Does welding on stainless rod cause substantial weakening 1

[saltoh101]

Hi

I have built a sculpture which is supported by a 1.5" diameter, 316 stainless rod. There is a square stainless tube welded to the rod and the welds are completely around the circumference of the rod. The rod (4 feet in length) supports a 7 foot long, 110 lb. dolphin sculpture. The 1.5" rod is sufficient to support the sculpture, but I have been hearing that the welds, around the circumference of the rod, have weakened the rods tensile strength. I am trying to gauge whether the welds have compromised the rod's ability to support the sculpture in adverse conditions. I am trying to ascertain whether the welds will cause the rod to snap or just bend should it be overloaded? Any thought would be appreciated.
Thanks

 

[MikeHalloran]

You need to engage an engineer locally to evaluate the installation.



Mike Halloran
Pembroke Pines, FL, USA

 

[EdStainless]

Probably not, but a detailed analysis is in order.

The reason that I say prob not is that the rod is likely annealed material to begin with. Since this is the lowest strength condition there is will not be a reduction in strength.
My concern is if the bracing system has created too much localized loading or bending moment.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[Metalguy]

Use a magnet to check if the rod is magnetic AWAY from the weld area. If not, you have annealed, low strength 316. If it IS magnetic, it is probably cold worked and stronger, so the welding could have weakened it.

"You see, wire telegraph is like a very long cat. You pull his tail in New York and his head is meowing in Los Angeles. Do you understand this? Radio operates the same way: You send signals here, they receive them there. The only difference is there is no cat." A. Einstein

 

[saltoh101]

Metalguy - I have always wondered about the magnetism of stainless. I usually use this to determine whether an unknown metal is stainless or not. What I have found is that most stainless has a low level of magnetism. So are you saying that if the magnet drops off it is annealed and if it barely hangs on it is cold worked?
Thanks

 

[saltoh101]

Attached is a photo of the scale model of the full size sculpture, to give you an idea what I am talking about.

 

[Metalguy]

Wow, beautiful sculpture!

There is a big difference in mag. between annealed WROUGHT 304/316 and cast/weld metal. When 316 is poured, it solidifies in two phases, the non-mag. austenitic phase and the very magnetic ferrite phase. The ferrite phase is usually ~10-20%.

HOT working usually converts the ferrite to non-mag. aust. Cold working will convert some of the aust. to magnetic martensite. Heavily cold-worked 316 can be strongly magnetic, so a good magnet will stick very well.

316 SS weld metal usually has ~5-15% ferrite in it to prevent weld cracking, so it will be slightly magnetic.

Do you know if the design required cold-worked 316 or not?

"You see, wire telegraph is like a very long cat. You pull his tail in New York and his head is meowing in Los Angeles. Do you understand this? Radio operates the same way: You send signals here, they receive them there. The only difference is there is no cat." A. Einstein

 

[EdStainless]

Most commercial austenitic stainless grades are not fully stable. When you deform them some transformation martensite forms. This phase is magnetic.
Often the rotary straightening of SS bar in enough to leave the skin slightly magnetic.
When you weld another transformation happens, some ferrite remains in the welds. We usually want this because it helps minimize weld hot cracking. The ferrite is magnetic also.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[IRstuff]

You didn't say what your "adverse" conditions are.

Nice sculpture, BTW.

TTFN

FAQ731-376

 

[saltoh101]

Thanks for the compliments... I have found out that the stainless rod
is cold drawn, not sure what that means. The adverse conditions could be winds of 40-50mph, however unusual here in San Diego .
I guess my biggest concern would be of the rod snapping,
if it just bends no one would get hurt, as it is up high.

Any and all comments/thoughts are welcome
thanks

 

[Metalguy]

Highly unlikely it would snap. Even cold drawn, it should bend like a pretzel. Do you know what the yield or ultimate tensile strength is?

"You see, wire telegraph is like a very long cat. You pull his tail in New York and his head is meowing in Los Angeles. Do you understand this? Radio operates the same way: You send signals here, they receive them there. The only difference is there is no cat." A. Einstein

 

[saltoh101]

So even with the welds around the circumference of the rod
it should still bend?

Per the web, 316 stainless has these properties:
- tensile strength is 85,000 pounds per square inch
- Tensile Yield Strength (.2% offset):
276 MPa / 40 kpsi
- Ultimate Tensile Strength:
621 MPa / 90 kpsi

 

[Metalguy]

Yes, it should bend right over before it breaks if there are no circumferential cracks present. The elongation of cold drawn 316 should be ~45%, which is a very large amount.

There is one thing that could be a problem. When 316/304/etc stainless is welded, it can be "sensitized" in the HAZ (heat affected zone), a narrow area right next to the weld metal. The amount depends mostly on the amount of carbon in the rod. Anything more than ~.035% C can easily have sensitization. The upper limit for C in 316 is .08%, and it readily sensitizes.

What happens is that the sensit. area loses its corrosion resistance, and something called SCC (stress corrosion cracking) can occur, especially if chloride ions (cl-) are present, such as seacoast area outside. If such cracking gets deep enough, the rod could suddenly snap off.

If there is a small extra piece of the rod available, a met. lab. can perform a test to measure the amount of C. Unfortunately there is no really good method to do this outside a lab. What should be done, if the rod is outside, is to have a simple dye-penetrant test performed on the
HAZ area if it's accessible.

There are other mitigation methods, but they all require access to the weld area.

"You see, wire telegraph is like a very long cat. You pull his tail in New York and his head is meowing in Los Angeles. Do you understand this? Radio operates the same way: You send signals here, they receive them there. The only difference is there is no cat." A. Einstein

 

[EdStainless]

It is generally bad form to have a weld all of the way around a support. Staggering or offsetting portions of the weld would be better.
This material has so much ductility that even if there was some surface cracking it wouldn't concern me too much. There will be a lot of bending before things break.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[stanweld]

The weakest link is the weld between the bar and square tube supporting the dolphin. Is this weld only a fillet weld or is it a full penetration weld with reinforcing fillet? The heat affected zone in the cold drawn bar will have lower strength than the cold drawn bar but so will the deposited weld metal and the heat affected zone in the tube if it were cold finished as well. Is the tube cold finished? What is its thickness? The weakest point appears to be at the toe of the weld at the tube side of the joint. If the weld is properly sized and the thickness of the tube is sufficient, failure will not occur.

 

[saltoh101]

It is a fillet weld and since the rod is 1.5" in diameter the penetration is definitely not full.
The rod is cold drawn but no cold finishing.
The 1.5" rod goes thru the full length of the square tube,
which has a .25" wall.
There is a .125" thick plate (with a 1.5" hole in it and the shape of the square tube)that abuts the rod/tube joint. So, there is a fillet weld between the rod and the plate and another weld between the plate and the tube; not sure if I desrcibed this correctly.
There is also another plate and welds where the rod exits the other end of the rod.
There are also a few plug welds half way along the 9" square tube.

Thanks

 

[stanweld]

I believe that your design would withstand a Force 5 hurricane together with a 7.5 earthquake.

 

[MikeHalloran]

Yes, but it might fatigue and fracture catastrophically under the influence of a gentle sea breeze.

As already mentioned, it needs a detailed analysis, which should include estimation of natural frequencies and at least an educated guess at the aeroelastic behavior of the multiple cantilevered objects.

This forum is not the appropriate vehicle for that.



Mike Halloran
Pembroke Pines, FL, USA

 

[saltoh101]

Mike - So how does one get a study done, ie.
- what type of engineer would be appropriate
- how does one find a qualified engineer in the San Diego area
- what would you estimate the cost of such a study to be

thanks

 

[MikeHalloran]

You need a permit to put it up, right?
Ask the folks there to suggest someone.
Or, now that we know where it is, maybe someone will speak up.



Mike Halloran
Pembroke Pines, FL, USA