GB2489586A

GB2489586A - Determining percentages by weight in a filter material

Info

Publication number: GB2489586A
Application number: GB1205420.1A
Authority: GB
Inventors: Dierk Schra Der
Original assignee: Hauni Maschinenbau GmbH
Current assignee: Koerber Technologies GmbH
Priority date: 2011-03-30
Filing date: 2012-03-27
Publication date: 2012-10-03
Also published as: DE102011006414B4; GB201205420D0; DE102011006414C5; DE102011006414A1; ITMI20120320A1

Abstract

The invention relates to a method and device for determining percentages by weight of substances such as triacetin in a cigarette filter material in the form of a filter rod, which is based in particular on celluÂlose acetate. The device comprises at least two measurement devices, a microwave measurement device and a high-frequency (HF) measureÂment device, a conveyor deÂvice for the filter material and an evaluation device. In the method, the filter material is conveyed through the measurement devices, wherein at least three measurement variables that are inÂfluenced by the quantities and the relative permittivities of the substances of the filter material contained in the mixture, are measured at at least one microwave frequency (f1),preferably between 1GHz and 30 GHz, and at at least one high-frequency (f2), preferably between 100KHz and 300MHz, wherein percentages by weight of the substances of the filter material are determined from the measured results.

Description

Method and Device for Determining Pçççzgçs_kyWeight in a Filter Mate nat The invention relates to a method for determining percentages by weight of a plurality of substances in a filter material in the form of a filter rod and/or an endless filter rod that is based in particular on cellulose acetate. The inven-tion further relates to a device for determining percentages by weight of a plurality of substances in a filter material in the form of a filter rod and/or an endless filter rod that is based in particular on cellulose acetate, said device comprising at least two measurement devices, one conveyor device for the filter material and one evaluation device.

For the production of filter rods that are cut to length from a previously pro- duced endless filter rod, the constancy of the material properties, and there- fore, the constancy of the filter properties is vital. Therefore, with the pro- duction of endless filter rods, value is placed on consistent quality and com- position of the rod material. These are checked during the production opera- tion. Measurements for quality assurance are either made online at the end-less filter rod in an endless filter rod machine, or offline at the cut filter rod, for example at so-called filter rod measurement stations.

Filter rods, discharged from the production process on a random basis or sys-tematically, run sequentially through filter rod measurement stations that typically comprise a series of two or more measurement devices. Discharg-ing occurs automatically, or based on a manual request, using pulses of air pressure, or with other suitable auxiliary means.

Filter material typically consists of a mixture of substances. Thus, in typical filters that are produced from a filter tow composed of cellulose acetate (CA), a plasticiser is added, typically triacctin or triethylene glycol diace- tate. As a result, the cellulose acetate fibres are bonded and the filter mate-na! is hardened. The plasticiser is finely atomised in a dosing device, and spread onto the outspread filter tow before the tow is formed into a cylindri-cal endless filter rod. The percentage of the plasticiser in the endless filter rod typically amounts to approximately 6% by weight. In addition, the end- less filter rod during the production thereof absorbs moisture from the sur- rounding air, if applicable also in case of climatisation in a climatic cham-ber.

The filter properties of the finished filter rods strongly depend on the per-centages by weight of the different substances of the filter rod, particularly on the percentage of the plasticiser. Therefore, the application of plasticiser in relation to the dry mass of the filter tow must be ensured to be as uniform as possible during the production of the endless filter rod.

Different methods are known for determining the mass proportion of plasti-ciser in an endless filter rod or in filter rods. European patent application EP 1 895 291 Al discloses a filter rod measurement station, in which the pro- portion of the mass of the plasticiser is determined in a filter rod measure- ment station, thus offline, in that a resonance frequency shift and a broaden-ing of a resonance curve are measured. The mass and the pressure drop of the filter rod are also measured. The proportion of plasticiser in the filter rod is then computed from these three measurement results.

These measurements, which, in addition to a measurement of the dielectric parameters in a microwave resonator, are also based on the measurement of mass and/or pressure drop of the cut filter rods, attain relatively low accu-racy and necessitate costly and machine-dependent calibration.

Online measurements are typically performed so that a microwave measure- ment is made using a microwave resonator at the completely formed cylin- drical endless filter rod, alternately with and without plasticiser. The differ-ence between the measurements provides information about the proportion of plasticiser in the endless filter rod. A disadvantage in this case is the fre- quent interruption of the filter production due to switching off the applica- tion of plasticiser in order to provide a reference for the differential meas-urement.

The document EP 1 197 746 Al discloses another measurement method per-formed at the endless filter rod, thus online. In this case, a difference is formed from two microwave measurements which are made before the appli-cation of the plasticiser at the open filter tow, and after the application of the plasticiser at the completely formed endless filter rod. The difference pro- vides information about the proportion of plasticiser. A further, correspond-ing differential measurement by means of microwaves before and after an application of plasticiser is disclosed in the document EP 1 480 532 BI.

These measurement methods which are based on a differential measurement of the microwave resonator measurements before and after the application of plasticiser also require comparatively costly calibration. They are tainted with a comparatively large measurement inaccuracy because the filter tow at the measurement location is non-uniform before the application of plasti-ciser. Consequently, the material density at this location fluctuates in the course of the conveyance of the filter tow, and the distribution of the filter tow mass in the resonator chamber changes continuously. The material den-sity and the material distribution are substantially lower after the formation of the endless filter rod, so that the basis of comparison of the measurements before and after the application of the plasticiser frequently fluctuates.

In contrast, the object of the present invention is to make it possible to de-termine percentages by weight of a plurality of substances in a filter material in the form of a filter rod and/or an endless filter rod, which is based in par-ticular on cellulose acetate, with a lower calibration expenditure than before, and with an increased measurement accuracy.

This object is achieved by a method for determining percentages by weight of a plurality of substances in a filter material in the form of a filter rod and/or an endless filter rod, which is based in particular on cellulose acetate, that is further developed in that the filter material is conveyed through a mi-crowave measurement device and through an HF measurement device, wherein at least three measurement variables, which are influenced by the quantity and the relative permittivities of the substances of the filter material contained in the mixture, are measured at at least one microwave frequency and at at least one high-frequency, wherein the percentages by weight of the substances of the filter material are determined from the measured results.

High frequencies (HF) in the scope of the invention are understood to be fre-quencies below the range of microwave frequencies. Therefore, according to the invention dielectric parameters of the filter material in the form of filter rods and/or an endless filter rod are measured at at least two very different frequencies. At least one frequency f1 lies in the microwave range, and the at least one other frequency f2 lies in the significantly lower high-frequency range.

The measurement in the microwave range can be a measurement in a micro- wave resonator. Different approaches are known for measurements using mi-crowave resonators. The shift and the broadening of the resonance can be measured as measurement variables for instance. For this purpose, measure- ments can be made at two frequencies lying near each other in the micro-wave range, for example, where relative changes therein provide information about a shift and broadening of the resonance in the case of presence of a dielectric material. These measured values make it possible, by inference, to determine the real and imaginary part of the relative perrnittivity of the filter material. The relative permittivity is also called the relative dielectric con-

S

stant or, is also known by the older term, dielectric constant.

The permittivity or dielectric conductivity is a complex variable that depends on the material, frequency and temperature. It describes the transmissibility of a material for alternating electromagnetic fields. The permittivity c or relative permittivity is the ratio of the permittivity of a medium to the vac-uum permittivity. It designates the field weakening effect of the dielectric polarisation of the medium. The real part is typically labelled as e', and the imaginary part, which is associated with the dielectric losses in the dielectric material, is labelled as c".

The present invention makes use of the effect that the progression of the relative permittivity for different substances differs depending on frequency.

Investigations have shown that the real part 8'Tr of the relative permittivity 6Tr of triacetin due to the slight polar character of this molecule at approxi-mately 10 MHz has a significantly larger value than at 5 GHz, whilst the real part S'CA of the relative permittivity 8CA of cellulose acetate has nearly the same value at both frequencies. The temperature and frequency dependent progression of the relative permittivity of water, a further component of the filter material is also known.

Therefore, as the dielectric variables, for example, magnitude and phase of the alternating electromagnetic field, or resonance shift and resonance broadening of a resonant alternating electromagnetic field, are measured in-dependently of each other at two very different frequencies, a total of four measurement variables are available, from which it is possible to draw con-clusions about the percentages by weight of cellulose acetate, plasticiser, and water, for example, in the endless filter rod, or respectively, in the filter rod.

In principle, three of the four available measurement variables are already sufficient. However, by utilising all four measurement variables and the re- dundancy obtained as a result, the accuracy of the measurement can be fur-ther increased.

In addition, compared to the present-day microwave measurement method, it is possible to determine not only two substances such as mass and water con-tent, but additionally further substances such as the content of plasticiser, for instance.

Because the measurement is made at the same endless filter rod, or respec-tively, at the same filter rod, at two different frequencies, a self-reference is already provided. Costly calibration measures concerning filter rod mass or pressure drop can be eliminated. In the case of the endless filter rod, the measurement is made at the formed endless filter rod, with plasticiser ap- plied, so that the shape and mass distribution is the same for both measure-ments at the two frequencies, thereby eliminating the consideration of a shape change as was the case with earlier differential measurements. in fact, the measured values can henceforth be compared directly to each other.

Preferably three of the four variables real part s' and imaginary part & of the relative permittivity of the filter material at the at least one high-frequency f2, and the real part c', and the imaginary part c" of the relative permittivity of the filter material at the at least one microwave frequency f1, are determined from the measured measurement variables. These dielectric variables are composed respectively from the corresponding variables for the different substances in the endless filter rod, or respectively in the filter rod, and have a functional relationship with the real and imaginary parts of the relative permittivities determined from the measured variables at the differ-ent frequencies for the filter material. This functional relationship can be determined in the scope of the in any case necessary calibration of a meas-urement system, in that the mass of the substances, thus for example, of the dry filter material used, and the plasticiser, as well as possibly the degree of moisture, are varied under precise control, and the measurement results are set in relation to these parameters. The percentages by weight of the sub- stances of the filter material can then be calculated in the context of the pre-sent invention even without the intermediate calculation of the previously named relative permittivities.

The at least one microwave frequency f1 used preferably lies in the range between 1 0Hz and 30 0Hz, particularly between 4 0Hz and 8 0Hz, and the at least one high-frequency f2 used preferably lies in the range between 100 kHz and 300 MHz, particularly between 1 MHz and 10 MHz. At the high-frequency f2 used, the water portion of the filter material has a very high real part a' of the relative permittivity, which at room temperature lies in the range of approximately 80, whilst the imaginary part a" is negligible in this frequency range. At the named microwave frequency range, the real part a' of the relative permittivity a of water decreases approaching higher frequen-cies, whilst the imaginary part a" passes through a maximum.

For cellulose acetate in both frequency ranges, the real part C'cA of the rela-tive permittivity is approximately 3.5, whilst the imaginary part E'CA is close to 0. For triacetin in the high-frequency range, the real part C'Ti of the rela-tive permittivity is approximately 7 and the imaginary part a"Tr is close to 0.

At 1 0Hz, the value of the real part 8'Tr for triacetin has already decreased significantly, while the imaginary part a",. increases to approximately 0.8.

In an advantageous further development of the method, the at least three measurement variables are measured in the microwave measurement device and in the HF measurement device, with substantially the same field geome- try and/or with substantially the same measurement volume and/or at sub- stantially the same temperature of the two measurement devices. The meas-urement devices can differ in the microwave range and in the HF range. That is, a microwave resonator can be advantageously used in the microwave range. The document DE 198 54 550 B4 for example, discloses a suitable microwave resonator. However, with the same or similar measurement vol-ume size no resonance field develops in the HF range. In this case, an HF measuring capacitor can be advantageously used, as described in WO 2006/069720 A2, for example. The contents of the documents DE 198 54 550 B4 and WO 2006/069720 A2 are included herein in full.

A substantially equal field geometry in this case means that the progression of the field lines in both measurement devices proceed with the same orien- tation to the longitudinal axes of the rod or endless rod shaped filter mate-rial, where, in particular an identical measurement volume is advantageous because in this case the measured values can be related directly to each other at both frequencies without, or substantially without, correction factors.

Therefore, the calibration is further simplified.

At microwave frequency, the measurement is made at the endless filter rod or at the cut filter rod, and at high-frequency the measurement is made at the endless filter rod or at the cut filter rod. Therefore, the scope of the inven- tion comprises combinations of the two measurements that are both per-formed online at the endless filter rod, both offline at the cut filter rod, for example in a filter rod measurement station. The measurements can be com-bined online and offline with each other. Thus, for example, the microwave measurement can be performed online at the endless filter rod and the meas-urement at high-frequency can be performed offline at the filter rod, or vice versa. Therefore, an endless filter rod machine already comprising a micro- wave measurement device, but without any available space for a supplemen- tal HF measurement device, can be supplemented with an offline HF meas-urement device, that is equipped with an appropriate filter rod measurement

station for example.

In an advantageous further development of the method, weight and/or pres-sure drop and/or diameter and/or hardness and/or temperature of the filter material are measured. In particular, the measurement of weight and/or pres-sure drop can be used for further refining and controlling the measurement, or determining the proportions of the substances of the filter material. These variables which are preferably measured offline at the filter rods, are further used for determining whether the cut filter rods have the desired filter prop-erties.

The measurement of the temperature of the filter material, which should be nearly constant during normal operation of an endless filter rod machine, can be used for refining the measurement of the proportions of the substances of the filter material, and for considering temperature-dependent changes of the dielectric variables of the filter materials and the substances contained therein. Thus, temperature drift-dependent distortions of the measurement results can be compensated. Therefore, the additional measurement results, particularly weight and/or pressure drop and/or diameter and/or hardness and/or temperature of the filter material, are preferably incorporated in the determination of the percentages by weight of the substances.

Advantageously, there is also foreign body detection. Metallic foreign bod- ies, for example, temporarily alter the resonance field very dramatically dur-ing passage through a microwave resonator. Therefore, they can be easily and reliably detected. Filter rods containing such foreign bodies are prefera-bly pneumatically discharged, and thus, excluded from further filter and cigarette production. Additionally or alternatively to this, there is preferably a detection of an additive, particularly a plasticiser, in particular triacetin.

In an advantageous further development of the method according to the in-vention, method parameters are changed during the production of the endless filter rod depending on changes of the measured values determined accord- ing to the invention. This relates particularly to the application of a plasti-ciser and/or climatisation and/or moisturisation of the endless filter rod. The -10 -previously named measurement values determined according to the invention are based primarily on the dielectric properties of the filter material and its substances, but also on the further measurement variables regarding weight, pressure drop, diameter, hardness and/or temperature of the filter material.

The control of the method preferably occurs independently of whether these measurement variables were determined online at the endless filter rod or offline in a filter rod measurement station, or in a combination thereof For the offline measurements, measurements are made at the filter rods selected at regular or irregular intervals on a random basis and/or systematically and/or regularly, wherein in particular, slow changes in the method parame-ters can be detected.

The object of the invention is also achieved by a device for determining per-centages by weight of a plurality of substances in a filter material in the form of a filter rod and/or an endless filter rod that is based in particular on cellulose acetate, said device comprising at least two measurement devices, a conveyor device for the filter material and an evaluation device, wherein the device is further developed in that the at least two measurement devices comprise a microwave measurement device and an HF measurement device, wherein the measurement devices are designed to measure at least three measurement variables at at least one microwave frequency and at at least one high-frequency, said variables being influenced by the quantity and the relative permittivity of the substances contained in the mixture of the filter material, wherein the evaluation device is designed to determine the percent- ages by weight of the substances of the filter material from the measured re-sults.

In the scope of the present invention, the term device also comprises a sys-tem, apparatus or arrangement with the appropriate components conveyor device, evaluation device, microwave measurement device and HF measure- -11 - ment device. In the case of a combination of online and offline measure- ments, a system or an arrangement, for example, is implemented in an end-less filter rod machine and a filter rod measurement station. In particular, in such a case, several conveyor devices are preferably provided, for example an endless rod conveyor in the endless filter rod machine and/or a pneumatic conveyor device for filter rods, along with a removal device for filter rods and for supplying removed filter rods to the filter rod measurement station.

The device according to the invention is preferably designed to perform a method described above according to the invention. The microwave meas-urement device is preferably a microwave resonator device, and the HF measurement device is preferably an HF measurement capacitor.

The microwave measurement device and the HF measurement device pref-erably have substantially the same field geometry and/or substantially the same measurement volume. This facilitates the comparability and applicabil- ity of the measurement results from the two measurement devices for deter-mining percentages by weight of the substances of the filter material.

Advantageously, the microwave measurement device is disposed in an end-less filter rod machine or in a filter rod measurement station, and the HF measurement device is disposed in an endless filter rod machine or in a filter rod measurement station. Therefore, the already described combination of online and offline measurements are comprised according to the invention.

In addition, there is preferably at least one measurement device for measur- ing weight and/or pressure drop and/or diameter and/or hardness and/or tem-perature of the filter material. In the case that the microwave measurement and the HF measurement are separated to an offline and online measurement, it is also advantageously provided to measure the temperature both at the endless filter rod in the area of the microwave measurement device as well -12 - as at a filter rod measurement station in the area of the HF measurement de- vice, in order to consider a cooling of the filter after removal from the end-less filter rod machine.

The evaluation device is preferably designed to detect foreign bodies in the filter material based on the measurement results and/or for controlling a dis-charge device.

The device according to the invention can be designed as a filter rod meas- urement station, or as a combination or arrangement of a filter rod measure-ment station with appropriate measurement devices and evaluation devices in an endless filter rod machine, or exclusively as an arrangement and combina-tion of appropriate measurement and evaluation devices in an endless filter rod machine. Within the scope of the invention, the evaluation device or ap- propriate control devices in or at the endless filter rod machine are prefera-bly constructed to use the measurement results of the device according to the invention, or respectively the HF measurement device and microwave meas-urement device thereof, in order to control the method parameters during the production of the endless filter rod, in particular endless rod drive speed, climatisation, addition of moisture and plasticiser application, so that an endless filter rod and subsequently filter rods are produced with uniform pressure drop and filtering properties within the set tolerances.

The features, properties and advantages named with the named subject mat-ters of the invention, i.e. the named method according to the invention and the named device according to the invention, apply without restriction also to the respective other subject matters according to the invention.

The invention is described below, without restricting the general idea of the invention, using exemplary embodiments with reference to the drawings, where reference is made expressly to the drawings with regard to the disclo-sure of all details according to the invention that are not explained in greater detail in the text. In the figures: Fig. 1 shows a schematic representation of the process of the method according to the invention, Fig. 2 shows a schematic cross-sectional representation through a mi-crowave resonator, Fig. 3 shows a schematic cross-sectional representation through an HF measuring capacitor, Fig. 4 shows a schematic representation of the frequency progression of dielectric variables of a triacetin and cellulose acetate, and Fig. 5 shows a schematic representation of the frequency progression of the dielectric variables of water at room temperature.

In the following figures, the same or similar types of elements or respec-tively corresponding parts are provided with the same reference numbers so that a corresponding re-introduction can be omitted.

Fig. 1 schematically shows a possible progression of a method according to the invention. In a first step la, an endless filter rod is conveyed to a meas- urement device, or in an additional or alternative step ib, a filter rod is re-moved and conveyed to a filter rod measurement station. In the method step 2, a measurement according to the invention is made in a microwave meas-urement device 14, for instance a microwave resonator. Therein preferably two variables, for instance the real part s' and the imaginary part c" of the relative permittivity of the endless filter rod or the filter rod, are determined at a microwave frequency f1, preferably in the range between 1 GHz and 30 0Hz, particularly preferably between 4 and 8 0Hz.

This is followed by conveyance to a second measurement device, and a method step 3, that corresponds to a measurement of dielectric variables at a high-frequency f2 in an HF measurement device. The high-frequency pref- erably lies in the range between 100 kllz and 300 MHz, particularly prefera- bly between 1 and 10 MHz, with a large separation in frequency to the mi- crowave frequency f's. The value and separation between f1 and f2 are se- lected so that the relative pcrrnittivity of the substances to be measured de-velop markedly distinctly. Using the measurements, the real and imaginary parts of the relative permittivity of the filter material can be inferred at the frequency f2.

Further method steps 4 and 5, for example, can then follow in which the weight of the filter rod, the pressure drop, the hardness or a temperature are measured, for example, wherein in the method steps 2 to 5, the measurement data for the filter rods are supplied to an evaluation device 6. These meas- urement results can be used individually and directly and/or already proc-essed by the evaluation device 6, and also for controlling the endless filter rod production. This is represented by the last open arrow in Fig. 1 beneath the box symbolising method step 5. Thus, further steps according to the in-vention can follow, in particular, control of the endless rod production method.

Fig. 2 shows a microwave resonator according to DIE 198 54 550 B4 that can be used in the method according to the invention and in the device according to the invention. An endless filter rod 11 which is shown as partially broken open and is moved in the direction of the arrow 15 is composed of a filter material 12 and a wrapper of a plug wrap 13, passes through a resonator housing of a measurement device 14, to which microwaves are supplied for the purpose of detecting the mass and/or the moisture and/or a plasticiser -15-percentage.

The resonator housing has a cavity 16 in the shape of a hollow cylinder, the interior 17 of which is symmetrical to the endless filter rod 11. For closure, a cover 18 is screwed to it. The hollow cylindrical hollow body 16 and the cover 18 are preferably composed of a material with a very small coefficient of thermal expansion. This results in good constancy of the geometry of the resonator housing and also good constancy of the measurement results. Al-ternatively or additionally, the temperature of the resonator housing can be controlled wherein the temperature thereof is recorded by a temperature sen-sor 19. A heating transistor 21 or a heat resistor, or another suitable heating means is controlled using the temperature sensor 19. Tn this way, the tem-perature of the housing is maintained substantially constant.

The interior 17 of the resonator housing is preferably vapour coated with a thin layer of gold, which reliably prevents the formation of corrosion that would adversely influence the measurement value consistency.

A protective pipe 23 that is preferably composed of a substance from the polyaryl ether ketone group (PAEK), e.g., of polyether ether ketone (PEEK) serves for the mechanical closure of the interior 17 with respect to the end-less filter rod 1, and from possibly conveyed dirt particles that would impact the measurement results. The protective pipe 23 is widened in the shape of a funnel at one of its ends 23a, at which the endless rod 11 enters into the resonator housing. Outside of the interior 17, the resonator housing extends in a tubular shape (lGa, 18a) outwards on both sides in the direction of the rod 11, in order to prevent the emission of microwaves from the resonator cavity. It can also extend in a tubular shape (lob, 18b) inwards to some ex-tent.

An antenna 26 insulated from the metal housing by an insulating ring 24

I

serves for coupling in the microwaves generated by a microwave generator.

An antenna 28 insulated by an insulator 27 serves to decouple microwaves which are to be supplied to an evaluation circuit, not shown. A suitable evaluation circuit can be found in the document DE 197 34 978 Al, for ex-ample.

Fig. 3 shows a preferred embodiment of an HF measurement device 31 in the form of an HF measurement capacitor. The HF measurement capacitor, which may also be an HF precision capacitor, is constructed substantially rotationally symmetric about the longitudinal axis, that is, the central axis of the endless filter rod 11. The endless filter rod 11 is guided through a central bore hole 32 of the HF measurement capacitor in the transport direction 15 which coincides with the longitudinal direction. The HF measurement ca-pacitor comprises two rotationally symmetric disc-shaped base bodies 33, 34, oriented perpendicular to the longitudinal direction, that are spaced apart from each other by means of an outer annular non-conductive limiting body 35, and which respectively have a central bore hole 32 for the endless filter rod 11.

An electrode 36, 37 of the HF measurement capacitor in the form of a metal- lie surface, for instance a metallic coating, for example by gold vapour depo- sition, is attached to each inner surfaces, oriented perpendicular to the longi-tudinal direction, of the base body 33, 34. The HF measurement capacitor is designed therefore as a plate capacitor with plate-shaped electrodes 36, 37 that are shaped like circular discs and are oriented perpendicular to the lon-gitudinal direction, and have a central bore hole for the endless filter rod 11.

In this arrangement, the field lines run substantially parallel to the direction of transport.

A field-filled cavity 38 is formed between the base bodies 33, 34 and is closed radially outward by the base body 35. The high-frequency field ex- -17 -tends into the central product cavity 39, and interacts there with the endless filter rod 11.

The electrodes 36, 37 have further conductive connections 40, 41 to external electrical connectors. The base bodies 33, 34 each have a tubular extension 42, 43, comprising the endless filter rod, projecting axially outward, having a metallic surface or respectively coating 45 on the inner walls that is expe-diently connected to the electrodes 36, 37.

Furthermore, there is a pipe 44 composed of non-conductive material extend-ing over the entire length of the sensor, directly surrounding and guiding the endless filter rod 11 that prevents contamination of the interior of the sensor by product residues.

The housing parts of the HF measurement device 31 are preferably composed of a non-conductive material with a small coefficient of thermal expansion in order to attain increased shape stability against thermal influences. For the same purpose, there is preferably a control device, not shown, for maintain-ing a constant sensor temperature.

Fig. 4 schematically shows the frequency-dependent progression of the real parts a' of the relative permittivities TT of triacetin and CA of cellulose acetate, as well as the imaginary parts a" of the relative permittivities of C"Tr triaeetin and "CA of cellulose acetate. The horizontal axis shows the logarithm of the frequency in arbitrary units, wherein the frequency f1 in the microwave range and the frequency f2 in the high-frequency range are em-phasised.

In the case of cellulose acetate, there is a constant value of approximately 3.5 for CA over the entire frequency range. In the case of triacetin, at the frequency of 10 MHz for f2, Tr has a value of approximately 7, whilst be-

S

yond that the value decreases, and at the microwave frequency f,, between 1 and 10 GHz for example, has decreased to slightly above the value CA for cellulose acetate. The magnitude of the imaginary part C"CA for cellulose acetate is substantially smaller and is essentially constant in the represented range. The magnitude of the imaginary part s"ç-for triacetin is also smaller than the associated real part, but increases in the represented range. This dif- ferent behaviour of the real and imaginary parts a" of the relative permittivi-ties of triacetin and cellulose acetate is used according to the invention for the purpose of determining the relative percentages by weight thereof in the filter material.

Fig. 5 shows the frequency-dependent progression of the real part a' and the imaginary part a" of the relative permittivity 8 for water at room tempera- ture. The real part a', that has a value of 80 at approx. 1 MHz, remains con-stant at 80 over a broad frequency range, namely approximately up to 1 GHz (1.0 x l0 Hz). Beyond this frequency, the magnitude of the real part a' of the relative permittivity decreases sharply, and approaches the value of 6.0 beyond frequencies of approximately 100 GHz (1.0 x 10" Hz). In the range of the sharp change in the value of the real part a', the imaginary part a" has a maximum at approximately 35.

Whereas the moisture in the filter material is generally low, however, due to the relatively large values of the real and imaginary parts of the relative permittivity a of water in comparison to triacetin and cellulose acetate, they should therefore advantageously be included in the determination of the per-centages by weight of the substances of the filter material.

The exact position of the changeable range and the value of the relative per-mittivity of water depend on the temperature of the water. The fluctuation factor with the typical fluctuations in operating temperature lying in the per-cent range with respect to the values of a' and a", can either influence the -19 -measurement or respectively the determination of the percentages by weight of the substances of the filter rod as a small uncertainty, or can be consid- ered using a temperature measurement. In addition, the temperature depend- ency of the relative permittivities of the other substances is preferably con-sidered as an uncertainty or directly based on a temperature measurement.

All named features, including those taken from the drawings alone, and indi-vidual features, which are disclosed in combination with other features, are considered individually and in combination as essential to the invention.

Embodiments according to the invention can be satisfied through individual characteristics or a combination of several characteristics. -20 -

S

List of referenceyxnbols la endless filter rod conveyance lb filter rod conveyance 2 microwave measurement 3 HF measurement 4 weight measurement pressure drop measurement I la endless filter rod 1 lb filter rod 12 filter material 13 plug wrap 14 microwave measurement device transport direction 16 hollow body lôa, 16b tubular extension 17 interior 18 cover iSa, I Sb tubular extension 19 temperature sensor 21 heating transistor 23 protective pipe 23a end of the protective pipe 24 insulation ring 26 antenna 27 insulation 28 antenna 31 HF measurement device 32 through hole 33, 34 disc-shaped base body limiting body 36, 37 electrode

38 field-filled cavity

39 central production space 40,41 conductive connection 42, 43 extension 44 pipe metallic coating d electrode spacing f1 frequency in the microwave range frequency in the high-frequency range real part of a relative permittivity imaginary part of a relative permittivity Tr real part of a relative permittivity of triacetin C Tr imaginary part of a relative permittivity of triacetin C'CA real part of a relative permittivity of cellulose acetate imaginary part of a relative permittivity of cellulose acetat

Claims

-22 -SCLAJMS1. A method for determining percentages by weight of a plurality of sub-stances in a filter material (12) in the form of a filter rod (1 ib) and/or an endless filter rod (1 la), which is based in particular on cellulose acetate, characterised in that the filter material (12) is conveyed through a micro-wave measurement device (14) and through an HF measurement device (31), wherein at least three measurement variables that are influenced by the quan-tities and the relative permittivities (a) of the substances of the filter material (12) contained in the mixture, are measured at at least one microwave fre-quency (f1) and at at least one high-frequency (f2), wherein percentages by weight of the substances of the filter material (12) are determined from the measured results.
2. The method according to claim 1, characterised in that at least three of the four variables real part (a') and imaginary part (a") of the relative per-mittivity (a) of the filter material (12) at the high-frequency (f2) used, and the real part (a') and the imaginary part (a") of the relative permittivity (a) of the filter material (12) at the microwave frequency (f1) used, arc deter-mined from the measured measurement variables.
3. The method according to claim 1 or 2, characterised in that the at least one microwave frequency (f1) used lies in the range between 1 GHz and 30 GFIz, particularly between 4 GHz and S GHz, and the at least one high- frequency (IT2) used lies in the range between 100 kflz and 300 MHz, particu-larly between 1 MHz and 10 MHz.
4. The method according to one of the claims ito 3, characterised in that the at least three measurement variables are measured in the microwave measurement device (14) and in the HF measurement device (311) with sub- stantially the sane field geometry and/or with substantially the sane meas-urement volumes and/or at substantially the same temperature.
5. The method according to one of the claims I to 4, characterised in that the measurement at the microwave frequency (f1) is measured at the endless filter rod (ha) or at the cut filter rod (lib), wherein the measurement at the high-frequency (f2) is measured at the endless filter rod (1 la) or at the cut filter rod (1 ib).
6. The method according to one of the claims ito 5, characterised in that in addition weight and/or pressure drop and/or diameter and/or hardness and/or temperature of the filter material (12) is or are measured.
7. The method according to claim 6, characterised in that the additional measurement results, in particular weight and/or pressure drop and/or diame-ter and/or hardness and/or temperature of the filter material (12) influence the determination of the percentages by weight of the substances.
8. The method according to one of the claims 1 to 7, characterised in that in addition there is foreign body detection and/or a detection of an additive, particularly a plasticiser, in particular triacetin.
9. The method according to one of the claims ito 8, characterised in that method parameters are changed during the production of the endless filter rod (ii a) depending on changes of the measurement values, in particular ap-plication of the plasticiser and/or climatisation and/or moisturisation of the endless filter rod (1 la).
10. A device for determining percentages by weight of a plurality of sub-stances in a filter material (12) in the form of a filter rod (1 ib) and/or an endless filter rod (11 a) that is based in particular on cellulose acetate, corn- -24 -prising at least two measurement devices (14, 31), a conveyor device for the filter material and an evaluation device, characterised in that the at least two measurement devices (14, 31) comprise a microwave measurement device (14) and an HF measurement device (31), wherein the measurement devices (14, 31) are designed to measure at least three measurement variables that are influenced by the quantities and the relative permittivity (a) of the sub-stances of the filter material (12) contained in the mixture, at at least one microwave frequency (f1) and at at least one high-frequency (f2), wherein the evaluation device is designed to determine percentages by weight of the sub-stances of the filter material (12) from the measurement results.
11. The device according to claim 10, characterised in that they are de-signed to perform a method according to one of the claims 1 to 9.
12. The device according to claim 10 or 11, characterised in that the mi-crowave measurement device (14) and the HF measurement device (31) have substantially the same field geometry and/or substantially the same meas-urement volume.
13. The device according to one of the claims 10 to 12, characterised in that the microwave measurement device (14) is disposed in an endless filter rod machine or in a filter rod measurement station, wherein the HF meas-urement device (31) is disposed in an endless filter rod machine or in a filter rod measurement station.
14. The device according to one of the claims 10 to 13, characterised in that additionally at least one measurement device is comprised for measuring weight and/or pressure drop and/or diameter and/or hardness and/or tempera-ture of the filter material.
15. The device according to one of the claims 10 to 14, charaeterised in that the evaluation device (6) is designed to detect foreign bodies in the filter material (12) based on the measurement results and/or to control a discharge station.
16. A method of determining percentages by weight of a plurality of sub-stances substantially herein described with reference to the drawings.
17. A device for determining percentages by weight of a plurality of sub-stances substantially herein described with reference to the accompanying drawings.