Camera Link Frame Grabber Configuration Guide: Matching Image Width, Height, Pixel Format, Tap Arrangement and Camera Output

A Camera Link frame grabber does not automatically understand every image stream simply because the industrial camera is connected through a compatible cable. The acquisition hardware must be configured to receive the exact image structure the camera is transmitting. Image width, image height, pixel format, bit depth, tap count, tap arrangement, Camera Link configuration, timing behavior and camera operating mode all need to agree. If even one important parameter is mismatched, the system may show no image, an incomplete frame, incorrect brightness, shifted lines, interleaved regions or an apparently corrupted picture despite a physically correct Camera Link connection.

This makes Camera Link frame grabber configuration an important part of designing and commissioning a high-speed machine vision system. Engineers searching for Camera Link frame grabber setup, Camera Link camera configuration, Camera Link image acquisition, frame grabber pixel format, Camera Link tap configuration, Camera Link image width and height, Camera Link camera output settings, MDR-26 Camera Link cable or SDR-26 Camera Link cable should treat acquisition setup as a matching exercise between camera output and frame-grabber interpretation.

Kyptec Automation® supports the physical connection between compatible cameras and image-acquisition hardware through its dedicated Camera Link Camera Cable collection. The current range includes MDR-26-to-MDR-26, SDR-26-to-MDR-26 and SDR-26-to-SDR-26 cable configurations. Once the correct physical connection is established, the frame grabber must still be configured to understand exactly what the camera is sending.

What Does a Camera Link Frame Grabber Actually Do?

A frame grabber receives the high-speed image stream generated by the industrial camera and converts that stream into image data that can be stored in host memory and used by machine-vision software.

Its job is not merely to provide a connector.

The frame grabber must recognize the timing and data structure presented by the camera, capture the correct number of pixels and lines, interpret the selected pixel format, reconstruct parallel tap outputs and place the resulting image into memory in the intended structure.

This makes the frame grabber an active part of the acquisition architecture rather than a passive adapter between the cable and the computer.

Camera Output Configuration Must Be Defined First

Frame-grabber configuration should normally follow the actual camera output mode.

Before changing acquisition settings, confirm how the camera is configured to operate.

Important parameters include the active image width, active image height, transmitted bit depth, output format, tap count, tap arrangement, pixel clock, trigger mode and supported Camera Link configuration.

The frame grabber should then be configured to expect that same output structure.

Trying to configure the acquisition board without first defining the camera output often leads to trial-and-error settings and difficult troubleshooting.

Image Width Must Match the Camera's Active Output

Image width tells the acquisition system how many horizontal pixels belong to each valid image line.

If the camera outputs 2,048 active pixels per line but the frame grabber is configured for another width, the acquisition software can divide the incoming stream incorrectly.

Pixels may wrap into the next line too early or too late.

The resulting image may show diagonal displacement, repeated sections, shifted edges or apparently torn geometry.

For this reason, the frame-grabber width should match the camera's actual transmitted width rather than simply the maximum sensor width.

Sensor Width and Output Width May Be Different

An industrial camera can have a sensor with one native resolution while transmitting a smaller active region.

A region-of-interest setting, cropping mode or camera-specific output configuration can reduce the transmitted image width.

The frame grabber should follow the transmitted output, not blindly use the full sensor specification.

This distinction becomes especially important when an OEM changes the region of interest to increase acquisition speed or inspect only a selected portion of the scene.

Image Height Must Also Match the Active Frame

Image height defines how many valid image lines are expected in a complete area-scan frame.

If the frame grabber expects more lines than the camera transmits, acquisition may wait for data that never arrives or handle the buffer incorrectly.

If it expects fewer lines, the resulting image can be cropped or incomplete.

The correct value should therefore come from the camera's selected active output mode.

For line-scan cameras, the acquisition concept is different because continuous lines may be assembled into application-defined image blocks rather than camera-generated area-scan frames. The frame-grabber configuration should reflect that operating architecture.

Pixel Format Must Match the Transmitted Camera Data

Pixel format defines how the incoming digital values should be interpreted.

A camera can transmit monochrome or other supported image representations at different bit depths depending on its design and selected operating mode.

The frame grabber must understand the format the camera is actually sending.

If the expected pixel format differs from the transmitted format, the acquisition system may place the data into memory incorrectly.

The image may still appear, but pixel values can be wrong or the structure may become unusable.

Stored Pixel Format and Transmitted Pixel Format Are Not Always Identical Concepts

A camera can transmit a particular number of meaningful bits per pixel, while the host software stores those values inside a larger memory container.

These are related but separate issues.

Engineers should distinguish between the format carried by the Camera Link connection and the memory representation used by the host system.

The frame grabber is responsible for receiving the transmitted structure correctly before any later software representation is applied.

Bit Depth Should Be Confirmed at Both Ends

If a camera is changed from one bit-depth mode to another, the acquisition configuration may also need to change.

For example, a machine validated using one output precision should not automatically be assumed to work correctly after the camera is switched to a wider transmitted format.

The amount and arrangement of data can change.

This is why production camera settings should be controlled rather than casually modified after commissioning.

Tap Count Is a Frame-Grabber Configuration Parameter

High-speed cameras can divide image output across multiple parallel taps.

The frame grabber must know how many taps are active.

A camera producing data through four logical taps cannot necessarily be interpreted correctly by an acquisition configuration expecting two.

All data paths may be physically present, yet the reconstructed image can be incorrect because the acquisition hardware combines them according to the wrong rule.

Tap count should therefore be documented alongside width, height and pixel format.

Tap Arrangement Matters as Much as Tap Count

Knowing that a camera uses multiple taps is not enough.

The acquisition system also needs to know how those taps correspond to positions within the image.

Different camera output architectures can divide pixel information differently.

Some may distribute alternating pixel positions. Others may divide larger sensor regions or use another supported mapping.

The exact interpretation should come from the camera's defined output mode.

The frame grabber must use the corresponding reconstruction arrangement.

Wrong Tap Arrangement Can Look Like Cable Failure

A tap mismatch can create images with repeated bands, alternating sections, reversed groups, interleaved pixels or displaced image regions.

These symptoms can appear dramatic enough that an engineer may immediately suspect the Camera Link cable.

However, if the pattern is stable and repeats consistently on every frame, configuration should be checked before replacing hardware.

A physical data-path fault is more likely to produce intermittent or changing symptoms, although final diagnosis should always evaluate the complete system.

Camera Link Base, Medium or Full Must Match the Camera Output Architecture

The camera and frame grabber must support the same Camera Link configuration used in the selected operating mode.

A camera configured for a wider output architecture cannot be expected to operate correctly through an acquisition path configured only for a narrower one.

Base conventionally uses one physical cable connection, while Medium and Full conventionally use two.

The frame grabber therefore needs both the correct logical configuration and the correct physical connections.

Cable count alone does not create the configuration.

Physical Cable Connections Must Be Mapped Correctly

Where two Camera Link cables are required, the system documentation should clearly identify which camera port connects to which frame-grabber port.

Two physically identical cable assemblies should not be considered arbitrary interchangeable links if the system architecture assigns different signal groups to the two connections.

OEM machine builders should label both ends of each cable and retain the mapping in the electrical documentation.

This prevents commissioning errors and future service mistakes.

Pixel Clock Compatibility Must Be Confirmed

The frame grabber must be capable of receiving the camera's operating pixel clock within its supported acquisition architecture.

A valid cable cannot make an acquisition device support a clocking mode outside its capability.

If the camera is switched into a faster mode, engineers should confirm that the frame grabber configuration supports the corresponding output conditions.

The most reliable validation is performed at the actual production operating rate rather than at a reduced laboratory setting.

Timing Signals and Frame-Grabber Configuration Work Together

Camera Link carries timing information that helps the acquisition hardware determine when image data is valid.

The frame grabber uses those timing relationships together with the configured image structure to build the intended image.

Timing signals alone do not replace width, height, pixel-format or tap settings.

Likewise, correct software dimensions cannot compensate for an acquisition path that is not receiving the required timing or data correctly.

Both levels need to agree.

Trigger Mode Can Change How the Frame Grabber Waits for Images

A free-running camera continually generates images according to its configured operating mode.

A triggered camera produces an image only after the required trigger event and camera sequence occur.

Frame-grabber acquisition should therefore be configured around the actual operating behavior.

If the software expects continuous frames while the camera is waiting for an external trigger, the system may appear to have stopped even though the physical connection is healthy.

Similarly, a trigger can reach the camera successfully while the acquisition side remains incorrectly configured for the resulting image stream.

Acquisition Timeout Is Often a Configuration Symptom

A frame-grabber timeout means the expected acquisition event did not complete within the configured interval.

That can happen for many reasons.

The camera may not have been triggered. It may not be generating the expected output. The frame grabber may be configured for the wrong dimensions or acquisition mode. The Camera Link configuration may not match. The physical connection may also be faulty.

Timeout should therefore be treated as a system-level symptom rather than immediate proof that the cable has failed.

Camera Configuration Changes Should Be Reflected in Frame-Grabber Setup

Changing one camera parameter can affect several acquisition assumptions.

Adjusting region of interest changes transmitted image dimensions.

Changing bit depth changes pixel representation.

Changing tap mode changes reconstruction requirements.

Changing the camera's output configuration can change the number of required physical Camera Link connections.

After any significant camera-mode change, the frame-grabber configuration should be reviewed systematically.

Camera Replacement Requires More Than Matching the Connector

Replacing a damaged or obsolete camera with another unit that has the same physical Camera Link connector does not guarantee identical acquisition behavior.

The replacement camera may use different default output dimensions, tap architecture, bit depth, clocking behavior or supported configuration.

The frame-grabber setup should therefore be verified before the replacement is approved.

Likewise, the existing cable should be checked against the actual endpoint format rather than assuming compatibility from the interface name alone.

MDR-26-to-MDR-26 Frame-Grabber Connections

Where both compatible endpoints require MDR-26, Kyptec Automation® provides the Kyptec Automation® Industrial Camera link Camera Cable: MDR-26 Pin Male to MDR-26-Pin Male Cable.

The product provides the physical camera-to-image-acquisition connection for compatible MDR-26 endpoints and is available in standard 2 metre, 3 metre and 5 metre lengths.

Once the physical connection is established, engineers must still configure the frame grabber according to the camera's actual output dimensions, pixel format, tap arrangement and operating mode.

SDR-26-to-MDR-26 Frame-Grabber Connections

Some industrial cameras use SDR-26 while the compatible frame-grabber port uses MDR-26.

For that physical arrangement, Kyptec Automation® provides the Kyptec Automation® Industrial Camera link Camera Cable: SDR-26 Pin Male to MDR-26-Pin Male Cable.

This cable provides the required physical endpoint combination.

It does not convert pixel format, tap arrangement, bit depth or Camera Link configuration.

The frame grabber still needs to be configured to interpret the camera output correctly.

SDR-26-to-SDR-26 Frame-Grabber Connections

Where both compatible camera and acquisition endpoints use SDR-26, Kyptec Automation® offers the Kyptec Automation® Industrial Camera link Camera Cable: SDR-26P Male To SDR-26P Male Type.

As with the other two connector combinations, physical connector matching and logical acquisition setup are separate engineering tasks.

Correct mechanical fit provides the required connection; correct frame-grabber configuration produces the intended image.

A Practical Frame-Grabber Configuration Sequence

A disciplined commissioning sequence can reduce unnecessary troubleshooting.

Begin by recording the camera's selected output mode. Confirm active width and height, transmitted pixel format and bit depth, tap count and mapping, Camera Link configuration, pixel clock and trigger behavior. Confirm that the frame grabber supports that output architecture. Connect the required Kyptec Automation® Camera Link Camera Cable configuration between the defined endpoints. Configure the acquisition hardware using the camera's actual parameters. Start at a controlled operating mode, verify the first correctly reconstructed image, then validate the system at full production acquisition rate.

This method isolates configuration from physical connectivity more effectively than changing several parameters at once.

The First Correct Image Is Not the End of Commissioning

A displayed image proves that many parts of the acquisition chain are functioning, but it does not automatically prove production readiness.

The OEM should verify that the complete frame is correct, pixel values are interpreted properly, tap reconstruction remains stable, triggered acquisitions complete reliably and the system maintains continuous acquisition at the intended speed.

If the machine uses multiple cameras, all channels should be tested together under realistic production load.

The final approved settings should then be frozen in the machine configuration.

Frame-Grabber Settings Should Be Part of the Controlled Machine Record

OEM documentation should preserve the camera output mode and matching acquisition configuration.

Useful records include active width, active height, pixel format, bit depth, tap count, tap arrangement, Camera Link configuration, cable mapping, trigger mode and approved software settings.

This makes future troubleshooting much faster.

If an image changes after software service, camera replacement or configuration restoration, engineers can compare the current values against a known working reference.

Cable Selection Should Follow the Defined Acquisition Architecture

Once camera and frame-grabber compatibility has been established, cable selection becomes much more straightforward.

The engineer can identify the connector at each endpoint, required number of cable connections and actual installed length.

Kyptec Automation® provides the three major endpoint combinations within its Camera Link Camera Cable category, allowing OEM buyers to match the physical connection after the acquisition architecture is defined.

For prototype, production and repeat OEM requirements, the Kyptec Automation® OEM Orders page can be used to discuss defined quantities and requirements, while the Contact Us page provides a direct route for technical and purchasing enquiries.

Frequently Asked Questions About Camera Link Frame-Grabber Configuration

1. Why does my Camera Link frame grabber detect the camera but show the wrong image size?

Detection or communication does not automatically configure the image dimensions. The camera may be outputting a region of interest or another active resolution different from the value expected by the frame grabber. Confirm the camera's actual transmitted width and height and make the acquisition configuration match those values before investigating the cable.

2. Should frame-grabber image width match the sensor's maximum width?

Not necessarily. It should match the width the camera is currently transmitting. A sensor can support a larger native resolution while the camera outputs a cropped region, reduced active area or another configured image size. Using the maximum sensor width when the active stream is smaller can produce incorrect image reconstruction.

3. Why do image lines wrap incorrectly on a Camera Link frame grabber?

Incorrect line width is a common cause. If the frame grabber groups more or fewer incoming pixels into each line than the camera actually transmits, the next line begins at the wrong memory position. This can create diagonal displacement, wrapping or repeated-looking patterns even when the physical Camera Link connection is stable.

4. Can changing camera bit depth require a new frame-grabber configuration?

Yes. Changing transmitted bit depth can alter how pixel information is represented and may require a corresponding acquisition-format change. A configuration validated for one camera output format should not automatically be reused after changing bit depth without confirming how the new data is transmitted.

5. What should I check if the Camera Link image has correct dimensions but wrong brightness values?

Verify pixel-format and bit-depth interpretation. The physical width and height can be correct while the incoming pixel values are being interpreted in the wrong format. Confirm the camera's selected output representation and make sure the frame grabber and host software store those values correctly.

6. Why can the same frame grabber work with one Camera Link camera but not another?

Two Camera Link cameras can use different output resolutions, bit depths, tap structures, pixel clocks and Camera Link configurations even when their physical connectors appear identical. The acquisition profile that works for one camera may therefore be unsuitable for another. Each camera output architecture needs to be verified.

7. Does a multi-tap Camera Link camera require a special frame-grabber setting?

Yes. The frame grabber needs to know how many taps are active and how their data corresponds to the image. Incorrect tap count or mapping can cause interleaved, repeated or spatially rearranged image regions even when all electrical connections are intact.

8. Why does my image become incorrect after changing the camera's region of interest?

Changing the region of interest can change the number of transmitted pixels per line and sometimes the image height or operating rate. If the frame-grabber configuration still expects the previous dimensions, it can reconstruct the stream incorrectly. Camera and acquisition settings should be updated as a matched pair.

9. Can the wrong Base, Medium or Full setting prevent Camera Link acquisition?

Yes. The camera and frame grabber must use compatible Camera Link configurations. A wider camera output architecture cannot be interpreted correctly by a frame-grabber setup expecting a narrower configuration. Where Medium or Full uses two physical connections, both required cables and the correct port mapping must also be present.

10. Why does a Camera Link acquisition work in free-run mode but fail when triggering is enabled?

Free-run and triggered operation follow different event sequences. In triggered mode, the camera may wait for a valid trigger before producing an image, while the frame grabber may need to be armed and waiting for the corresponding acquisition. Verify trigger routing, camera trigger mode and frame-grabber acquisition behavior before assuming the Camera Link cable is faulty.

11. What does it mean if the frame grabber reports a buffer size or image-size mismatch?

It usually means the acquisition software expects a different amount of image data than the camera is providing. Check active width, height, pixel representation and any padding or host-memory settings used by the acquisition software. Buffer allocation should be based on the correctly reconstructed image format.

12. Should I copy frame-grabber settings from another machine using the same camera model?

Only if the complete operating configuration is also identical. Two machines can use the same camera model but different region-of-interest settings, bit depths, trigger modes, tap configurations or Camera Link output modes. A controlled machine configuration is safer than assuming that camera model alone guarantees identical settings.

13. Can replacing a Camera Link cable change frame-grabber configuration?

Replacing a cable with the same verified endpoint configuration and installation specification should not change logical acquisition settings. If an image-format problem appears immediately after replacement, verify that the correct cable configuration and port mapping were installed before changing frame-grabber parameters. Kyptec Automation® provides clearly differentiated MDR-26-to-MDR-26, SDR-26-to-MDR-26 and SDR-26-to-SDR-26 products to help avoid endpoint ambiguity.

14. What should an OEM save as a known-good Camera Link acquisition configuration?

Save the camera output mode, active width and height, transmitted pixel format, bit depth, tap count and mapping, Camera Link configuration, trigger mode, relevant timing settings, frame-grabber acquisition profile and physical cable mapping. This creates a controlled baseline that can be restored during servicing or replacement work.

15. Where can machine builders source Camera Link Camera Cables after the frame-grabber architecture is finalized?

Once the camera output, frame-grabber compatibility, connector arrangement and required cable count are defined, OEMs can review the Kyptec Automation® Camera Link Camera Cable collection. The current portfolio includes MDR-26-to-MDR-26, SDR-26-to-MDR-26 and SDR-26-to-SDR-26 physical configurations with standard 2 metre, 3 metre and 5 metre length options, giving machine builders a focused source for the physical camera-to-acquisition connection.

Conclusion

Reliable Camera Link acquisition depends on more than connecting a compatible camera to a frame grabber.

The acquisition hardware must be configured to expect exactly what the camera is transmitting.

Image width determines how incoming pixels form each line. Image height defines the expected frame structure. Pixel format and bit depth determine how digital pixel values are interpreted. Tap count and tap arrangement tell the frame grabber how parallel image streams should be recombined. Camera Link Base, Medium or Full configuration establishes the required data architecture and physical cable count. Pixel clock and trigger behavior define additional operating conditions that the acquisition system must support.

A mismatch in any of these areas can produce no image, incomplete frames, incorrect pixel values, shifted geometry or apparently corrupted data even when the Camera Link cable is electrically healthy.

The strongest engineering approach is therefore to define the camera output first, configure the frame grabber to match it, connect the required physical endpoints correctly and then validate the complete system at full production operating conditions.

Kyptec Automation® supports that physical connection through its dedicated Camera Link Camera Cable portfolio, including MDR-26-to-MDR-26, SDR-26-to-MDR-26 and SDR-26-to-SDR-26 configurations for compatible industrial cameras and frame-grabber hardware. When a properly specified cable is combined with accurately matched acquisition settings, OEM machine builders gain a far more controlled and repeatable Camera Link image-acquisition architecture.