Cold bending of steel vs. Rockwell B hardness

[SMF1964]

I have a drain line in a base-loaded power plant that failed due to corrosion in the bend. That is, the wall thickness is near original in the straight sections but decreases rather gradually as you move into the bend (from either direction). If I take a hardness measurement of the 2¼Cr-1Mo steel pipe in the straight region, the average is 78 RB. If I take a hardness measurement in the bend region, the hardness is around 87 RB. The bend radius is not known exactly since the final failure was rather energetic but it appears that it was around 6" or so. Diameter is 2" nominal, with a 0.4" wall thickness and the corrosion reduced the thickness to less than 0.06". The drain line sees 1000°F steam when the unit is coming on-line. The line is over 20 years old.

My question is this: Can I make any approximations or conclusions regarding the amount of cold work associated with the bend based upon the increase in hardness in the bend?


http://flypicture.com/bin/?id=rt38kqvb

 

[metengr]

SMF1964;
Interesting pic. From the picture it appears that the corrosion was external and locally thinned the drain pipe until it could no longer sustain internal pressure? I didn’t see anything on the failed pipe ID surface.

Regarding your question, I would expect some reduction in hardness over time if the bend was done cold versus hot at 1000 deg F over a period of 20 years. So, to answer your question, if your hardness readings are correct, it would appear that the bending operation was probably done well below 1200 deg F. Typically, for a cold or even warm bend of this type you are looking at greater than 25% strain on the extrados.

 

[SMF1964]

The corrosion was all external - moisture collecting under insulation as this drain line is out doors.

I re-ran the hardness readings, so unless I did it wrong both times... ;-).

So 25% strain on the extrados woudl give me this sort of increase in hardness? Or are you saying typical bends produce 25% strain on the extrados anyway and my hardness data doesn't fit into the picture?

 

[metengr]

SMF1964;
Yes, 25% minimum strain from cold forming would increase the hardness above 87 HRb scale initially for this material - probably closer to the lower 20's HRc scale. My comment regarding the current hardness value indicates some lower temperature tempering of the bend at 1000 deg F service temperature over a 20 year period. So, the only conclusion one could draw is that the forming operation was probably done either cold or warm (relative terms) with no post bend heat treatment. Although, I don’t believe this hardness has any relevance to the failure.

 

[SMF1964]

The question was asked: Why is there only corrosion at the bend? Someone (higher in the corporate food chain than myself) suggested that cold working might decrease the corrosion resistance of the steel, all other things being equal. In light of the food chain relationship, I am investigating for variation in hardness.

 

[EdStainless]

you need to investigate the insulation and external protection. My guess is that that there was insulation and/or weather shield damage in this area.
If we are talking about general corrosion resistance of a Cr Mo steel I can't see this slight hardenss variation having any impact.

= = = = = = = = = = = = = = = = = = = =
Corrosion, every where, all the time.
Manage it or it will manage you.

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[metengr]

SMF1964;
Evaluate the pipe run and location of the line, it could be a low point that can collect water from external sources below a boiler or turbine. Check for any field welds beyond the tangent points of the bend – it could be that portions of this drain line were replaced over the years (I have seen this before). The external wastage is definitely general corrosion attack, and was probably occurring over a period of 20 years with nobody aware of it until a failure occurred. Elbows on vertical pipe runs seem to be a terminal point for collecting water drips or water leaks from above. Trying to tie the elbow hardness to corrosion resistance is a stretch of upper management imagination.

 

[SMF1964]

1. Yes there is damaged insulation in this area.
2. Yes this is a low point in the line that can act as a collection point for external moisture trapped by the insulation.
3. Yes there was significant chlorides in the corrosion produce/insulation mush that was on the surface of the pipe just above the failure (detectable and semi-quantifiable by SEM/EDS is, in my opinion, a significant level).

There is a collection of viable scenarios that produce this corrosion failure, but I needed to address the hardness question to satisfy Those Who Must Be Obeyed; something I'm sometimes neglectful of doing which may explain my current career path.