Run closer to spec on a line that never stops
Willowark builds measurement, inspection, and control systems for building products manufacturers — gypsum board, insulation, engineered wood, roofing, siding, flooring, pipe and extruded profiles. The work is inline gauging that lets a line run nearer to nominal instead of heavy for safety, surface inspection at line speed, downtime data reflecting what the equipment did, and automation at the packaging end.
The economics here are measured per hour, so we design around what a control loop can act on. Measurement goes where feedback still has authority — before the dryer, before the press, before the saw — and cross-direction profiles matter more than a centerline reading because most defects are lane problems. Everything ties back into the existing PLC or DCS rather than becoming a parallel system operators ignore.
A first project on a building products line is usually one gauge or one section of line: thickness or basis weight at one point of control, moisture profile off one dryer, or downtime capture where it hurts most. Plants here run mixed equipment ages — a press from one decade, a dryer control upgrade from another, a new saw and stacker — so the engineering is mostly getting reliable signals out of what exists and keeping new sensors alive in dust, heat, and vibration. Margins are thin and volume high, so giveaway is usually the first thing worth measuring.
Reviewed
Sound familiar?
If you've said any of these, we should talk.
“We run heavy to guarantee spec and it costs us material every single hour.”
Giveaway is a measurement problem before it is a control problem. We put continuous thickness and basis-weight gauging where the process can still respond, close the loop to existing controls, and move the target toward nominal once real variance is known.
“Moisture coming off the dryer varies across the width and we correct too late.”
We measure the cross-direction profile rather than a single point, using near-infrared or microwave sensing depending on material and thickness, and feed zone-level results back into dryer or press control. Correction happens while material is still in process, not after a lab sample returns.
“The line goes down and three people give three different reasons.”
We capture states, faults, and drive data straight from the controls with a consistent event timeline, so first-out fault and downstream cascade are distinguishable. Operators classify only what the system could not attribute, and the Pareto chart then reflects the plant.
“End-of-line inspection is one person with a flashlight and a good eye.”
We build surface inspection that runs at line speed: line-scan cameras, lighting geometry chosen for the defect type, and processing that maps defects to position so grade and cut decisions can use them. It stops routine defects reaching the stack.
“The lab result comes back an hour after the material is already on the stack.”
Lab tests still matter, but they cannot steer the line. We put continuous inline measurement in front of the lab, correlate it against your lab method so both agree, and use the inline value for control while the lab confirms. The lab sample becomes a check rather than the only number anyone has.
How this industry actually runs
The operation as we usually find it.
A building products line is continuous, fast, and expensive to stop. Material is mixed or formed, run through dryers, kilns, or a continuous press, cured, then cut to length, stacked, wrapped, and palletized, often at speeds where a minute of off-spec output is a measurable pile of scrap. The properties defining the product — thickness, width, basis weight or density, moisture content — are exactly the ones that drift with ambient conditions, feedstock variability, and dryer performance, so most plants build in a safety margin and give away material every hour of every shift. Moisture is the recurring villain: leave too much in and the product fails, dry too hard and you burn gas and stress the sheet, and the profile across the width is rarely uniform. Product standards come from ASTM and equivalent test methods, and grade decisions often still depend on somebody watching the line. Demand tracks housing starts, so a plant runs flat out for months, then faces changeovers between SKUs it barely ran last quarter.
Machine signals to the people who decide
Components:
- PLCs & sensors (counts, states, current)
- Legacy machine (dry contact / clamp)
- Edge gateway (normalize, buffer)
- Production dashboard (downtime, OEE)
- Alerts & reports (who acts, when)
Connections:
- PLCs & sensors to Edge gateway (EtherNet/IP, Modbus)
- Legacy machine to Edge gateway
- Edge gateway to Production dashboard (MQTT)
- Edge gateway to Alerts & reports
What we build
Starting projects that fit Building Products & Materials.
- Inline thickness, width, and basis-weight measurement with closed-loop control to reduce giveaway
- Cross-direction moisture and profile measurement feeding dryer or press zone control
- Surface defect inspection at line speed with position-mapped defect data and grade decisions
- Downtime, rate, and OEE capture pulled directly from PLC, DCS, and drive data
- Line tracking, speed synchronization, and cut-length optimization
- End-of-line automation: stacking, wrapping, labeling, and robotic palletizing
- Energy monitoring on dryers, kilns, and compressed air with cost attribution per line
- Grade and scrap tracking that ties off-spec output to the process conditions that produced it
Capabilities we bring
Working in Building Products & Materials?
Tell us the line.
What runs by hand, what is not connected, what you are trying to build. An engineer replies within one business day with whether and how we would approach it.
Common questions
What Building Products & Materials teams ask first.
Will measurement equipment survive our environment?
That is a design requirement, not an afterthought. Dust, heat, vibration, and washdown drive enclosure ratings, purge and cooling design, sensor standoff, and mounting stiffness, and getting them wrong is the usual reason a gauge that demonstrated well becomes a maintenance item.
Can you tie into the PLC and DCS we already have?
Yes, and that is normally the right approach. We read and write through the existing control system over EtherNet/IP, PROFINET, Modbus, or OPC UA rather than building a parallel controller operators have to reconcile against. Where a section is too old to expose data, a gateway brings it in.
How do you handle changeovers between products?
With recipes. Gauging targets, control limits, inspection thresholds, and cut logic are stored per product and selected together, so a changeover is a selection rather than manual retuning. Recipe changes are logged, because the usual cause of a mysterious quality shift is a setting somebody adjusted three shifts ago.
How do you validate an inline gauge against our lab method?
By running both on the same material until the relationship is understood. We take inline readings at the point of measurement, pull lab samples from the corresponding position on the sheet, and compare across the product range and across the width. Offsets are usually stable and correctable; where they are not, that is itself useful information about the process. The lab method stays the reference, and the gauge reports in the same units it does.
Our line runs around the clock. When do you install?
During whatever outage you already have, usually a scheduled maintenance day or a changeover. Most of the work happens before then: mounting frames, cable routes, and control changes are designed and bench-tested against a simulation so the on-line window is spent on mounting, wiring, and a verification run. Where a sensor can be installed without stopping the line, such as a non-contact gauge on a clear span of web, we do that instead.
Strategy. Software. Systems.
Engineering for Building Products & Materials.
Describe the problem in your own words. An engineer reads it — not a sales script — and tells you plainly what it would take.