How to find the root cause with VGraph
Detecting variations in the thickness of steel strips or cold-rolled material is a classic challenge for production and quality departments. Often, the problem only becomes apparent when the coil report comes back—but where in the production line is the fault? In this article, we show you how to use VGraph to go from symptom to cause: from uneven thickness to action, in real time and with statistical precision.
The problem: You see variation – but not why
In many rolling mills and strip processing plants, the problem is familiar:
- Coils are received with noticeable thickness variations along their length – exceeding tolerances or with deviating patterns.
- Quality reports indicate increased scrap rates, more rework, or complaints related to thickness.
- The production team does have measurement data, but it is often fragmented, unsynchronized, or lacks sufficient resolution to pinpoint where the fault occurs.
The solution: VGraph – synchronous, high-resolution process data + statistics
VGraph is the software solution that allows you to take control of process data at a level rarely achieved by traditional systems. Here's what it does:
- Records up to 16 different process parameters synchronously – for example, belt thickness, width, speed, selection force, and tensile stress.
- Adjustable scanning speed down to 1 ms – enabling very high resolution in data and capturing even rapid variations.
- Advanced statistical tools: histograms, Cp/Cpk values, and defect lists—to help quantify variation and track improvements.
- Diagnostics for periodic faults, e.g. built-in FFT analysis that can identify recurring thickness variations caused by, for example, rolling circuits, vibration, or unevenness.
- Flexibility in presentation and evaluation: customizable dashboards, ability to exclude unwanted data drop-offs to obtain more consistent statistics.
By combining real-time data and statistical analysis, it is possible to identify exactly where the variation occurs – and thus target the source of the problem.
Three steps for using VGraph for thickness variation
1. Plug in and start logging
- Ensure that sensors for thickness, width, speed, selection force, and tension are correctly connected and calibrated.
- Start VGraph and set the desired scanning speed (e.g., 1–10 ms depending on line speed).
- Start a test coil or inspection run where variation has been noted.
2. Visualize and analyze
- Display time series for thickness over the length of the coil and parallel parameters (e.g., rolling force, speed).
- Generate histograms and Cp/Cpk values to see the spread and possible imbalance in the process.
- Use error lists to mark and filter out cut-off data ends (drop-offs) so that the statistics apply to the “correct” coil section.
- If the pattern indicates recurring variation, activate FFT analysis to detect periodic faults and link them to machine components.
3. Identify the root cause and take action
- If thickness variation correlates with, for example, reduced selection power or increased band tension, then the component is in focus.
- If FFT shows a dominant frequency that matches roller rotation or bearing frequency: schedule inspection/maintenance there.
- Document before and after statistics (Cp/Cpk improvement), generate coil reports via VGraph, and integrate this into the QA system.
- Follow up on a number of coils to confirm that the measure has been effective: reduced dispersion, higher process capability, and fewer complaints.
Contact us
See how VGraph can reveal the cause of your thickness variation – book a demo.
Contact us at Vollmer Scandinavia for a free consultation and demonstration where we will go through how VGraph would work in your facility.