Rust/Corrosion and Material Strength 3

[joshsgamb93]

Are there any good sources/articles/studies done on the effect that rust/corrosion has on a material's strength? Specifically, I am interested the effect that rust has on a metal's strength, and if there are any equations that exist to determine the strength after a certain amount of rust/corrosion has occurred.

 

[jdonville]

Not aware of any. Safest to assume no effective strength for corroded material. There are relations to estimate the depth of corrosion for steel(and galvanized steel) after a given number of years of exposure. The ones that I am familiar with are for steel buried in soil. I believe that FHWA is currently working on a study to provide specific guidance for buried steel foundation elements and buried steel piping.

 

[phamENG]

jdonville is correct - assume zero strength in the corroded region. Nearly all of the corrosion that I've seen (apart from typical surface rust that's less than 1/32" thick) has a delaminating effect associated with it. Even if the Iron Oxide had some strength (think about what happens when you pinch it), there's little to no bond between it and the base steel.

Other metals, like weathering steel, aluminum, etc. develop very thin patinas that prevent further corrosion. If you need to take the strength of that thing layer into account, your design is waaaay too tight.

Best bet is to use a small hammer and wire brush to knock off loose corrosion, clean down to sound metal, and use an ultrasonic tester or (if accessible) a pair of calipers.

 

[joshsgamb93]

Thank you both for the insight.

So if you have a metal staircase with corrosion covering roughly 70% of it, is it safe to assume zero strength for the staircase and to not allow access through it? I seem to get pushback on that assumption, but I don't see it as an unreasonable assumption.

I work in a very old facility.

 

[JAE]

I'd second what phamENG stated.

Typically you clean the rust off and then measure what is left. It is then re-analyzed with the reduced section properties to see where you are at with strength vs. demand.

For a staircase, if the rust is in metal pans (i.e. gage metal type thicknesses) then rusting many times goes through the thickness in discreet areas and not in others. That makes the analysis even more difficult.

 

[jdonville]

If you don't want to do a formal measurement of the remaining sound material, you could load the elements with progressively modest load increments and measure the response. Decide in advance what your pass/fail criteria is and what remedial measures should be.

 

[joshsgamb93]

Yeah jdonville that is something I was already planning out for certain areas where the rust is a little concerning. Thank you all for your insight! If you come across any studies/equations I can read over just let me know.

 

[jdonville]

Sounds like you are dealing with atmospheric corrosion of exposed elements.

I don't think that there could be any meaningful studies or equations that are relevant as there are too many variables (temperature, humidity, chemical vapors, etc.). This would be a "rule of thumb" territory if at all. And since you have access to test the structure, nothing will beat empirical observation (and maybe some back-analysis) of the actual structural response to known loading.

 

[joshsgamb93]

I agree with your statement about empirical observation. Maybe I will initiate a study with some colleagues.

 

[Ron]

To determine the "strength" of a a structural component, you have to determine the loss of cross section due to corrosion. As difficult as is sometimes encountered, you need to compute the cross section of the element subjected to corrosion to determine its current capacity.

 

[dhengr]

Joshsgamb93:
Our language and words actually have real meaning, and if we want to pretend to be engineers, and to be understood and believed, we should use them carefully. Corrosion/Rust don’t normally change the real material strength, but as mentioned above they do affect the cross section strength because they reduce its material area over time. Material properties (strengths, etc.) and section properties (which lead to the section strengths, etc.) are not the same thing.

 

[GGedge]

Note what dhengr said.

The coverage of a member with corrosion product is an unreliable guide to section loss and therefore capacity loss. The corrosion product of steel occupies a much great volume than the original thickness of steel that has corroded. As a rough rule of thumb the ratio is about 3:1. To estimate the remaining capacity you have to remove the corrosion product and measure the thickness of metal remaining, that can be done with an ultrasonic thickness meter or with calipers if there is access to both surfaces. You need quite a lot of measurements to get a decent sample population and also a visual inspection to check for localised damage such as pitting or excess local loss where standing water occurs. Also note the worst case corrosion may be on less accessible parts, such as where water may collect and not readily dry.

Cheers

 

[r13]

As a seasoned structural inspector, I would suggest to perform a vigorous thickness reading on the planks and all structural elements, including the fasteners, then present the results/maps to a structural engineer. From the presentation, a competent engineer can back check the remaining load capability of the stairs at present state, and provide remedies to strengthen it, if necessary/applicable.

 

[joshsgamb93]

Thank you all for the valuable input! I do realize material properties and section properties are not the same and I guess I should've chosen my words more technically. I appreciate the input from all of you

 

[drbrainsol]

While a few posters above explained this using more elaborate terms, here it goes - just because something looks rusty it is not necessarily unsafe. Detailed examinations are required to assess whether the strength of a structure / element is affected.