the absolute worst Al alloy w.r.t. saltwater corrosion). Maybe 2xxx or 7xxx aerospace alloys or Kevlar, etc. are used for serious racers like for America's Cup where masts are frequently changed & owners don't worry about lifetime & cost.
Your engineering comments are appreciated, but the mast design & material are 'locked-in' by rules, and it appears that the masts are being used beyond the yield stress (there are even rules regulating bent masts).

To all: Re: &quot;This approximately 5 degree prebend....is about 1/2 way to the breaking point for Al 6061-T6 and indicates that work hardening to the UTS did occur.&quot;
I meant to say about 1/2 way from YS to UTS and then I wondered about necking and further, whether I was being presumptious; could the masts have been bent in the solutionized state and then aged? Comments?

 

[Metalguy]

From Alan-

&quot;The manufacturers are informing us that the masts should not be used in winds over 12 knots, until the masts are &quot;broken in&quot;. A good quality mast can easily survive 30-40+ knots, which is beyond the skills of most people in the class.&quot;

Too bad the boats move-makes it much harder to calculate the mast bending stress!

But I daresay that there would be a LOT more stress at 30+ knots than 13. No way would a slight/moderate increase in strength cope with it.

Y'know, all &quot;we&quot; need is to perform a decent failure analysis on a bent mast, and compare it with a chunk of a good one.

&quot;Mast breakin&quot; indeed. That's better than some of the BS I get over here in Italy, home to &quot;cheap wine and good cheese&quot;!!!

 

[kenvlach]

Metalguy,
Is that the EtOH speaking?

I did some thinking outside the box and figured it all out.
The sailors are breaking the masts. See what I wrote Jan 17 (1st) about preloading the mast by tightening the sail.
Now, if the sails are strong enough and the masts pretty thin, that's a significant stress. And some sailors are stronger than others, so differing preloads. Ergo, the stronger sailors better go sailing for a while in some very light breezes to loosen up the sails and maybe distribute stresses at the fittings before a little bit of additional wind load becomes 'the straw that broke the camel's back.'
Hence, 'breaking-in' is due to the human factor.
And, if they take the sails down, maybe have to repeat the process.

Alan, does this make nautical sense?
An inverse relationship between sailor strength and wind speed for the broken masts?

 

[AlanD]

To answer some of the questions

Wall thickness and outside diameters:
Top section 1.5-2.0mm wall thickness, diameter 51.0-51.5mm
Bottom section (standard rig) 2.5-3.0mm wall thickness, diameter 63.5-64.0mm
Bottom section (radial rig) 1.8-2.0mm wall thickness, diameter 63.5-64.0mm
No measurement obtained for the sleeve wall thickness or diameter, but I would estimate it to be 1.5mm, 55mm.

It is my belief that the wall thicknesses are not sufficient when the extrusion dies are new.

There are no wires etc in the sails and no class limitations on the amount of bend you can generate. However the real limitation is how for you can pull the end of the boom down to the deck, which is part of the way the mast is bent.

15 years ago there was a change in the cloth weight of the sails, which increase the stiffness of the sail. However the materials haven't changed since then. However I believe the problem existed before this change and the wind strength we sail in also has not changed.

Even as the Australian measurer, I do not have access to a construction manual. I definitely know that the alloy is either 6061 or 6063 with a T6 ageing process, but no specific knowledge of the YS, UTS or hardness specifications.

There are 3 types of mast, depending on the sail area. The standard rig and radial rig bottom sections are mentioned above. The radial section is shorter and has a sleeve in it to permit a better sail, if a cut down standard section is used, the bottom section is too stiff to permit a fast sail shape to be achieved. The cut of the sail can achieve in achieving a fast sail shape; the reduced length of the mast did not permit enough bend for the sail to achieve this.

The 4.7 mast is smaller again and is designed for the use by young teenagers or very light adults. In order to achieve a fast sail shape, the designers found it necessary to pre-bend the mast, as even a thinner wall thickness was inadequate. Only the 4.7 mast has a pre-bend, the other two are both straight.

Part of the characteristic of a fast sail shape, comes back to the balance and having the tip of the mast back far enough (a bit hard to explain).

&quot;a) Do people try to gain some advantage by cheating a little on this rule?
Or maybe bend & then straighten???
b) There must be some enforcement of this rule, and there must be some tolerance due to original variation as manufactured? ½ degree? What happens to out-of-tolerance masts?&quot;

There is no deliberate cheating to my knowledge on this rule. When people bend their masts, it becomes fairly obvious and other competitors will comment. Most competitors will attempt to straight the masts, particularly top sections that straighten easily. We don’t have a tolerance measurement unfortunately, my general rule is if I can observe that the mast is not straight, I will ask the competitor to straighten it.

The 4.7 bottom section is bent after ageing, using pipe bending equipment, however as the sail is so small; they do not appear to bend accidentally like the other mast sections.

&quot;Alan, can you tell us the radius of curvature of this bend or more simply, the length of the non-straight portion of the mast?&quot;

The top section will bend more than the bottom section and they will be a bit more distortion where the two sections join.

All mast/boom sections are anodized where they are produced, the bend in the 4.7 rig done at the laser builder, so post anodizing. I haven't ever checked the thickness of the layer, but it would be about the standard for commercial anodizing (not hard anodizing).

&quot;To all: (If anodized masts) The anodize could crack and offer lots of crack initiation points if the mast flexed significantly during racing, but how relevant that is for overloading-type failure? My guess is less important than wall thinning.&quot;

I've never observed any cracking of the anodized layer caused by ordinary flexing or when the masts have been slightly bent. However when the masts have been badly bent or dented, yes the anodized layer has crazed. I don't think the anodized layer will significantly effect the properties of the mast.

I did my thesis on the corrosion aspects of anodized aluminum copper alloys, the corrosion issues are extremely significant. The way aircraft avoid the problem is by cladding the parts in a layer of pure aluminum.

I apologise if this post doesn't make to much sense, my mind is else where; we are having significant bushfire (wildfire) problems in Australia. Where I am is ok, but I have friend in trouble.

 

[AlanD]

&quot;Alan, does this make nautical sense?
An inverse relationship between sailor strength and wind speed for the broken masts? &quot;

I see what your aiming at, but most of the sailors will not treat there masts diffenrently irrespective of their strength. I'm one of the stronger people and as I do not believe in this &quot;break in&quot; concept, I will just use a new mast or sail in the conditions presented on the day. I will however aim to purchase a heavier mast section. A heavier mast section should have a thicker wall section.

 

[kenvlach]

Alan,
Thanks, and congratulations on being correct:
“It is my belief that the wall thicknesses are not sufficient when the extrusion dies are new.”
Also, there is no reason to speculate about work hardening below the yield stress; the YS is clearly being exceeded.

From your information, I think we can accurately say that the masts are failing due to gross overlading. The combination of inadequate mast wall thickness, preloading by tightening of sail for racing advantage, and wind loads can cause stresses beyond the YS and in some cases, beyond the UTS. The description of masts that are bent and straightened shows clearly that the YS is exceeded. Work hardening occurs via plastic deformation during both the bending and the straightening processes.
The variations in mast thickness, tightening of sails, unnoticed small but significant bending (remember, the YS is reached at only 0.4% deformation) and probably similar variations in thickness and tightening of fitting materials no doubt made it more difficult to figure out.

I’d like to point out that this bending and straightening of masts, with work hardening occurring both ways, can explain the rationale for ‘break-in,’ although probably not one that the mast mfrs. would admit to. Due to sail preload and then wind loading, the thinner masts bend when the YS is exceeded in a light breeze. If this had happened in a strong wind, the UTS would have been exceeded at the highest stress concentration in the mast, and the degree of plastic deformation (which I’ll call ‘necking’ for simplicity, although there may also be buckling) would have decreased the mast cross section enough that rapid necking and gross failure would have occurred. During the straightening, some recovery of the cross section occurs (at least, any buckling is removed), and the work hardening results in this section being stronger than other. During the next sailing where bending occurs, the next most stressed section ends up getting work hardened. After enough such bending and straightenings, the YS approaches the UTS for all of the most stressed sections of mast, and the mast is now ‘broken-in.’

I’ll repeat my request, “Will some mechanical or structural engineer please do computer modelling for the mast; Alan has provided the necessary info.”

Ken V.

P.S. When you mention purchasing a heavier mast section, do you mean taking along a micrometer and tape measure to select a standard bottom mast of 3.0 mm wall and 64.0 mm diameter, i.e., at the upper tolerance limits?

 

[AlanD]

Obviously once the masts have been bent and straightened again, they do work harden. However the manufacturers are claiming that if we just sail with the mast in light winds, that the masts will work harden and this will prevent them from bending in the first place, it this point I don't understand metallurgically.

I usually take a set of kitchen scales actually. Weighing them is a fairly accurate way of measuring, for our purposes. I just take the heaviest sections in stock.

Thanks for your help.

 

[kenvlach]

Alan,
It's been interesting, kind of using the Socratic Method to bring you to an answer that you really knew.
But there may be further developments.

I've got a subtly to refine the 'break-in' rationale in my last posting. Recall that the YS for longitudinal tension is slightly higher than for longitudinal compression [or better, see the Al 6061-T6 curves in MIL-HDBK-5H]. When the mast flexes near the nominal YS value, it is possible for the inner, compression side to begin plastic deformation while the tension side is still elastic. To further make this case, I point out that the radius of curvature on the compression surface (aft) is smaller than for the tension side, so the stress and strain are both higher. For the case of just slight plastic yielding in compression and none in tension, it is possible for the compression side to work harden a bit and then, when the stress is reduced or removed, the tension side pulls the mast nearly straight. Then, as I explained above, this section is slightly stronger, so the next bending and straightening cycle work hardens the next highest stressed area. Ad infinitum.

So, my 'break-in' rationale can explain the mfrs. claim. Scary thought, isn't it? Or, as Alice said, &quot;It gets curiouser and curiouser.&quot;

 

[aluminumphil]

Just a few comments about aluminum (and aluminium!).

1. All aluminum alloys can be work hardened - heat treatable or not.

2. Aluminum does not have a yield strength. Under stress, it starts to deviate from the Hook's law line right from the get go. The &quot;yield strenth&quot; is the 0.2% proof stress - that is, it has already accumulated 0.2% of non-reversable strain at that point.

Regarding the discussion about aircraft. I served on the US Air Forces MIL-HBK-5 committee for many years. All aluminum structure on an aircraft has a life expectancy and must be replaced at some point. Low cycle and high cycle stresses inflict damage at stresses well below the &quot;yield strength&quot; which is why old aircraft are retired. In recent years, advances in aluminum metallurgy have produce many damage tolerant aluminum alloys. Unfortunately, none of them like sea water.

 

[Metalguy]

I thought the aircraft industry used a plastic strain of 0.02% for their def. of YS (10 times less than others). Perhaps they have abandoned this?

Your comments on cyclic stresses are why I brought up the subject of fatigue cracking here even tho the masts don't fail via fatigue.

But maybe Kenvlach is onto something here. Will be interesting to see just how much increase in YS (0.2%) 6061-T6 will give during cold work. I don't have any decent alum. books with me.

 

[kenvlach]

Alan & Metalguy,
I'm not a mechanical engineer, but I think that what aluminumphil said about YS broadens the stress range to which my proposed mechanism applies. So, even more likely that low-cycle work hardening occurs.
Agree?

 

[Metalguy]

Ken,

I suspect there is a only a slight increase in YS given the very small strain involved. The differences in wind speeds would seem to indicate a large increase was required, even tho the boat is not anchored.

I'll still go with the idea of defective masts, either because of thin/marginal thickness or a problem with the YS/HT.

A good met. lab. would be able to tell what is going on.