Get the right blend; here's a better way to compare gravimetric blenders.

To choose a gravimetric blender that's right for your operation, here's what you should look for in blend accuracy - more specifically, blend homogeneity.

Accurate blending of resins and additives has a major effect on the properties, processability, and cost of plastic products. Before choosing a blender, processors must carefully evaluate what degree of blend uniformity they need and the potential cost savings they could realize from more accurate use of expensive additives. Although weight-based, or gravimetric, blenders have become the standard of accuracy, they don't all cost or perform alike. Not only are there differences between batch and continuous units, but also between different models of the more commonly used batch gravimetric types.

One goal of a recent R&D program at Process Control Corp. was to develop the best possible way to determine blender accuracy. The research demonstrates that processors can get more practical value from assessing batch-to-batch blend homogeneity than from the long-term blend accuracy that is typically cited by blender suppliers. This R&D program also shows that an improved blender design can raise the standard of blend homogeneity for a cost-effective gravimetric batch blender.

Accuracy vs. homogeneity

Blend accuracy is determined by measuring the amount of each material present in the resulting blend over a specified time period. Typically, the accuracy of batch blenders is given as a percent accuracy over a large number of batches - sometimes 10 or more. Blend accuracy of 0.1% is commonly claimed by blender makers, but this can be misleading because it is measured over long time periods and a relatively large volume of material. Since blender controls seek to optimize long-term blend accuracy, a cumulative measure of accuracy over many batches tells you little about dispensing errors within a single batch.

Blend homogeneity is closely related to blend accuracy and may relate better to product quality. Homogeneity is simply blend accuracy measured over shorter time periods, which may be more significant for a given process. A processing machine typically takes in blended material in discrete "bites" as the extruder screw turns. Material in each flight then melts, mixes, and exits the machine. In general, there is limited mixing forward and backward from one flight to another, so the consistency of the end product is heavily influenced by blend variations over much shorter time periods than those typically used to test blend accuracy.

How blenders evolved

In the early days of plastics processing, blends were produced in large batches by hand-measuring ingredients and putting them together in something like a huge cement mixer. This was labor intensive, and accuracy varied, but homogeneity was great because of thorough mixing.

In the late '60s and early '70s, suppliers introduced automated feeding. Operators didn't have to measure ingredients by hand anymore, but these systems measured by volume. So again, blend accuracy suffered.

It was with the first weight-based blenders in the late '70s that accuracy improved. Gravimetric batch blenders have a number of material hoppers, each with an individual metering device at the outlet. The operator enters a desired blend recipe, specifying the percentages of each ingredient, and the blender calculates from the batch size what weight it needs of each ingredient.

Materials all feed into a common weigh hopper. The main ingredient is dispensed first until its target weight is reached, then each additional material is dispensed one at a time. Then the entire mass drops into the mixer, which runs for a preset time. Because each material is added one at a time in layers, mixer design is critical to creating a homogeneous blend.

The first gravimetric batch blenders were a big improvement in labor and accuracy, but they were bulky and expensive. Also some blends separated if ingredients were of dissimilar particle size, density, or shape. So these blenders weren't suitable for all applications.

The mid-'80s saw the introduction of continuous gravimetric blenders designed to mount over an extruder or molding machine. Eliminating blend separation problems, continuous blenders gained acceptance in critical applications where material cost savings justified higher equipment cost. Continuous blenders meter ingredients simultaneously and continuously in percentages by weight, without relying on mechanical mixing to homogenize the blend. Continuous gravimetric blenders thereby provide the ultimate in homogeneity to meet tight specs at high output rates.

In the late '80s, smaller, less expensive batch gravimetric blenders were introduced that could be mounted atop an extruder or molding press. The goal was to bring the benefits of automated blending to processors whose applications didn't have the volume or critical specs to justify a continuous gravimetric unit.

To reduce cost and simplify operation, most of these newer batch models use air-actuated valves of a gate or bladder type to open and close the dispensing outlet. Also, the newer batch blenders usually don't mix materials as aggressively as earlier, larger models, so they have poorer short-term blend homogeneity and may be unacceptable for some uses.

Improving blend homogeneity

As part of a program to optimize blend homogeneity, Process Control Corp. obtained a new commercial model of a gravimetric batch blender from a leading manufacturer and set it up according to the manufacturer's directions. This model was studied to establish a baseline of blend homogeneity. We then undertook an R&D program to design a gravimetric batch blender with improved blending Capabilities. Through numerous design iterations, which investigated changes in chamber geometry, agitator, mixing speed, and mixing duration, we developed a new series of batch blenders with enhanced mixing for improved blend homogeneity. Called the Guardian Series, these new blenders will be introduced in January. The experiments used to develop and evaluate the new blender may suggest to processors procedures they could use, or data they could request from suppliers, to compare blender models.

Blended materials from both the commercially available blender and the new prototype model were sampled from a plug-flow tube. Consecutive small samples of blends were taken at intervals, timed to match the way material would be ingested by successive flights of an extruder. We chose a sample size of 0.125 lb to match the "bite" of one screw flight of a 3.5-in. extruder processing 600 lb/hr at 80 rpm. The individual ingredients in each "bite" were then manually separated and weighed to determine the percentage of each ingredient in the blend. Blend homogeneity in the extruder was then graphed flight by flight and analyzed over time. Different particle sizes were also examined.

We also estimated how much mixing the extruder did. If the extruder did no mixing forward and back between screw flights, any flight-to-flight variation would show up as a difference in the final product. Conversely, if the extruder mixes well, materials traveling over many flights would be homogenized and reduce blend variation. We checked with several extruder manufacturers, who said little mixing occurs across more than 10 flights of the extruder, so we conservatively assumed that an extruder mixes over 20 screw flights. Thus we averaged every 20 flights' worth of material to define a meaningful measurement of blend homogeneity.

These averages show a major improvement in blend homogeneity for a batch blender with enhanced mixing. Results for a leading current-generation gravimetric batch blender and for Process Control's new developmental Guardian series unit are compared in the graph on p. 66. The benchmark commercial blender has a two-sigma homogeneity (two standard deviations divided by the mean) of 50% vs. 6% for the Guardian Series. The difference in performance means processors could see a potential reduction by as much as 50% in the amount of costly additive needed.

Rob Northrup is v.p. of sales, William Dannelly is director of sales and marketing, and Eileen Silva is marketing manager for Process Control Corp., an Atlanta-based maker of gravimetric blenders and other auxiliaries for materials handling and scrap reclaiming.