The forming station of a spiral submerged arc welded (SAWH) pipe mill is where raw coil stock becomes a pipe — and where the accumulated dimensional uncertainty of the entire production batch is either controlled or allowed to propagate downstream into weld defects, diameter variation, and rejection. The strip enters the forming head at an oblique angle, is pre-bent, and coils helically onto the mandrel while simultaneous submerged arc welding joins the advancing edge to the previously deposited seam. The entire process is continuous and relatively fast, which means any developing misalignment of the strip position — a strip running narrow on one edge of the milling zone, a crescent-shaped plate head causing lateral drift, or uneven delivery-roll pressure from a variable-width coil splice — builds on itself with each coil revolution until the forming geometry is visibly wrong or the welder encounters an edge condition it cannot bridge cleanly. By that point, the off-spec pipe section may already be several meters long.
Manual strip width measurement with a tape or caliper, performed periodically by an operator standing in a high-noise, vibration-heavy environment near the forming mandrel, cannot realistically catch these developing deviations early enough to prevent defective pipe. The measurement interval is too long, the working environment discourages prolonged presence at the measurement point, and the results are not fed back into any automated comparison or alarm logic — they exist only in the operator's memory until they find an opportunity to make a correction. The automated strip width measurement and deviation warning system described in this article replaces manual periodic checking with continuous, three-station grating-ruler measurement, real-time PLC processing, and tiered visual-and-audible alarm output — bringing the detection latency from minutes down to the sub-second range and connecting that detection to a standardized response workflow.
Why Strip Position Accuracy Defines SAWH Pipe Quality
The quality of a spiral welded pipe is mechanically constrained by the consistency of the strip forming process in ways that are not always obvious from the finished product datasheet. Pipe outer diameter, wall thickness uniformity, seam weld joint preparation geometry, and the absence of surface imperfections such as uplift ridges and twist distortion all trace back to the strip's behavior in the 1–2 meters of travel between the delivery rolls and the weld point. Four specific defect modes arise directly from strip positional deviation:
The strip edge drifts laterally away from the milling cutter's designed cutting line. The edge bevel prepared for submerged arc welding is either not cut or cut at the wrong angle, degrading fusion zone geometry at the weld joint.
The strip edge drifts into the cutter by more than the design allowance. Excess material is removed, reducing the strip's effective working width and causing the pipe's nominal outer diameter to be undersize after forming.
Uneven delivery-roll pressure caused by a laterally drifted strip creates a local buckle or uplift ridge on the strip surface in the forming zone. The ridge disrupts the smooth helical seam and can cause poor weld penetration at the raised area.
A combination of lateral drift and non-uniform roll-pressure distribution causes the strip to enter the forming mandrel at an angle diverging from the designed helix angle. The resulting pipe section is geometrically distorted, producing out-of-roundness and misalignment of the weld seam.
Strip width variation across a coil or at a coil-to-coil splice causes the pipe diameter to deviate from nominal, with the forming geometry producing an asymmetric pipe end that cannot be squarely cut or field-welded without additional preparation.
Lateral strip position error at the weld point causes the strip's leading edge to be at a different radial height than the previously deposited seam, creating a step at the weld joint. API 5L and GB/T 9711 both impose strict limits on seam misalignment; consistent violations lead to batch rejection.
The root cause connecting all six defect modes is the same: the strip's actual lateral position deviates from its designed position at one or more points in the forming path, and the deviation either goes undetected until visible in the product or is detected but not corrected fast enough to prevent a run of defective pipe. Crescent-shaped plate head geometry (月牙弯, a common coil characteristic where the plate head curves laterally from the nominal strip centerline), variable-width coils at splices, and uneven inter-coil material properties are all sources of the initial perturbation — none of which the forming mill can prevent at source, but all of which the measurement and warning system can detect and prompt the operator to correct.
System Architecture: Three Measurement Stations, One Integrated Response Loop
The key design insight of the system described here is the placement of measurement at three discrete points in the strip travel path rather than at a single location. A single measurement point gives the operator a width reading but cannot distinguish between a strip that is running centered but narrow (a material issue) versus a strip that is full-width but has drifted laterally in the feeding direction (a positional deviation issue) — yet these two conditions call for different corrective responses. Three stations that span the pre-milling, post-milling, and post-pre-bending stages provide both the magnitude and the spatial pattern of any developing deviation, enabling a more targeted diagnosis.
Measures the original raw coil strip width as received from the steel mill before any edge processing. Establishes the baseline material width, identifies incoming width variation across a coil, and detects crescent-bend conditions at coil heads that will cause downstream positional drift.
Raw material width RS485 to MCGS1 (distance > 30 m)Measures the strip's working width after the edge-milling step that prepares the bevel for submerged arc welding. The difference between Station 1 and Station 2 readings confirms whether milling is removing the designed width of material or whether off-milling or over-milling conditions are developing.
Working width 4–20 mA analog to PLC AI moduleMeasures the strip's width and lateral position immediately before it enters the forming mandrel. This is the most operationally critical measurement station — any positional deviation detected here directly predicts what will occur at the weld point. It also confirms whether the pre-bending operation has introduced any width change or lateral shift.
Pre-forming position 4–20 mA analog to PLC AI moduleHardware Components
Each station uses a SICK grating ruler sensor with an individual beam pitch of 0.3 mm. The operating principle is straightforward: the sensor's transmitter projects a curtain of parallel light beams across the full width of the strip passage; the receiver on the opposite side registers which beams are blocked by the strip and which are open. Strip width is calculated as the count of blocked beams multiplied by 0.3 mm — so a strip blocking 2,000 beams has a measured width of 600 mm. Because the transition from blocked to unblocked at each strip edge occupies exactly one beam in a perfect measurement, and there are two edges, the maximum systematic measurement uncertainty is 2 × 0.3 = 0.6 mm. In practice, a residual 0.5 mm systematic sensor bias was identified during commissioning and removed by calibration offset, bringing measurement accuracy to within 1 mm — above the production process requirement. Strip position information (lateral deviation) is derived from the left-edge and right-edge blocked-beam counts independently, rather than from the total width alone, allowing the system to distinguish centered-but-narrow from full-width-but-offset strip conditions.
Control System Architecture
Three-Level Alarm Logic: Green, Amber, Red
The system's response to a measured deviation is not binary (alarm / no alarm) but uses a three-tier graduated logic that matches the urgency of the operator response to the severity of the deviation. Each grating sensor station has independently configurable thresholds set according to the specific process requirements at that station — the tolerance for positional deviation at the post-loading station is wider than at the post-pre-bending station, reflecting the fact that a given lateral offset at Station 1 becomes more consequential as it propagates toward the weld point.
The Environmental Challenge: Debris, Dust, and the Pulsed Air Blow Solution
Laboratory-quality measurement accuracy is straightforward to demonstrate on a clean strip in a controlled environment. The conditions at an operating spiral pipe mill are substantially different: the strip surface carries rust scale, iron fines, and milling chips; these particles are continuously dislodged by the forming and milling operations and settle on the lower grating element of each sensor array. When a sufficient accumulation of iron chips or scale bridges one or more light beams at the lower edge of the grating, the sensor registers these as part of the strip edge — producing a falsely elevated width reading and potentially triggering false alarms that, if frequent, cause operators to reduce their confidence in the system and begin ignoring its outputs.
The solution implemented was pulsed compressed-air blow-off of the lower grating element at each of the three sensor stations. Rather than continuous airflow (which would create its own disturbance and add to compressor load), the PLC controls three solenoid valves that operate on an intermittent duty cycle, producing short blasts of air at intervals timed to keep the grating clear without interfering with measurement cycles. The solenoid valve status — open or closed — is displayed on the MCGS operator interfaces, so that an operator can immediately see if a valve has failed to open and take corrective action before debris accumulation begins affecting measurement quality. This relatively simple mechanical countermeasure was the difference between a system that performed reliably in the factory environment and one that generated enough false alarms to become operationally ignored.
What This Means for Pipe Dimensional Quality and Weld Consistency
The measurable outcomes of the system relate to both process stability and product quality, though the connection between the measurement system and the final pipe's dimensional conformance runs through several intermediate steps and cannot be reduced to a single number. What the system directly controls is the operator's response time to a developing strip deviation — it transforms a situation that previously required either continuous human vigilance or periodic manual inspection into one where the machine itself monitors the strip position and alerts the operator only when action is required. This has two effects: the mean time from deviation onset to operator response decreases substantially (from the order of minutes to the order of seconds), and the minimum detectable deviation decreases from what an operator can visually or manually detect to 1 mm.
For SAWH pipe, the downstream quality implications of this improvement are most visible in seam weld consistency and in the reduction of the frequency of off-spec pipe sections at coil splices and at the head and tail of each coil (where crescent bends and width transitions are most likely to occur). Stable strip working width also directly supports submerged arc welding parameter stability: the submerged arc process is sensitive to joint preparation geometry, and maintaining the designed bevel angle and root gap within tolerance throughout the coil run requires that the milling station be cutting the designed amount of material from each edge — which in turn requires that the strip be consistently centered on the milling line. A measurement system that detects the first millimeter of drift from the designed position before it has grown into a milling error that affects weld geometry is therefore a quality assurance tool for the weld as much as it is a forming process tool.
Frequently Asked Questions
Q: How does spiral welded pipe (SAWH) compare to LSAW and ERW pipe in terms of the forming process requirements that this kind of measurement system addresses?
The three main pipe-forming processes — SAWH (spiral/helical submerged arc welded), LSAW (longitudinal submerged arc welded, via UOE or JCOE press), and ERW (electric resistance welded) — each have distinct forming geometries that create different quality-control requirements for the strip or plate material. LSAW pipe uses a flat plate that is pressed into a U-shape and then an O-shape before welding; the forming step is a discrete press operation on a single plate, and width variation in the plate is accommodated by the press tooling within limits — the strip position drift problem described in this article does not apply in the same continuous-feed sense. ERW pipe uses a narrow coil strip that passes through a series of roll stands to progressively form a tube; strip width control is important for weld seam geometry, but the strip feeds in a straight line along the mill axis rather than at an oblique angle, making positional deviation a simpler one-dimensional problem. SAWH pipe's oblique-angle continuous feed is what makes strip lateral position deviation both more consequential and harder to catch manually — each degree of lateral drift in the feeding direction translates into a helical misalignment that compounds with each turn of the coil, and the physical distance between the detection point (an operator watching the strip feed) and the effect point (the weld) can be several meters of pipe wall. This is why the three-station continuous monitoring architecture described here is more suited to SAWH production than to LSAW or ERW.
Q: What API 5L or GB/T 9711 pipe specifications does spiral welded pipe typically meet, and does the strip width tolerance directly affect the API 5L diameter tolerance for the finished pipe?
Spiral welded pipe is produced to API 5L (Specification for Line Pipe) in grades from Grade B through X70 and sometimes X80 for specific project qualifications, as well as to the Chinese national standard GB/T 9711 (which aligns closely with API 5L technically). Large-diameter SAWH pipe is also produced to SY/T 5037 for oil and gas gathering lines and SY/T 5040 for other pipeline applications in the Chinese market. The connection between strip working width and finished pipe outer diameter is direct and geometric: in the SAWH forming process, the pipe outer diameter D is related to the strip working width W and the helix angle α by D = W / (π × sin α). If W deviates from its design value, D deviates proportionally. API 5L specifies outer diameter tolerances that depend on the nominal pipe diameter and the manufacturing process — for pipe with D above 508 mm (20 inches), the OD tolerance is typically ±0.5% of the specified outside diameter. For a nominal 1,016 mm (40-inch) SAWH pipe, this corresponds to an OD tolerance of ±5.08 mm, which translates via the geometric relationship to a working strip width tolerance of roughly ±8 mm at a typical 65° helix angle. The 1 mm measurement accuracy of the system described here is therefore well within the margin needed to detect the strip width deviations that would cause the finished pipe to approach its OD tolerance limit.
Q: What is the significance of using three grating sensor stations rather than one, and could the same result be achieved with a single sensor at the most critical position (post-pre-bending)?
A single sensor at the post-pre-bending position (Station 3) would provide the most operationally critical measurement — the strip's position immediately before the weld point — but it cannot tell the operator what is causing a deviation it detects, which limits the speed and accuracy of the corrective response. If Station 3 shows a 3 mm lateral drift, the operator does not know from that measurement alone whether the drift originated from incoming strip width variation (a material issue that requires slowing the line and inspecting the incoming coil), a milling-position error (a cutter alignment issue that requires a specific mechanical adjustment), or a delivery-roll pressure asymmetry (a roll-load balancing issue). With all three stations active, the diagnostic picture is much clearer: if Station 1 reads correctly but Station 2 shows a drift that is carried through to Station 3, the milling station is the source. If Station 1 already shows the deviation, the incoming coil material is the root cause. If Stations 1 and 2 read normally but Station 3 shows a drift, the pre-bending process is introducing the offset. This three-point differential diagnosis capability is what justifies the additional hardware cost and integration complexity of the three-station design — it does not just tell the operator that something is wrong, it tells them where to look first.
Q: How does the 4G cloud remote access capability benefit a steel pipe manufacturer's maintenance and quality operations, and are there data security concerns with connecting production PLC equipment to the internet?
The operational benefits of remote access to production PLC systems are clearest in two scenarios: (1) troubleshooting and program modification by engineering personnel who are not physically present on the production floor — for a system like this one, where the PLC program logic, alarm thresholds, and sensor calibration offsets may need adjustment during initial deployment and after major maintenance events, the ability to download a modified program or update a threshold value from an engineering office without requiring the engineer to cross an active production floor reduces both response time and personnel safety exposure; and (2) remote condition monitoring by the equipment supplier or system integrator during the warranty period or under a maintenance contract, who can review logged data and advise on parameter adjustments without a site visit. The data security question is real and relevant: connecting a production PLC to the internet via a 4G gateway creates a network path that, if inadequately secured, could expose the PLC to unauthorized access. The cloud gateway hardware used in this implementation controls access through IP address management, user authentication, and VPN-equivalent tunneling, and the 4G SIM card connection provides network isolation from the local factory LAN's broader exposure surface. For facilities with stricter cybersecurity requirements (common in critical infrastructure pipeline projects), the remote access configuration would need to be reviewed against the applicable industrial control system security standards, such as IEC 62443, before deployment.
Q: Does this type of strip width and position monitoring system affect the pipe mill's ability to process coils at the head and tail ends, which are typically the most problematic portions for strip alignment?
Coil head and tail sections are disproportionately represented in SAWH pipe quality rejections for exactly the reasons this system is designed to address: the plate head is most likely to have a crescent bend (月牙弯) from the coil set; the width at the plate tail may not match the nominal width of the coil body; and the coil-to-coil splice itself creates a brief period of non-steady-state forming conditions. None of these problems is prevented by a measurement system — the crescent bend is a material characteristic, not something that can be corrected by better measurement. What the system changes is the response speed and certainty with which the operator detects that a head or tail condition is causing a positional deviation. Without the system, an operator typically identifies a coil-head deviation only when it has progressed far enough to be visible — either as strip-edge lift at the milling cutter or as a perceptible change in the forming sound. With the system, the deviation is flagged at 1 mm resolution and at each of three stations as the plate head passes through them in sequence, giving the operator time to apply a pre-emptive correction (for example, adjusting the delivery roll loading asymmetry before the head has fully entered the forming mandrel). The practical effect is a reduction in the length of off-spec pipe produced at each coil head and tail — the crop-cut length required to remove the affected section decreases, improving yield.
Summary
The strip width measurement and deviation warning system described in this article addresses a fundamental quality consistency challenge in spiral welded pipe production: the continuous, high-speed nature of the SAWH forming process means that strip positional deviations compound rapidly from their point of origin to the weld station, and manual detection methods cannot provide early enough warning to prevent the run of defective pipe that follows. The system's three-station grating ruler architecture (measuring at post-loading, post-milling, and post-pre-bending positions) combined with Siemens PLC300 processing, three-level visual-and-audible alarm output, and 4G cloud remote access achieves 1 mm measurement accuracy — exceeding the process requirement — in the demanding environmental conditions of an operating pipe mill, including the surface contamination challenge managed by pulsed air blow-off. The direct outcome is a reduction in the frequency and length of off-spec pipe resulting from strip deviation, which translates to more consistent submerged arc weld geometry and reduced crop-loss at coil head and tail transitions. For technical enquiries about ShunFu Metal's spiral welded pipe products and production capabilities, visit gaslinepipe.com.
