Not logged inAmerican Welding Society Forum
Forum AWS Website Help Search Login
Up Topic Welding Industry / Technical Discussions / crack in welding of ss 316
- - By sharif Date 04-12-2003 07:39
we weld a 3/4" nipple (ss316) to casing of pump (mat.ss410) with electrode E-309 but after 3 days that pump worked we observe that a crack produced in HAZ area of nipple .
please explaine for reason of it.
If we weld ss316 to ss410 material with electrode E-309L or Inconel-182 then we must do Stress Releive or not?
Parent - By chall (***) Date 04-12-2003 14:04
309 would be my selection for filler metal.

I think a repair is in order.

First you should make sure there is no other influence on the assembly that may be contributing to the situation (external load, excess heat, excessive vibration, etc.). If you think there might be an external cause, that has to be addressed before attempting a repair or you will be scratching your head again in a few days.

Make sure the area is thoroughly cleaned to sound metal, verified by PT. Redo the weld and verify by PT. We would keep the interpass below 350F and not perform PWHT.

Make sure the weld design is suitable for the application. A fillet weld reinforcement is probably required.

Charles Hall
Parent - - By GRoberts (***) Date 04-14-2003 14:45
Type 410 stainless steel is martensitic and should recieve a PWHT after welding. If the pump body is cast material, there is a good chance it is either ASTM A217 CA15 or ASTM A352 CA6NM. It is very highly reccommended to PWHT both of these after any welding no matter what filller metal is used because of the brittle HAZ. If this same attachment is going to take place frequently, it is common to order the case with a 309L overlay in the area to be welded to that gets a PWHT at the factory, then the 309L can be welded to in the field without PWHT because the HAZ in the pump case is already tempered.
Parent - By chall (***) Date 04-14-2003 15:08
Thanks for the correction, I had 409 on the brain when I sounded off.

Charles Hall
Parent - - By Niekie3 (***) Date 04-15-2003 19:22
Hi

I will agree with GRoberts regarding the welding procedure. I do however believe that this does not really give a reason for the failure you experienced, because it was in the 316 material's HAZ. - Not the 410 HAZ.

The typical problem that one finds when welding austenitic S/Steels such as 316, is hot cracking in the weld metal itself. HAZ cracking in austenitics is very rare. As such, I wonder if your failure is not from another source, rather than the actual welding operation. A typical failure mechanism may be fatigue, especially if you have much vibration and / or something that is relatively heavy attached to the nipple.

Perhaps you can tell us some more about the operation of the pump and the nipple. The answer may lie there.

Regards
Niekie Jooste
Parent - - By kk_chiong (*) Date 04-20-2003 14:25
Niekie,

Could you explain further above the hot cracking of SUS316 weld bead. I did experience this problems once when I weld 308 material onto mild steel (fillet join) and immediately after the welding the whole weld bead was crack all the way for the whole weld bead. The length of the weld is about 10" and the mild steel plate is having 1" thickness on both side. Both size of the plate is approx. 1" x 10" x 12". The throat head is about 4mm. No preheating is carry out before welding. I was surprise that the crack line exist on the weld bead as the 308 ss is having some elongation. Please advice the cause lead to this problem. I still in doubt above this incident.
Parent - - By Niekie3 (***) Date 04-21-2003 08:46
Hi KK

When a weld is busy solidifying, it tends to solidify along "fingers" called dendrites. These dendrites have a "solidification front". Behind this front you have solid metal, while you have liquid "mushy" metal infront of this solidification front. Only those constituents that are "compatibile" with the solidified weld metal will be absorbed into the dendrite. Those constituents not compatible with the solidified weld metal tends to get pushed ahead of the solidification front. Typically such constituents are low melting point constituents such as S and P. Finally all these low melting point constituents are left along the grain boundaries, as these are the last parts of the weld to solidify. As the now solidified, but very hot weld starts to cool down, shrinkage stresses are introduced. The grain boundaries with the layer of low melting point constituent (usually situated in the centre of the weld) can then no longer take the stress and seperate from each other. (Hot cracking is also often called centre line cracking.)

In the case of fully austenitic weld metals, the solidified weld metal is incapable of absorbing much of these impurities. For this reason, filler metals are usually designed to have at least 4% of ferrite in their structure because ferrite can absorb a great deal more of these constituents during solidification.

If you look at the fillers designed to weld "problem steels", (e.g. 312 filler) then you will see that they are so designed to have up to 60% ferrite in them, without forming martensitic zones.

Getting back to your problem welding onto the mild steel: While it could have been hot cracking, it sounds to me more likely to have been due to martensite formation in your weld bead. I say this because when welding 316 or 304 to C/Steel you would normally use a 309 filler and not a 308 filler. Using a 308 filler will almost certainly result in a structure that contains excessive amounts of martensite. Especially when welding thick plates without pre-heat, this weld will be very brittle.

Hope this answers your question

Regards
Niekie Jooste
Parent - - By kk_chiong (*) Date 04-21-2003 13:51
Niekie,

Your explanation is clear but I still have some question and I will try to get them from metallurgical book. Do you know any good metallurgical book which I can read to dip up the relevant information. Your kind advice is very much appreciated.

Regards,

KK Chiong
Parent - - By Niekie3 (***) Date 04-21-2003 19:06
Hi KK

The book that I have that is quite informative regarding this type of problem is "AWS Welding Metallurgy Volume 2, by Linnert".

Hope this helps

Regards
Niekie Jooste
Parent - By kk_chiong (*) Date 04-24-2003 15:22
Niekie,

Thanks for your info and willingness to elaborate further on my question. Appreciated for your help.

Regards,
KK Chiong
Parent - - By GRoberts (***) Date 04-21-2003 22:18
KK,

A book I am currently reading has a lot of information on weld solidification and cracking is "Welding Metallurgy" 2nd edition by Sindo Kou. It describes weld metal solidification as being either dendritic, as Niekie points out, or cellular, planar, or equiaxed dendritic, the mode of solidification has to do with the cooling rate, travel speed, which in turn affects the temperature gradiant at the solid/liquid interface. I wish I had read it when I was qualifying a subarc weld with NiCrFe-3 (Inco 82). The grains were too big, so I had to switch to a 1/16" diameter wire to get adequate ductility, but it can also affect hot cracking, as Niekie also pointed out. Kou also agrees with the fine grain = reduced hot crack thinking.


Niekie,

Your answers in this thread are spot on. I agree with your assesments 100%. You have a good way of making the complex understandable.
Parent - By kk_chiong (*) Date 04-24-2003 15:21
Robert,

Thanks for your further explaination and I will try to get hold on the books as recommended. Thank you very much.

Regards KK Chiong
Parent - - By kk_chiong (*) Date 04-21-2003 14:00
Niekie,

When I go thru again your explanation, it seem that the molten weld pool solidification is similar to the solidification of cast iron in casting process. The different may be the different alloying metal or constituents is having different solidification temperature. Therefore, different constituent tend to solidify first while others constituent solidify later according to the cooling temperature. If the cooling rate is too fast, do you think that before the higher melting constituent solidify completely, another constituent already start solidify and what is the effect of the type of solidification to the finishing solidify metal. Look towards your further advice and guidance.

Thanks & Regards,
KK Chiong
Parent - By Niekie3 (***) Date 04-21-2003 18:55
Hi KK

Yes, welds are small castings! Welding problems and casting problems are very similar to each other.

Regarding your question concerning cooling rates, I believe that the answer can be too complex to really answer it here. Typically, the faster the cooling rate through the solidification range, the finer the dendrite (and therefore grain) structure will be. In addition, if the cooling rate is high enough, not only will the dendrites start to form (nucleate) from the base metal, but will also start to nucleate within the melt itself. This will only happen if the moulten metal is "supercooled", because it takes a great deal of "driving force" for this to happen. Under these circumstances, the segregation mechanism will probably have less time to take place, but will none the less take place. It may reduce the grain boundary layer, but will probably increase the shrinkage stresses. As such, I am not sure that you will have any great benefit either way. This is only speculation on my part, as I have not seen studies looking at the effect of solidification rates on hot cracking. It may be an interesting exercise to try.

Regards
Niekie Jooste
Parent - - By Niekie3 (***) Date 04-21-2003 19:10
Just had a look at my AWS welding metallurgy book, and they recon that with a decrease in grain size, hot cracking decreases because the amount of surface area of the grains increase. This allows the contamination rich material to be spread over a larger area, reducing the concentration of the contaminant. Based on this I would think that increasing the cooling rate would reduce the hot cracking tendancy of an alloy such as 304 or 316.

Regards
Niekie Jooste
Parent - By kk_chiong (*) Date 04-22-2003 06:34
Niekie,

Thanks For your effort and info, there really help me to understand a bit more about the cooling of molten weld pool. Thanks.
Regards,
KK Chiong
Parent - By brande (***) Date 04-20-2003 04:58
The 309 would normally be a good choice, but seeing it is not holding up well, you might want to try 312. If the 312 does not do the trick, some procedure modification might be in order.

Good Luck

brande
Up Topic Welding Industry / Technical Discussions / crack in welding of ss 316

Powered by mwForum 2.29.2 © 1999-2013 Markus Wichitill