low ferrite in weld of Duplex steel 1

[smart143]

My welder has welded a test coupon for PQR in Duplex (UNS S31803) of 30 mm thickness with double welding .The electrode used was E2209-15 of dia 3.2 mm, the ferrite reported on MTC of electrode is 36%. I did not measured ferrite content during & post welding in weld, the coupon was subsequently solution annealed at 1140 deg C for 75 mins.
The test result reported by Lab shows ferrite in weld only 25% whereas HAZ shows 51% & PM shows 59%. Now we need compliance as per NACE MR0175 where in ferrite in weld required is between 35-65%.
What could be the reason for low ferrite in weld, the heat input during welding was 0.8-1.0 kJ/mm.
What I will need to do to meet the 35-65% ferrite level. I do remember that ferrite in weld of austenitic tends to reduce post solution annealing, does same phenomenon occurs in Duplex steel?

 

[pradipgoswami]

Solution anneal may reduce "ferrite" in Duplex S.S.Could you post the following details:-

- Exact Nominal chemistry of the filler metal, as recorded in the MTR
-Calculate the "Cr" and "Ni"quivalent as per DeLong Diagram.

How the solution anneal was carried out following holding @ 1140 Deg C, water quenching, forced air cooling or still air cooling??

Appreciate your input.

Thanks.

Pradip Goswami,P.Eng.IWE
Welding & Metallurgical Specialist
Ontario, Canada.
ca.linkedin.com/pub/pradip-goswami/5/985/299

 

[birken]

Ferrite measurement would have been taken prior to heat treatment.
The limits apply to as-welded deposit.

Using constitution diagrams to predict ferrite number for as-welded deposit is the way to go. This would give some idea of ferrite number before applying heat treatment.

Birken

 

[NickClark]

The overmatched chemistry of 2209 is designed to produce 35-65 without any pwht. Its rare to pwht duplex welds in my experience. Maybe the increase in austenite forming elements in the weld metal is promoting transformation more than expected. Its also possible you have Nitrogen pick up if you used it in the shielding gas. Ferrite variation from cap to root will help explain the effects of gas or dilution.

Nick

 

[pradipgoswami]

The following attachment is the phase diagram for Duplex stainless steels. The approximate composition of 2209 WM is indicated.Below is the extract on metallurgy of DSS:-
-DSS solidify primarily as ferrite at approximately 1425°C (2597°F) and [highlight #E9B96E]partially transform to austenite[/highlight] at lower temperatures by a solid state reaction.
-If the cooling rate is rapid, very little ferrite will transform to austenite resulting in an excessive ferrite phase at room temperature.
-Consequently, the cooling rate of duplex welds must be slow enough to allow the transformation of approximately 50% of the ferrite to austenite.

SMAW consumables had approx 36% ferrite, as welded ferrite readings are not available.Take the worst case scenario of 35% ferrite in the weld,when this alloy is soaked @ 1140 DegC for 75 minutes,some ferrite will start transformation to austenite.If this alloy is rapidly cooled or quenched, austenite will remain untransformed , causing higher austenite percentage in the weld.

The sugegsted ways to solve this problem:-
- if solution anneal is called for, perform the same @ relatively lower temperature 1050 DegC , shorter time, 30 mins.
- forced air cool(sufficiently fast upto 800 Deg C, to avoid sigma) and see the results
-otherwise weld a fresh coupon with welding consumables having better ferrite:austenite ratio (45:55 or 50:50)and solution anneal as suggested above.

The comments from the originator would be appreciated.

Thanks




Pradip Goswami,P.Eng.IWE
Welding & Metallurgical Specialist
Ontario, Canada.
ca.linkedin.com/pub/pradip-goswami/5/985/299

 

[smart143]

Dear All,

Thanks for your input, below is the chemistry reported in MTR.
C:0.04,Si:0.65,Mn:1.0,P:0.02,S:0.002Cr:23.2,Ni:9.1,Mo:3.1,Cu:0.11,N:0.17

FN:36%

Apology for wrong soaking temp in above write up, the actual soaking temp was 1040 deg C & cupon was water quenched.

Appreciate further input from all of you.

 

[weldtek]

smart 143,
Some things additional to the need for ferrite verification prior to any PWHT;

1. Be aware that values for measured vs calculated ferrite may vary greatly.
2. There's a substantial difference between Ferrite number and % ferrite at the higher values. They're considered to
be the same up to about 10, but over that the gap widens considerably.
3. The welding technique can greatly influence as deposited ferrite. SMAW is especially prone to nitrogen pickup and
that's a ferrite killer.
4. In my experience, SMAW electrodes often run on the low side - around 25% - 30% ferrite. Your calculated value
is on the low side of the spec, but on the high side of what I've found to be the norm. The experience of others
may vary on this.

 

[smart143]

Thanks weldtek for this information, after careful reading I noticed that Lab has reported avg FN as 25.49 whereas NACE MR0175 calls volume fraction between 35-65%.
Can you put some more light how this FN 25 can be converted to volume fraction?

 

[birken]

smart143,
The exact ferrite number of the filler with WRC 1992 diagram comes out to be 34.09 FN.

Reduction in FN is unavoidable during any heat treatment. To attend this matter and meet the specifications, an appropriate filler may be considered.

You need to predict FN for the weld joint (considering exact composition of base metal and consumable) and then leave some margin of reduction during the heat treatment. That would help in meeting the specifications.

I had the chemical composition of duplex S31803 (recently purchased) and I used that with chemical composition of filler you provided. Predicted FN of the weld with WRC 1992 came out to be 66.41 FN (see attached file)

This shows a large reduction in FN because of solution annealing putting question on this treatment. As you mentioned, the coupon was water quenched.

Birken

 

[weldtek]

smart 143
I've attached some info regarding conversion of % ferrite to FN.

 

[NickClark]

The responses are getting a little confusing. Duplex steels solidify as 100℅ferrite the transform partially to austenite on cooling based on two things. 1- cooling rate and 2- proportion of austenite forming elements. If your ferrite is too low you either need to reduce Ni or N etc or increase the cooling rate. As you are water quenching it suggest nitrogen pickup or severe over matching. Were you using nitrogen backing or shielding?

I still don't know if you need to pwht at all..

Nick

 

[OGMetEngr]

NickClark has it right. Increase cooling rate (to prevent as much austenite formation) or use filler material with less austenite forming elements. Duplex shouldn't be PWHT at all, unless it is fully solution annealed.

 

[smart143]

I reviewed the 2nd para A7.3 for welding of Duplex stainless steel in MR0175/3.It states

" The ferrite content in weld metal root & unreheated weld cap shall be determined in accordance with ASTM E562 and shall be in the range of 30% to 70% volume fraction."

Can any body put insight on interpretation of "unreahated weld cap".Does this mean that ferrite is to be checked before PWHT of weld & the section does not talk any thing about ferrite of PWTH DSS material.

 

[XL83NL]

Im not sure when you want to PWHT DSS, however, "unreahated weld cap" may refer to ferrtie determination in the weld cap before reheating it.
Reheating, e.g. by means of "TIG dressing" has influence on the ferrite content, and may give the weld a ferrite content outside the desired range.
This is why TIG dressing of DSS welds is often discouraged. Refer to e.g. API TR 938-C, annex C.

 

[metengr]

Unreheated weld cap means the last layer of weld metal deposited to complete the cap. Post weld heat treatment is not performed for this material unless it is subjected to a full solution anneal after welding.

 

[smart143]

Thanks metengr,

Solution annealing is required on weld as I am fabricating A928 DSS pipe.

Today I conducted 2nd trial on weld cupon & post solution annealing result is shocking.

ferrite on PM is also gone down to 25-29%, weld ferrite is also 25%.

This time I kept interpass temp as 100 deg C max & heat input between 1.0-1.2 KJ/mm.
base metal & weld consumalbes are same as first cupon. before PWHT (solution annealing) ferrite in weld was avg :32%

cycle used as follows

heating rate: 200 deg C/Hr

soaking temp : 1050 deg C

soaking time: 30 mins

Qenchning: water immerson

 

[bhollands]

I have had the same issue welding cast CD3MN material. While the HAZ and Base metal were both at 50% ferrite the 2209 weld metal would not provide a ferrite above the 30% minimum. We tried a stepped solution anneal, dropping from 2150F down to 1800F prior to quenching without success. I went to a E2959-15 that is usually used for superduplex in order to meet the NACE requirement.

 

[kumkumvijay]

I share my opinion as below:
1. Due to high possibility of ferrite formation the filler metals for duplex ss has high amount of Ni which is austunite stabilizer. If you get low ferrite in weldmed & alos in PM means, i suspect the heat treatment furnace may not reach the proper temp or time specified.

If there is no issue with furnace on reaching temp or soaking then the chemistry of the weldmed need to be verified for sufficient ferrite stabilizer.
Nitrogen is a strong austunitie stabilizer and nitrogen pick will directly affect the ferrite content so shielding gas need to be taken care.

The hand book can be useful to refer the parameters.

http://www.kskct.cz/images/materialy/en/avesta.pdf

 

[kumkumvijay]

Also high nitrogen content in parent metal increases the precipitation of austenite during casting & welding and prevent high ferrite content in the rapidly cooled welds. So please check the nitrogen gas content of PM also in a gas analyser.

 

[EdStainless]

If you are planing on annealing then you should use matched filler, not overalloyed filler.
The 2209 filler is designed for as welded use. You need to actually use 2205 filler.
What shield gas are you running? Ar with some He and N? You may be running more N than you need.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube