Machining Aluminium- Centrifugal force and yield strength 1

[sydneyjongleur]

Hi Guys

I am machining(turning) an aluminium alloy ring, just under 1.0m diameter x 150mm high. The ring starts with a wall section of around 20mm and it gets taken down to about 3-4mm.

The issue that I am having is that after the operation, whilst still on the machine, the ring grows by 0.75mm on diameter over the full length of the ring and also grows on length by around 0.3mm.This only ever seems to happen with aluminium.

My initial thoughts is that this is caused by centrifugal force and that the yield strength of the material must have been reached for the ring to expand by this amount. Can anyone confirm or tell me otherwise if this is correct.

I have worked out the centrifugal force based on a 20mm section and then with 3-4mm section to get an answer in Newtons(800 and 140 respectively), however, I am struggling to decide how to relate this to the yield strength of the alloy. If I can get to the bottom of it I may be able to slow the machine down hence reducing centrifugal force.

The alloy is precipitation hardened and the yield strength is around 400 Nmm^2.

I may be completely on the wrong path here so any guidance would be appreciated.

 

[MikeHalloran]

Measure the temperature.


Mike Halloran
Pembroke Pines, FL, USA

 

[hydtools]

More likely temperature affect.

Aluminum thermal coefficient of linear expansion is 22.2x10^-6 m/m C. At about 1.0 m dia, .22mm/10degC

Ted

 

[sydneyjongleur]

The machining operation is flooded with coolant so the ring itself is never that hot, although the temperature at the cutting edge can be hot. When the ring is removed from the machine and is sitting back at room temp the part is 0.75mm bigger than what it should be.

 

[GregLocock]

Is the ring a casting?

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[chicopee]

Are you machining the ring from the inside? If so yo may be stretching it with the cutting tool.

 

[BrianE22]

Measure the length and diameter after the part has cooled but is still clamped in the chuck (or fixturing). Unclamp it and measure it again.

 

[MikeHalloran]

Given the precipitation hardening mechanism, I suppose some size change could be expected just from that.

We could look it up if an actual alloy spec had been revealed.


Mike Halloran
Pembroke Pines, FL, USA

 

[dhengr]

Aluminum is soft enough and with a low enough yield strength and modulus of elasticity, that it is a particularly good material for spin forming. Google spin forming, aluminum. And, talk with a real smart machinist who works with aluminum, and in these dimensions, and with the machining process you are using. You are probably doing some spin forming on this part with your machining process and tooling, and should be able to make some adjustments in the process to get your desired final sizes.

 

[MikeHalloran]

Yeah, spin forming.

As in, do the parts meet spec when you use new/sharp tools?

Are you using tools optimized for aluminum?



Mike Halloran
Pembroke Pines, FL, USA

 

[Walterke]

Just out of curiosity, where would a part like this be used?

NX 7.5
Teamcenter 8

 

[sydneyjongleur]

The ring is a forging used in the aerospace industry- alloy 2618a. The ring gets machined from both the inside and outside. With regards to unclamping and checking again, the size is pretty much the same on and off the machine.you must be thinking how is that possible that you cut a diameter to size and then it's 0.75mm bigger in the machine before un clamping. Well say for instance you machine a diameter at the top of the ring near the start of the process by the time the operation is finished and material has been removed from other areas it has changed size. I am experienced in machining various steels, nickels, titaniums etc and agree there is distortion and size changes due to the thin wall nature of aerospace components and I would say that is possible that size could change if removing material from elsewhere on the ring. However the location diameter grows 0.75mm as well. As previously discussed the length of the ring is 150mm approx with a 5mm long external location diameter. This obviously does not get machined and as an experiment we opened out the location diameter on the fixture by 0.75mm and when we tried to remove the part after maching we couldn't get it off ie. location diameter grew 0.75mm
I don't have much Experience with aluminium , and certainly not at this size and wall section so I am trying to understand what's going on . I have a finishing operation aftermthis which takes the wall section down further and i am expecting some more material movement although hopefullymnot as much. I will look into the spinning of aluminium

Is it just stress relieving itself from stresses created from the ring rolling forging process as a lot of material has been removed.

The tools we use are sharp positive rake carbide inserts recommended for aluminium.

Hope this all makes sense

 

[Cockroach]

You say that the calculation has given an answer of 800N and 140 N, centrifugal force. This would be a radial force so divide by the interior surface area, pi X ID X width. That would be the radial pressure experienced by the piece.

I would use Von Mises Hencky equation Sigma=sort(3) P K^2/(K^2-1)for K=OD/ID. This is the wall stress experienced by the piece. Then your factor of safety is that stress divided by yield strength. If the result is less than one, then you are stressing the ring by turning. A number greater than one, unlikely this is the problem.

It could be thermal, even coolant wouldn't completely have the piece at ambient temperatures. Do the calculation based on a ring of equal circumference and see what temperature is required to grow the ID by 0.75 mm. This is a small amount, I suspect a few degrees.

Good luck with it.

Regards,
Cockroach

 

[Tmoose]

The castings for the Duramax engine blocks required multiple rough cuts to create nice true bores. A whole new machining sequence had to be developed. Initially of course the spindle builder (us!) was accused of designing a wimpy spindle.

I can believe Machining an un-stress relieved forging could be responsible for The change in diameters.
One easy test is to repeat the machining of an inside diameter or bore painted with dykem immediately after the first cut. The dykem will be untouched if the part distorted during that cut.

the change in length of a solid ring is interesting and harder for me to understand.

Is any kind of stress relief called for in the forging processing?

http://www.keytometals.com/page.aspx?ID=CheckArticle&site=ktn&NM=89

If there are no stress relief options then rough machining steps can help a lot. A sequence will have to be developed.

 

[Walterke]

So the item you're trying to create is a tube-like object diameter 1000mm with a 4mm thickness and a length of 150mm?
Is the inside diameter also too big, or is that one ok?
What is the order in which every step of the turning takes place?

NX 7.5
Teamcenter 8

 

[sydneyjongleur]

Cockroach- the issue I have with it being thermal is that it remains this size out of machine in a temp controlled environment. Could you expand on the Von Mises Hencky equation this is not something that I am familiar with- the id of the part is 800 and the width is 3mm. The yield strength of material is around 400Nmm^2

Tmoose- no there is no stress relieve allowed on this material unfortunately. I will look at changing the method of manufacture and will probably be able to overcome most of the problems, however, if I could understand the reasons why there is so much movement then I can go about overcoming it. This is the only material that I have experienced such a change in diameter. The strangest one for me is what I said in an earlier post, the location diameter which does not get cut, grows by 0.75mm also- This is why is was thinking centrifugal forces etc( or maybe it is just stress relieving itself)



Walterke-It is actually around 800mm diameter and finishes up about 3mm wall thickness and 150mm long. It is contoured and has pockets etc so the wall section varies but generally no more that 4mm. both internal and external get bigger. The diameters internal and external are machined before the pockets so I can understand how they may move etc. As I keep saying though,the bit that gets me is the location diameter grows 0.75mm. To try and explain, this is an external diameter and is held to 0.002" on the previous operation, is 5mm long and has a groove directly above it locate clamps on. So the part is located in the fixture, clamped and it still grows 0.75mm??. We basically had to jack the part off with great difficulty. As an experiment we opened out the loc dia on our fixture for the next part and after the operation took the part off of the fixture and sized the loc dia and it had grown by 0.75mm???? Either it is thermal, stress relieving itself or the centrifugal force due to surface surface speed.

 

[BrianE22]

Can you post a picture of the setup? I've had parts get loose in 3 jaw chucks when the wall gets thin but I haven't seen the part get clamped "tighter" as material gets machined away.

 

[dhengr]

There is so much unsaid here, about the way the rough part is made, and then the entire process of machining the finished part, that you should really get all your suppliers and people in your own shop involved in this, since we don’t know their/your secrets. The size (dia.) and wall dimensions 3mm thick x 150mm wide (with some pockets?) certainly make this a difficult part to deal with. It seems to me a big disadvantage if you can’t stress relieve this before machining. That certainly could be part of the problem. For example, does the forging process induce a compressive residual stress at the o.d. and a tensile stress at the i.d., in equilibrium in the rough? Then you machine the i.d. away, and are left with predominantly compressive residual stress near the o.d. This would tend to make/cause the part grow in dia. or circumference, as it relaxed or self-relieves and re- equilibrates. Maybe knowing more about the residual stress distribution in the rough part and taking equal amounts of material from the o.d. and i.d. would keep things in equilibrium.

 

[TVP]

You definitely need to understand the residual stresses in this type of part. Forging + quenching during heat treating + massive material removal during machining + no stress relief = distortion/dimensional variation.

Residual Stress Analysis of Aircraft Aluminum Forgings

http://www.asminternational.org/portal/site/

http://www/AsmStore/ProductDetails/?vgnextoid=50cd29e36c228110VgnVCM100000701e010aRCRD

Residual Stress in Aluminium Alloy Forgings

http://sites.google.com/site/temfemguy/research/residual-stress-in-aluminium-forgings

Applications: Cold-Compressed Aluminum Forgings

http://www.lanl.gov/contour/forgings.html

Effect of Forging and Heat Treating on Residual Stress in Aluminum Forgings

http://www.ms.ornl.gov/mplus/projects/effectofforging.html

 

[Cockroach]

I'm pretty sure you are seeing some form of accelerated creep. Clearly the measured deformation is the result of processes that change the dimensions of the specimen. There is also the added uncertainty with material, the precipitation hardness of aluminum is meant to increase the spheriod size of impurities internal to the matrix as a means to mitigate dislocations. Clearly your machining process has added to the overall stress.

So it could be a complex combination of things, how you are chucking the raw material, depths of cuts turning 20 mm thick wall to 2 - 3 mm, imparted stresses from cutting thereof, etc. My suspicion is that several other commentors have given you some form of verbal solution, no need for repetition here.

Kind of hard to comment on or provide a numerical solution to the phenonema without more input datum. But I think that would fail to adequately address your situation. You are removing a lot of material and ending up with a very thin ring of soft material age hardened for strength. Machining stress are most likely the source of your deformation. You need to consider Poisson's Ratio in two planes, diameter and length in order to figure out the way material flows to each term in the strain relationship.

Kind of a neat problem, I suggest holding back on your cut depth and possibly stress relieving the ring intermittently during your manufacturing process. Perhaps jigging this across the span of length would help stabilize the OD somewhat.

Regards,
Cockroach