I have a metallurgy questing that I 3

[OZWelder]

I have a metallurgy questing that I was hopping you may be able to clear up for me.

I have a few Duplex 2205 S31803 welds that have been subject to ferrite testing. The testing was conducted using a Fischer Feritscope MP-30. The client specification stipulates that a ferrite percentage (%) of between 35-65 is acceptable, I have a number of welds that have returned a value of between 65.6 to 67.5. The client specification also notes that the ferrite percentage is determined my means of metallurgy examination, this is however impossible to do on the production welds as we obviously can not cut the production welds up to do metallurgy examination. Would you be able to give me an indication as to whether the values given from the Feritscope and the values from metallurgy examinations are the same and if I should be concerned about 0.5-1.7% excess ferrite. From a QA point of view I look at is as though I have a client specification that give 35-65, I have a report stating welded with vale higher then 65 so these are a DNC as even id 0.1 high, out of spec is out of spec. However I understand that sometimes there is a lot of wiggle room in these specifications and I would be really happy to get away with the high values we have on a few butts. The high values have been found in the weld metal, HAZ and in the fitting some welds in all 3 locations some only in weld metal and some only in the HAZ.

I thank you very much for your time.

 

[Guest102023]

Measurement by metallographic is expensive and labor intensive, and not necessarily representative of the surface FN. Fischer Feritscope is the gold standard for ferrite measurement and the only instrument I would consider.

Out of spec is indeed out of spec, but there are a couple of arguments you could make:

1) At that high FN, your excess is probably within the measurement error of the Fischer.
2) Are the high measurements only found on the outer surface? The inside surface (which I presume is the process side) is the only side that matters.
3) Sometimes the consumable will not produce the specified FN; I have experienced this myself with covered electrode from a top tier manufacturer. We worked around that by replacing SMAW with pulsed GMAW, with very good results. A good duplex SS tech spec will have FN requirements that vary according to the process/filler metal. Corrosion testing should be the final arbiter.
4) To make the outliers disappear you will want to average a set of say, 10 readings to represent one measurement. However I think you will find that readings are very consistent within a weld bead.

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[EdStainless]

There are two reasons for these limits.
The first is that when the A:F skews too far either way the chemistry of the phases shifts and one phase will end up having noticeably lower corrosion resistance than the other one.
The second issue is that the ferrite has lower ductility, a distinct DBT, and higher hardness than the austenite. So you end up with different mechanical properties.
I did a large 2205 job recently and required the ferrite be limited to 35-60%, and every test came in 35-50%.
We used our lab samples (metallography) form the PQR as calibrations for the Fischer. And surface finish matters also.
So were you using 2209 filler?
Very low interpass temp? (<150F)
Nitrogen in the weld and backing gas? (2-3%)

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[EdStainless]

I'll add that in our PQRs we tested; A923 low temp impacts, A923 corrosion, metallography for phase balance, and Fischer values (corrected from FN to actual ferrite %).
This increased the confidence that actually had good weld metallurgy.
And it gave us options going forward to use various tests as the cases allowed.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[weldstan]

Ed and brimstoner have given you great info. Use it in your further evaluation.

 

[OZWelder]

Thank you so much for the replies, greatly appreciated.

Edstainless, the PQR I run for this job had A923 corrosion testing, ferrite examination, macro, impact testing to -60C, all passed no problem.

The wire I am using is 2209LSi and a 2% nitrogen gas mix for the back purge and the welding torch. GTAW is the process.

WPS I wrote had max interpass temperature of 120 deg C. Root pass max heat input of 1.15Kj/mm for root pass, 0.95Kj/mm cold pass, fill and capping.

The Fichte scope samples for the weld were taken in four location on each weld 1@ 0deg 2@ 90deg 3@180 deg and 4@ 270 deg. What I was allowed to do in the end was add the four value together and get the average of the four.once I did this all values were within the 35-65% range.

The problem I have is the welders as soon as I turn my back I highly expect them to not follow the inter pass temperature and they all struggle on the 2” pipe to weld with in the heat inputs.

Thank you all so much for your time.

 

[EdStainless]

Some of my passes were 1.2kJ/mm, slightly higher input is better than too low.
Sounds like everything is covered.
On one job we required welders to take a photo after each pass, with their phone showing date and time and their bare hand on the weld. It worked. Of course anyone found cheating is kicked off of the job also has an impact.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Guest102023]

Ha ha, I will steal your technique for temperature control EdS. Because imposing special controls on welders is like herding cats.

The test methods in A923 evaluate the same problem in different ways. Precipitation of 'detrimental intermetallic' phases also embrittles the steel, so if you fail one test you are likely to fail the other.

I'm surprised to see you doing impact testing S31803 @-60°C; usual is -40°C. That's a big difference even in the absence of embrittlement. You quote S31803 but these days it will come dual certified with S32205, with more restricted composition.

"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[OZWelder]

Hello brimstoner,

I can assure you we did toughness testing to -60 deg Celsius. At first I though the client had made a mistake between deg F and deg C. But after a bit of back and forth we had to conduct impact testing to -60C, they also would not accept lateral expansions, the acceptance was 50J average 40J individually. Some of the material was dual certified to S32205 some only to S31803.

We achieved around 60J in the HAZ I think around 65-80J weld metal using manual GTAW, we also ran 1 PQR using TIP TIG semi automatic GTAW, with this TIP TIG process and a constant heat input of 0.93Kj/mm and max inter pass temp of 120C we achieved impact results of over 100J in weld metal and HAZ at -60C. Gas for purging and torch was 2% nitrogen in argon.

The hard part is trying to keep the welders in check during production regarding interpass temperature. 10% of all production welds are monitored using digital data logger but what happens once I walk away is another question.

 

[EdStainless]

We wrote the RM spec to require S32205 with A923, it was welded pipe.
You have A:F on the RM? What was it?
It should be 40-50% ferrite.
The final annealing temp will control this.
I have seen welds with F as low as 30%.
They will have great impact and strength, but the corrosion resistance of the A is somewhat suspect.
Our impacts were actually done at -50, half size samples were all >100J.
A923 requires 54J weld and 34J HAZ as minimums. A bit too generous.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[kingnero]

Can't offer much mmore than what already has been mentioned. But the 120°C interpass on your WPS is on the high side! What was the PQR temp?

On one job we required welders to take a photo after each pass, with their phone showing date and time and their bare hand on the weld.

This is a genius solution.

 

[Guest102023]

BTW, 60°/140°F is the bare hand tolerance temperature (BHTT). Your domestic hot water temperature should not be set higher than this, but not just for your hands. Copper plumbing is significantly more susceptible to erosion corrosion above this same temperature. (I've been trying to tell my landlord to back off the setting, which is far too high. I don't want to have to tell him 'I warned you' or miss three days of hot showers.)


"If you don't have time to do the job right the first time, when are you going to find time to repair it?"

 

[kingnero]

Doesn't it need to be higher in order to prevent legionaire's disease (or whatever it's called in english? I believe treshold is somewhere around 70°C)