• Font Size:
  • A
  • A
  • A

Feature Articles

Web Scanners Revisited

by Nello Zuech, Metropolis Informatics S.A - AIA

 

The underlying technology used in web scanners has improved so now comprehensive high-speed surface inspection of products manufactured in continuous sheets meets the expectations of even the most demanding applications and customers. In this article we are referring specifically to the inspection of unpatterned sheets as opposed to patterned sheet products that are also manufactured in continuous sheets or webs, such as printed materials. Interestingly, some take issue with the term – ‘‘web scanner.’‘ They suggest that this term reflects a system that only looks at a sample of the continuous sheet as it passes while the scanner head physically moves across the web perpendicular to the direction of travel. Often the actual process data being monitored are moisture or beta rays indicative of thickness.

Machine vision-based web scanners, on the other hand, physically inspect 100% of the surface of the sheet looking for physical, geometric or reflectance defects, both random and repeating. Today’s systems not only detect the defects but also analyze them for classification purposes offering the possibility to correlate specific defects to specific process upset and signal operators to take corrective action quickly to reduce or avoid scrap. Most systems come with standard features including scrolling web map, realtime display of defect images and statistics, alarm outputs, archiving of defects and roll data and reporting functions. 

In North America there are about 20 companies competing in the machine vision-based web scanner markets. These markets include primary metals, paper, plastic, film, coated products, glass, electronics and non-wovens. Not all companies compete across all these markets. Some have focused on one or two of these markets.

Representatives of all the companies known to compete in the North American market were asked to contribute to this article. Regrettably, only four companies took the time to answer our questions.

  • Erhardt and Leimer – Gunter Dorn, Herbert Oberstaller
  • ISRA VISION SYSTEMS – Joe Paumi, Peter Ackroyd, Johannes Giet
  • Lasor/Systronics (recently acquired by ISRA VISION SYSTEMS, now ISRA SURFACE VISION) – Werner Goeckel
  • Wintriss Engineering – Vic Wintriss

1. Are there some industries that have embraced web scanners more than others? Why is that?
[Vic Wintriss]  Global economics plays an important part in this respect.  Industry specific interest waxes and wanes depending upon global demand and competition. The U. S. paper market, for instance, is presently in the doldrums.  Textiles are difficult, and this area of the market has probably been slow to embrace web scanning systems.  On the other side of the coin, there are some industries that could use web scanning technology for quality control, such as the baking industry, that have not yet discovered the applicability of web scanning technology to its products.

[Gunter Dorn, Herbert Oberstaller] All industries with high quality requirements are going to use more and more web surface inspection systems. Food packaging, pharmaceuticals, hygienic products industry, and all kinds of coating processes are taking advantage of this technology.

[Werner Goeckel] Some industries such as paper, steel and nonwovens have been more receptive to web inspection either because the end users demand inspection or they want to find the defects early in the process before significant value is added. In case of nonwovens there are functional reasons for inspection because such things as thin spots and/or holes create problems for the end users. In the case of paper and steel the process managers want to find the defect and analyze for cause to eliminate.

[Joe Paumi] High volume producers with high machine costs and high value added products with high liability have always lead the adoption of vision systems on web products. Examples include coated papers, non-wovens, high-end plastic films such as PET, PVB, Polyamide; coated metal like lithoplate and sensitized films like X-ray film. Fortune 500 companies were the typical early adopters as they had the resources to integrate the systems into the processes. The ROI was there and large!

[Peter Ackroyd] High cost materials and health and safety issues demand 100% automated inspection.

[Johannes Giet] Manufacturers want to produce perfect products. Meanwhile Automatic inspection is a standard practice in most industries. During the last years the development of sensor technology used for application area has advanced from laser scanners and phototransistor detectors to CCD and today CMOS cameras. In the paper, pharmaceutical and hygiene markets web scanners or surface vision systems are a must. Besides these markets, in several countries like Japan zero defect levels are so important, that they do not accept anything else.


2. What are critical web scanner system performance criteria?

[Gunter Dorn, Herbert Oberstaller] Equal sensitivity over the whole detection range, easy handling, interfacing, long term spare part supply, System provider including mechanical and software integration. Worldwide partner.

[Werner Goeckel] Web scanners need to not only find the defect, size the defect, identify location on the web but also be able to image the defect. The system software needs to be operator friendly and the stored information easily retrievable. More customers are asking for marking systems so that master rolls can be sent to work stations to slit out defects and salvage the roll. This increases the yield of first quality production for the processor.

[Joe Paumi]  Whether we are discussing camera-based or laser-based systems, the critical factors are resolution, line speed, and processing power. All three variables are continuously being pushed to their limit. Smart inspection systems have made great progress in approximating the human eye/brain combination, but they have decades of development yet to go. Typical systems measure 50 to 100 defect features in order to help ID the defect correctly. Most often there is no question about detecting a defect, the development now is in proper identification.

[Peter Ackroyd] Speed, resolution and defect analysis.

[Johannes Giet] 

  • Inspection Data Input (camera data rate)
  • Process Limits: Web Speed vs. Line frequency
  • Resolution and defect repeatability
  • Classification performance/false rate
  • Maximum defect image storage rate in defects/seconds
  • Detected defect displayed in real – time (within 0.01seconds)
  • Easy to use GUI (Graphical User Interface)
  • Comprehensive statistical data retrieval functions

The end-user is looking for low-priced systems that are easy to install and easy to use.

[Vic Wintriss]   Web speed, web width, defect size and contrast ratio.


3. What changes have been taking place in the technologies that are the basis of web scanner systems that has resulted in improved performance?

[Werner Goeckel] The major changes are increased speeds of the cameras and PC's. This allows for faster line speeds and ability to handle more data. The pixel arrays of the cameras have also increased which results in fewer cameras being needed. This depends to a large degree on the scan rate required for the particular application.

[Joe Paumi] Larger camera linear arrays that operate at higher speeds, coupled with ever increasing PC power, yield higher performance and keep the system prices stable. Prices today are the same dollar value as 20 years ago, but the processing has increased 100 fold. The cost per pixel is going down as well as the cost per processing operation.

[Peter Ackroyd] Camera technology and data throughput.

[Johannes Giet] Web inspections have become more prevalent, even on smaller machines and see more because the systems are more accurate and more sensitive. In these systems cameras check the web and not only document defects but analyze them as well. The defects can be classified in real time and additional information such as uniformity, strength properties, and formation can be stored in a standard database for long-term archiving. A decisive point is the data provided by web inspection systems are no longer used for quality analysis purposes but more and more to optimize the process.

Using intelligent modules or cameras had the result that the processing power of the inspection system is distributed where it is most needed. The basic decisions concerning the defect detection and classification are made in these modules/cameras. The defect results are sent to a standard PC, which combines the information to cover the whole width of the web. This PC also hosts an operator interface, reporting and net communication tasks  - and all this at affordable prices.

[Vic Wintriss] New lighting techniques, smart cameras. 

[Gunter Dorn, Herbert Oberstaller] Faster and more light sensitive cameras. Simpler and more reliable software architecture with LINUX as operating system, networking and access via Internet.


4. Where do you see breakthroughs coming in the technologies that are the basis of web scanner systems that will result in further improvements in the near future – next three years?

[Joe Paumi] Wider and faster camera arrays and the introduction of color processing and continuous PC processing speed improvements.

[Peter Ackroyd] Increasing amounts of data from the material and the ability to throughput the analysis to the user interface.

[Johannes Giet] Global demands for quality control will mean that automated inline color inspection and analysis will become more and more essential. The smart camera technology is further revolutionizing the surface vision market. On the other hand we will benefit from advances in the LED lighting from color and brightness perspective and from IT-developments like Gibit-Ethernet, powerful and standardized network connections and for the camera consumer market. Technology advances (CMOS, LED, SMART-Cameras, IT-Advances) continue to drive down the cost and make systems more plug-and–play. The web inspection system is tied into the plant information network so that information on defects and specific rolls can be stored and recalled.  Technological advances are extremely important in this market and rapid progress in the fields of computing and electronic imaging has facilitated key developments.

[Vic Wintriss] Sophisticated lighting systems, faster and higher resolution cameras, and smarter cameras. 

[Gunter Dorn, Herbert Oberstaller] Even more light sensitive and faster CCD chips. Automatic calibration, self classification, expanded software algorithm for defect classification, expert systems not only detecting defects, but also show the reason and close the control loop to the machine.

[Werner Goeckel] There will continue to be increases in speeds in both the cameras and PC's which will lead to the ability to handle even faster line speeds and get better imaging, even at high line speeds. Lighting will also continue to improve both in intensity and lower cost. More customers in the future are going to be looking for process control solutions in addition to finding the defects. This will be a challenge because of the variety of manufacturing processes. But similar to the thickness measurement systems that provide die control, more processor want automated controls to ‘‘ fix’‘ the problem. This will require greater integration of the inspection systems to the process control systems and ability to provide process analysis software.


5. As a supplier of web scanners what are some challenges you face in marketing web scanners?

[Vic Wintriss] Most potential users of web scanning systems are not particularly knowledgeable in the art and therefore find it difficult to fully articulate their requirements until after a system has been installed.

[Gunter Dorn, Herbert Oberstaller] Develop new markets where surface inspection systems provide the best return on investment for our customers.

[Werner Goeckel] Too much competition for available business. Because there is not a large capital requirement to enter the business there are too many small one to five man integrators in the business. I do not see this changing. The vision system manufacturer's that are able to provide more complete process solutions will be able to differentiate themselves from the small integrators. The successful vision supplier will have to find ‘‘ niche ‘‘ markets where more is required than just the hardware. Also those companies that have true global service capability will have an advantage. 

[Joe Paumi] The downsizing of support staff in most user companies puts a higher demand on the robustness of the inspection systems. High Travel costs both pre and post order, eat away at margins and cause the small competitor to fail. Failing companies hurt the reputation of the industry.

[Peter Ackroyd] A significant challenge is creating a value proposition in a particular application. Providing a technical solution falls short….there must be a ROI that can be demonstrated through scrap reduction, reduced labor, increased throughput or process improvement.

[Johannes Giet] A crucial part is to identify areas where a supplier has an edge over his competitors. It is difficult for the customer to differentiate. A decisive point in the future is the on-going service and support.


6. What are your thoughts on the future of web scanners?

[Gunter Dorn, Herbert Oberstaller] Due to increasing requirement of production quality this market is going to grow, 20 % each year.

[Werner Goeckel] The web scanner market will increase at a nominal rate in revenues but prices will continue to come down. This will reduce margins forcing vision supplier to offer additional capabilities with the systems to maintain margins. This is particularly true in the software area. Vision manufacturers will also need to reduce hardware costs and increase manufacturing efficiencies to stay competitive.

[Joe Paumi] Cameras (with color) will ultimately win out over laser-based technology. System will be providing full color images of defects, as is common today in grayscale. Higher resolution at faster speeds goes without saying. Pricing will remain constant as the technology improves.

[Peter Ackroyd] 100%, objective inspection of web products will continue to drive web scanner market and indeed increase demand as manufacturing globalization levels the labor costs and drives the automation of product inspection.

[Johannes Giet] In some years we will utilize smart cameras on a chip with an imager offering 10,000 pixels and more resolution and integrated computing power capable of performing the image processing and analysis and all this and at an acceptable price point.

[Vic Wintriss] I firmly believe that the optical web scanning industry is still in its infancy and is entering a period of explosive growth along with supplier consolidation.


7. What advice would you give to a company investigating the purchase of a web scanner?

[Werner Goeckel] The advice I would give is to first look at what they wish to accomplish. Forget about who has what technology and analyze the expected end result.  Then look at who has the technology to do the job and can provide the service support. I would look at the company’s history of doing similar jobs and talk with some users of the company’s product. I would also ask the vision supplier how long he expects to support the technology and what kind of service is available. Does the vision supplier support his own installations or hire outside contractors to do the servicing. Is the system expandable and do I get new software upgrades as they are introduced? At what cost? Last I would look at the financial stability of the company.  How long have they been in business, are they growing and profitable?

[Joe Paumi] Go with a company that has staying power, one that is making money, and has a good global support structure and reputation. Don’t buy on price alone.

[Peter Ackroyd] Look for a supplier that provides engineered solutions as opposed to an integrator.

A good solution requires:

  • Complete understanding of the web application including significant testing of samples.
  • Selecting the correct illumination approach.
  • Optimum selection of cameras.
  • Applying correct image enhancement to isolate defects and
  • Flexible defect analysis to properly classify defects.

[Johannes Giet] Before investigating it is essential to get an overview about the typical defect spectrum, the defect sources and costs and the costs of rejects and returns.

    1. Simply - What do you want the system to do?
    2. Crucial - Selecting a vendor who has the experience to assure that all the parameters and requirements are considered.

[Vic Wintriss] It is vital that a company investigating the purchase of a web scanning system be able to concisely put into writing their requirements document.  Probably the best way to accomplish this is to install a demonstration system so that they can understand the capabilities of optical web scanning in their own environment and realize the full potential of web scanning as it applies to their product.

[Gunter Dorn, Herbert Oberstaller] A lot of small companies are starting in this market and disappear very soon again. These companies are mostly system integrator, component manufacturer, or software provider. There is almost no company providing complete system solutions (hardware, software and integration). We would suggest looking for a reliable and innovative partnership. A company with a certain size, combining own development of hardware and software with mechanical design department and own manufacturing. Long term spare part supply guarantee and software support.


 

 

 

 

 

 

 

Comments:

There are currently no comments for this article.


Leave a Comment:

All fields are required, but only your name and comment will be visible (email addresses are kept confidential). Comments are moderated and will not appear immediately. Please no link dropping, no keywords or domains as names; do not spam, and please do not advertise.

First Name: *
Last Name: *
Your Email: *
Your Comment:
Please check the box below and respond as instructed.

Search AIA:


Browse by Products:


Browse by Company Type: