Creep on Carbon Steel 1

[Hansac]

I have a furnace, tubes with material A106 Gr. B. It has been about ten years. I got the circumferential reading, and compared it to nominal circumferential reading. Tube is NPS 125 mm (actual OD = 141.3 mm). Calculated that cirmferencial length is 443.91 mm. I used this to compare against the measured and got values ranging 1.60%-2.95%. I do not have actual circumferential reading for the tubes before commissioning, that is why I am using calculated nominal.

My questions:
1. How much creep can carbon steel take before retirement?
2. Is 2%-4% a magic number? Meaning does it apply to all variants of steel (low allow, carbon, stainless)?
3. Do people retire material if creep exceeds 4%?

Thank you, in advance, in gratitude.

Hanafi Ali,
Mechanical Engineering by training,

 

[metengr]

My questions:
1. How much creep can carbon steel take before retirement?

Your tube OD (circumferential) measurements should be used as a guide in this case because you did not have actual OD data to begin with, as you pointed out.

From my experience with boiler tubing in service, when you approach about a minimum of a 4% increase in OD, as measured from nominal dimensions, its time to begin thinking about tube replacement OR rely on other methods to evaluate in-service creep damage like ID oxide scale thickness testing and tube sampling.

2. Is 2%-4% a magic number? Meaning does it apply to all variants of steel (low allow, carbon, stainless)?

No.

3. Do people retire material if creep exceeds 4%?

No.

Recommendation;
In lieu of relying solely on tube deformation measurements, I would recommend you look into tube ID oxide scale thickness testing, along with tube sampling or surface replication (in service nondestructive testing to evaluate the microstructure of the tube metal).

The ID oxide thickness test method is offered by most large boiler OEM's and some industrial metallurgical testing labs that cater to Power Generation. The method involves measuring the actual oxide thickness on the tube ID surface along with wall thickness. The testing can be performed on multiple tubes within the furnace to obtain a representative sample of the tube circuit. The data (oxide thickness and wall thickness) are used in a computer program that predicts remaining useful service life (the program uses the Larson Miller parameter to estimate creep rupture life). This is an effective tool that can trend the creep life of the tube circuit.

Tube sampling is an effective method to evaluate the actual tube metal microstructure. Tube sampling normally is used to corroborate the ID oxide thickness testing. Tube sampling can also be used for creep testing, as well. Surface replication can also be used to evaluate the condition of the tube metal from exposure to elevated temperature service.

So, you have several options to gather data and make an informed decision to run or replace.