电耳说明书2[1]

ELECTRIC EARINSTALLATION COMMISSIONNING & MAINTENANCE MANUALSUMMARY1 －PURPOSE (3)2 －MILL LEVEL MEASUREMENT PRINCIPLES (4)2.1 －Foreword (4)2.2 －Comparison between the different level methods (5)2.3 －Noise characteristics of mills (7)2.5 －Application of the M.S.C. 2000 on tube mills in direct firing (8)3 －DESCRIPTION OF THE EOUIPMENT (9)3.1 －General presentation (9)3.2 －Microphone (11)3.3 －Audio signal converter ....................................................... 错误！未定义书签。

3.4 －Power supply (12)3.5 －Display (13)4 －INSTALLA TION OF THE EQUIPMENT (13)4.1 －Microphone (13)4.2 －Transmitter (14)5 －PRIMARY TESTS BEFORE START UP (14)5.1 －Installation check (14)5.2 －Audio signal converter (15)6 －START UP CALLBRA TION (15)6.1 －Calibration of the booster (15)6.2 －Calibration of the transmitter (15)7 －NOISE CORRECTION MEASUREMENT (16)8 －SETTING OF THE CONTROL LOOP (17)9 －PERFORMANCE TESTS (18)10 －ANNEXES TO THE MANUAL (18)1 －PURPOSEThe purpose of this present manual is to describe the equipment, Installation and commissioning procedure of the mill noise level system called “ELECTRIC EAR”. This system can substitute other mill level equipments such as deep tubes P system or mill motor power system.2 －MILL LEVEL MEASUREMENT PRINCIPLES2.1 －ForewordA prerequisite for the correct use of horizontal mills (tube mills), for direct firing of pulverized coal, is that the amount of coal inside the mill must be controlled to a "constant level" in order to maintain the air/coal ratio as constant as possible and also to be sure that the grinding is optimum.From the begining of use of this kind of mill several systems have been envisaged (some of them "accrobatic"), but in fact three types of measurement are today under competition．They are :- the measure of the mill motor power consumption- the probe method (deep tubes system)- the measure of the noise emitted by the mill．Each of these measures has its own advantages and it is very interresting to compare the different solutions in order to adopt the best one．In final we have to keep in mind that the answer will be given by the mill grinding behaviour.Just looking from the "outside" we can make a first classification:- The power measurement is not directly in contact with the process, but a failure in the transmitter needs anyway to stop the motor (thus the mill). The power is in fact an indirect measure method．- The probe method is the only direct method of measurement of what happens in the mill; nevertheless the measure is directly in contact with an hostile process (coal dust, turbulent atmosphere，cascading of steel balls. probes plugging risks...).- The noise measure method is not in contact at all with the process but it is also, like the power, an indirect method. It sqhall be noticed that is the only method where the maintenance is very easy and does not require the ginding circuit to be shut down.STEIN INDUSTRIE using at first the Probe method has now developped a sophisticated noise measurement system and the associated level control based on the "electric ear", the generation of this king of equipment is called the M.S.C. 2000 (Mill Sound Computer). We will see further that this measurement can also be used to assess whether a mill is "optimized" or not.2.2 －Comparison between the different level methodsTo better evaluate the advantages of the noise measurement system，it is well to understand how the grinding takes place and to compare the evolution of the different types of measure signals according to the different phenomena which can happen in a mill.Starting from anempty mill andfilling upprogressively thethree signals ofpower, noise anddp given by theprobes levelsystem can be seenin the figures 3,4 and 5.Figure 3Curve power versuscoal amount (L)Figure 4Curve noise versuscoal amount (L)Figure 5Curve delta P versus coal amount (L)It is possible to explain the reason of the shape of each curve given in the figures 3 to5 by looking of what happens inside the mill and mainly between the balls. The figure6 gives four major steps in the mill filling up.There is no coal. Balls are grinding betweeneach other. There is a ball excessive wear.Raw coal is introduced and begings to fill thefree space between the balls. The density ofthe mixture ball + coal increases (Powerbegins to increase and noise smoothes).Grinding is still not efficient and ball wear isimportant.There is an optimal coal filling and thegrinding is good by attrition. Coal acts as alubricant between the balls smoothessignificantly the noise, power begins todecrease as the grinding medium gravitycentre gets closer to the rotating axis.There is now too much coal, balls are now"floating" in the coal. Attrition and grindingare very bad, on the other hand the wear ofballs is very low.Figure 6－Phases of mill filling upFigure 7－Evolutionof power, noise and dpaccording to the steps ofmill filling up2.3 －Noise characteristics of millsThe noise evolution curve in figure 4 gives the noise emitted by the mill versus the coal amount it contains. In comparison with the deep tube signal it can be seen the difference of sensitivity between the two signals.When looking to the noise detected when the mill level is controlled by an other means i.e. the deep tube, it can be seen that the noise changes when load (mill output) changes.This is due to the fact that the noise depends not only of the coal amount in the mill but also of the average granulometry in the mill which changes according to the resident time in the mill.The figure 9 shows that a mill can be considered as a vessel with aFigure 9 －Resident time demonstrationraw coal input and pulverized coal output. For example taking a mill of rated output of 100 t/h in which we have about 17 tons of coal the resident time at full load (full output) is:17 / 100 = 0.17 h that is about 10 minutesAt half load the resident time is the double, it means that the coal has more time to begrinded and therefore the granulometry is lower than at high load.The change of granulometry has an influence on the noise and can be summarized as follows:* when at high loads the particules of coal are bigger and the noise is smoothed,* at low load the coal particules are smaller and the shocks between the balls give more noise.The figure 10 gives a rough idea of the evolution of the noise when the mill load changes.Figure 10 －Dependance of noise versus mill outputAs a demonstration of this influence of the mill output on the noise the figure 11 here below gives a chart record showing this phenomenon.Figure 11 －Recording noise versus mill output2.5 －Application of the M.S.C. 2000 on tube mills in direct firingThe control of the "level" on a mill in direct firing using the noise emitted by the mill is very interesting as it is a solution giving more flexibility than the other means such as dp or power (see paragraph 2.2).As said before the noise signal is representative of the coal amount in the mill but it isalso influenced by the mill output. The figure 13 shows the family of curves of the noise signal versus the coal in the mill the parameter being the mill output.Figure 13 －Noise characteristic curveslIn order to control correctly the coal 1evel in the mill it is then necessary to compensate the noise in fonction of the mill load. For this purpose the good parameter to be used is to take the primary air flow through the mill which represents the coal output.3 －DESCRIPTION OF THE EOUIPMENT3.1 －General presentationTo obtain an audio signal which is representative of the coal amount in the mill andwhich remains reproducible and accurate, STEIN INDUSTRIE has developped an audio signal converter whose basic structure is given in figure l1. The audio signal from the microphone is processed as follows:-Initial wide audio bandwidth preamplification to bring the millivolts coming from the microphone into an acceptable level to operate the filter.-Filtration by a switched-capacity numerical filter where central frequency and bandwidth are adjustable. The setting of this filter is made in laboratory and does not need to be touched during mill commissioning.-Final amplification and impedance adaptation before rectifying.-Rectifying of the signal from AC sonic compounds into DC smoothed signal db．-Linearization of the signal to increase the sensitivity of the signal in the operating control area．- A potentiometer can adjust the output current reading 20 mA when there is no coal.- A data sampling circuit can be used for making analogy signal digitalization.- A microprocessor can analyze and deal with the sound signal.-An current signal generator of 4 –20 mA can be used for displaying the coal amount of mill.The M.S.C. 2000 noise transmitter is presented in standard racks 19 inches, including the following modules:- A power supply accepting AC v01tages 220 V ±20% (50 or 60 Hz).-The audio signal converter-An LCD digital display giving the output signal from the converter or the isolating module with a switch allowing a display either directly in mA or in percent.The figure 12 here below gives the presentation of the rack front.Such equipment as presented delivers a signal suitable for many applications such as: -The monitoring only: in this case the signal is used for recording, for indication and for alarm threshold relays (very high level or very low level)-The control of the level in a constant flow mill: this is the case of a pulverizing plant for indirect firing or other purposes. In that case the signal delivered by the noise transmitter is used directly in a conventionnal controller-The control of the level in a tube mill in direct firing by using the noise signal available from the MSC 2000 through a load correction signal (pA flow) in order to deliver a “corrected noise” which can be considered then as a“1evel”of Coal in the mill．3.2 －Microphone- Reference : EN 5463A- Typical impedance : 8 ohms- Output signal(mill empty of coal with balls only) : 3 mV- Outline dimensions : see figure 13Figure 13 －microphone3.3 －Power supply- Reference : EN 5809 A- size : European 3-Input voltage : 220 V ±20% (AC)- Output regulated voltage : + 32 V ; +/- 15 V ; + 5 V (DC)- Max rated output current : 400 mA- Mains consumption : 20 V A- Presentation : see figure 153.4 －DisplayThis display allows to read the output signal coming from the noise transmitter that is the audio signal converter.- Reference : EN 5658 A- size : European 3- Display : LCD- Display mode : % or mA- Display digits : 3 1/2 + dp- Presentation : see figure 16Figure 16 －Output display module front view4 －INSTALLATION OF THE EQUIPMENT4.1 －MicrophoneThe microphone shall be placed close to the mill as defined on the drawing in annex 1．It must be installed as close as possible to the mill shell with a sufficient gap to avoid it to be touched or destroyed by the bolts which will pass close to it．The cabling between the microphone and the transmitter shall be made with shielded cable．On the microphone side the shield of the cable must not be connected and the two wires of the cables are to be connected on the terminals 1 and 3 of the microphone plug (See cabling details on figure 13)．Before inserting the plug on the microphone please check that the weldings of the cables are correct by made, “dry weldings” giving a risk of bad contacts even inoportune disconnection while mill running．The microphone shall be fixed strongly on a post or a support．It shall be ensured that the front face Of the microphone is a little bit inclined facing the mill shell but hot too much inclined to avoid pulverized coal deposits on that face．4.2 －TransmitterThe transmitter is to be mounted near the mill in a suitable place far away from risks of shocks．It can be fixed on a wall or on a post.The access for wiring, cabling, commisioning and maintenance is made by the front door．The cabling is shown in annex 2.The mains ( using three wires) is connected to the terminal plugs ref. 1, 2 and 3. The plug 3 corresponds to the earth connection and must be always wired．The micropohone signal coming with a shielded cable is linked to terminal plugs 6 and 7．If the cable is a twisted pair + shield，the shield must be connected together with the wire on the terminal 7 and in that case the shield must be free on the microphone side. In case of a coax. the wiring must be as shown on the drawing．5 －PRIMARY TESTS BEFORE START UP5.1 －Installation checkFirst ensure that the equipments involved microphone, transmitters,… are correctly installed and wired．5.2 －Audio signal converterBefore any energizing of the assembly, it is vital to check the configuration of the audio signal converter PCB.6 －START UP CALLBRATION●Start the mill main motor●If the mill was filled with coal previously run the mill blowing through primaryair to remove the coal during a suficient time in order to ensure that the mill is empty of coal. If not, pass to the next step after a waiting period of l minute.6.1 －Calibration of the booster●Check that the power supply EN 5822 is energized (LED D2 on this module shalllit．●Adjust the VR potentiometer and make the LCD indicator reading zero.Reason of this operation: this adjustment allows to get the maximum signal but remaining at the limit of the saturation of the signal.6.2 －Calibration of the transmittera.Switch on the power supply on the rackb.Turn potentiometer P1 “Input level” to lit the LED “Input 1evel correct”, if the signal is too strong the LED “Input level saturation” can also lit. In this case turn slowly back to switch off this LED, just the LED “Input level correct” shall lit.c.Switch the display EN 5658 A on position “mA”.d. Turn slowly the potentiometer P4 “Calibration level” in order to read 20.0 mA on the display. Than wait few minutes just to make sure that the reading stays constant．If needed readjust with smooth actions on potentiometer P4REMARK: In case the microphone is not correctly located or if the cable length between the microphone and the audio converter board is too important, itcould be that the actio n of p1 is not enough to lit the “input level correct”LED, then give more signal on the interface EN5819 A module by turningone step more the “interface output level” switch (position 1 to 2 forexample).7 －NOISE CORRECTION MEASUREMENTa)Perform the filling of the mill in accordance with the operating instruct ionsb)Stabilize the mill at minimum load (the measurements of pressures temperatures and feeder speeds must be constant)c)Take e reading in the control room and complete the reading sheet for the mill in question (sea annex 4)d)Increase the air flow by one step (mill load)and wait for the mill to stabilize (15 minutes minimum), then take a reading. Repeat this operation up to maximum load.e)When the mill load is maximum, perform the same tests but decreasingf)Calculate the function generator as defined below:-calculate the scale of sound evolution as a function of load as defined in the example below：Calculation of F (x) withlevel setpoint at 22.75 ％Sound scale：70％The corrected sound must be constant (22.75％)，whatever the load，thus Corrected sound = Raw sound xKScale of F(n) with level setpoint at 22.75％Attention，this example is given for information purposes in order to facilitate the understanding of the sound correction curve. It is quite clear that the results obtained will be totally different.In order to find the max signal with the mill empty and a primary air flow quite nul it is necessary to fix a K = l for air flow below 30％. This 10w 1imit of air flow of 30 % is given as an example；for a better reading it is always recommended to fix the bottom 1imit a 1ittle bit lower than the maximum operational turn down in order to help more the operators during the mill filling up on start up or emptying during shutdown．It is also possible to built the curve of the noise according to mill load (see annex5)and then to establish the correction curve as shown in annex 6 an order to maintaina compensated noise independant of the load．As already said above these curves shown are only examples．For a tube mill under commissioning these curve shall be established according to this method．8 －SETTING OF THE CONTROL LOOPWhen the correction of noise according to the mill out put is established then the control loop for the mill level auto control using the electric ear shall be introduced in the analog control loop system.The first step is therefore to establish the compensation of the noise.Then the procedure consists to check that the corrected noise is nicely independant of the mill output by making some load charges using an other method to maintain the level．Finally to place the level control on auto with the electric ear.9 －PERFORMANCE TESTSOn commissioning of the electric ear under auto-control the following tests shall be made in order to ensure that the control system using the noise works properly．-Test of maintaing the correct level at constant load. loads between minimum and maximum (make at least three points i.e. 40％,70％, 100％). duration of each test > 1 hour under steady conditions10 －ANNEXES TO THE MANUAL。