BLP-GL· Pack segment station 03 · Automated on both line grades
Liquid Cooling Plate Gluing Station BLP-GL
The adhesive nobody sees again. BLP-GL lays a thermally conductive structural bead on the liquid cooling plate and a structural sealing bead on the enclosure floor, against a stored recipe, with a record attached. Three quantities controlled; three conditions under which a part stops.
on the part as it sits
per segment, live
on the located part
In the line, on the fixture, with a record
The Station and Its Two Workpieces
BLP-GL is a liquid cooling plate gluing station built into the line, not a bench dispenser: it takes the part on its fixture, photographs it, dispenses against a stored recipe and hands the part on with a record attached. The same head glues both parts: the cooling plate with a thermally conductive structural adhesive, the enclosure floor with a structural and sealing bead.
Two servo-driven metering axes push the two components at a ratio between 1:1 and 2:1, set by recipe. Path speed runs up to 300 mm/s. Parameters and the pressure curve behind them go out over PROFINET, filed by enclosure number.
Process and quality engineers
Specifying a pack line for energy storage or EV pack assembly, and answerable for that bead months later.
Plant engineers replacing a hand step
A hand-dispensed bead that has started to cost at final test. The station is quoted on its own into a running segment.
Quoted two ways
Fifteen stations · one segment
Liquid Cooling Plate Gluing Station at Station 3
The pack segment opens with enclosure and cooling-plate cleaning, then plate air-tightness, then this station, then module insertion, then module fixing. By the time the modules are in at station 4 the adhesive is structural, thermal and out of reach.
Before a part is allowed in
The cooling plate arrives already leak-tested at station 2: 200 kPa of dry air, held 60 s, a pressure drop of 1 kPa or less to pass. That gate sits before the glue on purpose; it is the first of the three hold conditions on this page.
Automated on both line grades
One of four automated stations in a segment that is otherwise hand work: plate air-tightness, gluing, whole-pack air-tightness and pack EOL. Automated because the thing it produces cannot be inspected afterwards, so the process record has to stand in for the inspection. The whole-pack gate later in the segment is a different test with a different threshold, written up on the pack EOL testing page.
| Pack segment station | BLP-P8S | BLP-P15 |
|---|---|---|
| 01 · Cleaning | Manual | Manual |
| 02 · Plate air-tightness | Auto | Auto |
| 03 · Gluing, BLP-GL | Auto | Auto |
| 04 · Module insertion | Semi-auto | Semi-auto |
| 05 · Module fixing | Manual | Manual |
| 13 · Whole-pack air-tightness | Auto | Auto |
| 14 · Pack EOL | Auto | Auto |
Both grades run the pack segment with 12 to 14 people; the four automated stations are the ones whose result cannot be checked by hand afterwards.
Hold condition 01 · the gate before the glue
Air-Tightness Before the Adhesive
A leak found on a bare plate is a plate that gets replaced. The same leak found after the bead is down is a plate with cured adhesive on it, inside a box that has already been scheduled for modules. The leak test is placed where a failed part is still cheap.
| Same leak | Found at station 2 | Found after the bead |
|---|---|---|
| What fails | A bare plate | A glued plate inside a scheduled box |
| What it costs | One plate, swapped on the fixture | Cured adhesive removed by hand, a box re-planned |
| Who decides | The test, before the head moves | Someone at station 4, in a hurry |
| What is recorded | A pass, before the glue | Nothing until final test finds it |
Same head · a different job on each part
Two Workpieces, One Metering Head
The cooling plate: a thermal and mechanical interface
The bead has to be present, continuous and in the amount the thermal design assumed, because the heat path through the pack was designed on paper before anyone glued anything.
- Large, thin, never perfectly flat, and delivered with its own supplier's tolerances.
- Located to 0.2 mm, so the path runs on the plate that is on the fixture rather than on the nominal one.
- Volume judged per segment, live, so a corner and a long straight each get the quantity the design assumed.
The enclosure floor: a structural and sealing bead
Continuity along a defined path carries as much weight as total quantity. Enclosures are fabrications, and between batches the floor is the least repeatable surface this station touches.
- Path committed to 2 mm on the located part, written as a promise about where the bead lands.
- Segment volumes follow the flange run, the corners and the long straights of the floor.
- Out of reach once module insertion starts at station 4; the record is the inspection.
Position · volume · time
Three Controlled Variables, Three Hold Conditions
Measured, commanded, clocked. Right place with the wrong quantity, or right quantity left open too long: neither reaches station 4.
Position
±0.2 mmVision locates the part as it sits on the fixture, before the head moves.
Volume
5 %Dispensed against commanded, per segment, judged live.
Time
Open timeStored with the adhesive recipe; starts when the bead is finished, stops when the modules go in.
Six figures · what each governs · when it is judged
Liquid Cooling Plate Gluing Station Figures
Two describe geometry, the rest describe substance and evidence. The pair most often asked about is 0.2 mm and 2 mm, and the gap between them is the point: the station measures the part to 0.2 mm so that a 2 mm path can be a promise about the bead instead of a hope about the fixture.
| Figure | What it governs | When it is judged |
|---|---|---|
| ±0.2 mmPhoto positioningGeometry | Where the station believes the part is, against the part on the fixture, not the nominal one. | Every part, before the head moves. |
| ±2 mmDispensing pathGeometry | Where the bead is committed to land on that located part. A commitment to your bead design, not the camera's error. | Every bead, on the located part. |
| 5 %Dispensed volume | How much adhesive went down against how much was commanded for that segment. | Live, on every segment of every bead. |
| 300 mm/sPath speed, up to | The travel envelope the path is planned inside; the figure your cycle time is calculated from. | Set by recipe. |
| 1:1 to 2:1Mix ratio, settable | The two-component ratio the twin-servo head holds. | Set by recipe with the adhesive, fixed for the run. |
| PROFINETTo MES, by enclosure number | Where the parameters and the pressure curve go. | At acceptance, against the criteria sheet both sides sign. |
Hold condition 02 · the pressure window
Volume as the Controlled Variable
This is the mechanism the rest of the station is built around. What gets commanded is volume: not a dispense time, not a head pressure, not a number of seconds an operator learned to trust.
Each segment of the path carries a commanded quantity from the recipe for that part number, and dispensed volume is held to 5 percent against that command.
Pressure is what proves it. The head records a live pressure curve while the bead is laid, and the recipe carries the window that curve has to stay inside. A blockage, an air inclusion, a cartridge running down or a mixing tube starting to set all have a shape in that curve before they have a shape you could see on the part.
Operator
Selects the recipe that matches the part, releases parts that passed, handles parts the station held.
Process engineer
Reads the curve, decides what the held part gets, and signs that decision into the record against the enclosure number.
Quality engineer
Sees the criteria the window came from, written before the line shipped and signed by both sides.
What a hold looks like on the floor
The part stays on its fixture; the screen shows the recipe, the segment and the curve against its window. The process engineer's disposition and any rework go into the record.
Drawings, adhesive, target takt
Two Drawings, One Adhesive, One Takt
Send the enclosure drawing, the cooling-plate drawing, the adhesive you intend to use with its mix ratio, and the takt you are building to. That is enough to say how this station would be configured for your pack and what it would be quoted as inside the segment.
Hold condition 03 · the clock in the machine
Open Time on a Clock
Adhesive has a working life once it is mixed and laid, and the pack segment has a buffer, a break, a stopped conveyor and a module that was not ready. Those two facts collide at station 4, and this station is where the collision is managed.
Open time is stored with the adhesive recipe. The clock starts when the bead on that part is finished and runs until the modules go in. If it expires, the part is held.
A stored timer does not get busy
Shift discipline works on a normal day and fails on the day the line stops, because that is the day everyone is busy with the line. The clock is in the machine for that reason.
Bead not yet cured
The plate or enclosure is cleaned off and re-glued at the station. The process engineer judges that call and records it.
Recorded against the enclosure numberBead already cured
The held part goes to the rework position, the adhesive is removed by hand, and the part re-enters the station as a fresh part.
A part that went round twice is visible as oneThe stoppage that has to be recoverable
Mixing Tube Changed In Line
A two-component adhesive begins to set inside the mixer as soon as the head stops. The mixing tube is changed in line, without taking the station out of the segment, so a stoppage becomes a tube change rather than a queue of held parts.
- 01Bead path
- Taken from your drawings
- 02Segment volumes
- One command per segment
- 03Pressure window
- Per adhesive, signed by quality
- 04Mix ratio
- 1:1 to 2:1, fixed for the run
- 05Path speed
- Up to 300 mm/s
- 06Open time
- Per adhesive, timed by enclosure number
Every number lives in a recipe
Recipes Held by Part Number
Operators select recipes. They do not write or edit them. That is a deliberate restriction, the same one placed on module pressing and on busbar welding, because a line where three shifts can each nudge a parameter is a line where nobody can explain a bad month.
Training follows those lines after site acceptance: operators, process engineers and maintenance separately, because the three need different things from this station. The same metering control runs on the smaller dispensing machine on the cylindrical and consumer pack lines, worth knowing if you build both kinds of pack in one plant.
Three trainings, not one
PROFINET · filed by enclosure number
The Record That Leaves the Station
Over PROFINET, filed by enclosure number, the station hands your MES five things. The set is chosen so it can answer the questions that get asked after the fact.
| What leaves the station | The question it answers |
|---|---|
| The recipe number it ran | Which recipe was this part built to |
| Commanded and delivered volume, per segment | How much adhesive actually went down |
| The pressure curve behind the bead | What the pressure was doing at the moment it went down |
| The open-time stamps | How long the bead was open before modules landed on it |
| Any hold event, with disposition and rework record | Whether this part had been round the station before |
The interface and the data are provided
Mapping them into your MES is a scoped piece of work agreed with your automation and IT people during the project, said plainly rather than letting a protocol name stand in for an integration. What is not negotiable is that the data exists and leaves the station.
- Filed by
- Enclosure number
- Leaves over
- PROFINET
- Read by
- Your MES, and pack EOL at station 14
- Mapped by
- Your automation and IT people, scoped in the project
A format change reaches this station as three things
A New Enclosure, a New Path
What does not move
- The 0.2 mm photo positioning.
- The 2 mm path commitment on the located part.
- The 5 percent volume accuracy, per segment.
- The record that goes out over PROFINET.
These describe the station, not the last part it ran. What a change of cell, module or pack costs across the whole line is on the compatibility and changeover page.
Every figure becomes a line on a criteria sheet
The Liquid Cooling Plate Gluing Station at Acceptance
The engineer who set the parameter writes the criterion and quality signs it, before the line ships, so nobody arrives at acceptance arguing about what the number meant.
FATWorks acceptance, in a prebuild bay
The segment runs with your enclosures and cooling plates, and the sampled packs are built with the recipe this station will run in your plant.
- Recipes called by part number, not typed in.
- Volumes held to 5 percent, segment by segment.
- Pressure curves in window, and hold events behaving as written.
SATSite acceptance, at takt
Continuous running at takt in your plant, with your own parts and adhesive, against the same criteria.
- Open time timed on the live line, with its buffers and breaks.
- Record fields arriving in your MES by enclosure number.
- Operators, process engineers and maintenance trained separately after the run.
The full sequence, what each side brings and what the acceptance documents contain are on the line acceptance page. The laser and vision process engineers who write these recipes, and the prebuild bays the run happens in, are described on the about page.
What to send
Drawings, Adhesive, Target Takt
The two drawings say where the bead runs, on which face, and whether one station has to serve both parts on your line. Add the adhesive with its mix ratio and the takt you are building to, and the reply describes the liquid cooling plate gluing station you would actually be buying, not the one in the datasheet.
If the volume does not justify it
If your volume genuinely does not justify an automated station, the reply says so, and you still get the criteria you should be holding a hand-dispensed step to.
Your enquiry, ready to send
If your email client opened, it is already in the message. If nothing opened, use one of the two routes below.
Start with the two drawings
Send the Two Drawings
Four inputs to our engineering team and the configuration, the segment count and the record fields come back within 24 hours. Behind it: 26 engineers and 40+ line segments into 12 countries, on the about page.