Measuring Anilox Volume: FQC’s Gauge R&R Study

In measuring anilox volume, questions inevitably arise about how repeatable and reliable instruments really are because of the sheer number of methods available. A Flexo Quality Consortium (FQC) committee was formed to explore that question. Statistics contained in this report were analyzed and compiled by Professor Tim Claypole, MBE, of Swansea University. Kern Cox, lecturer, of Clemson University, coordinated the measuring.

Anilox volume measurement is critical for consistent color, repeatable print results, press productivity, ability to match proofs and, ultimately, customer satisfaction. It is important to note that this study is not designed to choose or recommend one method over another but, rather, is purely intended to report the results of the Gauge R & R study. Each method has its place and advantages in the pressroom and anilox manufacturing facility.

Repeatability: The study aimed to determine any variation in measurements from a single operator and instrument measuring the same roll—under the same conditions.

Reproducibility: With each instrument, the effort was made to determine the variation between different operators measuring the same engraving.

Instruments: We selected the most commonly used instruments to test:

• Microscope or Gravure Scope (Unitron)
• Capatch
• Liquid Volume Measurement (LVM)
• Troika – Anicam
• Microdynamics – 3DQC
• Testex Press-O-Film tape, also referred to as Microfax

Influences on Measurement

There are a number of outside variables that must be considered, including the experience of the person in using the instrument and the associated effect on the measurement. To determine this, two experienced and two novice operators were tested. The novice operators were trained on the day of the measuring.

Another variable considered was whether diameter or curvature influences the result? Three very different roll diameters were chosen to answer this question.

The study also asked whether coarse engravings measure differently than fine engravings? Fine, medium and coarse engravings were employed. Volumes were chosen based on original engraving specification from multiple anilox suppliers—some rolls wear more than others. Choosing the original specification as nominal does not infer the nominal volume is completely accurate. It represents a reference point with minimal bias toward any system.

Anilox

Based on the above parameters, rolls available at Clemson University that met the criteria were employed. All were closed-cell, 60-degree engravings. There were three circumferential measurements on each roll, multiplied by four operators, totaling 12 readings:

Kern Cox, Clemson University, outlined, in detail, the procedures and coordinated the entire exercise. This is an example of the form to capture data:

Volunteer companies participating in the measuring:

• Clemson University; two graphic arts students participated as novices
• Harper Corporation of America
• Microdynamics
• Pamarco Global Graphics
• Praxair Surface Technologies
• ARCS –Anilox Roll Cleaning Systems Inc.
• Provident Group/Troika Systems

Measurement Results

Claypole analyzed the data, extrapolating the following results and summary.

Reliability: While all systems show the same general trends within the mean volume graph, the microscope system has much higher results with the higher volumes.

This becomes much clearer, considering the difference to nominal volume of the system (see Difference to Nominal graph). The microscope system produces a significantly higher volume at the higher nominal volumes. The four most consistent systems (Anicam, 3DQC, Microfax and Liquid Volume) all report a volume lower than the nominal for the largest volume. This suggests that the nominal volume is over estimated, or the anilox may be worn.

The Capatch system shows the next largest difference to the nominal value. The other systems have a spread of less than 1 bcm between them and are all within +/-1 bcm of the nominal. The 0.0 line on the vertical axis represents nominal.

Note each system interprets volume differently. The instruments are not calibrated to each other. And, the question remains: what is the true nominal?

Standard Deviation

Consistency is critical for the use of these systems as a quality control tool. This can be compared from the standard deviation of the results of each system as shown in both the table and graph (see “Standard Deviation” graph). The microscope and Capatch system are the least consistent with the highest standard deviation. Overall the 3DQC produces the most repeatable results, followed by the Anicam and Microfax and then the Liquid Volume method. Note the Microfax are measured by one of the highly consistent electronic tools.

Reproducibility: One concern when measuring reproducibility is having an operator independent of the results. Comparing the results of the novice and expert users alleviates this apprehension. In both the mean and standard deviation, it is the systems with the most variables that show the biggest difference between operators. The novice operators of the microscope have a tendency to produce a lower estimation of the volume of the anilox, whereas with the Capatch and the Liquid Volume method the difference is random.

There is no trend in the standard deviation (SD) of the novice compared with the expert operators. Both novice and expert produce the same variability with the Anicam and 3DQC system.

The implications for quality control or consistent readings are best illustrated by calculating the standard deviation range about the measured mean for each system. A representative sample for a mid diameter roll is illustrated in the chart MID Diameter Roll 1 with a nominal volume of 12.3 bcm.

The consistency of the microscope and Capatch systems can be improved by averaging more measurements on the same roll. However, even with nine measurements, the consistency of the microscope would only be improved to one third of the spread shown.