Casting in aviation
Rather than further thread jack MP' thread about the effect on Young's modulus of casting defects, I thought it better to start a specific thread about casting in aviation.
Foundry Joe brought up a good point on forging verses casting in aviation...
That is a great question, I spent years tracking down something similar, I could never get my test bars to consistently meet the published values
Ultimate Tensile Strength Yield Strength Elongation
Typical Handbook 34,000 PSI 24,000 PSI 3.50 %
Handbook values came from ALCOA during WWII and were called out as "typical"
Results reported are actually “what a good foundry can achieve on a good day” not average and definitely not a minimum
These results may or may not have been from separately cast test bars and not excised from actual castings
The lawyers got involved and no company now accepts responsibility for these results due to fear of lawsuits if a design fails
Auto designers decided to find out what they were actually getting in their parts
This data was from excised samples from production castings made at high quality foundries, suspension parts, parts driving down the road
Industry was surprised to see huge scatter within part and across parts
Strength is pretty good, buy look at elongation as high as 17% and as low as zero….porosity, oxides and shrink are killers for elongation and fatigue strength
That is why forgings are used in Aerospace, no inclusion defects....
Values are approximate
Ultimate Tensile Strength Yield Strength Elongation at Break
Max 51,000 39,000 17%
Avg 41,000 32,000 8%
Min 36,000 25,500 0.7%
So for design Yield strength would be very low after safety factor is allowed, how high would you go - 15,000?
Unfortunately I don’t have YM data, but you would have to down rate, maybe around 25% of the published values?
- - - Updated - - -
Since it came up, in aviation, we require additional factors of safety with castings, due to the inconsistencies inherent in the process.
For example, for a casting used in critical structure of a small aircraft, the requirement 23.651 would require a factor of safety of 1.5, in addition to the standard factor of safety of 1.5 for structural elements. So 2.25x calculated worst case stack up of design loads. At the least, critical castings must receive 100% visual inspection and dye penetrant inspection, plus X-ray inspection of the structurally significant internal areas, with at least one casting load tested to those factors of safety.
That casting factor can be reduced, by testing more specimens to obtain statistical data to demonstrate the actual process used is more robust.
Your original question was of the achieved Young's modulus, and the effect of casting defects on such. In aviation (and I would guess the automotive industry is similar), you normally assume the Young's modulus of the homogenous material, and then insert the worst case (-3 sigma) defects which your process produces into your FEM model, and assess the effect on crack initiation and propagation, rather than assuming a lower Young's modulus and designing with it.
- - - Updated - - -
Do you have any experience in how 3D metal printed parts are being treated in aviation? What unique defects do have have or are they considered like a forging?
- - - Updated - - -
This is a bit off topic, but why would a forging be more structurally sound then a casting, because isn't the metal that is used in the forging cast at some point. Is it because there is better control of turbidity and inclusions in the casting of the stock used for forging?
Forgive my ignorance, the only way to learn is by asking questions (and failure )
My experience is as an authorities certification inspector, so I only critique/accept/reject and audit the substantiations I see, rather than having real practical experience of having done the work myself. None of the projects I have been involved with have used 3D printed parts yet.
3D printing technology is too new to be covered directly by the FAR's/CS's. Also the rulemaking world is moving away from such proscriptive rules, to more flexible results orientated style, so it is unlikely that specific factors will ever be defined with new technologies.
At present, these will be manufacturer specific processes. So each one will have a difference substantiation which will be held closely as company IP. If you were to buy an off the shelf machine like a Deckel Lasertec, and make a turbine casing, as they demonstrate in their marketing video, you are still going to be a long way from getting an approval to fly that casing. You would need to do all the engineering to substantiate the process and the resulting parts, and also need to do an STC to demonstrate to the authority that the certification requirements are met, including all the required engine testing.
The manufacturer probably provides a substantiation of their material properties, of their process control and of the resulting variability in accordance with FAR/CS 2x. 307 and 2x.619. Basically they will have done a lot of work up front to develop the process, to determine which process variables effect the outcomes, and have destructive tested large number of samples to substantiate that process control work. The outcome is a set of material properties which the component design engineers can use in their design assumptions.
For the specific aircraft part being certified they will have test and inspection requirements tied to the design, to ensure the process is robust and that variations in the process are identified and non-conforming parts rejected. They will probably test and cut up ever x part off the printer, at least initially.
(a) Compliance with the strength and
deformation requirements of CS 23.305 must be
shown for each critical load condition. Structural
analysis may be used only if the structure
conforms to those for which experience has shown
this method to be reliable. In other cases,
substantiating load tests must be made. Dynamic
tests, including structural flight tests, are
acceptable if the design load conditions have been
The factor of safety prescribed in CS 23.303
must be multiplied by the highest pertinent
special factors of safety prescribed in CS 23.621
to 23.625 for each part of the structure whose
strength is –
(2) Likely to deteriorate in service
before normal replacement; or
(3) Subject to appreciable variability
because of uncertainties in manufacturing
processes or inspection methods.</I>
While the airframe requirements do not have specific proscriptive requirements for forgings, the engine requirement of FAR33/CS-E do. Basically on top of all the process control from ore selection, through melting and casting the billet, heat treatment, machining, coatings etc, you also need to design in projections on your forgings to create multiple test coupons which can be cut up and analyzed to ensure each forging is free of forging laps and strain induced porosity etc. Some of those test coupons will also be stored for the life of the resulting parts to enable analysis years later if a failure occurs.
While you are right that everything starts with casting molten metal including forgings, the foundries like Precision Cast Parts Inc doing aviation critical parts have slightly better process control than we have at home For example, due to hard alpha inclusions in the 2 time vacuum melted cast billets used for manufacturing engine compressor disks in the seventies, the industry moved to three times vacuum melted billets, and improved inspection techniques. You can imagine the plant necessary to melt and cast a couple of hundred kg of a precise titanium alloy under a vacuum.
Forging is used to improve the material properties of the base alloy. Forging the casting in the die aligns the grain giving significantly improved strength and fatigue resistance. Basically forgings are used because you can make parts of a given strength and fatigue life lighter.
Last edited by rotarysmp; 01-11-2017 at 09:09 AM.
On the forging note, I'd add that as the compression takes place density is increased depending on the forging.
With Aluminium I'd also suggested it can be done cold as well as hot.
PS thanks Mark for this post
Thanks for including that bit on United Airlines 232, that is a really amazing story. Respectfully, the situation could have easily killed everyone on board if not for some quick thinking and excellent piloting.
I have also done an audit in a German auto engine plant, and the level of process control which is now standard in these industries really is impressive. In my home casting, I only just made a pyrometer to control temp somewhat, and my friend made an argon lance to try and improve the gas porosity. Long way from what is today industry standard. That is why am so impressed by Jeff's home casting technique.
Another project I'm involved in at work has a company restoring an old 1920's light aircraft motor. Made in Germany, with only a very limited number made, the reduction gearbox housing has terrible porosity. Looks like worms got in and ate whose all through it. I am surprised it got used.
Casting at the high end
I was at an aerospace foundry and can vouch for the extreme level of process control and inspection.
I saw them shoot over 100 preliminary x-rays on a rough casting. The techs found a defect in a critical area and immediately carried the casting to a band saw and cut it in half.
I asked how much it cost and it was over $70,000. They said it removed any temptation from upper management to try to save the casting in a Material Review Board.
Trust in God, Pratt and Whitney, and these guys!
My last job was at a brand new high pressure vacuum die cast plant making heat treated auto parts.
We collected over a terra byte of data every day. Each part had a laser bar code serial number traceable to all process and inspection data.
The robots would automatically scrap parts if the process was out of tolerance. The digital x ray could see a defect the size of a pencil tip
It was a treat to operate at that level, it was absolutely spotless, no smoke, and just a bit of noise from air coolers, nicest plant I ever worked in
We also had high speed fiber internet into the plant and also throughout the plant - that ruins you for anything less...so fast, everything was just instant
I am also very impressed with what the skilled artisans are doing here on the forum, some of these castings are world class