Nonwovens and fibre plants: uneven webs and lab-only quality

The web comes out cloudy and uneven along the line

What happens. The scanner shows basis weight wandering up and down along the roll: thin patches, heavy streaks, a cloudy look when the web is held against the light. The causes sit upstream: fibre dosing into the card or airlay, changing fibre lots, humidity, the fill level of the feed chute. Operators react when the trend line leaves the band.

What it costs. Rolls that fail the customer’s minimum weight or strength are downgraded or scrapped. To stay safe, most lines run a higher average weight than the order needs, which is fibre paid for and given away.

Why the usual fix fails. Tighter tuning of the feed controller helps for one fibre and one speed, then oscillates when either changes. A second scanner shows the problem more precisely but corrects nothing, because the correction still comes long after the cause.

How a fixed-price remote diagnosis checks it. From your scanner records, feed and chute signals, line speed and fibre lot changes I look at which disturbances explain the swings and how long they take to reach the scanner. The fixed-price remote diagnosis tells you whether a feed controller that acts on the cause, not on the scanner alone, could hold the weight closer, and what that would take.

Basis weight varies across the width, so the whole web runs heavy

What happens. Across the width the web is heavier in some lanes and lighter in others, often thinner at the edges or in a fixed band. The profile changes with speed, product and fibre blend. The adjustments for the cross profile are set by hand after a look at the scanner picture.

What it costs. The thinnest lane decides whether the order is met, so the whole web is made heavier than needed. That is raw material in every roll, and slit rolls from the weak lanes are the first to be rejected.

Why the usual fix fails. Hand adjustment works for the product on the line today; after a change the old settings no longer fit. Neighbouring adjustments influence each other, so correcting one lane disturbs the next. Automatic profile loops that treat each zone on its own tend to fight each other.

How a fixed-price remote diagnosis checks it. From scanner profiles, actuator positions and product changes I estimate how each adjustment acts on its own lane and on its neighbours. The fixed-price remote diagnosis answers whether a profile control that handles all zones together could flatten the profile enough to lower the average weight, and what data and changes it would need.

Fibre quality is known only from the lab, and few people can steer the line

What happens. In fibre spinning, fineness, strength and elongation are measured in the lab hours after the fibre was made. In between, a small number of experienced people read temperatures, pressures, draw ratios and bath values and decide when to intervene. When they are on holiday or retire, the line runs more cautiously.

What it costs. Off-spec fibre made while the lab result was still pending, cautious set points that cost throughput, and a dependence on a handful of people the plant cannot replace quickly. Engineers who can model such processes are hard to find.

Why the usual fix fails. More lab samples shorten the gap a little and add cost. Writing the experts’ rules into the control system captures the normal cases but not the reasoning behind the unusual ones.

How a fixed-price remote diagnosis checks it. From your process history and the matching lab results I check whether the lab values can be computed from data the line already records, and how reliably. The fixed-price remote diagnosis tells you whether such a computed quality value could guide operators or a predictive control, and what it would take to get there.

Background: Computed measurements

Energy prices swing, and the energy cost per tonne of fibre is a guess

What happens. Fibre plants use a lot of steam, heat and power in spinning, washing, drying and recovery. Energy is bought as a site total, and how much each step uses per tonne of fibre is estimated rather than known. When prices rise, management asks for a cost-down plan, and the numbers behind it are thin.

What it costs. Without a clear picture per step, savings are argued, not shown. Investment goes to the most visible equipment, while the plant keeps its usual operating mode whatever energy costs that day.

Why the usual fix fails. An energy audit lists the consumers once and ages quickly. More meters show consumption but not how it depends on throughput, product and operating mode, which is what decides where savings are possible.

How a fixed-price remote diagnosis checks it. From your meter readings, production logs and the main set points I build a picture of energy per tonne for each step and how it moves with load and product. The fixed-price remote diagnosis shows where the operating mode, rather than the equipment, drives the cost, and whether running the plant against energy prices could lower it.

Background: Economic predictive control