Terminal addressing / BLP-AD — module segment, station 11 of 17

A Module Reaches the Welder in One of Three States. The Welder Can See Two.

Two questions have to be answered before any energy goes into the part: where each terminal is, and whether it is the terminal the program thinks it is. One vision set answers both, in line, at 200 to 300 millimetres per second.

Send your module drawing See what it costs at each station

⚠️ The third column below is not a machine state. It is what you have when nothing measured — and it is the only one of the three that a welding station cannot tell apart from the first.

The three states, and who knows about each

At the moment of welding	Established, correctAddressed at station 11	Established, reversedAddressed at station 11	Never establishedNo station asked
Who knows which pole is which	The line does, before anything is fitted over the terminal	The line does, and it stops the module here	Nobody. The module looks identical to the first column
What happens next	Busbar plate and CCS go on, then the joint is made	Held, with the terminals still bare	The joint is made anyway
Where it surfaces	Nowhere — there is nothing to surface	Here, on a bare terminal, before hardware	After the joint existsWhy that is a different problem →
What it has cost by then	Nothing beyond the cycle it took	One module, held, and the cycle	A joint that has already been made

≤±0.2 mmAddressing position
200–300 mm/sMotion speed
1Vision set, two questions
11 / 17Station in the module segment

What each of the three states costs, and who finds out first

Two of them are ordinary industrial events with a known price. The third is what makes this station a line item rather than an option, because it has no price until much later.

Established and correct

The vision set has returned a position for every terminal and accepted the polarity of each. The two stations after it inherit both answers. Nothing has been spent and nothing has to be looked at again — which is why this state is invisible in a project plan and easy to undervalue.

Established and reversed

The module is held here, with the terminals still bare and nothing seated over them. It is the cheapest moment this can be found, and finding it is the reason a polarity judgement sits inside a positioning station rather than in a separate inspection somewhere downstream.

Never established

No station asked the question. The stack is banded, the plate goes on, the joint is made — and from that point the question is no longer which pole this is. It is what to do about a joint that already exists, and that argument is made once, on the equipment page.

A quotation almost never separates these three. It says vision inspection and leaves you to assume the first state is what you are buying. The useful question is narrower: at which station number does the asking happen, and what is sitting on top of the terminal when it does.

Assembled module with terminals facing up before hardware is fitted

Detail of a prismatic cell terminal on a banded stack

⚠️ The three images are the same module at three moments. Only the first two are moments at which a question can still be asked cheaply.

Station eleven of seventeen, and the last one that reads

Its place in the sequence is not an administrative detail. It is the reason the figures on this page mean what they mean, and it can be read straight off the published station list.

1234567891011121314151617

Module segment — 17 stations, and this page is number 11

09 Pressing and steel banding — the stack stops being cells and becomes one object
10 Polarity and insulation test — an electrical question, with its own instruments
11 Terminal addressing and laser cleaning — this station, and the terminals are bare
12 CCS and busbar plate fitting — hardware goes on top of the terminals
13 Busbar laser welding — the joint is made

Read that downward and the boundary is obvious. Everything above station 11 is the line finding out what it has; everything below it is the line acting on what it found. This station is the hinge — and the last place where a terminal can be looked at directly rather than through something bolted over it.

The full station list, with what each one is entitled to assume from the one before, is on the module assembly line page. It is not restated here.

Module segment seen along its length, upstream of the welding station — Rendering — stations 09 to 13 in one view

In-line station with an overhead vision head above the carrier

Take this to any supplier

"Which station number does your vision addressing run at, and what is fitted over the terminal at that point?"

Banded module on a carrier arriving at an in-line station

A weld cannot ask whether the terminal under it is the right one. The last station that could ask is this one.

Before

The line is reading the module: what it is, where its terminals are, which way round they sit.

After

The line is writing on it: hardware seated, energy applied, geometry created. None of that asks a question.

Two figures and one architecture, and what each is a criterion for

⚠️ Published here for the first time on this site. The second column deliberately does not repeat the figure — it says what the figure can be held to.

What is published	What it is a criterion for
Addressing positionWhere a terminal was found	≤±0.2 mmThe position the next two stations are entitled to assume. It is a criterion about where a terminal was found on this module — not a promise about where your cell sits inside the stack, which is settled upstream by pressing and banding. → Where that is settled
Motion speedHow fast the head travels the module	200–300 mm/sThat this is a station rather than an inspection. A check that cannot keep up with the line becomes a sample, and a sample answers the question for some of your modules. → section 07
One vision setPosition and polarity, one pass	Both questionsThat both answers come from the same part at the same moment. Split them across two devices and a module can pass one check and miss the other, with nothing in the line that notices. → section 11

±0.2 mm appears twice on this site, describing two different machines. Here it is a terminal position on a banded module. On the cooling-plate gluing station it is photo positioning before a bead is laid. Same number, different machine, different question — worth saying out loud rather than letting you find it.

We do not publish a cycle time for this station, or a count of terminals per module. Both follow from your series configuration and your module drawing, and a figure quoted without them is a figure quoted against somebody else's module. → section 12

The speed is a window, not a setpoint. Where inside it your module runs is settled with your drawing — the same discipline the busbar welding station applies to its own speed range, and for the same reason.

Overhead vision and tooling above an in-line station

Terminal row on a banded module, seen from above

Station standing in line with carriers passing through it

Why this happens before the busbar plate goes on, not after

This is the part of an addressing station that decides whether its accuracy figure means anything, and it is settled entirely by which station number the station carries.

01 · Why this and not the tolerance
Every supplier quotes a positioning figure. Almost none of them says where in the sequence it is taken
A tolerance is a statement about a measurement, and a measurement is only worth what the thing it was taken from is worth. On this line that part is published rather than implied: station 11, after banding and before CCS fitting.
02 · What goes wrong if it moves later
Measure after station 12 and what you are measuring is hardware
Once the busbar plate and the CCS are seated, the terminal has something over it. A vision set looking at that assembly can return a position for what it can see — and what it can see is the part that was just fitted, not the terminal underneath it. The tolerance is unchanged; what it describes is not.
03 · What we do instead
One pass, bare terminals, both questions asked together
The vision set returns a terminal position to ≤±0.2 mm and judges polarity in the same pass, while the stack is banded but still open. Laser cleaning carries the same station number and works on the same exposed terminal — one station, one moment of access.
04 · Why the speed figure is the evidence
200 to 300 mm/s is what makes it a station instead of a sample
A check that has to interrupt the line to happen will eventually be done to fewer modules than the line makes — that is not cynicism, it is what happens to every check that costs takt. The speed range is the published reason this one does not have to be rationed.
05 · If a supplier offers the opposite
Ask for a station number and an object, not a tolerance
"Which station number does your vision run at, and what is on top of the terminal at that moment?" ⚠️ It is one email, the answer is a number and a noun, and it tells you more about the machine than its datasheet line does.

None of this makes the tolerance larger or smaller. It decides what the tolerance is a tolerance about — and that is a property of the sequence rather than of the camera. Which is why the station number is printed beside the figure on this page instead of living in a separate document.

Station enclosure open above a module on its carrier — Rendering — the moment the terminals are still reachable

The shortest version of this page

"An addressing figure taken through the hardware that was fitted over the terminal is a figure about the hardware."

What the welding station is entitled to assume

That it is told where each joint is rather than having to look for it. The busbar station's own criteria, and what it does and does not check, are on the busbar welding station page — not repeated here.

Control cabinet holding the vision and motion programs

Two hundred to three hundred is a window, not a setpoint

Ranges make buyers suspicious, and usually they should. This one is a range for a reason that can be stated in a sentence, so here it is stated.

Automated station moving over a module on the line

What actually moves at that speed

The vision head travelling the module, taking the terminals in order. The figure describes that motion — not the takt of the line, which is set by the slowest station in the segment and is rarely this one.

High-automation module line running at full length

Why a range instead of one number

Where inside it a module runs depends on your terminal layout and how many positions there are per module. A supplier who names a single speed before seeing your drawing has named it for a different module.

Prismatic cells and modules staged beside the line

What the range actually buys you

Room to stay in line. A station that keeps up checks every module; one that does not becomes a sampling plan — and a sampling plan answers the polarity question for some of your modules.

Line layout showing stations in sequence

What it does not tell you

Cycle time for this station. That needs your position count, and we would rather leave it blank than print a figure that had to be guessed. It is the first entry in the list of deliberate blanks. → section 12

Takt for the whole segment is a different subject with a different owner: it is published per configuration on the tier selection page, and nothing on this page should be read as a takt claim.

An accuracy figure is a promise about a datum, not about a part

This is the one argument on this page we did not work out here. It is made once on the conveying page, and it applies to every accuracy figure quoted anywhere on this line.

Where the figure is measured from

Position is returned with respect to the carrier the module is standing on, and to the way that carrier is stopped and lifted at each station. Both of those belong to the conveying and positioning page, which is where the argument is set out.

Why it is repeated on this page at all

Because ±0.2 mm quoted without a datum is a number rather than a criterion, and this page publishes that number for the first time. Borrowing the argument as evidence and linking to where it lives is the whole of what we do with it here.

What it means when you compare quotations

Compare what each figure was measured from before you compare the figures. Two stations quoting the same tolerance against different references are not quoting the same thing, and the difference will not appear until commissioning.

There is a version of this page that prints ±0.2 mm in large type and stops. It would be shorter, and it would be worth less, because the number on its own cannot be checked by anybody reading it.

The same fault, found at three stations, with three different prices

A wrong or unread terminal position does not get expensive gradually. It gets expensive at one specific moment, and the columns below sit on either side of it.

At the moment it is found	Station 11Here — addressing	Station 13At the busbar weld	Station 16Module end-of-line test
Terminals still bare	Yes — nothing has been fitted over them	Plate and CCS are already seated	And welded over
The joint exists	Not yet — two stations away	It is being made now	It was made three stations ago
Who can still act cheaply	This station, by holding the module	Partly, and it depends what has been welded	The module is finished by then
What has been consumed	One cycle	The busbar hardware and the joint	The module, plus a test slot on a gate that had other work to do
Where the argument lives	This page	Busbar welding station	End-of-line testing

We do not restate why a completed weld is a different kind of problem from an unfitted one — the equipment page makes that argument once, and it is the reason this station is worth carrying its own station number instead of being folded into the one after it.

Module passing between two stations on its carrier

Found here, the module is still an assembly of parts.

Found at the weld, part of the answer has already been written in metal.

Test and inspection position at the end of the module segment

Found at test, what is left to decide is what to do with a finished module.

Four things about your module decide what this station does

None of them is a yes-or-no. Each changes the program, the fixture or both — and all four are things you already have on a drawing.

Prismatic cell showing how its terminals present

How the terminals present

Flush, recessed or shrouded. The vision set has to see them, and how they sit is a property of your cell rather than of this station.

Cells staged before entering the module segment

The module envelope

Published once, with the blueprint module, on the energy-storage pack line page. Not repeated here — what matters at this station is only that your module is inside it.

Cells arranged in series inside a module frame

Your series configuration

It sets how many positions the vision set visits on each module, and therefore where inside the speed window this station ends up sitting for your product.

Line bay with stations positioned along the carrier route

What we need from you

The module drawing, with terminal positions and polarity marked. That single file is what turns everything above into a program, and it is the first thing we ask for.

Banded stack Addressed — here CCS fitted Welded

Why a person is the wrong instrument for this particular check

This is not the usual argument about people making mistakes. It is narrower and easier to check: one of the two answers this station produces cannot be handed over by a human being at all.

Polarity is a yes-or-no, and a person can do it

One module at a time, in good light, with the terminals bare, an experienced operator will be right nearly every time. That is genuinely true — and it is the half of the argument most equipment pages leave out, because it does not flatter the machine.

Position is not a yes-or-no. It is a coordinate

The station after next does not need somebody to be satisfied; it needs a number it can move to. A person cannot hand a welding program a coordinate to a fifth of a millimetre, and no amount of care changes that. It is a difference in the kind of output, not in the quality of the work.

And both answers have to come from the same instant

Split them — a person for polarity, a machine for position — and you have opened a gap between two checks on the same module, with nothing in the line that notices when one of them was done on a module the other never saw. One vision set, one pass, both questions is a smaller claim than a fully automatic line, and it is the one that matters here.

Put the three together and the conclusion is narrower than the usual one: this check is not automated because people are unreliable. It is automated because half of its output is a number that has to be handed to another machine.

⚠️ Automation levels on our lines are labelled per station, AUTO / SEMI-AUTO / MANUAL, and the pack segment is mostly hand work. Nothing on this page should be read as a claim about the rest of the line.

Manual working position beside a line — Rendering — where hand work genuinely belongs

The distinction worth keeping

"The question is not whether a person is careful. It is whether the answer can be written down as a number and passed to the next machine."

Semi-automatic position on a module line

Four things you will look for here and will not find

Each one is blank for a different reason, and the reasons are not interchangeable. Two are gaps in what we have been given, one belongs to another page, one should never be guessed.

Not on this page	Why, specifically
Laser cleaning parameters	Station 11 cleans as well as addresses. No figures were released to us for the cleaning half, so none are printed here. Naming one would be inventing a specification, and it would be the easiest thing on this page to invent.
The rework route for a held module	Not published. What happens to a module stopped at this station is a plant procedure agreed with you, and a route printed in advance would be a route written for somebody else's plant and somebody else's people.
Check frequency or sampling plan	Not applicable by design — and worth saying rather than leaving blank. At 200–300 mm/s the station runs on every module that passes it, so there is no sampling interval to publish.
Station 10 test thresholds	Owned by the end-of-line testing page, which holds every test threshold on this line. Station 10 is an electrical test with its own instruments; this one is a vision station, and the two are not the same check twice.

Two of these four are gaps we intend to close and two are things this page should not carry. Saying which is which costs us the appearance of completeness and buys you the ability to check what is left — which is the trade this whole site is built on.

Station being positioned during installation — Rendering — the cleaning half, with no figures released

Offload position beside a module line — Rendering — where a held module would wait

Module line seen along its length — Rendering — every module, no sampling interval

Four questions to put to any supplier quoting a vision addressing station

Including us. Each one has a short factual answer, and a supplier who needs a paragraph to reach it is telling you something on its own.

At which station number does your vision run, and what is on top of the terminal then?

A number and an object. If the answer is "during welding", the position is being taken through hardware. Ours is station 11, and the answer to the second half is nothing — the terminals are bare.

Does the same device judge polarity, or is that a separate check?

If it is separate, ask what happens to a module that passes one and misses the other. Ours is one vision set answering both questions in the same pass on the same part.

What is the positioning figure measured from?

A tolerance without a reference is a number. Ours is returned against the carrier the module stands on, and the carrier is described on the conveying and positioning page.

Can it run at line speed, or does the line wait for it?

A check that stops the line becomes a sample eventually. Ours is published as a speed range — 200 to 300 mm/s — and deliberately not as a promise about your takt, which the slowest station sets.

All four are answerable in a sentence by anyone who has actually built the station. They are also the four questions a datasheet is least likely to answer, because none of them is a specification line — every one of them is about where the specification was taken.

Operator panel and cabinet beside an in-line station

Why these four and not a longer list

A longer list of questions is easier to write and harder to use. These four each have a wrong answer that is recognisable without any battery knowledge — which is the only property that makes a question useful to somebody comparing three quotations on a Friday afternoon.

Semi-automatic module line, stations in sequence

Each subject on this line is stated once, and here is where

This station borrows four arguments from four other pages and repeats none of them. The links below go to the page that owns each one.

The reason this page is short on adjectives is that everything it could have borrowed is one click away instead. A station page that restates the four pages around it looks thorough and gives a buyer nothing new — and it creates four places where one figure can quietly drift into two.

Module and pack line seen along its full length — Rendering — one line, nine machines, one figure each

Standard-tier module and pack line in its bay

⚠️ If you find the same figure stated twice anywhere on this site with two different meanings, that is a defect and we would like to know. ±0.2 mm appearing on two machine pages is the one case we already know about — see section 05.

Station enclosure open for access during commissioning

Send the module drawing and we will mark the addressing positions

Terminal layout, series configuration, and how the terminals present. Three inputs, and what comes back is the position count for your module and where inside the speed window it lands.

What comes back

The number of positions the vision set visits on your module, where that puts you inside the speed range, and this station's line item written the way it would appear in a quotation.

If your drawing is not final

Send it anyway and say which parts are still moving. Terminal layout is one of the few inputs where a draft is genuinely useful, because the answer changes shape rather than value.

Go to the enquiry form WhatsApp

[email protected]

Phone / WhatsApp

+86 15327155363

Address

Wuhan, Hubei, China

Response

Reply within 24 hours