I read with interest your article titled “Optical Sorting Equipment” in the Aug./Sept. 2005 edition and feel a big disservice is given the readers as all optical technology is not more or less equal as indicated by the comparison chart in the article. It seems that the most effective comparison was missed and that is the detection and ejection of identified items.
Comparing optical sorting technologies really means comparing each manufacturer’s ability to detect and eject different objects. In the past some manufactures had advantages by being able to detect more types of polymers or by having systems that could change the sorting tasks performed more easily than the others. Today, almost all manufactures offer similar systems (from a distance) and the difference lies on the precision of the detection and ejection.
The detection precision depends on the number of scanpoints in a certain measurement area. More scanpoints means more precision. It works like a digital camera — if it has few pixels, the picture is blurry; if it has many, the picture is clear. Manufactures which have scanners with many and small scanpoints will “see” the objects more clearly: the boundaries will be clearer, the labels will be clearer and the sorting unit will know exactly the material type, color and shape of the object. If there are few and large scanpoints, there might be three different materials and colors into a single scanpoint and the sorting unit will not be able to “decide” correctly.
TiTech is able to make up to 160,000 scanpoints per second (right picture). This means up to 2.5 million single wavelength measurements per second with the NIR scanner. Scanners with lower resolution will not detect the objects with the same precision (picture on the left). Our competitors are able to reach from 3,000 to 25,000 scanpoints per second (or 300,000 single wavelengths measurements per second).
The other important component is how precise the ejection is. One way to eject the objects is to hit them right on their gravity center. However, it is not always easy to calculate it and hit it right. Therefore it can happen that the object will flip when shot, instead of being ejected. This negatively affects the purity and the capture rates. Another way to eject the objects is by shooting on the whole body of the object. TiTech shoots all the area within the object boundaries. This secures higher precision. In order to do so, there are four types of valve blocks available: depending on the object’s size and weight, the most appropriate one will be chosen. Our high resolution valve block, for example, has a distance of 11 mm between the nozzles. This means that we are not only able to detect objects down to 12 mm but also to eject them.
Achieving high purity and capture rates are the most important parameters when it comes to evaluating optical sorting systems. As shown in the graphic, a very small decrease in the capture rates can mean a lot of money over one year:
A “small” two per cent increase in the capture rate can mean more than US $120,000 extra revenue per year for a sorting plant with input of 15,000 tonnes/year of clear PET (assuming US$ 0.20 per lb of clear PET).
Capture and purity will also determine if manual or optical quality control will be required, adding significant costs to the sorting tasks. Of course a customer must also compare the reliability of the systems, the service coverage of the supplier and many other aspects. But having the most advanced technology is definitely the most important parameter to be compared.
Adapted from a letter to the editor submitted by Don Holliday, who represents Van Dyk Baler/Lubo USA and TiTech products in Canada. Contact Don at email@example.com