Packaging machines: narrow seal windows, higher speeds

Seal temperature stays fixed while film and speed vary

What happens. The sealing jaws are held at a set temperature. When the film changes, the speed rises or the machine warms up, the same temperature gives a seal that is too weak or burnt.

What it costs. Leaks, rejected packs and complaints from food packers. Operators raise the temperature to be safe, which wastes energy and damages thin film.

Why the usual fix fails. A temperature controller keeps the jaw at its setpoint, but the setpoint is not what matters; the heat that reaches the seal is. Retuning for each film is manual work.

How a fixed-price remote diagnosis checks it. From jaw temperatures, speeds, dwell times and the leak-test results of past runs I estimate the state of the seal and check how an adaptive heating control would follow film and speed. The fixed-price remote diagnosis shows whether it would hold the seal over the range of products you serve.

Background: Control that adapts to drift and wear

Recyclable mono-material films have narrower sealing windows

What happens. Packers switch to thin mono-material films and heat-sealable papers that can be recycled. They have a narrow window between a seal that is too weak and one that burns, and they leak where layers pile up at folds.

What it costs. Rejects and slower lines during the switch. Machines that ran the old film at full speed are slowed for the new.

Why the usual fix fails. Raising the jaw temperature closes the leaks but burns the film. Flexible jaws and new tools help, but still need the heating and pressure to be matched to the film.

How a fixed-price remote diagnosis checks it. From jaw logs, film data and leak-test results I check how a model-based control of jaw temperature and sealing parameters would work within the narrow window. The fixed-price remote diagnosis shows whether your machines can seal the new films at the speed the customer expects.

Background: Control that adapts to drift and wear

Faster machines run on old cam profiles

What happens. A machine is sped up, but its motion profiles for cams, pushers and feeders are the old ones. Vibration rises, parts bounce, and the drives hit their limits.

What it costs. Vibration and wear in gears, cams and bearings, and a speed that stays below what the mechanics could do.

Why the usual fix fails. Smoothing the profile by hand costs time per product. A stiffer mechanism is heavier and does not suit retrofit.

How a fixed-price remote diagnosis checks it. From drive records and a model of the mechanism I compute motion profiles that respect the drive limits and the natural frequencies. The fixed-price remote diagnosis shows how much speed is available for your machine without changes to the mechanics.

Background: Optimal paths and setpoint plans

Every format is set up by hand

What happens. Changing from one pack size to another means adjusting guides, positions, speeds and temperatures. Operators know the settings of formats they have run, and look up or guess for new ones.

What it costs. Downtime at every changeover, and a start-up with rejects until the settings are right.

Why the usual fix fails. Recipe tables store what has been run, not what a new format needs. A new operator adds time and errors.

How a fixed-price remote diagnosis checks it. From machine and recipe data I check how settings relate to pack size and film, and whether a model could suggest the settings for a new format. The fixed-price remote diagnosis shows how much of the changeover can be taken over by settings from a model.

Background: Replacing hand-tuned parameter tables by a model

Web tension swings at splices

What happens. When a new reel is spliced in, the diameter, the inertia and the tension step. The web swings, goes slack or tears, and the machine has to be slowed or stopped.

What it costs. Breaks, waste of film at each splice and downtime for threading.

Why the usual fix fails. A tension controller tuned for a steady reel handles the splice badly. Slowing down for every splice costs output.

How a fixed-price remote diagnosis checks it. From tension, speed and drive logs around splices I estimate the web tension and the reel diameter and check how control with that estimate would behave. The fixed-price remote diagnosis shows whether the splice can run without slowing the line.

Background: Computing unmeasured states for control

Print mark registration drifts

What happens. The print marks on the film must meet the cut or seal position. Over a run, the film stretches, the temperature changes and the registration drifts away from the mark.

What it costs. Waste: packs cut through the print and the machine runs with a margin. Operators correct by hand.

Why the usual fix fails. A mark sensor and a simple correction react after the error has appeared. Higher speeds make the delay more visible.

How a fixed-price remote diagnosis checks it. From mark sensor, drive and tension records I identify the sources of the drift and check how a model-based register control would reduce it. The fixed-price remote diagnosis tells you how much waste the control could remove and which signals it needs.

Background: Control that adapts to drift and wear

Products differ from batch to batch and cause jams

What happens. The product to be packed changes in size, stickiness or shape from one batch to the next. The infeed that worked yesterday jams today.

What it costs. Jams, stops and operators who stand by the machine to clear them. Output is below the nameplate.

Why the usual fix fails. Setting the infeed for the worst product slows everything. Stricter incoming checks move the problem to the supplier but do not solve it.

How a fixed-price remote diagnosis checks it. From stop records, sensor signals and product data I check which product properties correlate with the jams and whether the handling could adapt. The fixed-price remote diagnosis shows what adaptive handling would need to see on your machine.

Background: Control that adapts to drift and wear

Servo axes are tuned on every machine at commissioning

What happens. Each machine leaves the factory with servo axes that have to be tuned at commissioning, because tolerances in mechanics and cables differ from machine to machine.

What it costs. Time at the factory or at the customer for each machine, and results that vary with the person.

Why the usual fix fails. Taking over the values of a reference machine gives an approximate start. Auto-tune functions give safe but slow results.

How a fixed-price remote diagnosis checks it. From the response data that each axis can record in a short test I identify its dynamics and compute settings. The fixed-price remote diagnosis shows whether automated tuning would give you the same quality on every machine with less commissioning time.

Background: Replacing hand-tuned parameter tables by a model

Heaters and vacuum pumps are always on

What happens. Sealing heaters, tunnels and vacuum pumps run continuously, also during stops, waiting for material and changeovers.

What it costs. Power for equipment that does not work, and wear on pumps and heaters.

Why the usual fix fails. Switching off by hand risks long warm-up and delays the restart. A timer does not know about the schedule.

How a fixed-price remote diagnosis checks it. From the power logs and the machine state records I check how much of the consumption falls in stops and which units could follow demand safely. The fixed-price remote diagnosis shows how much a demand-led control would save and what the machine controller needs.

Background: Economic predictive control

Conveyors: accumulation and gaps in the product flow

What happens. Products arrive in bunches, then in gaps. The infeed of the machine is starved or blocked, and the line speed has to follow.

What it costs. Throughput below the machine’s capability and stops caused by upstream behaviour.

Why the usual fix fails. A buffer conveyor smooths some of it. Adjusting the speed of every conveyor by hand needs constant attention.

How a fixed-price remote diagnosis checks it. From the sensor and speed logs of the conveyors I check how line-flow control that coordinates their speeds would have kept the feed even. The fixed-price remote diagnosis shows how much throughput the flow control could add without new equipment.

Background: Optimal paths and setpoint plans

Line stops are not explained

What happens. The line stops many times a shift, mostly for short periods, and the operator’s log says only “jam” or “sensor”. The cause behind remains unknown.

What it costs. Lost output across a year, and effort spent on the loudest problems rather than the biggest ones.

Why the usual fix fails. A counter for stop minutes gives a total, not a cause. A study by an engineer on site looks at a few shifts.

How a fixed-price remote diagnosis checks it. From the machine’s stop records, alarms and process signals I look for patterns: what precedes the stops, which ones repeat, which depend on product or time of day. The fixed-price remote diagnosis shows the main causes in order and which of them are control problems.

Customers abroad call for help and service travels

What happens. A machine runs at a customer in another country and stops. The customer calls, the service engineer needs a day or more to get there, and the cause is a setting or a sensor.

What it costs. Travel cost, time of experienced engineers and a stopped line at the customer, who judges the maker by the downtime.

Why the usual fix fails. Phone support without data is guesswork. A remote-access link helps only if someone looks at the right signals.

How a fixed-price remote diagnosis checks it. With logs, alarms and settings that your machines record, I diagnose problems remotely and tell the service team what to do. The fixed-price remote diagnosis shows what a remote diagnosis can find on your machines and which data you should store to make it possible.

Shrink tunnels are set for the worst pack

What happens. The tunnel heat and the belt speed are set so that even the biggest, densest pack shrinks fully. Smaller packs get more heat than needed.

What it costs. Energy, and film that is overheated: weak seams, hazy film, deformed products.

Why the usual fix fails. Setting by product from a table needs good tables for every product. Measuring each pack’s temperature in the tunnel is hard.

How a fixed-price remote diagnosis checks it. From tunnel temperatures, belt speed and energy logs I estimate the pack temperature as it moves through and check how much heat can be saved per product. The fixed-price remote diagnosis tells you how far the tunnel can be run on computed pack temperature.

Background: Computed measurements

Glue amount varies with temperature

What happens. The hot-melt glue flows differently as its temperature changes in the tank, the hose and the nozzle. The same pump speed gives a thicker or thinner bead.

What it costs. Rejects with weak or open glue seams, and glue that is wasted by overdosing for safety.

Why the usual fix fails. Raising the amount for margin costs glue and can smear. Checking the bead by eye finds it after the fact.

How a fixed-price remote diagnosis checks it. From temperature, pump and line speed logs and the reject records I check how the bead depends on temperature and speed and how adaptive dosing would hold the amount. The fixed-price remote diagnosis shows how much glue and how many rejects control could save.

Background: Control that adapts to drift and wear

Wider and faster lines bring new dynamics

What happens. The next machine generation is wider and runs faster. Longer rollers, heavier reels and more axes in one machine change how it swings and how its parts interact.

What it costs. Controller settings from the last generation no longer fit, and the first machines on site need many corrections.

Why the usual fix fails. Scaling up the old settings gives a start that has to be corrected on the real machine. A bigger controller does not change the dynamics.

How a fixed-price remote diagnosis checks it. From design data, the last generation’s measurements and the new machine’s drawings I build a dynamic model and see where problems will arise. The fixed-price remote diagnosis shows what control the new generation needs before the first one is built.

Background: Predictive control of process plants

Faults are found at the customer

What happens. Software and sequences are tested in the factory on a machine that is not the final one. Faults in format changes, rare product types or the interplay with the customer’s line show up at the customer.

What it costs. Delays at the customer, repeated trips and a damaged reputation.

Why the usual fix fails. A longer factory test covers more cases, but not all of them. Simulation of the signals alone proves the logic, not the behaviour.

How a fixed-price remote diagnosis checks it. From the control programme, drive data and machine drawings I check which behaviour a dynamic simulation must cover to find the faults before shipping. The fixed-price remote diagnosis shows which of your typical on-site faults a simulation could catch.

Commissioning experts are overloaded

What happens. The few engineers who can commission the machines are on the road most of the year. Everyone waits for them, and what they know is in their heads.

What it costs. Delays in the start-up of machines, travel stress and the risk of losing one of them.

Why the usual fix fails. Training new people takes years. Written instructions cover the cases that were written down.

How a fixed-price remote diagnosis checks it. From machine records and the settings that your experts chose I derive how the settings follow from the machine and the product, and document them as a model. The fixed-price remote diagnosis shows how much of the experts’ know-how can be written down as model-based settings that others can apply.