Crack during cold rolling of thick plate

[MFQC]

During rolling(cold rolling) of a 125 mm thick plate of SA-299 , plate encountered cracking which was started from the surface of the tapered edges. We usually perform beveling and tapering by the means of thermal cutting (before rolling). At the start of the crack , we could find some martensite structure which was accumulated by the surface defect caused by thermal cutting.
Rolling was at the beginning stage.( edge bending)
Hardness of the thermal cut surfaces was at the range of 400-500 HBN and martensite had 3mm depth. For the entire process we performed mechanical beveling which is more time and cost consuming.
By 170 Celsius degree preheat, the surface hardness decreased to about 300HBN. In this hardness again we had martensite formation.
Would you please help me in this material issue. Do we have to strictly insist on mechanical beveling, or any other solution exist for this problem?

 

[DriveMeNuts]

Does the cracking occur in tappered edges that are at the leading edge of the plate going into the roller, or the side edge which is being rolled?
What is the orientation of the cracks?
The dimensions of the bevel?

 

[EdStainless]

You either can mechanical bevel, or hot cut them and then wet grind the bevel to remove the roughness and martensite.
These edges should be free of martensite and smooth for good cold forming.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[0707]

In my point of view is a too heavy thickness. I think you have to insist on mechanical bevelling. Your thermal cutting was done with plasma? Did you have perform thermal cutting for bevelling in those thickness before?

luis

 

[r6155]

Thermal cutting – Flame heating :the heating operation is much more of an art than a science. Should be applied with care and only by an operator who is experienced or is under close supervision.

For 125 mm the bevel should be type double “J” . Do you think to use narrow gap welding?
Consult a welding engineer.

Any photo?

Regards

 

[MFQC]

Dear DrivemeNut
crack started from the tapered side of the plate (which was for circumferential welds) and propagated to the rollers direction till complete break happened.

Dear Edstainless
By the means of machining we can assure about the thickness which will be removed. But with grinding even though we perform hardness test, we can not be sure about the total removal of matensite in all the bevel surface.

Dear 0707
our routine process for cutting thick plates of SA-516 gr70 with the thickness even more than this ,is flame cutting. But this was the first time we flame cut SA299.
thanks

 

[gtaw]

Have you thought of water jet cutting and prepping the edge? There will be no heat affected zone, dross, oxides, or distortion to contend with.

Best regards - Al

 

[r6155]

See ASME II SA-20 9.3 Edge Imperfections and TABLE A1.14 Visible Edge Indications Extending Approximately Parallel to Rolled Surfaces.

9.3.7 Fabricators should be aware that edge cracks may
initiate upon bending a sheared or burned edge during the
fabrication process. This is not considered to be a fault of the
steel, but is rather a function of the induced cold work or heat
affected zone.

Regards

 

[EdStainless]

I have seen people macro-etch (with acid) in order to assure martensite removal.
Water jet would be a better way to go.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

Had a look at SA-299, which is P-No. 1 Group-No. 3 in ASME IX.

1) Compared with SA-516 it has significantly higher hardenability, as indicated by CE.
2) Plate thickness > 1" permits even more Mn and Si, raising hardenability and exacerbating the problem.

My rule of thumb when thermal cutting and gouging is that you should preheat the same as you do for welding. With this material at this thickness, I would recommend more. (Remember that ASME non-mandatory appendices for preheating are fairly lax.)
So when going from SA-516 to SA-299 you need to raise your game - I guesstimate not less than 200°C (400°F) preheat.

With respect to subsequent welding operations, the problem isn't the joint design but the fact that you have a lot of hard material to remove first.


"Everyone is entitled to their own opinions, but they are not entitled to their own facts."