SCC of 316L in steam 30bar 300degC

[jaapsin]

I am investigating a severe SCC type failure in flanges of 316L used for only 10 hours on steam 30 bar , 300degC.
All innner surfaces are black (magnetite oxide?).
The 8 # flanges completely broke off (with typical branched scc type cracks)near the weld with a CF3M casting in the HAZ , however many almost equal spaced longitudinal staight cracks are also found on inner surface of flange.From the bottom of these longitudinal fractures scc branched cracks grow. No cracks in the casting of CF3M stainless. All piping is carbon-steel no cracks reported.
Boiler feedwater analysis shows 50 ppm Cl- , seems high?
Any tips where to start looking for the reason?
-boiler feed water chemistry wrong?
-material: inclusions?
-surface imperfections due to forging? because of equal spaced cracks along the flange.
-forging not done properly , stress?
-welding residual stress too high, there is no stress relieve heattreatment?
-why no crackks isn the casting CF3M?
How to solve the problem? I must use stainless flanges to keep standard product.

regards Jaapsin, application engineer

 

[Metalguy]

What kind of "typical branched" scc do you have--intergranular or transgranular? Probably going to end up being trans. SCC from the Cl, but that's just my first guess. Also sounds like the stress level is WAY up there!

 

[jaapsin]

it is transgranular SCC.
jaapsin

 

[unclesyd]

What standard did you pruchase the flanges and were they inspected prior to use? Are you talking cracks in the bore of the flanges?

It sounds like you may have two problems.

I thinks the Cl level is too high to feed a boiler.
But you shouldn't have carry over of the Cl's into the steam.

What was the quality of steam at this point?

If you have magnatite (black iron oxide) you have some heavy corossion, probably with O2, going on.

 

[stanweld]

Did this pipe undergo a high pressure steam blow or other cleaning method prior to operation and if so what were the parameters?

 

[Metalguy]

Unclesyd,
"If you have magnatite (black iron oxide) you have some heavy corossion, probably with O2, going on."

Have to disagree with this. Magnetite formation is the desireable oxide formation in boilers, as it indicates fairly low oxygen-otherwise you get hematite (red/brown, of course).

Mag. becomes very protective above ~250 deg. C.

 

[unclesyd]

Metalguy,
I partially agree with your post, but want get into that aspect for now. I was questioning the jaapsin's formation of the black coating, whatever it maybe, in ten hours. Magnetite, being a mixture of oxides, should form very slowly in a new system. A quick test is scrape off a small amount on a new part, not near the failure, and check it with a magnet.

I mentioned O2, but it might have something else as the high Cl level in system is bothersome.

jaapsin,
Was any cleaning done on the boiler prior to startup?
What was the spec's on the flanges?

 

[stanweld]

One reason that the casting may not have been attacked is that CF3M castings often have considerable ferrite, which provides considerable resistance to SCC. Ferrite is also contained in the deposited ER/E316(L) weld metal and may also be the reason that the weld metal had no cracks since cracking apparantly was only seen in the fully austenitic wrought alloy flange.

The chloride content is also way out of range for normal boiler feed water.

 

[jaapsin]

thank you folks;
meter is pressure tested befire delivery 1.5 times rating of 600 lbs 6 inch flanges.ASME code
the longitudinal cracks are in the bore of the flange
black coating very thin cannot be scraped off
the meter with the broken flanges was mounted 3 days before taking into service , no pressure testing done in the line.
jaapsin

 

[andyenergy]

jaapsin -

I have some information that might be of use. In my industry (UK nuclear) we have seen SCC in many unexpected materials especially associated with low alloy steels in feedwater. I can't really go into too much detail, but we control lubricants and sealants very carefully indeed. SCC in 316 stainless in feedwater whilst not common does crop up from time to time and appears to be associated with lubricants and the like. I'm not really an expert in the corrosion field but our criteria for things to be worried about are (from internal guidance documents):

Copper - limit of 1000 ppm comes from sealant materials and the catalytic release of suplhur from those materials.

PTFE - limit of 500 ppm - dissociates above 280 C giving rise to possible SCC.

Halogens - limit of 100 ppm - mostly everyone knows about halogens and SCC

Graphite - 10,000 ppm - can cause electrolyitc corrosion and possible SCC resulting from organic acids resulting from hydrolysis of oils but little is known about this.

Lubricant we do not like at all are

Poly butyl cuprysl (PBS)
Molybdenum disulphide (MoS2)

and those containing Cu, Pb Cd, Zn, In.

Have any of these components been coated, or has thread lubricant been liberally applied in the area? It seems like a very short time indeed to ceate such cracking - where the components kept dry and covered before use or did they sit in pools of water outside before installation?

keep us informed of your investigations

 

[stanweld]

jaapsin,
It appears that the black coating was only on the flange. If so, it is hard to conceive that the coating occurred during operation unless the flange is downstream from the meter casting. You should definitely analyze the coating to determine if it is only a high temp oxide or a contaminant from the meter as indicated by andyenergy.

I would also check with the manufacturer for sources of contamination and any transportation/storage records for the equipment that could have led to contamination.

Also check the forging for proper manufacture. Reforging billets may or may not have been used in manufacture. The flanges may or may not have been forged. I've seen many manufacturing and quality problems associated with so-called forged flanges. Many of the stainless steel,six-inch forged flanges that I have seen manufactured have almost no hot work performed during the forging operation and the only real hot work seen was in hot rolling of the reforging billet. I have also seen "forged flanges" merely machined from bar stock.