Zoom Lens for EV Battery Weld Inspection: FOV, Magnification and Working Distance

Battery pack welding is inspected twice on the same part: the whole seam has to be located and checked for position, spatter and missing weld, and then individual beads have to be examined for porosity, cracks and undercut. Those two jobs want different fields of view, and a battery module is usually already clamped in a fixture when the check happens. That combination - two magnifications, one stand-off, no room to move the camera - is what makes a motorized zoom lens the practical choice on a weld inspection station, and it is also why the selection has to start from the seam rather than from a lens catalogue.

Why weld inspection on a battery pack resists a fixed-focal lens

A fixed-focal lens forces a compromise: set the field of view wide enough to catch the full seam and every bead occupies only a few dozen pixels, which is not enough to see porosity; set it narrow enough to resolve a bead and the station has to move the camera to dozens of positions to cover one module. Moving the camera is the expensive part - it costs cycle time, it adds an axis that has to be re-homed, and on a laser weld cell it also means the optics travel into an area where spatter and fume are present. A zoom lens keeps the camera fixed and changes magnification instead, so the same optical axis covers the overview pass and the detail pass. The trade-off is that magnification is now a variable the recipe has to control, which is a software problem rather than a mechanical one, and the rest of this page is about getting the optics right so that the software has something repeatable to control.

Start from the seam, not from the lens

The first number to write down is the width of the region that has to be visible at once - for a tab-to-busbar weld that is the footprint of the weld plus the heat-affected zone, and for a module-level pass it is the pitch between two welds. Call that width W. The field of view the lens must deliver is FOV = W × (1 + margin), where the margin covers fixture tolerance and part-to-part position variation. Once FOV is known, the required magnification follows from the sensor: magnification = sensor width / FOV. On a 2/3-inch sensor (8.8 × 6.6 mm) the arithmetic is short enough to do on paper, and the table below is that arithmetic for the two zoom ranges POMEAS builds for this kind of station. Every value in it is derived from the sensor size and the magnification, not a separate datasheet figure.

Two readings matter here. If the overview pass needs more than roughly 12 mm of width on a 2/3-inch sensor, the 6.5X series runs out of range at its low end and the 12.5X series is the one to look at, because it starts at 0.58X and delivers 15.17 × 11.38 mm there. If the detail pass needs to resolve features at the 10 µm scale, neither series should be chosen from a table alone - the resolution at the exact magnification you will run has to be read off the datasheet for that model, because resolution changes across the zoom range and does not follow a single rule of thumb.

Working distance decides whether the lens fits the cell

On a weld station the lens is not alone: the laser head, the fume extraction and the clamping all compete for the space above the seam. Working distance is therefore a hard constraint rather than a preference. The 6.5X motorized body (PMS-LZ-650104) is specified at 82 ± 2 mm working distance as a bare lens, and the 12.5X family (PMS-LZL-12100 and its variants) at 77.4 ± 2 mm. Both accept a 2/3-inch sensor and use a C-mount, so the camera side is not what limits the design. Front attachments and video tubes change the working distance and the magnification together - the 6.5X body is commonly paired with a 1X video tube and a 0.5X front attachment when a wider view or a longer stand-off is needed - so the figure to design around is the one for the exact stack you will build, verified against the datasheet of each element rather than the bare lens alone.

Depth of field and the shape of a weld bead

A bead is not a flat target, and the check for undercut or a convex cap is a check on height as much as on width. Depth of field therefore has to be compared with the bead height, not with the flatness of the surrounding plate. In the 12.5X family the datasheet gives 2.75 mm of depth of field at 0.58X with a numerical aperture of 0.025, falling as magnification rises; the 6.5X body is specified from 1.9 mm at 0.7X down to 0.1 mm at 4.5X. At the low-magnification end there is ample depth for a full module pass, and at the high-magnification end the depth is smaller than a tall bead, which is exactly when focus has to be set on the feature being judged - the bead crown for porosity, the toe of the weld for undercut - rather than on the plate beside it. This is also the reason a weld recipe should record its focus position together with its magnification; changing one without the other is the most common cause of a station that passes its first article and fails an hour later.

Reflective stainless and aluminium change the lighting, not the lens

Battery tab and busbar materials are usually nickel-plated steel, aluminium or stainless, and a weld bead on any of them is a curved mirror. Ring light alone tends to produce a bright band along the crown and a dark toe, which reads as a false undercut. Two arrangements work: a diffuse dome for the overview pass, where the goal is position and spatter rather than surface detail, and coaxial illumination for the detail pass, where the light has to come back along the optical axis to keep the toe and the crown evenly lit. POMEAS builds the 6.5X body in both a plain motorized version and a coaxial version (PMS-LZ-650105), so the choice can be made per station rather than by adding an external coaxial unit in front of the lens, which would change the working distance again.

Putting magnification into the recipe

Once the optics are fixed, magnification has to become a number the PLC or the vision software can command, otherwise the two-pass idea collapses back into an operator turning a knob. The motorized zoom bodies use a closed-loop DC servo driven over a 9-pin RS-232 interface, single or dual axis, which means the magnification can be commanded as part of a recipe and read back afterwards. In practice a weld inspection recipe ends up with three fields that belong together: the magnification for the overview pass, the magnification for the detail pass, and the focus position for each. Reading magnification back matters more than commanding it, because the check that a station stayed in calibration is a comparison of commanded and reported magnification after a few hundred cycles - the method for that is the same repetition test used for any motorized zoom axis.

Integration checklist

• Seam width and module pitch measured on a real part, with fixture tolerance added, before any lens is quoted.
• Working distance for the complete stack - body, tube and front attachment - checked against the cell layout, not just the bare lens figure.
• Depth of field at the detail magnification compared with bead height, and the focus reference defined on the feature being judged.
• Lighting chosen per pass: diffuse for position and spatter, coaxial for crown and toe.
• Magnification and focus stored in the recipe and read back on a schedule, so drift shows up as data rather than as a rejected batch.

What this page does not cover

Weld appearance is one of two optical jobs on a battery pack. Dimensional work on the electrode and the tab - coating width, tab position, burr - is a measurement task with a different error budget, and it is covered separately in the guide on telecentric lenses for battery electrode and tab measurement. If a single station has to do both, the usual arrangement is a telecentric measurement head for dimensions and a motorized zoom head for weld appearance, sharing a motion platform rather than sharing one lens. For the wider question of which zoom body fits an application, the manual versus motorized comparison and the zoom lens catalogue are the next reads; the two bodies named on this page are the PMS-LZ-650104 6.5X motorized zoom lens and the 12.5X motorized zoom lens.

Frequently asked questions

What field of view do I need for a battery weld seam?

Take the width of the weld plus its heat-affected zone, add the fixture and part-position tolerance as a margin, and that is the field of view for the overview pass. Divide the sensor width by that figure to get the magnification it implies: a 15 mm requirement on a 2/3-inch sensor works out at about 0.59X, which is at the bottom of the 12.5X range and below the bottom of the 6.5X range.

Can one zoom lens cover both seam position and porosity checks?

Yes, provided the two passes are treated as two recipes rather than one compromise setting. The overview pass runs at low magnification where the field covers the seam and the depth of field is large, and the detail pass runs high enough to resolve the features that define porosity, with its own focus position. What has to be verified is that the resolution at the detail magnification is small enough for the defect you must catch.

Should weld inspection use a telecentric lens instead of a zoom lens?

Only if the question being asked is dimensional. A telecentric lens holds magnification constant with object distance, which is what makes it right for measuring tab position or coating width on a part that is not flat. Weld appearance - spatter, porosity, undercut, bead shape - is judged by eye and by texture, and there a zoom lens with two selectable magnifications covers far more of the part per image.

How do I switch magnification between weld inspection recipes?

The motorized bodies take magnification commands over a 9-pin RS-232 interface driven by a closed-loop DC servo, so the recipe can carry the overview magnification, the detail magnification and the focus position together, and the reported magnification can be read back after the move to confirm the axis arrived where it was sent.

Next step

Send the seam width, the defect size that must be caught, the stand-off your cell allows and the sensor you intend to use through the contact page, and the two-pass magnification pair can be worked out against the datasheet of the specific body rather than against a general table. If the station also has to measure dimensions, say so - that changes the head count, not just the lens.