What's new
What's new

Metallurgy for vibration resistance

BCreekDave

Plastic
Joined
Mar 30, 2018
Location
California
Looking for recommendations for a type of steel tubing that is particularly resistant to fracturing due to vibrations. I have an ultrasonic welding application (20khz) where the anvil in this case is a piece of thin wall steel tubing. The wall of the tubing is cracking at about 1000 to 5000 cycles. We are currently using 316 SS but have tried with the same results 304SS. The tubes are about 24 inches long and the weld region on the tube is ~8 inches long. Not averse to starting with a solid rod and gundrilling, if necessary to get the right metal alloy.

Thanks,

Dave
 
No matter what material you use, you are going to have a problem. Did some work a while back with an engineer that specialized in ultrasonics. Hardened steel, titanium, aluminum all had the same problem.

Tom
 
A more complete description of the tube and whatever is welded to it would help. If the material must be stainless, the reason would help. If the tubing is not seamless DOM, it should be.

If the tube is to allow clearance for a projecting part of the US welded assembly, perhaps a secondary plate can go on top of the tube structure, with an elastomer between the top plate and tube. The US welding parameters might have to be adjusted to account for the changes, but isolating the tube could fix the issue.
 
What is the size of the tubing, what wall thickness? Have you tried annealing it? We annealed cable fittings that were swaged on and eliminated cracking during the swaging process. The fittings were made of 303 Se, but that material may not be a good match unless you drill it from solid.
Maybe with the material you are using annealing every 500 cycles?
 
Some directions to look for..
Sandvik 2205 duplex stainless
Aermet 100
Carpenter custom 625
Rene 41

If weight or modulus of elasticity is part of the problem titanium or beryllium(Yuck!) alloys could be something to look for
 
More info would help us help you.A sketch of your set-up, the direction, frequency and intensity of the US vibration.

Clearly your application stresses the tube beyond its fatigue limit. This is significantly lower than the yield point for most materials. It is also highly sensitive to surface finish and internal cleanliness of material, so choosing a higher strength material often does not buy a similarly large increase in fatigue strength.

I would first look at your overall design and geometry, to see if stress could be reduced, then consider small details of support, load application, surface finish and material cleanliness, lastly look at stronger material.....austenitic stainless is not anywhere close to top of list.
 
Monel, inconel, or Mp35n might be other alternatives as well. I remember zirconium 702 and 704 being a softer version of titanium.
 
More info would help us help you.A sketch of your set-up, the direction, frequency and intensity of the US vibration.

Clearly your application stresses the tube beyond its fatigue limit. This is significantly lower than the yield point for most materials. It is also highly sensitive to surface finish and internal cleanliness of material, so choosing a higher strength material often does not buy a similarly large increase in fatigue strength.

I would first look at your overall design and geometry, to see if stress could be reduced, then consider small details of support, load application, surface finish and material cleanliness, lastly look at stronger material.....austenitic stainless is not anywhere close to top of list.
The tube is actually a forming tube in a vertical form-fill-seal machine. A polyester felt strip is wrapped around it and our product flows through the inside of the tube. Some of these tubes have lasted millions of cycles without failure, but the wall thickness in those is much thicker (+0.10 inches) When you get <0.07 inches the failure rate goes up drastically. This is with the incumbent 304 or 316SS. Thinking perhaps a 4130 or 4340 chrome-moly tubing may be better? Perhaps a A513 or a 17-7SS? In this particular application my wall thickness will be ~0.03.
 
This is with the incumbent 304 or 316SS. Thinking perhaps a 4130 or 4340 chrome-moly tubing may be better? Perhaps a A513 or a 17-7SS? In this particular application my wall thickness will be ~0.03.
4130 or 4340 should be better if it does not need to be stainless steel. You really need to specify VAR grade (vacuum arc remelt) like 300M (4340 VAR grade)to have less inclusions

Sandvik 2205 would probably have 2-3x higher stress limit vs 316SS. Wtih Other exotics that I mentioned the fatique limit can be even questimated 5x of the 316 stainless but you are probably going to run to availability and cost problems.
 
4130 or 4340 should be better if it does not need to be stainless steel. You really need to specify VAR grade (vacuum arc remelt) like 300M (4340 VAR grade)to have less inclusions

Sandvik 2205 would probably have 2-3x higher stress limit vs 316SS. Wtih Other exotics that I mentioned the fatique limit can be even questimated 5x of the 316 stainless but you are probably going to run to availability and cost problems.

Those are the kind of tips I am looking for. thank you.
 
Would it make sense to make it easy to remove and anneal it from time to time before the cracks start. I am just guessing here that this would actually stop the fatigue cracks if done often enough.
Bil lD.
 
Maybe. Depends on the actual cost of the tube. If they are cheap enough we will make them quick-change and just pitch them. Right now, it is tube welded to a flat plate. Going to design a clamp system to eliminate the welded joint. then the setup folks can change it out at the machine. Still need to increase the life span by a few orders of magnitude.
 
Welds are stress relieved I take it?
Have you thought about shot peening? Builds residual compressive stresses that can extend fatigue life immensely. There have also been studies done that show re-shot peening before the onset of cracks will essentially reset the fatigue life. If having it preformed outside works out you might look into flap peening.
 
Thin stainless tube is always hard, anneal it should help, but can you just switch to mild steel, or a chrome moly tube? 316 is a bitch when it comes to work hardening, it happens fast and is a seriously common problem. Boat tanks made out of stainless frequently fatique crack do to the liquid sloshing, alu or steel holds up far far better in thoes applications, like your magnitudes better.

Other option may be ditch metal, use a composite, they can resist fatigue loads far far better.
 
Welds are stress relieved I take it?
Have you thought about shot peening? Builds residual compressive stresses that can extend fatigue life immensely. There have also been studies done that show re-shot peening before the onset of cracks will essentially reset the fatigue life. If having it preformed outside works out you might look into flap peening.

The welds are about 8 inches away from the area where the ultrasonic energy is focused. It always cracks at the area where the horn contacts the felt material, never where the plate is welded to the tube.
Shot peening sounds good but I would like to get the metallurgy as optimized as possible and then see if shot peening adds significantly to this.
 
Other option may be ditch metal, use a composite, they can resist fatigue loads far far better.

Beat me to it.

I was going to say some Polycarbonate or Carbonate Fibrous material. But really, even UHMW is basically indestructible in a shock zone. (albeit, Machining a Tube of UHMW with .03 wall would also be near impossible.)

R
 
Composite maybe, but a plastic tube would probably not stand the heat for too long. During the US weld operation, the felt is melting (being welded) and locally exceeds 400F for short duration. Not to mention, but maybe I should have, that the US weld occurs with about 700 to 1000 lbs clamp pressure. A thin wall plastic tube would be crushed.
 
Composite maybe, but a plastic tube would probably not stand the heat for too long. During the US weld operation, the felt is melting (being welded) and locally exceeds 400F for short duration. Not to mention, but maybe I should have, that the US weld occurs with about 700 to 1000 lbs clamp pressure. A thin wall plastic tube would be crushed.

Maybe, but there are materials like ®Torlon that are designed for that, I think rated for working at +200ºC and a 25,000 lb. Tensile strength, and no melting point. Just an idea, sometimes I get stuck thinking my way and no other way is possible, because it doesn't like the same music as me or doesn't appeal to my appetite. Makes no sense right?

R
 








 
Back
Top