Packaging
Bends and transfers where a container turns sideways and stops the whole line.
Teal Neural
The belt has stopped and the line keeps pushing.
Jams, gaps and build-ups on belts and conveyors. What turns a ten-second stoppage into a twenty-minute one.
The jam is seen when it is already a pile.
A belt stops over something trivial: a container on its side, a badly stacked box, a product that turns sideways. If somebody sees it at the time, it is sorted in ten seconds.
If not, what is behind keeps coming. And what was one badly placed container becomes a pile that has to be taken apart by hand, with the whole line stopped.
The difference between the two situations is thirty seconds of attention. The same thirty seconds nobody has.
You watch the point where it always happens.
Everyone in the plant knows which bend causes trouble. You put the rule there: if product stops moving for x seconds, it alerts.
The alert goes wherever it is useful: a screen, a beacon in the area itself, or the mobile of whoever is nearby. In a jam, a local alert is worth more than an email.
And you are left with the figure: how many jams, at which point and at what time. With that you decide whether the problem is the bend or the format that was changed in March.
What it tells apart.
- Product stopped at a point on the belt.
- Build-up upstream of a jam.
- Gaps: belt running empty when it should be carrying product.
- Product fallen off the belt.
- Belt stopped with product on it.
- Belt speed and mismatches between sections.
- Following one particular product along the route.
- Silos and hoppers: level, leaks and blockage at the discharge mouth.
- Belt misalignment and rubbing against the structure.
Lo que Teal Neural entiende
5 detecciones grabadas.
Producción y línea de fabricación.
Teal
Teal
Teal
Teal
Teal
Cada una es una grabación del sistema funcionando. Pasa el ratón por encima —o tabula hasta ella— para verla en movimiento.
A real case.
In a drinks plant, the bend into the case packer jammed an average of nine times a shift. Each jam cost between 4 and 25 minutes depending on how late it was spotted.
With the alert at the time, the average time to clear it fell to under a minute. The number of jams stayed at nine — that is a mechanical problem and it was fixed separately — but the time lost went from two hours a shift to nine minutes.
Where it is used.
Parcels and baggage
Sorting belts, build-ups and items that have fallen off the belt.
Bulk materials
Belts carrying ore, grain or waste, and jams in the discharge hopper.
Automated warehouse
Transfers between conveyors and accumulation points.
Following a product end to end.
On a long line with diverts and accumulators, knowing where a particular order is is difficult: you know it went in and you know it came out.
With visual identification, an item can be followed from one section to the next without carrying a label. That is useful for two things: finding it when somebody asks, and measuring how long it really takes to travel the line.
And travel time is what tells you whether the problem is belt speed or the waits at the diverts, which are usually 80% of it.
Silos and hoppers.
The discharge mouth of a silo is one of the places that jams most and is seen worst: it is up high, it is dirty, and nobody climbs up to look until material stops coming out.
A camera there tells you whether material is coming out, whether it has bridged and whether there is a leak at the seal. And the rough level of the silo comes from the same image without fitting a probe.
The misaligned belt.
A belt rubbing against the structure wears on one side and ends up breaking. Before it breaks, it rubs for weeks.
That sideways drift is visible in the image and gets compared with how it was. It is exactly the kind of thing that sets off no alarm and costs you a new belt.
Casos grabados
Funcionando, en sitios de verdad.
Botellas atascadas - Optimiza tu línea de producción con Infinity Neural
Optimización de la Producción Frutícola con Teal Neural
Monitorización de manómetros con Teal Neural
Teal Neural: Optimizando la producción industrial
Teal Neural: Optimización y control en cintas de transporte
What it does not do.
It does not clear the jam. It alerts, and if needed it can stop the upstream belt so that no more piles up; taking the container out is still done by a person.
And it needs to see the belt. A covered section, or one inside a housing, cannot be watched this way.
And what sits alongside it.
Other things the same family understands, on the same cameras.
What people ask us.
▸Can it stop the belt automatically?
It can send the command if there is something to send it to: a relay, a PLC input, a control system. What gets done with that command is decided by whoever knows the line.
In many installations what goes in first is only the alert, and the automatic stop is added later, once you know how many false alerts there are.
▸How many false alerts does it give?
In the first few days, as many as it takes to tune it. That is why it is left running without alerting anyone for a week: that tells you whether the threshold is set right.
A system that over-alerts gets ignored in three days, so tuning that is part of the installation and not an extra.
▸And if the belt is covered?
Then it cannot be seen and there is nothing a camera can do. You cover the visible sections, which tend to be exactly the troublesome ones: the bends and the transfers.
▸Does it tell a jam from a normal stoppage?
Yes, because they look different in the image: in a normal stoppage the belt is still and empty, or with product in order; in a jam there is product piling up.
Get started
Do you want this on your site?
This is already recorded working. What is left is to see whether your camera allows it: where it is fitted, what angle it has and what can be seen from there. If it will not work, we will tell you.