Effect of Creep damage on Room Temp Strength?

[SMF1964]

Assume, for the moment, that you have a piece of low alloy steel that has experienced elevated temperature service to the extent that it has creep damage (random microvoids at the worst - metallography showed no aligned microvoids or microcracks visible). Now take that piece of steel and pull it at room temperature until it fails. Can I expect the creep voids to significantly contribute to a reduction in yield and tensile strength?


Thanks.

 

[swall]

First of all, what was the original heat treat condition, if any, of the alloy steel? What was the creep temperature exposure? Or is this strictly a hypothetical question?

 

[redpicker]

Assuming the exposure to the creep conditions did not affect the hardness of the alloy steel, I don't think you will noitce a change in yield strength at all, and any effect on the tensile strength should be minimal.

The presence of microviods and other creep damage may affect the %Elongation, fracture toughness, and/or fatigue resistance, but you would probably not see any change in the room temperature tensile strength or yield strength values obtained from a simple tensile test.

 

[metengr]

Yes, I believe it should. Presuming that the material is not thermally embrittled, the presence of microvoids will result in lower mechanical properties. During a tensile test, pre-existing microvoids will induce local yielding and tensile rupture. In addition, the pre-test material most likely had degraded properties from exposure to elevated temperature service like spheroidization damage.

Why is this so? Because I had seen this firsthand with Cr-Mo superheater boiler tubing after we suspected exposure to creep damage in service. We ran tensile and creep rupture testing.

 

[SMF1964]

Excellent - thanks for the input. And now, for some more information:

swall: The original heat treatment is normalized. Sort of. These tubes are pack chromized, with a 20 mill thin layer of chromium diffusion coating applied to the outer surface of the tube. The tubes are packed into a retort with chromium or ferrochrome powder and baked for 24 hours or so at 1900-2000°F and allowed to cool. This produces a decarburized layer that is roughly 50% thru wall. Microstructure of the decarb layer is extremely coarse grains that get finer as you move towards the ID surface (I'm thinking it's the influence of increasing carbon, which interferes with grain growth during this treatment). Microstructure of the rest of the tube is ferrite + pearlite, with 'sloppy' pearlite lamellae often seen in low alloy boiler tubes (1¼Cr-½Mo).

These tubes were exposed to temperatures in excess of 1000°F for only 3 years before we had two creep failures and five other tubes with visible bulges and microcracking. The tube I tested did not contain any microcracking or aligned microvoids in the metallographic sections I took.

There are a number of issues being considered here: The relevant one (to this thread) is the mechanical properties at room temperature and at 1000°F. At room temperature, the mechanical properties are essentially unchanged between a new tube and the service exposed tube (both tubes were from the same steel heat and chromizing pack). At 1000°F, however, a tensile test (not creep test) showed a drop in the yield strength of 30% and a drop in the tensile strength by 20% between the new tube and the service exposed tube. The microstructural differences are not readily apparent in terms of spheroidization (metengr) although the service exposed tube does show a more disorganized pearlite lamellae (the colonies remain clearly defined).

redpicker: No significant difference in % elongation was observed either at room temperature or at 1000°F.

I am trying to reconcile the difference in high temperature tensile strength and the lack of difference in room temperature tensile strength and was thinking that the microvoids (if present) would have an impact at high temperature but not at low temperature.

 

[metengr]

SMF1964;
Your additional information makes perfect sense to me because what you have is damage from stress rupture over a 3 year period versus pure creep rupture. The information that I had provided in the post above was from superheater tubes that were exposed to "long term" overheat (300,000 operating hours) not 3 years at higher operating temperature.

The mechanical test results on creep damaged, long term overheated boiler tubing resulted in degraded mechanical properties. In your case, the isolated random creep cavitation in a pearlititc structure would not have an effect, it would be like having small, rounded inclusions in the steel. Depending on the size and density of the cavities, I could see where no effect would be seen at RT.

 

[SMF1964]

metengr: That explains why there was no effect on the room temperature strength, but why the significant difference at 1000°F? The service exposed steel has a yield/tensile strength of 25/37 ksi and 24/34 ksi (two tests) while the non-exposed steel has a yield/tensile strength of 21/51 ksi and 28/51 ksi (again, two tests). Room temperature tensiles were consistent at 61 ksi.

Could the difference be just the deterioration of the carbides that would be too fine to show up in light optical metallography?

 

[metengr]

Yes.

 

[metengr]

SMF1964;
What is your original tube material? I had thought more about your question, and your reported values of 51Ksi at 1000 deg F for new boiler tube material seems rather high. I had reviewed the data in EPRI "Boiler Tube Failure Metallurgical Guide", and for new ferritic boiler tubes, the UTS and YS at 1000 deg F are as follows;

Low Carbon Steel @1000 deg F
23 Ksi UTS
14 Ksi YS

Med Carbon Steel @1000 deg F
27 Ksi UTS
18 Ksi YS

T2 (0.5% Cr-0.5% Mo) @ 1000 deg F
38 Ksi UTS
19 Ksi YS

T11 @ 1000 deg F
45 Ksi UTS
19 Ksi YS

Assuming you have a high strength alloy steel boiler tube, I selected the data for T91 as comparison;

T91 @1000 deg F
55 Ksi UTS
45 Ksi YS

 

[SMF1964]

The tube material is SA-213 grade T11 (1¼Cr-½Mo). I double checked the numbers and, yes, it is 51 ksi tensile strength.

 

[unclesyd]

A little off topic but we are getting close to having an evaluation tool for such problems.

http://www.deform.com/ht_brochure.pdf