Views: 0
You get a call about a batch of tube that cracked on the customer's bending jig, and the first question is always the same: was the weld seam bad, or did the mill drift out of setup? We look at the seam first, because that is where most disputes start. A steel pipe mill can hold tight dimensions for hours and still ship a weak bond if heat input, squeeze pressure or scarfing slipped.
We wrote this to show where we place inspection on the line so a defect gets caught while it can still be fixed. We cover what we check before the welder, how we read the seam, where eddy current, ultrasonic and hydrostatic testing belong, and what goes into the acceptance test. If a figure is not on our published pages, we mark it to be confirmed in writing rather than guess.
Blame usually lands on the HF welder, but on a steel pipe mill our quality team starts further back. Strip that is not flat, dirty edges, or a strip end join that wanders will fight the forming rolls before current reaches the weld point. We watch uncoiling, flattening and strip end joining first, because a stable strip is a precondition for a stable seam.
Our published process runs coil loading, uncoiling, coil peeling, strip flattening, strip end joining and strip accumulating before forming and H.F. welding. Each step feeds the next, so an early fault shows up later as a seam defect that looks like a welding problem. We treat the impeder and ferrite core as setup items; our technical articles note a ferrite core length allowance of 15 to 40 mm. The impeder and ferrite core range is where we point buyers who want that adjustment.
Three things make a longitudinal weld on an ERW line: heat at the edges, pressure from the squeeze rolls, and removal of the bead. Our HF welder works on the skin effect and proximity effect of high-frequency current, so the edges heat to welding temperature while the bulk of the strip stays cooler. We do not publish welder power in kW or frequency in kHz; those are to be confirmed in writing against your strip and speed.
After the squeeze rolls close the bond, inner and external scarfing cut away the flash. A dull scarfing insert or blade leaves a ragged bead that hides a lack of fusion, or gouges the wall into a thin spot. Then come seam ironing, intermediate annealing, air and water cooling, before any electrical test. Testing hot gives false readings, so we place our checks after cooling rather than straight after the weld box.
In the published process, Eddy Current Testing, or ET, sits after water cooling and before sizing. That is deliberate. ET is fast and contactless, good at picking up longitudinal seam discontinuities and surface-breaking flaws while the tube still moves at line speed. Put it after sizing and you inspect a tube that has already been worked, and the signal changes.
We treat ET as a screening step, not a final verdict. It flags a suspect length so an operator can react during the run. Our online NDT, ET and UT equipment is scoped around that job: catch the drift, log it, keep the suspect pipe traceable. A common mistake is to assume one tester covers everything. ET has limits on certain flaw orientations and wall thickness, so we pair it with slower checks downstream.
UT and hydrostatic testing answer different questions, and buyers often ask us to drop one to save cost. We push back. UT, placed after rough straightening in our process, uses sound to look for internal flaws and lack of fusion through the wall. A pipe hydrostatic testing machine, near the end after marking and cutting, pressurises the finished tube to prove it holds without leaking.
One checks the material, the other checks the sealed tube under load. Passing a hydro test does not prove the bond is sound inside the wall; a clean UT scan does not prove the pipe holds pressure at the ends. We scope both where a standard or end use calls for them. We do not publish test pressures, hold times or acceptance classes as marketing numbers; those are to be confirmed in writing and verified in the acceptance test. Describing EN, ASTM, JIS, DIN, GB or API as standards a buyer may specify is not a certification claim, and we do not make one.
A test result that cannot be tied to a coil, a shift and a length of pipe is just a number. We build the record so you can follow a failed length back to the strip it came from and the setup in force at the time. That means logging ET events, UT scans, hydro results and scarfing checks against one identifier, not on separate sheets.
On a steel pipe mill the useful record is boring and consistent: time, coil reference, size, operator, settings that drifted, and the disposition of any suspect length. Disputes are settled with documents, not memories. Our published scope includes a thorough quality inspection process and performance testing before shipment, so the log travels with the line. If you sell into a standard-driven market, ask how the record is exported and who owns it, and ask us to walk a sample through before you sign.
When a seam fails, our first move is not to turn up the power. We look at the strip edges, the impeder position and the squeeze roll setup, because a cold weld, an overheated weld and a misaligned seam can all look like bad bonding to a tired operator. The table below is the order we actually check, not a catalogue of every fault.
A gray, oxide-lined seam points to too little heat or poor edge preparation; a brittle, burnt seam points the other way. Open seam or edge offset sends us back to the forming rolls and the entry guide. A bead that tears tells us to inspect scarfing. We fix the cause upstream before we touch a setpoint, then adjust one thing, run a length, re-test and log it. Chasing three variables at once gives a line that passes in the morning and fails after lunch.
Weld seam fault-finding order | ||
Symptom | Likely location | What we check first |
|---|---|---|
Weak bond after bending | Strip edges or heat input | Edge condition and impeder position |
Brittle or burnt seam | Weld heat setting | Heat input and line speed |
Open seam or edge offset | Forming and entry guide | Roll alignment and strip tracking |
Ragged internal bead | Scarfing | Scarfing insert condition and shaving |
Repeated ET alarms | Upstream setup drift | Strip join, flattening and cooling |
Before shipment we run a performance test, and we would rather you define what it shows than accept a generic pass. Ask to watch the line produce the size you actually buy, at the speed you will run, from the strip grade you will use. A test on easy material at low speed proves the machine, not your order.
Ask to see ET, UT and hydro results on that run, plus the scarfing and seam annealing setup behind them, and how a suspect length is marked and removed. Ask for the performance test report and the inspection record together. Finally, ask what we will not put in writing: we will not quote a scrap rate, a warranty period or a delivery time we have not agreed, and we will not claim certification to EN, ASTM, JIS, DIN, GB or API. A supplier honest about those boundaries is the one whose test results you can trust on a steel pipe mill.
Weld seam quality is mostly decided before the electrode and confirmed after the cooling water. We look at the strip, place ET, UT and hydro testing where each can catch a real defect, and keep a record that ties a result to a coil. That is the whole method, and it is less glamorous than a brochure claim about perfect welds.
If you are weighing a steel pipe mill, judge it by the inspection sequence and the documents, not a headline speed. Shijiazhuang FAITH Machinery Co., Ltd. publishes its process and model range for reference, and we scope testing and acceptance criteria around the standard you name. Send us your strip grade, size range and target standard, and we will put the inspection plan in writing. We would rather mark a test to be confirmed in writing than invent a number that fails on your floor.
In our published sequence ET comes after water cooling and before sizing, so it screens the whole seam at line speed before sizing changes the signal. We treat it as an early warning, not a final pass, and pair it with UT and hydrostatic testing downstream. The exact tester configuration is to be confirmed in writing against your size range.
No. A hydrostatic test proves the finished tube holds pressure without leaking, while UT looks inside the wall for lack of fusion and internal flaws. Passing one does not clear the other. We scope both where your standard or end use calls for them, and we do not publish test pressures or acceptance classes as marketing numbers.
Usually the strip, not the welder. Dirty edges, a strip that is not flat, a wandering strip end join, or a misplaced impeder will all produce a weak bond with the welder unchanged. We check the uncoiling, flattening, joining and impeder setup first, then adjust one variable at a time and re-test.
Ask for the performance test report and the inspection log together, tied to a coil, a shift and a size. It should show ET events, UT scans, hydro results, scarfing checks and how suspect lengths were removed. We keep a thorough quality inspection process and test before shipment, so a plain reproducible log beats a dashboard you cannot verify.
We check the scarfing tool on a schedule and read the chip it produces, because a dull insert leaves a ragged bead that hides a bad seam or gouges the wall. There is no universal hourly figure we publish; the interval depends on your strip grade, wall thickness and speed, so it is to be confirmed in writing during commissioning.
English
