焊接工程

Alright team, we're qualifying a new WPS for the reactor shell welds — SA-516 Gr.70 carbon steel, 85 mm thick. The joints are double-V groove, and we're using SAW for the fill passes. This PQR has to pass ASME Section IX and the client's supplementary requirements.

Test coupon is ready — two plates of SA-516 Gr.70, 85 mm thick, 300 mm × 600 mm, double-V groove with 60° included angle, 4 mm root face, and 2 mm root gap. Preheated to 120°C as per the preliminary WPS.

I've verified the preheat with a contact thermometer — 120°C uniform across the entire joint area, 150 mm beyond the groove edge per code. Hassan, what consumables?

Root pass is SMAW with E7018, 4 mm electrode, at 160 amps. Fill passes are SAW — EM12K wire, 4 mm diameter, with neutral bonded flux F7A6-EM12K. Welding current for SAW is 650 amps at 32 volts, travel speed 450 mm/min. Heat input calculates to 2.8 kJ/mm.

2.8 kJ/mm is good — below 3.0. Above that, we'd risk grain coarsening in the HAZ. Hassan, check interpass temperature after every pass — must stay below 250°C. Let it cool if needed. Park Min-Jun, why is interpass control critical for thick-section carbon steel?

Two reasons. First, high interpass temperature reduces the cooling rate, which leads to larger grain size in the HAZ — that reduces toughness. Second, if interpass gets too high, the weld metal can lose its mechanical properties — the yield strength can drop below the minimum. For SA-516 Gr.70, we need to stay within 120-250°C.

First root pass is in — I'm measuring interpass at 148°C. Go ahead with the first SAW pass, Hassan.

Running SAW pass 1... pass complete. Slag removal looks clean — no surface porosity.

After the last fill pass, we'll do a dehydrogenation treatment — hold at 250°C for 4 hours before the PWHT. This prevents hydrogen-induced cold cracking in thick sections. Then PWHT at 620°C ± 15°C for 2 hours per inch of thickness, so 85 mm is about 3.3 inches — that's 6.6 hours, minimum.

(after full welding and PWHT)

Alright, NDE is done. RT shows no cracks, no lack of fusion, no slag inclusions exceeding the acceptance criteria of ASME Section VIII. UT confirms full penetration throughout. I'm releasing the test coupon for mechanical testing.

(three days later, reviewing test report) Mechanical test results are in. Two tensile specimens — both broke in the base metal, not the weld. That's what we want — weld is stronger than the base metal. Four side bends — no cracks, no openings exceeding 3 mm. Charpy impact at -29°C: average 78 Joules in the weld metal, 92 Joules in the HAZ. Both well above the 27 J minimum. This PQR passes.

Excellent. Maria, you'll take these PQR results and write the WPS — reference PQR number, list the essential variables: base metal, welding process, filler metal, preheat, PWHT, electrical characteristics. Any change to an essential variable requires a new PQR. Non-essential variables can be revised on the WPS without requalification.

On it. This WPS will cover all reactor shell longitudinal and circumferential seams. We'll qualify the welders to this WPS using the WPQ test — a test coupon in the 3G position, bend and visual test only. Per ASME, if they pass 3G, they're qualified for all positions except 6G pipe.

Oscar, I've got bad news. Weld joint SP-127 on the 12-inch high-pressure steam line — RT shows a defect. I'll pull up the radiograph.

Let's see it. (studying the film on the viewer) There — a linear indication running along the center of the weld, about 40 mm long. That looks like incomplete penetration. The root pass didn't fuse properly. Lin Wei, what do you see?

Agreed — it's lack of penetration. And look here, there's also a small cluster of porosity — three gas pores, each about 1.5 mm. Combined with the LOP, this weld is rejectable per ASME B31.3 — the client specification says no linear indications and porosity maximum is 2 mm individual.

Diego, this is your weld. Any idea what went wrong?

(looking embarrassed) I think the root gap was tight — maybe 1.5 mm instead of the specified 2.5 mm. The fitter rushed the fit-up. And I was welding the root pass with GTAW at 90 amps — WPS says 100-130 amps. I should have stopped and reported the poor fit-up, but I thought I could compensate with the higher heat. Clearly I couldn't.

Diego, I appreciate your honesty. But you know the rule: never compensate for a poor fit-up with welding technique. If the joint prep isn't right, you stop, you call me, we fix the joint. A rejected weld costs us a day of schedule and NDE rework. A weld that fails in service... that could kill someone.

Alright, let's focus on the repair. Per ASME B31.3, we can repair this weld once. We need to gouge out the defective area — carbon arc gouging, then grind to clean metal. The repair cavity has to be checked with PT to confirm all defects are removed before re-welding.

Diego, you'll do the repair yourself — it's your weld, you fix it. But this time, I'm going to watch the root pass. And Lin Wei will verify the joint prep before you strike an arc. Sergey, can you mark the defect location on the pipe so we know exactly where to gouge?

Already marked with a paint pen — 40 mm section between these two lines. Gouge through the full weld thickness in this area. I'll do a guide shot with RT after gouging to make sure all the defective metal is gone.

(after gouging, grinding, and PT check)

PT shows no indications. The repair cavity is clean. Joint prep is now correct — root gap is 2.5 mm, verified with a gap gauge. Diego, this time, GTAW root at 115 amps, SMAW for the fill and cap. And I want to see the root pass before you continue. Understood?

Understood. Thank you for giving me the chance to fix it.

(after repair welding and re-RT)

Re-RT is clear. No linear indications, no porosity. Weld SP-127 repair is accepted. Oscar, here's the RT report and the repair record for your files.

Good. Diego, this goes in your training record — not as a punishment, but as a learning point. Every welder makes mistakes. The best welders only make each mistake once. Let's make sure the rest of the crew learns from this too. I'll cover the fit-up quality expectations in tomorrow's toolbox talk.