Test report on the inline color measurement of recyclates
Sensor Instruments: The increasing use of recyclates in the plastics industry means that efficient product control is assuming ever-greater importance. Whilst laboratory-based random quality control was sufficient in the past, the industry is increasingly considering the deployment of 100% control, especially in terms of the color gradient. Sensor Instruments has launched a number of inline systems to perform this function. Providing production staff with information about recyclate production at a very early stage, this covers not only the trend in the color of the recyclates, but the color of the recyclates in combination with their temperature, whilst monitoring the product flow level. The product flow level is important for exact determination of the recyclate color, whilst also providing timely information about a possible blockage of the sieve immediately after the vibrating feeder. The data is also made available for quality assurance via a digital-serial interface (Ethernet).
1. Preparatory measures
First, the current Windows® software SPECTRO3 MSM DOCAL Scope V1.4 was installed on the panel PC. The SI-PPC-500-15” panel PC was positioned immediately after the extrusion system for inline color measurement at the vibrating feeder. Then, the SPECTRO-3-FIO-MSMANA-DL (SI inline system) color sensor array was calibrated using a white reference surface (a RAL 9003-P plastic card served as the white reference surface) and to the relevant recyclate. Calibration was performed inline, i.e. the sensor array did not have to be removed from the system for this purpose and calibration was performed on the moving recyclate flow. This meant that the random position of the individual granules could be averaged out. The optimal working distance to the KL-D-0°/45°-85-1200- D-S-A3.0 color sensor front end of 85mm was ensured using a deflector plate. A light spot diameter of type 20mm already performed optical averaging of the random position of the plastic pellets. Averaging over a settable time window of between 5 and 60s can be set using the SPECTRO3 MSM DOCAL Scope V1.4 software. A period of 15s has already emerged as the ideal time window during the first tests. With a recyclate flow speed of 50mm/s to 100mm/s, this results in a measuring length of 750mm to 1500mm. Before EN_PI_2023-10-11_SI_Test report on the inline color measurement of recyclates.docx 2/15 performing the actual color measurement of the various recyclates, the temperature of the recyclate flow at the measuring position and the housing temperature of the color sensor front end were measured.
The surface temperature of the recyclate flow was measured as c. 75°C at the measuring point, whilst the housing temperature of the color sensor front end was c. 66°C. As there are no optoelectronic or electronic components at the front end of the KL-D0°/45°-85-1200-D-S-A3.0 color sensor, the temperature of the housing did not constitute a problem. A black, a white and a blue masterbatch were prepared for the upcoming test measurements. The idea was to start with black colored recyclate. Then, the feed of black masterbatch should be stopped and a white masterbatch be added with a time delay. After stopping the addition of white masterbatch, a blue masterbatch should be added after a certain time. In addition, the black masterbatch should be added after a certain time, and the supply of blue masterbatch should be reduced.
2. Inline color measurement on the vibrating feeder
The inline color measurement could begin: Start of recording: 10:21. End of recording: 11:12. In addition to the date, time and the actual L*a*b* color values of the recyclate, the deviations dL*da*db* were recorded in relation to the black recyclate using the SPECTRO3 MSM DOCAL Scope V1.4 software on the SI-PPC-500-15” panel PC:
A diagram was generated showing the L*a*b* color values as a function of time and a diagram showing the course of dL*da*db* in relation to the black recyclate sample (L* = 29.34 a* = -0.04 b* = -1.66).
10:22 Stop black masterbatch feed 1<
0:29 Add white masterbatch
10:33 Stop feeding white masterbatch and add black masterbatch
10:45 Stop feeding black masterbatch
10:50 Add blue masterbatch
10:55 Add white masterbatch
10:59 Stop feeding white masterbatch and add black masterbatch
11:04 Stop feeding blue masterbatch
The following graph shows the dL*da*db* course in reference to the black recyclate (L* = 29.34 a* = -0.04 and b* = -1.66):
The following two graphs show the course in approximately double resolution. The pictures of the individual recyclates have been integrated into the diagrams:
3. Examining the recyclate samples with the SPECTRO-3-0°/45°-MST SI laboratory device
In order to establish a reference to the color measurement systems already available in the laboratory, it was first necessary to make injection moulding plates from selected recyclate samples (a measurement with the laboratory color measurement system requires the production of injection moulding plates). After ascertaining the L*a*b* color values from the injection moulding plates using a laboratory color measurement system, the recyclates matching the injection moulding plates were presented to the SPECTRO-3-0°/45°-MST SI laboratory device (a color sensor array identical to the inline system) for user calibration; the L*a*b* values determined using the laboratory color measurement system were assigned using the SPECTRO3 MSM DOCAL Scope V1.4 software. The calibration wizard made the user calibration very easy.
Calibration of the SPECTRO-3-0°/45°-MST, SI laboratory device was performed using the recyclate samples shown in the picture. These were placed one after the other at the correct distance under the color sensor front end of the SI laboratory device and guided by the calibration assistant. To avoid being dependent on the random position of the pellets in the light spot during calibration, the respective calibration sample was continuously moved under the sensor head within the intended and settable time window. Once calibration had been completed, the calibration factors were stored in the color control electronics.
This was followed by color measurement of the 30 different recyclate samples using the SPECTRO-3-0°/45°-MST SI laboratory device. Here, too, the measurements were carried out using SPECTRO3 MSM DOCAL Scope V1.4.
In addition to displaying the dL*da*db* values in relation to an entered reference, the color values L*a*b* incl. the color deviations dL*da*db* were saved in a file which is editable using Word® and Excel®. It was also possible to choose between different representations during color measurement recording: trend display (DOCU); numerical color value and color value deviation display (C SPACE); display of the current color value in the color space inc. the color tolerance windows (C SPACE 3D) and display of the raw data (XYZ).
While the displays on the monitor provided the operator with valuable information regarding the trends of the color values of the respective recyclates, the evaluation of the stored data is of very great importance for quality assurance, since not only the dL*da*db* color deviations from an L*a*b* reference saved in the TEACH table are stored, but also the L*a*b* color values together with the date and time and information as to whether the color of the recyclate is still within the respective color tolerance range (color number). Below are the diagrams created using the Excel® file. The samples taken during production were marked with the time at which the respective recyclate sample was taken.
L*a*b*- values of the various recyclate samples, marked with the time of sampling during production
dL*da*db* values of the different recyclate samples, marked with the time of sampling during production
4. Optimization measures
In principle, the color measurement works very well at the current measuring point, but some changes should still be made. For example, a greater volume of dust is generated whilst changing the big bag, whilst the lighter plastic flakes settled on the surface of the sensor array. It would certainly be advantageous to cover the sensor array. A very rough pellet flow surface was observed at the measuring point. In contrast, the plastic granulate flow near the sieve had a flat surface even without the use of a deflector plate. However, it was also found that the height of the pellet flow increased when the sieve was blocked or partially moved. In order to be able to make a correct statement about the color value of a recyclate, the distance of the recyclate surface to the front end of the sensor should be kept approximately constant. The level of the recyclate flow can be determined by attaching an additional laser distance sensor. For example, a warning lamp can be activated if an adjustable height tolerance window is exceeded or not reached; in parallel, the height of the recyclate flow is also displayed by the evaluation unit (panel PC: SI-PPC-500-15“) and saved in a file. Preliminary tests with a type L-LAS-LT-130-SLP laser distance sensor have delivered very good results on different coloured recyclate surfaces. The laser sensor also has a serial interface and provides digital outputs (0V/+24V) that inform whether the height of the recyclate flow lies within or outside an adjustable tolerance window. Another important parameter is certainly the temperature of the recyclate flow, as this also influences the color of the recyclate. An infrared camera was used to take temperature measurements on the recyclate flow; the surface temperature of the pellet stream was measured as c. 75°C. It is now planned to position a pyrometer together with the laser distance sensor near to the color sensor. The temperature is displayed on the control unit integrated in the SI-PPC-500-15” and the temperature is recorded by a panel PC. Exceeding an adjustable temperature limit can be indicated by a warning lamp.
The following section shows the individual sensors integrated in a protective housing. The laser distance sensor and pyrometer are located in the immediate vicinity of the color sensor head. The deflector plate has been replaced by a heat and abrasion-resistant plastic plate, which enables containment of the heat transfer from the recyclate flow to the sensor head. A fan is provided to ensure sufficient air circulation to prevent heat build-up inside the cover housing.
The sensor head can be positioned at different positions across the vibrating feeder with the help of the aluminium profile frame. The sensor array can be calibrated whilst installed, during operation, so it is not necessary to remove the sensor array. A mechanical bracket is provided with which to insert the white reference (for example RAL 9003-P), which can be mounted during operation. The calibration card bracket can be removed again once the white balance has been performed. Calibration is then performed on the recyclate currently under production; a sample taken from the current recyclate is used to produce an injection moulding plate for measurement of the color values (L*a*b*) using the laboratory color measurement system. After entering the color values into the software of the panel PC, the calibration process can be completed and the COLTEM-85 inline color measurement system can be used to determine the color and temperature of the recyclate.