Patent Grant
6846447
The present invention relates to methods and apparatus for monitoring resin-loading of wood materials during assembly for manufacture of engineered wood product. More specifically, this invention is concerned with calibration and use of spectroscopic instrumentation for quantitatively measuring resin-loading during on-line assembly of composite wood product; and, maintaining that resin-loading within manufacturing standards, in particular, during assembly of particleboard (PB) and medium-density fiberboard (MDF).
A primary object provides for calibrating spectroscopic instrumentation for accurate and prompt measurement of resin loading during assembly of wood materials prepared for manufacture of composite wood product.
A related object provides for calibrating use of spectrometers using selected wavelength ranges within the full visual (VIS) and near-infrared (NIR) wavelength region of electromagnetic radiation.
An important object provides for monitoring resin-loading of wood particulate and fiber materials during on-line assembly, respectively, of particleboard (PB) and medium-density fiberboard (MDF).
A specific object is to provide on-line measurement of resin-loaded wood materials as assembled prior to heat and pressure bonding treatment.
A further object provides for feedback of resin-loading data to maximize continuing on-line assembly within manufacturing standards and to optimize resin usage.
The above objects and other contributions of the invention are considered in more detail in describing the invention in relation to the accompanying drawings.
Contributions of the invention involve uncovering opportunities for potential improvement in the assembly and manufacture of engineered wood product. Recognizing that producing quality composite wood product requires a consistent rate of application of resin was a major factor in determining that consistency in resin-loading can best be achieved by quantitative-analyses during assembly of the raw-wood material. And, also, concluding that in order to ensure product homogeneity, minimize out-of specification product, and optimize adhesive resin use, accuracy in measuring resin-loading should be carried-out while the wood material is traveling during in-line assembly.
Specific embodiments are described working with furnish-type wood-material. An adhesive resin in liquid form is atomized and directed into a chamber forming a fluidized bed for in-line movement of the wood material; and, resin application is carried out during that movement in the fluidized bed while moving at an in-line assembly rate. Specific examples of wood material include particulate wood for manufacture of particleboard (PB) and wood fibers for assembly of medium-density fiberboard (MDF). Resin-loading of the wood material is monitored following resin application in an assembly line, prior to heat and pressure bonding treatment to form commercial engineered wood product.
The above procedures provide early on-line detection of out-of-specification material, if any; and, in addition, the invention provides for adjustment of resin-loading, if any is required, to be carried out promptly to bring the resin-loading within manufacturing specifications. Promptly maintaining desired manufacturing specifications minimizes or eliminates losses in material and production rate.
It is significant that non-invasive on-line monitoring of resin-loading takes place without disturbing furnish-like wood materials during assembly. Resin measurement involves use of spectroscopic measuring instrumentation utilizing electromagnetic radiation which is absorbed by the wood material. Further, accurate measurement of resin-loading is dependent on calibration, as disclosed herein, of that instrumentation.
The term “resin-loading” is used interchangeably herein with “resin content”; and, resin-loading of particulate and fiber “furnish” is measured and indicated as a weight percentage in relation to the weight of the wood material.
For resin-loading of particleboard (PB), liquid resin is atomized for application to sawdust-type wood particulate. After non-invasive measurement of resin content, the resin-loaded particulate is then subjected to heat and pressure, in a press layout which cures the resin resulting in bonding, to produce an extended-surface-area mat, which can later be cut into commercial sizes. The present non-invasive and continuing measuring contributions are particularly advantageous for manufacturing installations carrying out bonding in a continuous-in-line press.
In
Relating the material of interest to the measuring technique for quantitative analyses, it was discovered that, spectroscopically, resin-loading is related in a linear manner to absorption of radiation within a selected electromagnetic radiation wavelength range; and, calibration of spectroscopic instrumentation could accordingly be verified. That calibration can be initiated in an assembly line at station 22.
During calibration, in summary, wood materials of the type to be assembled are selected and accurately-predetermined resin-content reference-source test-samples are established. With furnish-type wood material, resin-loaded determinations are established by making static chemical-test measurements of the test sample materials for accurate resin-loaded weight in relation to weight of furnish. Preferably, reference-source test-samples, for calibration purposes, are resin-loaded in an incrementally progressive manner; such as: zero percent, about four percent, about eight percent, and above about twelve percent.
Calibration of spectroscopic instrumentation, within a selected range of visible (VIS) and near infrared (NIR) irradiation is carried out at station 24; calibration steps are described in more detail and graphically presented in relation to later FIGS. It has been found that the wavelength bands for moisture-content (MC) of both the wood and the resin do not effect accurate measurement of resin content; and, MC wavelength bands can be removed when the instrumentation is used to measure resin content. It should be noted that non-absorbed energy reflected by the wood materials is used to measure resin content. Calibration data, from the instrumentation used for measuring resin content, is directed from station 24 over interrupted-line 25, for use in the assembly line for the wood materials.
At station 26 in the assembly line, resin-loaded particulate is supported on an in-line conveyance system for continuing assembly of a mat of particulate, which is moving toward ultimate bonding by heat and pressure. Non-invasive capabilities for measuring of resin-loading are significantly important during assembly of wood particulate. VIS/NIR spectroscopic instrumentation for carrying out continuing non-invasive measurement is located at station 27.
A lamp with full-spectrum-light is used to illuminate the wood materials; the “red” portion of that light is absorbed by the resin. Non-absorbed radiation energy, reflected by the wood materials, is measured by the instrumentation sensor. The sensor is positioned a selected distance from the resin-loaded materials for laboratory determination of a calibration model and, also, for calibration of instrumentation being used on an assembly line. The sensor is placed above the moving materials. The illuminated and VIS/NIR irradiation covers a selected area of about three to about twenty four square inches; VIS/NIR radiation is selected in a wavelength range, made available by the invention, which penetrates the resin-loaded blend by about two to five mm.
The return reflected VIS and NIR energy is collected fiber-optically and is computer processed at station 28 to enable graphical presentation of absorbed radiation based on reflected non-absorbed radiation; instrumentation calibration data for measuring resin-loading is received, as referred to above, over interrupted-line 25 from station 24. That measurement of resin-loaded percentage weight is indicated at station 30. During production, adjustment of resin-loading percentage weight, if any is required, can be directed, manually or automatically, over interrupted-line 31 to resin-loading station 21. The rate of resin-input can be adjusted, or the assembly line-speed can be adjusted, in order to maintain uniform resin-loading; preferably the rate of resin-input is utilized.
The resin-loaded particulate at station 32 continues toward bonding at station 34. If desired, a secondary sensor can be positioned at station 35, after heat and pressure treatment, to measure resin-loading of the bonded product; station 35 can provide added confirmation that the resin bonded product is within desired manufacturing objectives.
Referring to
Other than that pre-bonding compaction of fibers, any differences in measuring resin-content between the furnish-like materials of wood particulate and wood fibers are inconsequential for purposes of carrying out the measuring methods of the invention, although the finished composite wood products are significantly different. Various types of wood for particulate for PB, or for fiber for MDF, can be used when carrying out measurements in accordance with the invention.
During calibration in a lab or on-line, accurate resin-loaded reference-source test-samples are used for establishing a calibration curve for the instrumentation. For spectroscopic quantitative analyses, a relationship between content of the material and transmittal or absorption of the radiation is the basis calibration of instrumentation. It was determined that a linear relationship exists between absorption of radiation and the amount of resin-content; and, that is used for calibration of the instrumentation. The calibration equipment of
A rotatable turntable 37, schematically illustrated in
Absorption, within a selected range of VIS/NIR radiation, by the wood material is utilized. The resin-loading is determined by measuring non-absorbed VIS/NIR energy reflected by the wood material. Calibration is carried out by graphically-representing “resin content” of pre-established reference-source test-samples; based on the linear relationship of resin content to absorbed VIS/NIR. The calibration curve correlates that linear relationship, as shown graphically in later FIGS, for measurements of actual resin-loading on-line.
A unique advantage during calibration, and for on-line monitoring, is the non-invasive nature of resin-loading measuring of the furnish-type wood materials. That is, calibration of spectroscopic instrumentation, and on-line measurement of resin-loading of wood materials during assembly, can proceed without disturbing the wood materials which are resin-loaded. Of course, when on-line measurements indicate that resin-loading is not within manufacturing standards, resin-loading adjustments are made; but, again, those measurements are made without disturbing the actual assembly of furnish-type wood material.
Absorption of radiation by the resin-loaded materials can be measured by selecting a wavelength range, from the full-scale wavelength region of 350 nanometers to 2500 nanometers (nm), which provides for penetration of the wood material. A wavelength range of 400-2200 nm satisfactorily covers the above-mentioned full scale region. However, significantly, it has been discovered that acceptable standards can be maintained by other than use of full-scale, or near full-scale, wavelength regions.
It has been found that selection can be made from multiple ranges of wavelengths; an individual wavelength range can be selected from the following: (i) 350-1050 nm, (ii) 1000-1800 nm and (iii) 1000-2500 nm; those ranges and the type of sensors for measuring resin loading in each range are later tabulated herein. The resulting advantages from discovery of those multiple ranges can be important in relation to the size of composite-wood manufacturing installations.
During calibration, direct measurements are made of reference-source test samples wherein the amount of resin has been accurately pre-established. Selected peaks in the spectrum are used in calibration, which brought out that moisture content (MC) absorption wavelength band could be eliminated. That calibration data is correlated with measurements in which those samples are obtained from raw-wood material for the on-line assembly process. The apparatus of
The resin-loaded wood materials are illuminated with a full visible light spectrum lamp as indicated at 36 in
Example I involves VIS/NIR spectroscopic measurements to determine resin-loading of particles which are to be made into particleboard (PB). Resin-loading for calibration can be varied from 0% to above about 12%; initial moisture content (MC) can be 6%. Spectral data were collected using instrumentation and software as tabulated later herein. The instrument sensor head is connected to a fiber optic bundle. The sensor head is mounted about seven inches above the particle/resin blend. The size illumination and measurement area can be between about a three inch diameter to about a twenty inch diameter on the moving support for the test samples.
Example I was carried out using a rotary table. For calibration, test-samples having known pre-established resin contents were measured. It had been determined that resin-loading is in linear relationship with absorbed radiation of the selected wavelength range(s). The resulting measurements of the pre-determined test samples establish that a linear relationship between absorbed radiation and resin-loaded percentage weights, which is the basis for the linear calibration curve shown later herein.
The turntable simulates a rate of movement on a continuous conveyor system. Although the height of the furnish was not strictly controlled in the lab embodiment, a straight edge can be used to roughly level the surface where the VIS/NIR beam contacts the furnish. The bulk of the measurement was made on 6% initial moisture content (MC) PB furnish.
Example II utilizes VIS/NIR spectroscopic instrumentation for determination of resin-loading of medium density fiberboard (MDF) fibers which are compacted as described earlier. Resin-loading was varied from zero to above about twelve percent; MC was varied from six to ten percent. Spectral data were collected with a sensor head connected to a fiber optic bundle. The sensor head was mounted at a distance of about seven inches above the compacted form. The illumination measurement area covered about a three to about a twenty square inch area. Example II was supported using a rotary turntable. A wavelength range was selected in the 350-2500 nm spectral region. Measurements again verify that moisture content has no significant effect on measurement of resin-loading.
The resin used in Example II was the same UF resin utilized in Example I. Measurements were made on compacted MDF placed on a turntable; rotation at ten rpm was established. As above noted, the turntable was rotated to simulate the travel rate of a continuous conveyor system. The bulk of the measurement was made on six percent MC wood material; but measurements were also carried-out on compacted MDF containing ten percent MC.
Resin levels were pre-established at 0%, 4%, 8% and about 12%. Atomized resin was fed into a blending chamber and applied to the wood fibers from a fluidized resin bed established in the chamber. A Carbon-Nitrogen (CN) static chemical analyzer was used to establish actual resin content of referenced-source test-samples. The sensor head was suspended above the pre-press form at a distance of about seven inches; fiber height of the pre-press forms was not closely monitored, which is similar to production conditions. Non-absorbed reflected VIS/NIR spectral data were obtained and processed by computer to provide resin-content percentage weight.
Calibration based on resin absorption, by excluding water absorption wavelength regions of 900-1000 nm, 1400-1450 nm and 1900-1950 nm, is shown in
The resin is selected in a fluid, preferably liquid, state so as to facilitate atomizing for fluidized bed application; some powdered resin can be used provided atomized fluidization is not disturbed. Wax, for water resistance of the composite product, and some other additives, can also be utilized with furnish-type wood materials; provided they can be added without disturbing desired atomizing for the resin-loading fluidized-bed operation of the invention.
Monitoring and measuring processes can also be carried out either including or excluding the wavelength regions measuring moisture content of the resin or wood materials, i.e., 900-1000 nm, 1400-1450 nm and 1900-1950 nm. Accurate results are obtained in absence of measurement of the water wavelength bands.
Spectroscopic Equipment
Source:
Analytical Spectral Devices, Inc. (ASD)
5335 Sterling Drive, Suite A
Boulder, Colo. 80301
Instrument Name: FIELDSPEC PRO.
Sensors (Arranged in Increasing Order of Price)
Software Processing Equipment
Source: ASD
For collecting spectra data
For measured Data Processing
Resin
Detailed information on materials, values, apparatus and software have been described for purposes of disclosing the invention. However, the above disclosure can position others to make modifications in those specifics, without departing from the valid scope of protection of the present invention; therefore, the scope of patent protection should be determined from the appended claims, interpreted in the light of the above disclosure.
The present application is a continuation-in-part of U.S. Ser. No. 10/294,296 filed Nov. 14, 2002 entitled “NIR SPECTROSCOPIC MONITORING OF RESIN-LOADING DURING ASSEMBLY OF ENGINEERED WOOD PRODUCT”, which is incorporated herein by reference.
| Number | Name | Date | Kind |
|---|---|---|---|
| 3196072 | Wirtz | Jul 1965 | A |
| 4508772 | Churchland et al. | Apr 1985 | A |
| 5104485 | Weyer | Apr 1992 | A |
| 5619038 | Parigi et al. | Apr 1997 | A |
| 6461743 | Tanzer et al. | Oct 2002 | B1 |
| Number | Date | Country | |
|---|---|---|---|
| 20040094853 A1 | May 2004 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10294296 | Nov 2002 | US |
| Child | 10338069 | US |
