EP1346491A1 - Controlling a processing plant comprising one or more objects - Google Patents

Abstract

Controlling a process plant comprising one or more objects where the objects comprise means for receiving standard codes with associated parameter value via the data communication means and for changing a parameter in the object, e.g. the real value of the process parameter, the goal for the process parameter or an alarm condition, the object furthermore comprises means for sending standard codes with associated parameter value via the data communication means reflecting a parameter in the object, e.g. the real value of the process parameter, the goal for the process parameter or alarm condition.

Description

Controlling a processing plant comprising one or more objects

The present invention concerns a method for controlling a process plant including one or more objects.

Prior art:

Controlling processes and plants have particularly been developed in the last century. The requirement of rationalisation drives this development, and humans are substituted with automatic equipment with integrated control when this pays.

In the beginning, control was individual and with feedback from the human senses which interfered if any undesirable situations arose.

With electronics, the control became more advanced, and controls were built for the individual task/tasks. This meant that there was almost no flexibility, and extensions or changes implied a new control or a total rebuilding.

With the computer, one got flexibility and easy access to extensions. In the case of control units, the computer is viewed in two ways: As a control adapted to a certain industry but with flexibility within the industry, or as a universal control expressed in the PLC.

Over time, the control jobs have become larger, caused by energy saving and environment improving requirements. Just think of e.g. a car.

Almost all control tasks contain controlling electrical power many times greater than those handled by the computer. Therefore, a control (e.g. PLC) almost always appears put together with an switchboard providing for controlling large powers and providing possibility of manual interference via switches and control lamps. Between the control and the switchboard there are typically a lot of control wires. When control unit and switchboard are installed on the site, the electrician has to connect a lot of wires with both control signals and power for the external components. For this, a comprehensive documentation is required, which most often has to be made individually for each single task. This is also the case for the program read into the PLC which therefore is different almost every time.

Switchboards are still made by hand, and with the great increase in the cost of trained workers, the switchboard is clearly the most expensive part compared with the control unit itself which has experienced strongly reduced prices through the years.

Many of the control signals below require great speed in order for the process to run, and this makes demands on the control program which in some cases does not have enough hardware for the job. For the local electrician it is almost impossible to service this type of facility today. He cannot take the control unit apart into smaller pieces and test them without communicating with the PLC or the like. By this, there is a great risk that he accidentally changes important control parameters, and therefore he prefers not to be engaged with the matter.

The situation becomes worse as more and more demands are made to the facilities and the processes. For a business exporting to less developed countries, these problems are very serious, and it is often necessary with several solutions with different technological level.

In general, the solution and the invention consist in one or more objects connected to an operating face. All these elements together with examples of objects will be explained below.

The invention differs from the prior art by comprising a method which is peculiar in including one or more objects, where for each object there is a pre-defined parameter depending on the function of the object, in which:

- the object controls the process parameter so that it falls within the predefined parameter, - the predefined parameter is communicated to the object as a value which by data processing on the object is translated to a control parameter,

- a control parameter is communicated between a control unit and one or more objects by data communication on a data bus, - the control parameter is structured as a signal containing objectaddress.code.data- type.data.check. wherein:

- objectaddress defines the object concerned,

- code defines the object function concerned,

- datatype defines whether they are control data, object data or log data, - data is the actual control parameter, e.g. temperature, moisture or parameter inquiry,

- check is a check signal,

- when the control parameter is parameter inquiry, the object communicates the actual parameter value to the control unit on the data bus.

Hereby is achieved that the controlling of single process parameters is decentralised. After the object having received the predefined parameter or parameters, it controls the process parameter by itself. This makes it simple to provide for sufficiently powerful hardware as an object is only to control a limited number of process parameters. Furthermore, it is simple to change the goal for the process parameter as this only requires that the control unit communicates a new parameter to the object on the data bus. The use of data bus limits the wiring as the same cable may be used for communication with a large number of objects. Expanding the plant with new objects is a simple operation as new objects are only connected to the existing data bus. Even thoughh the control of the single process parameters is decentralised, it is easy to get an overview of the different process parameters by letting the control unit inquire to the desired objects about the actual parameters.

An embodiment of the invention comprises a method which furthermore includes that the predefined parameter is communicated to the object as a value which by data processing on the object is translated to a control parameter between 1% and 100%. This serves to standardising the codes so that the communication takes place in a uniform way whereby connecting different objects is facilitated.

An embodiment of the invention comprises a method furthermore comprising one or more objects communicating control parameter on the data bus to at least one other object.

Hereby, process parameters that are to be regulated in relation to each other, may be readily controlled correctly without involving a central control, as for example the operation panel, by letting one object transmit a control parameter to the object controlling the dependent process parameter. An object may also by itself inquire about e.g. an independent process parameter at another object, and change its own process parameter depending on the reply.

An embodiment of the invention comprises a method further including communication without loss of data on the data bus with collision, where

- the data bus is initiated by providing it with a first predefined voltage,

- a transmitting object sends data on the data bus by attempting to give the data bus a desired voltage level by, in the case of a first type of bit, e.g. 0, attempting to draw the voltage level on the data bus to another predefined voltage level, and in the case of a second type of bit, e.g. 1, to let the data bus keep the first predefined voltage level,

- the transmitting object measures the voltage of the data bus in order to verify whether the voltage level is at the desired level,

- the transmitting object continues transmission if the desired voltage level is present on the data bus, but discontinues the transmission if the desired voltage level is not present on the data bus.

This method is advantageous by increasing the effective band width on the data bus. Even thoughh two objects try to transmit simultaneously, the object lastly transmitting a bit causing the voltage level of the data bus to deviate from that previously defined will be able to complete the transmission without any break or delay. The other object will only try to transmit a little later. An embodiment of the invention comprises a method, which further includes handling of alarm in a control of a process plant comprising one or more objects where:

- an alarm object makes an inquiry to other objects about the real size of a process parameter,

- the alarm object compares the returned value with a predefined alarm limit,

- if the alarm limit is reached, the alarm object issues an emergency order to the object in question.

An independent alarm object may thus operate independently of the single modules, so that even a total breakdown of one or more objects will not prevent the alarm signal to be detected. If the alarm limit is reached for a given process parameter, the alarm object communicates an emergency order to the object concerned. The alarm object is thus not directly involved in the controlling of the single process parameters, but leaves to each single object to correct the process parameter by transmitting an emergency order. The alarm object may advantageously be coupled to signal transmitters so that manual intervention may be ordered. Several alarm objects may be coupled for ensuring alarm even thoughh one alarm object is out of service.

An embodiment of the invention includes a method further comprising power supply to objects with control lamps in a process plant wherein:

- the object is supplied a voltage which is considerably greater than necessary for driving a processor in the object,

- a control lamp is placed with the possibility of being short-circuited by an electronic switch, e.g. a transistor, in series with the processor of the object in the part of the circuit where the voltage is substantially greater than necessary for driving the processor of the object,

- the object turns on the control lamp by opening the electronic switch,

- the object turns off the control lamp by closing the electronic switch.

This method is particularly advantageous when there are several objects on the same data bus as it reduces the power consumption by using control lamps. Other power consuming units may advantageously be supplied in the same way as the control lamps. By using a higher voltage for the object, it will also be possible to use a higher voltage on the data bus which makes the communication less noise sensitive.

An embodiment of the invention differs from the prior art by including an object for a process plant, where the object includes a processor, data communication means, inputs and/or outputs for reading and controlling a process parameter and a predefined object address.

An embodiment of the invention differs from the prior art by including an object including means for receiving standard codes with associated parameter value via the data communication means and for changing a parameter in the object, e.g. the real value of the process parameter, the goal for the process parameter or an alarm condition.

An embodiment of the invention differs from the prior art by including an object which includes means for transmitting standard codes with associated parameter value via the data communication means reflecting a parameter in the object, e.g. the real value of the process parameter, the goal for the process parameter or alarm condition.

By using objects for controlling, a high degree of decentralised control is achieved. Furthermore, it will be simple to add new objects to the plant. Even objects for controlling process parameters that are unknown presently may easily be integrated in an existing plant as the objects communicate with standard codes. An object may also control other objects by communicating a suitable standard code with associated parameter value. Thus it will also be simple to integrate new objects at any time which either by themselves control independent process parameters or control dependent process parameters. It will be simple to ensure sufficient processor capacity for controlling each single process parameter as this only requires that the selected processor has sufficient capacity for controlling the relatively few process parameters connected to an object. The processor may e.g. be an mcu, micro controller. The algorithm itself for controlling the specific process parameter is implemented in software so that the same kind of hardware may be used for controlling quite different process parameters.

An embodiment of the invention differs from prior art by comprising an object which is supplied with a voltage that is substantially higher than necessary for driving the processor of the object, and where the object includes a control lamp supplied with power in series with the processor of the object.

Hereby is achieved that the power consumption of the object is largely not dependent on possible control lamps. This is particularly suitable as it is desirable to carry conductors for data communication, alarm condition and power supply jointly. Since the cable is preferably to have as small dimension as possible with regard to price and ease of assembly, this will demand a low current in the cable. Furthermore, the possibility of communicating with relatively high voltages on the data bus is achieved, whereby the noise sensitivity is reduced.

An embodiment of the invention differs from the prior art by including an object where the object includes an output for power supply to a unit reading or affecting the process parameter.

If this unit has a sufficiently small power consumption, such an output will make it very simple to install new units for controlling process parameters. The cabling will be reduced to the cable for the object, as there will be no need for separate power supply cables.

An embodiment of the invention differs from prior art by comprising an object where the object comprises an alarm input, means for preventing change of a parameter in the object when the alarm input is in a pre-defined state.

Hereby, unintended changes of an object parameter may be avoided. An unintended change may e.g. occur by a manual mistake or a noise pulse. The alarm input is connected to a separate conductor the condition of which may be changed. Then is will be suitable that the condition only can be changed by direct intervention from the user so that mistakes in software, processor or other electronics are not given any possibility of changing critical parameters.

An embodiment of the invention differs from the prior art by comprising an operation object, the operation object comprising a processor, data communication means and an operation panel. The operation object comprises means for manually selecting and transmitting standard codes with associated parameter value via the data communication means, and for receiving and displaying received standard codes with associated parameter values on the operation panel.

The operation object shares many properties with the object which will make cheaper the production and provide many of the same advantages that are associated with the object, e.g. easy integration and cabling. With the operation panel is selected both the object or objects to be communicated with and different parameters from the objects are transmitted or received. All the objects may be operated from the operation object so that a total overview from a single place over the whole control of the process may be provided.

An embodiment of the invention differs from prior art by including an operation object where the operation object is supplied with a voltage, which is substantially higher than necessary for driving the processor of the object, and where the object comprises one or more control lamps supplied with power in series with the processor of the object.

Hereby are achieved the same advantages as mentioned above for the object with regard to power saving and noise immune data communication.

An embodiment of the invention comprises use of an operation object as described above for controlling one or more objects as described above according to a method as described above. Industry Analysis:

The problem may be solved by assuming an object oriented view. This means that within a industry it is mapped which functions there are to be in order that a control may function. Examples of functions are: • Start/stop/speed and monitoring of a belt conveyor

• Start/stop/speed and monitoring of a pump for cooling a coolant

• Start/stop/speed and monitoring of a ventilator

• Positioning of a spindle or cylinder

• Measuring a temperature • Measuring a contamination

Functions and Objects:

When the basic functions are found, one may develop and build units fulfilling the function. Such a unit is called an object. The object thus performs the desired function on the basis of the properties, methods and events belonging to the object, and their settings. Thus there may very well be several objects for the same function where some are more able than others. However, there is a certain minimum which the object must be capable of. The control may now perform its task by giving general instructions to the associated objects which now by themselves provide for their execution. Instructions are always delivered as a value between 0 and 100 %, even also when an associated object only has ON/OFF properties. In this way, the control is always to give the same instructions irrespectively of the kind of object associated therewith.

Example: An example will illustrate this. A belt conveyor is to run with a certain speed depending on the previous belt. If too much comes on the belt, the speed is reduced and compensation is to be made. Start and stop of the belt has to occur softly with certain periods of time. In case of failure and overload, alarm is to be set off, and an alarm lamp is to flash at the belt. There is a minimum and a maximum speed for the belt out of con- sideration to the material. With the present solution, one will put in some control routines in the PLC ensuring correct speed of the belt. The speed control of the belt may be with frequency control, with DC motor or with a mechanical variator. They all require different PLC programs, control signals and different structure of the switchboard. There will be a speed sensor on the belt which is connected to the PLC via the switchboard. There will be a connection from the contactor of the switchboard and to the motor of the belt conveyor. There will be a connection so the alarm lamp may flash.

With an object oriented solution, an expression for the belt speed is defined in general. The control transmits the desired speed, and the object conveyor belt attends to maintaining this speed. Min and max values are located in the object itself, and so does the speed control and the alarm lamp. The only connection between the control and the object thus become a data line. On this the control regularly transmits the speed requirement via a certain code. If the control wants to know the actual speed, or another of the control parameters, it may ask via certain codes. These codes are common to all object of the class belt conveyors. This is also the case with the data format.

In case of operational problems, the operator or the electrician may communicate with each single object via the operation panel of the control. He may set the object to manual mode and set the speed to a certain value. Now he can check whether the object reacts correctly. If this is the case, he has to find the problem in another object. If it is not the case, he may concentrate on finding the failure in that object. Note that he has not operated the PLC yet. He has only selected one object and tested its reaction to a control signal.

As most conveyor belts have about the same requirements irrespective of the industry concerned, most of the codes and values will coincide. This means that if you have learned to know a conveyor belt, you may know them all irrespectively of their manufacturer and size. In the above case, control signal and codes will be identical irre- spectively which regulating principle is used (frequency control, DC motor or mechanical variator). Now, it is also possible to make a complet guide for the conveyor belt, including the possibilities it may have. One may envisage a conveyor belt which cannot be speed controlled but may be controlled by intervals, so that the average speed become as desired. This will not change the control signal from the control, but there will be other codes to be set and other limitations in these settings.

If two conveyor belts are to run in parallel with the same speed, they are just given the same control signal (^function). If one is contingent on the running of the other, the control has to inquire concurrently on the speed of the first, and if it is not running, then set the control signal for the second to stop.

The Control Panel:

From the example it is seen that the control only has to take care of the general and the co-ordination matters. The objects are taking care of all "technicalities". This makes the control program much smaller. Therefore, it is much easier to write and to test it.

This means faster "time to market" when new demands arise. The solution to a new demand from the market may also be a new object, or just expanding with a new method, property or event for an existing object. The changed object may still be used instead of the older objects as it just has something added. This reduces the spare part stock. Finally, an object many years old may still be used in a new control as long as its properties etc. suffice. Recycling is therefore realistic with the environmental advantages implied therein.

The control may become a control panel and the object in the following. The control panel itself consists of some buttons, some display and a couple of data lines. This makes it cheap, and the customer may pay for having a spare part in stock. Also, it is easy to make panels with another outlook, so that individual firms may differentiate on outlook and possibly operation. The real know-how is now located in the settings used in the object for the different tasks. Integrated Objects:

When a conveyor belt object is built, it will natural to make is completely finished so that it only has to be connected to the data line and to electric power. This means that the switchboard has disappeared. Instead, there are a few decentralised boards with fuses. Motor contactor and protective circuit switch switch are integrated in the control box for the conveyor belt. Hereby, one may also from the control enquire about their condition via certain codes.

Discrete Objects: In some cases, the object wished to be viewed is not available. The object may then be build the object with other objects. This will most often require that the objects have an inherent possibility of linearising. It may be a table converting the control signal from outside to another internal control signal. If a spindle motor e.g. has to be on 10% when the control signal is 30%, this has to be indicated in a table or a formula for a graph. The table is the most user friendly. An example illustrates this:

There is need for exhaust that changes its air volume from 0 to 100% through a combination of rotational speed regulation of the fan and regulation of a control damper. Two objects are connected to the same function, and thereby they receive the same control signal. There is used rotational speed regulation till 55% air output, and damper regulation from 55 to 0%. The table may also be used for correcting possible warped characteristics. Examples of tables appear in table 1.

Documentation: The documentation for such a plant with maybe scores of conveyor belts etc. are now limited to a cable plan with indication of a data line to all objects and the necessary electric supply (Phase, Nolt and Ampere). The local electrician may easily provide this, and he can also dimension how many and how large fuses he wants to use. In some cases, he may reuse old fuses and electric systems. There is no need for an elec- trie diagram for the electric switchboard or a diagram for connections to external parts of the switchboard, as control wires are not present, and the objects are associated to the different functions via a number. Table 2- 4 show examples of code overviews, here conveyor belts and a crusher. Values which are determined at the running-in, are here shown with another print type. The table is part of the documentation which the maker of the conveyor belt provides. All code oversights are gathered in one section in the documentation binder. Besides that, the maker of the conveyor belt provides an electric connection as shown in Fig. 1. All electric connections are gathered in a section in the documentation binder. The coupling of the objects is shown in a cable diagram which also contains suggestions for object numbers. Fig. 2 shows an example of this. Finally, the simple control pro- gram is added to the control unit (the PLC) in the binder. The documentation is then complete. The work with this is a fraction of the present work where electric diagrams and switchboards are drawn up through many days.

Operation: The operation interface is to have minimum the following possibilities (buttons):

Code Here is chosen the code for which data is requested

Function Here is chosen the function for which data is requested

Object Here is chosen the object with the unequivocal number received by the object in this control unit

Data Data for the selected object and code may here be changed

Section Here the section/zone, where the function or the object is located, is selected. If only one section exists, the button may be omitted.

The button may also be fields on a display where e.g. a mouse may activate the field.

Code and function may also be translated to clear text if the display allows this. Besides the above, there are buttons for changing values, e.g. + and - buttons, or numerical keyboard. Fig. 3 shows an example of a control forming part of object oriented control.

If it is desired e.g. to know how the protective circuit switch in the crusher is set, the code oversight for the crusher (Table 4) is found. At the top it is seen that it has object number 8.3. Now, the object button is pushed and with + and - buttons until right Data display shows 8.3 (see Fig. 3). From the code oversight is seen that the protective circuit switch current is code 68. Now, the Code key and the + and - keys are pressed until Code display shows 68. The Data display to the right will now show the actual value. If valid data do not appear, the selected object does not have this code. If no data comes, the object or the data network is faulty.

Manual Operation:

The system also enables an improved possibility of manual control. In the present systems, one may choose operational state on the switchboard (aut-0-man). Often there is a safety switch on the conveyor belt itself, but it is not possible to choose between aut and man operation as well as resetting a triggered protective circuit switch is not possible. All possibilities are present on the object itself but apart from that, one may choose and control the object via the control.

New Objects:

In hindsight appears the remarkable fact that one may add a new object to the control without the control knowing this initially. This is due to the standardised structure with Object-Code-Data structure. The control may be used for reading and changing values. This is also the case with the alarm and data objects as these are operating on the basis of the provided object numbers and code numbers. It is also seen, that a control panel without any control program temporarily may be inserted instead of the normal control panel, and with this fault finding in all objects may occur via check of all code settings and manual control. This effectively prevents that a service electrician changes settings in the normal control panel by accident.

Objects in Plural Levels:

In very large control jobs, it will be possible to stratify so that a control unit with all its objects may be viewed from the outside as one object as a part in a greater whole. Here, the codes will only show general data for this (composite) object. The composite objects are, however, to fulfil the requirements for industry selection and a minimum of properties, methods and events. Overview:

As accessory, a unit showing all objects and their actual value and state of operation (automatic or manual) may be made. The unit may also be portable so that it may be take along to an office or out to one of the objects. The same unit may be used in all control systems as it only shows the created objects and associated functions and their values. The unit may also be made so that it sends data to teletext pages of a TV. The user may then, with the remote control (and during the commercials), see how the condition of the control is.

In the above, the invention has been described in a general way. In the following, it will be describe in more detail with reference to the Figures.

General Functions Communication

Structure

As described above, it is a requirement that the control can communicate with the objects. This requires intelligence in the objects and a data line to communicate on. The intelligence is a cost increase compared with traditional analogue technique when speaking of simple sensors or 0-10V controlled motors. Therefore, it is significant that the requirements to the data line are small. Due to the structure of the functioning, the requirement to speed is reduced. However, it is still difficult to find faults in a data bus that maybe extend over many hundreds of meters. It may be done more easily by divi- sion into two: Input data and output data. Data coming from sensors are input data, and data for e.g. a spindle motor or belt conveyor are output. Hereby, one may perform fault finding by testing firstly whether all input are correctly present in the control. If this is the case, the fault is to be found in the output data line. This means two independent data lines called D- and D₀. 3 Service Lamps

As a further aid in fault finding, 3 control lamps Line, Send and Error are used. A green lamp is lit if the line is high, i.e. if the line has connection and there is voltage on and will therefor flash when there is traffic on the line. A yellow lamp is lit when the objects transmits data. If data are not correctly transmitted, the object will resend within a short period, and a highly flashing yellow lamp will therefore indicate that there are problems with discharging data. A red warning lamp is lit when fault exists. If the data line is low over a certain time, the error lamp is lit as a sign of the object is supplied with electric current but is without any connection with the other objects. If there are several objects with the same number, the Error lamp flashes rhythmically over a certain period of time. If the Error lamp is lit and if there is communication, one may ask for the cause via an operating panel.

Integrated Supply By simplifying the work of the electrician, 24 N is to be supplied together with the data line. Power supply to objects with small power consumption may advantageously be made centrally, and an uncomplicated voltage is 24N DC. There will be a voltage drop in the return line of several volts, normally requiring a 2-tiered balanced data line. If, however, a voltage fluctuation of 24V is used on the data line, it will be im- mune to voltage drops up to 12N in the return wire. The drawback with 24N fluctuation is a greater energy consumption if the driving impedance is to be e.g. 50 Ohms. With the low speed on the data line, this provides possibility for the drive impedance not to be equal to the impedance characteristic to the cable. A value of 2200 Ohm will be a fair compromise. With regard to noise considerations, the cable should be shielded.

n:n Network with Collision without Loss of Data

In all bus-oriented systems, collision will occur, i.e. two objects transmitting at the same time. This will destroy the transmission, and the receivers will recognise that check sum or frame is not correct and therefore will ignore the data received. It is waste of time to be forced to discard a transmission. If instead all objects are synchro- nised and check is made on bit level, it may be achieved that data, even by collision, are not wasted.

This occurs in that all objects may draw the data line low. If two or more objects si- multaneously draw the line low, the line continues to be low. If there is an object drawing the line low while the other transmitting object does not drawing low because it desires to transmit high, there is a conflict. But, if all objects are made to check if the transmitted level is also present on the data line when the level is received, the object may see if a disturbance has occurred. Is this the case, these objects stop the transmission, and the only object may thus finish the transmission without being defective. However, this demands that all transmit simultaneously, and this is achieved by all synchronising their transmission time at any initiation of a reception. Transmitting and receiving timers may also not be independent which is normal. The principle also means that the built in transmitting and receiving modules in microcomputers cannot be used but reception and transmission may occur as a part of the software.

Protocol

The format on the data line may be indicated as object.code.datatype.data.check. As datatype is seen control data, log data and object data. Control data may be divided into order or inquiry/reply. This will require two byte for 4 combinations. If 2 byte are used for data+data type, the largest data then become 2¹⁶/2²-l = 16383. If the largest data are defined to a lower number, e.g. 15103, with the data value it may be shown whether the received reply is a reply (data is under 15103) or an inquiry (data are over 15103 = no data). In this way it is achieved that an object can check whether there are objects with the same number on the data line. If an object recognises a reply with the same object number as itself, there are more with the same number. The object can. react as described under 3 Service Lamps.

Object number 0 is reserved for broadcast, meaning that all have to react to this. Power Generator as Pull-Up

As data signals with a large voltage fluctuation (24N) is used, and excess energy consumption is unwanted for communication, it is very limited how small the pull-up resistance on the data line can be. A cable has a certain capacity relative to earth, and this will limit the range of the data line as the voltage does not become high within the time for one bit. The voltage proceeds according to the formula v(t) = N_s x (l-e^"Vτ), where V_s is the output voltage and τ is the time constant for pull-up and cable capacity. If a power generator is used as pull-up in place of a resistance, the high voltage is achieved more rapidly. An numerical example illustrates this:

Output voltage is 24N and the capacity is 0.5μF. As pull-up is used 2400 Ohm, why τ is 2400 x 0.5 μ = 1.2 ms. The current is thus 10 mA at low, and the power generator is given the same value. For the power generator it applies that v(t) = 1/C x i x t, where C is capacity and i is current. The voltage 22N is thus reached after 22V x 0.5μF/10mA = 1.1 ms. For the resistance applies t = 1.2 ms x -ln(l-22/24) = 3 ms.

The solution with the power generator may thus handle almost 3 times as long cables as with a pull-up resistance. Note that usual pull-up resistances may be used until the cable network become extended so much that problems arise. Then a unity with a power generator may be inserted to solve the problem.

Plural Power Supplies

In large plants, several power supplies may be inserted, either in parallel or decentralised. If there is e.g. 300 m between two groups of damper motors, one may insert a power supply at each group.

If the power supply is equipped with two outputs, each with a fuse, and the 24V supply of the objects is divided into two, it is ensured that a short-circuit in an object or a cable does not remove the current to all objects. Overview Data Line

The data line (first version called the VE bus) therefore consists of following elements: a shielded cable with 4 conductors. One of the conductors is thicker than the others. There is +24 V to carry the large current. The 3 thin conductor are named Di, Do and Al and function as indicated above. The display has a large copper cross- section and is therefore used a common minus/return. In order to facilitate the work of the electrician, there are used different colours with a symbolic value:

+ 24V red known from accumulators Di white there is an i in the word white

Do orange there is an o in the word orange

Al black there is an a in the word black

Code/Data Access All the objects have a code/data table. All properties and methods are associated with a certain code. Upon enquiring on the object with a code, the object returns object data corresponding to the actual value of the property or the method. This also means that any apparatus that may transmit data as described under the protocol, may set and read an object, both present and possibly new future ones with new functions.

Security Systems

Most control systems are to be monitored by an independent alarm system. This will normally imply that extra cables are to be laid out for collecting these informations. However, it is possible to make a 100% independent alarm system by inserting an ex- tra conductor in the communication cable between the objects. The conductor is called

Al. The alarm system keeps this wire on 24 V whereby all alarm settings are locked. In order to change an alarm setting, Al is to be kept low while the value is read by the object affected. Is the value low, this is a sign that changes are allowed.

Al may be brought to low in several ways. The operating units may be designed so that pressing the button for data change results in a low Al line. This has to occur around the MCU in order to avoid that it can do this by mistake. On Fig. 4 is shown a diagram, e.g. of a circuit for controlling Al. Here it is seen that there is connection from the Data button to the inverter 40106. When data button is pushed, BC337 receives current via the resistance of 22k. The current is amplified, and the Al line is drawn low. When the Data button is released, the resistance of 2M2 performs a time delay together with the condensator 0.33u. The time delay is to be large enough that the order can be processed by the relevant object in time.

Another possibility is to insert a key switch. When the key is turned, the Al line is laid to 0 Volt, and there may be recorded in all alarm objects. When the change has been performed, the switch is turned back to neutral position, and now changes cannot occur. The key switch may also have third position where it lays the Al line to + 24 V. Hereby changes in alarm objects cannot occur, not even via the operating units.

Common Minus/Return May Give Rise to Problems Using common minus/return may give rise to problems, if the objects are connected to different earth conductors with different voltage. In mountain areas where it is difficult to get earth connection, the neutral wire is often used as protective conductor. The problem may be due to the occurrence of different voltage. If they are now connected with a relatively small cable, there will run a strong current in the return wire. This may imply disturbances in the communication. Therefore, the objects and the power supplies are thus adapted so that they are not directly connected to earth. Instead, con- densators for high frequency earth and PTC for forming an average level are used. The method is shown in Fig. 5. The use of the Kirchoff circuit laws shows that the voltage on minus/return become (Earthl + Earth2 + Earth3)/3 = the average of the voltages. If the resistance is a PTC standing 230V AC, it will be able to protect itself in case of failure in the protective conductor where it receives a high voltage.

Common Properties/Codes

In order to facilitate understanding and operation, table 5 shows universal codes. Table 6 shows units determined for the technical control code 48. Protecting Settings

Besides the protecting possibility found in code 100, the objects may also have a built- in switch locking change of settings. If the switch is pushed to Set, one may change the properties and methods of the object. If the switch is set back again, the object will not accept changes any more. Hence, they are protected.

In a similar way, calibration values from the production are protected with a soldering bridge. If calibration is to occur, the soldering bridge has to be soldered for this to occur. After finished calibration, the tin is removed from the soldering bridge again, and calibration data are fully protected.

Reduction of Current Consumption

With about 250 objects on the same network and consumption of current for relays, contactors and motors, the power consumption of the object itself is of significance.

Power Supply to Control Lamps

If the object has e.g. 5 control lamps, each to have 15 mA, this will require 5 x 15 = 75 mA in all. For 100 objects, it is 7.5 A, which is some current. If the current is also to be able to run on backup batteries, these have to be very large.

A saving possibility is to utilise the voltage loss from 24V to the normal 5 V of the MCU. This may be done as shown in Fig. 6. When the base on BC546 is zero, no current runs in BC546. Hereby, no current runs in BC558 either, whereby the current to 78L05 regulator and MCU is forced to run through the light diode, causing it to be lit.

If the base is set to + 5V, there will run a current of (5-0.7)/47k = 91 μA. The current is amplified in BC558, whereby the collector-emitter voltage will go below IN at a current of 20mA. Hereby, the lamp is turned off. From this it is seen that the extra current consumption for a lamp will be 91μA, being far from 15 mA.

With a reasonable crystal frequency of 10 Mhz, a modern MCU like Motorola MC68HC908JK/JL may suffice with 11 mA, whereby the consumption of the object itself will be under 15 mA. For 100 objects this is only 1,5 A which is easier to make and distribute.

Power Supply to Contactors In objects where a contactor is used, this will require some current. Modern types like

Danfoss CI110 may suffice with 100 mA, but with many objects this will, however, mean a large current. By moving the AUX switch back to the MCU it may be detected whether the contactor is pulled. If this is the case, one may supply rectangular voltage to it and utilise the inductance of the contactor. A mean current of 33 mA is sufficient to hold the contactor.

Often a control lamp is desired for showing that the contactor has been pulled. Light diodes for 100 mA are not available, and the voltage loss of about 2V may be critical when connecting. Instead, the light diode is inserted in series with the one-way diode, whereby it only receives the 33 mA and simultaneously it does not steal voltage from the contactor coil. The principle appears from Fig. 7.

Power Supply to Motors

24V motors are large consumers of current. A strong spindle motor may easily have a consumption of 4 A. This rapidly induces large voltage drops in the cable between the power supply and the motor.

Now, DC motors have the beneficial property that the tractive force (the torque) is proportional with the current. And since a DC motor with a consumption of 4 A has an inner resistance under 2 Ohms, only 8 V is necessary for it to run. However, it is slow but in applications where this is not a problem, one may advantageously transform the 24V to 8V.

For this purpose, a buck converter may be used which with a composition of MOSFET transistors, Schottky diodes and iron powder cores attains an efficiency of 85-90% with obligatory noise suppression. The inductance of the motor is thus not used, and there is not extra wear on the brushes. The current in the input therefore is reduced to 4 A/85% x (8V/24V) = 1.57 A. Hereby thinner cables may be used, or motors may be controlled over greater distances.

If one has many motors on a long, thin cable, the motor is allowed to stop and wait if the voltage on the supply is too low, e.g. 20V. Hereby, the current in the supply cable is reserved to the objects that disconnect as the last. When these motors are ready, the voltage rises, and the next motors start and decide which are to continue.

Objects

Temperature Sensor

The temperature sensor is to measure the air temperature and to transmit it when asked on code 48.

A possible structure appears from Fig. 8. The temperature sensor may be analogous or digital. In the example, a digital LM74 is used, communicating with the MCU via a serial bus. LM74 is calibrated to an inaccuracy of 0.5 degree, why adjusting in the production may be omitted. If a greater accuracy is desired, a further calibration may occur via code 123 = correction and code 124 = sign for correction. The sensor may be set to manual via code 01, whereby it no longer responds to the measured temperature.

This may result in production losses if one forgets that the sensor is in the manual setting. Therefore, a time-out for manual setting is inserted. A new order via code 01 postpones the time-out period. In Fig. 8 the circuit 81 for communication on the data bus and the processor, here an MCU 82, is seen.

Emergency Sensor

In certain applications for ventilation, it is to be ensured that fresh air is always supplied. In stable ventilation it is also animal heat that requires supply of fresh air. If the ordinary control fails, there is to be a back-up system to take over. This is to be com- pletely independent of the ordinary system. Normally, the temperature is used as parameter. If this becomes too high, it is a sign that too little air is supplied. If it becomes hot outside, this is reflected inside also, and a non-existing fault is recorded anyway. Therefore, an outdoor temperature sensor is inserted, and when it becomes too hot, it is not the temperature, but the temperature difference which is of interest. If it becomes too great, this is a sign of too little ventilation.

The method, however, has some drawbacks: 1. It is difficult to understand and test. 2. Measuring correct outdoor temperature is very difficult, sun influx and heat radiation from the ground may give deviations of several degrees, particularly in hot climates.

3. It is the body heat from the animals which is heating the room. If the outdoor temperature is over 30 degrees, and the stable is without any insulation, the animals can- not heat up the stable as they are giving off water vapour instead of heat.

If instead it is measured whether air is sucked out of the stable when the temperature has exceeded the desired temperature, a failure in the exhaustion may easily be ascertained. In stables with tunnel ventilation (air being sucked in at one gable and blown out at the other gable) this principle does not suffice. If the air sensor is placed close to an exhaust, and the emergency openings in the side walls of the stable are not closed, the animals in the opposite end of the stable will not be ventilated. In order to safeguard this situation, the air sensor is to be positioned close to the end, where the air enters, and directly measure the air speed longitudinally.

As this speed is in the range 0.5 to 3 m/s, the sensor has to cope with this. Stable air is filled with dust, contains ammonia, and is regularly disinfected and washed. This excludes traditional methods like wind wheel anemometer and the hot wire method. The problem with zero point operation is also great as the sensor may be in position for several years without service. Instead, relative temperature measurement is used. A temperature sensor as described above is added a heater. Both are encapsulated in re- sistant plastic. An MCU may read the temperature and steplessly control the power in the heater.

Firstly, the temperature is measured without heat addition. The value is stored, e.g. 28 degrees. Then heat is applied, and the heat is regulated so that the sensor achieves and keeps a surface temperature of e.g. 10 degrees. After some time, the sensor has become 38 degrees. When the temperature is stable, the supplied power is stored, e.g. 300 mW. The heat is turned off, and the sensor cools for a sufficiently long period. Then it is repeated.

The supplied power expresses the air speed, since a great speed cools strongly, why greater power is to be supplied for maintaining the overtemperature. The correlation has previously been laid into the MCU as a table, and the corresponding air speed is calculated and provided at inquiry.

Dust deposition on the sensor will reduce the cooling. This will correspond to a lower speed. In practice this will mean that if dust cleaning 4 times a year fails, a fail message will be given off at some time, even thoughh the failure is not real. This prevents a passive emergency function.

A deficient heater will cause the MCU to supply maximum voltage and hence believe that the speed is too high. At the start of the heating period it is therefore checked that the sensor really gets a heating.

The sensor may also be equipped with a code indicating maximum temperature, where the monitoring goes from temperature to air speed. In addition, a code indicating the lower limit for the air speed. If the measured temperature exceeds maximum temperature and the measured air speed is lower than the lower speed for the air speed, the emergency sensor transmits a state of emergency and calculates a relative deviation to be changed on the controlling objects. The problem is just that the system is not independent. A disturbance may theoretically provide a transmission changing maximum temperature to a very high value and therefore in reality disconnects the monitoring. This may occur even though there is a check sum in the protocol. Therefore, the emergency sensor may read the Al line, and change of the settings of the emergency sensor may only take place if the alarm line is low.

Damper Motor

Damper motors are used for positioning something. A universal way is to control ac- cording to 0-100%, where 100%) e.g. is defined as open state. As the damper motor control has built in intelligence, it is obvious to introduce some new methods.

The motor control has four buttons:

Aut The motor runs according to orders from outside via the data line Close The motor runs until end stop in direction towards 0%

Stop The motor has stopped Open The motor runs to end stop in direction towards 100%

Transformation of the voltage from 24 V to e.g. 8 V is mentioned in the introduction. If one has a buck converter, it is also obvious to use it for soft start whereby large pulses are avoided. And if one has variable voltage to the motor, it will be obvious to make a proportional band around a desired position so that the speed decreases, the closer the motor comes to the desired position. If integration is also used, the motor is controlled quite precisely on the desired position.

If the supply voltage fall below a certain value, the motor has to stop. This occurs for relieving the supply. Hereby, other motors may finish their work as also described in the introduction.

Adjusting the working range of the motor may be performed by pressing simultaneously the Stop and Open buttons for some seconds when the motor is in the 100%) position. In the same way, the 0% position is determined by pushing Stop and Close buttons for some seconds while the motor is in the completely closed position. It is thus possible to adjust the motor while you are standing right beside it. The values are stored in the memory and do not disappear by power failure. Normally, a potentiometer is used for determining the position of the motor. However, it is a problem for spindle motors with long travel, as known from stables, where the walls are a curtain. If the motor is equipped with an impulse switch giving a number of pulses per revolution, the object can count these pulses. Since the object also knows the direction of the motor, it may count up if the motor is moving towards 100%, and down if the motor is moving towards 0%. If a power failure occurs during operation of the motor, pulses may be lost. Therefore, the object is provided with an energy reserve that may supply the MCU until the motor stops. While this energy reserve is consumed, counting is continued. When the reserve is used, the counter is moved to the part of the memory that remain after a power failure. Due the small power consumption, the energy reserve may be a condensator of lOOOμF.

Other properties are motor voltage and max. motor current. If the motor current is over the limit for a certain time, we are speaking of overload, and the motor is stopped. Then, there is waited some time before a new attempt is made. It may e.g. be the case with a curtain which for periods is pressed so hard against the wind that it cannot be drawn up without destroying cords and cord wheels. Here, the object will try again, and if the attempt is made in a period where the wind is weaker, success will appear.

If the object cannot reach the desired position, it will give off an alarm. This will also be the case if the relay changing polarity on the motor does not do it. It will imply se- rious damage that the motor shuts while the object believes that it opens.

Since the object has a buck converter, the supply voltage to it may also be an unregulated voltage. Therefore, the object has two terminals: + 24V from the VE bus and + 24-48 V to the buck converter and thereby to the motor. If one does not have a local power supply, the two terminals are interconnected with a short wire. Many of the features to be controlled by the damper motor are non-linear. This may give problems with the control which may be to excited in one working range and too slow in another. If the object is provided with a table, the non-linear may be changed to something linear. The table is seen in codes 10 to 20 in Table 7.

For a concrete object VE 161-40, all methods, properties and events are seen in table 7.

Exhaust Control Exhaustion with ventilator is controlled most economically according to the

Low/Normal method. This means that the ventilator runs on reduced speed at low power whereas the damper, via the built-in damper motor, regulates the air output. If there is need for more exhaustion than the Low position can provide, shifting is performed between Low and Normal in a time proportional way.

This requires that a damper motor object is furnished with a relay for starting and stopping the ventilator and a second relay for shifting between Normal and Low.

The object is controlled as normal with a control signal of 0-250, corresponding to 0- 100% exhaustion. When the value in code 47 is reached, the damper motor regulation stops, and regulation only occurs by shifting between low and normal.

If code 46 is set to 0, the ventilator is stopped, while the damper regulation continues unchanged. This is called natural ventilation.

In some cases one will insert a measuring wing measuring the real exhaust. The object has connection to a measuring wing, and the object will now attempt to get the measured value to be in accordance with the control signal. If the object performance is indicated as 10000m³/h, and the control signal is 40%, the damper motor will work until the measuring wing calls 4000m /h.

Table 8 shows properties, methods and events for a VE132 exhaust object. Relay Control

Controlling an object requiring a connecting or disconnecting switch. The object has the operating buttons: Aut The relay operates according to orders from outside received via the data line Off The relay is off On The relay is constantly on

The object may be set to work ON/OFF or proportional in time. If proportional in time, there are three other properties: Cycle time, minimum on time, minimum off time.

In place of a relay, a contactor may be inserted whereby 3 -phase equipment may be controlled.

The tasks could be electric heating, lighting, motors with built-in protective circuit switches, pumps with built-in protective circuit switches etc.

Motor Control Motors may be single-phase or triple-phase. The object is identical with the relay control, but with power transformers for measuring the motor currents.

When the current exceeds maximum motor current, the exceeded amount is added to a variable. The exceeding is calculated as relative exceeding in 2nd power. If the current is under maximum motor current, it is subtracted. If the sum of exceedings becomes too great, too much surplus heat has been deposited in the motor, and the relay/contactor is released and the motor stops. New reconnection may only occur when a certain cooling time has passed. Reconnection may occur automatically or manually by pushing the button.

Contrary to traditional protective circuit switches based on the thermal principle, here the disconnecting value and the cooling time may be set. If the motor works under easy conditions, it may be allowed to disconnect some time before in order to achieve better protection. Instead of the contactor, one may use solid state switches like triacs.

The state of the object is also to be stored in case of power failure so that reconnection cannot occur after a short-time power failure.

Fig. 9 shows an example of the structure of a 3-phase motor control. It appears that the motor current is transformed to a voltage with neutral at half the supply voltage for the MCU. The MCU transforms the voltage to a number with reference in this artificial neutral point. The deviation is accumulated in a variable, and at regular intervals there is calculated an average expressing the magnitude of the current. By very large currents, the power transformer will go into saturation, and the measured current becomes slightly less than the maximum value. The power transformer is therefore to handle the start current of the motor without being saturated.

Due to the hysteresis of the core in the power transformer, it will no be linear at small currents. This is compensated by inserting 3 adjusting points, e.g. 0, 1, and 4 A.

By emergency condition (normally power failure), the relay may be allowed to fall off, if the task of the objects is not critical, in order to save current from the back-up batteries.

Connection to a PC

Modern control systems have to be connected to a PC for greater overview and operat- ing comfort. The protocol of the VE bus is adapted to the conditions in which the objects are mounted. No standard protocol on a PC fulfils these requirements, neither RS232, RS484 nor RS 422, since none can bear the high common mode voltages occurring. The n:n principle is also unknown for PC. Therefore, a data converter is inserted between the VE bus and the PC. The data converter has connection to the Di and the Do lines and converts the 5 byte to 11 byte necessary for enabling data transfer to the PC. At the same time, the format is changed to RS232, but this may also be USB or a parallel port. If Di/Do works at 600 baud, the speed to the PC has to be 600 x 2 x 11/5 = 2640 baud. The closest standard value is 4800 baud.

In practice, it always difficult to couple two data network together, including finding the right connection on the PC. Therefore, the data converter is equipped with three control lamps:

Di Is lit when the Di line is high and flashes when traffic occurs

Do Is lit when the Do line is high and flashes when traffic occurs PC Is turned off if the RS232 cable does not have any connection to a PC Flashes, if the connection is OK, but the program is not running or is using another port

Is lit when the connection is right, port is right, and the program is running

Flashes quickly when data traffic occurs.

In the protocol between the data converter and the PC, it may be determined by a bit which data line is to receive the actual transmission.

Alarm and Emergency Opening

The system has an object associated with it for taking care of the following tasks: Alarm Wrong values, objects are not responding

Power failure One or more power supply objects have fault. Establish emergency condition Emergency opening The conditions are wrong, a fault must exist somewhere.

Transmit emergency orders

As describe previously, the emergency sensor may give orders if fault occurs in the normal system. But if the system consists of several sections with each their objects, the emergency sensor will produce emergency control of all objects and thereby disturb sections where the operation is OK.

At the setup, the alarm object receives information about which emergency sensors are associated with which sections. Then information about which objects are to be acti- vated in case of an emergency condition. Hereby it may be made selective. The alarm object is to use two informations from the emergency sensor: Temperature and air speed. Therefore, both code 48 and code 50 are used. If the limits are exceeded, both alarm and emergency condition/emergency opening are released. Emergency condition prevents the activated objects from going towards 0%. Emergency control is a relative control. The transmitted control value is to be added to the actual value of the object. The example in Table 9 shows how.

If fault continues to exist, the emergency control is repeated with fixed intervals. Hereby, all associated object will eventually get the control value 100%.

Alarms are universal. There is specified: Which object The object number, e.g. 55 Which code The code in the object to be checked, e.g. 49 Over/under Whether alarm is to be released at exceeding or going below

Alarm value The threshold value to release the alarm

From this appears that the alarm object also may handle future objects which were not provided when the alarm object was manufactured.

During the setting up, the alarm object also gets a list of the object numbers of the power supplys. It may then regularly enquire on code 112 = the alarm code. If it is below a certain value (but not zero), an emergency condition is released as the power supply is not in order.

Alarms may be formed by connecting bells, sirens, or phone call machines to one of the alarm outputs which are relay switches free from potential. The power for sirens and bell may be taken from + 24 V of the VE bus. Logging

Logging may advantageously be made on a PC, but in some cases a PC is not sufficiently reliable. It is commonly known that a PC may stop or does not start correctly after power failure. Therefore, a logging object is provided in the system.

The log object may be set to log a certain value in a certain object with certain intervals. The log object has a real time clock used for chronological determination of the loggings. The data logged are stored on the same medium used by digital cameras: Flash RAM.

In order not to get too many redundant data, particularly time data, the time is indicated as a relative value in relation to time loggings.

If the logging unit is chosen to minutes, every 4th hour a time log will be recorded, consisting of year.month.day. time. Every logging then consists of realtime.object. code.data. It appears that a logging will occupy 1+1+1+2 = 5 byte. A time log will occupy the same. An 8 Mb flash RAM may therefore hold 8M / 5 = 1.6 loggings.

When inquiring about data, this occurs on a separate data line which also may be a direct connection to the internet. One may ask on an object and a code and a starting time. The MCU in the log module will then search the memory in order to find matching data. The relative time indication is substituted by an absolute time indication, so that data may be read immediately.

When the memory is full, the oldest part is deleted and new data are placed here.

One may also activate a concentrator function whereby a full memory will result in concentrating the most frequent data. If one has e.g. logged some data every minute, the average may be stored over 4 hours.

The log module may be set up to log all alarm and emergency condition activities so that a failure situation may be analysed later. Fig. 10 shows the general structure. A relatively cheap MCU MC68HC908GP32 is connected to the two data lines of the VE bus. Via three latches, the address is set up to the flash RAM, and data are recorded or read immediately.

Spraying

In pig production, it is a requirement in several countries that the fertilised area is sprayed at fixed intervals. For this may be used relay object controlling an electromagnetic valve. Compared with the present central control, the relay object may be placed in immediate proximity of electromagnetic valve and water conduit. Therefore, with the on and off buttons it is easy to check whether the plant is working as supposed. After finish of the testing, the object is reset to Aut.

This will, however, imply certain advantages to make a real spraying object. It does not need to receive orders from a control panel, but may independently spray with certain intervals recorded in its codes.

Informed about an object and some codes, one may let it follow a centrally positioned clock, and provided with the object number of the temperature sensor of the room, one may increase the frequency of sprayings with increasing temperature.

Cooling

Here is used evaporative cooling as example. Cooling may be spray cooling controlled proportionally in time. Therefore, here is used a relay module, either 1 -phase or 3- phase.

By cooling where water trickles through a pad, a pump has to be controlled. It may be a motor control as well as a relay object. However, there are considerable improvements associated with making a real cooling object: The object can be expanded by connecting electrodes in the water reservoir, whereby the water level may be controlled by an electromagnetic valve. If a motor is also connected for emptying the reservoir, the following advantages maybe achieved:

Controlling the water level becomes very precise, materials for the reservoir are saved etc.

Bleed off via the emptying motor may be established by opening a little from time to time.

Emptying water may be performed when no cooling demand exists any longer. Hereby, algal growth is limited and the service life of the materials is increased.

With three electrodes, the water level is determined very precisely. Hereby, the two other electrodes may be used for conductivity measurement, and bleed off may occur based on a real measurement and not as an assumption.

2-speed Control

By controlling motors with a load, the torque of which decreases with the square of the rotational speed (pumps and ventilators), the control form with low and normal speed may be used. The low speed is determined by several factors, but will often be in the range 25 - 35Hz, the normal being 50Hz. Therefore, a stepless regulation until e.g. 35 Hz is to be used, after which time modulation is performed between 35 and 50 Hz.

The 35 Hz are typically formed by a frequency control controlling a 3 -phase motor. At

35 Hz, the motor voltage is to be correspondingly lower. Instead of 400V, the voltage is to be 35/50 x 400 = 280 V. If using a single phase motor instead, the voltage is to be 35/50 x 230 = 161 V. This corresponds to a peak voltage of 161 x 2 = 228 V.

Fig. 11 shows the diagram for an arrangement which may frequency control a single phase motor with very simple means. The supply voltage is 3 x 230/400V. The diodes form a full bridge 3 -phase rectification, which means that the voltage over Tl and T2 will vary between 488 and 563 V. The curve is seen as the uppermost curve in Fig. 12, designated differens. When the motor current runs back to the neutral conductor, it is the positive voltage, which is available when the motor is to have a positive voltage, and correspondingly the negative voltage when the motor is to have a negative volt- age. The actual voltages are seen as the curves Plus Volt and Minus Volt in Fig. 12.

Fig. 12 is the required voltage for 35 Hz shown also as the curve Res. Ul volt. Note that in some places it lacks a little in order to be a perfect sine curve. It is where the positive or negative voltage cannot suffice. The small irregularities will be evened by the inductance of the motor and will therefore not have any significance.

If a lower speed is required, the voltage is also to be lower, and the irregularities correspondingly less.

From this appears that one may establish a frequency control of a single phase motor which goes up to 35 Hz with a supply of 3 x 230/400 V 50Hz. Consequently, this may also be done at 60 Hz, where correspondingly 35 Hz then become 42 Hz.

Compared with the traditional frequency control, the following advantages are achieved:

There is no great capacity after the rectifiers, and therefore the phase currents are not large charging currents. This means a better power factor. Simultaneously, the power losses in the rectifiers are considerably less, and the necessary common mode filter in the input may be designed for a somewhat lower current.

There is only one transistor totem pole as opposed to three in the normal. This provides savings, both for transistor, cooling, drive and logic.

The current is not to run back via a totem pole, whereby the heat loss is halved. In normal operation, the motor is coupled to one of the three phases and thereby receives the current by-passing the converter. Therefore, there is no loss when full power is demanded. If there are several motors on the task, they may be distributed on the three phases at normal operation so that one will always run if there is power on one phase.

There are not any problems with harmonic currents in the network either. At 35 Hz, the motor only absorbs (35/50)³ = 0.343 of the power at 50Hz. Correspondingly, the current in the phase is 34.3%. If the 3rd harmonic current is e.g. 3 A when operating at 35Hz, this will only count as 33% since the current at 50Hz will be 9 A, and this will be sine shaped.

Hereby, one may avoid inserting a cost-increasing link for power factor correction.

Operation Panels

For general control of several objects, one inserts an operating object. This contains software for the general control and buttons and display, so that settings and readings may be performed. On the operation panel, one may also choose a certain object after which operation of the chosen object may occur. Fig. 3 shows an example of an op- erating panel.

Buttons for code and data are seen. The display above shows the selected code and opposite the data button the associated data. Below, the buttons function and object are seen. Function is discussed in the section concerning method. If function or object is pressed, the display above the function buttons shows the actual function, and the display above the object button the associated object.

Change always takes place in the same way. Push the button for the value that is to be changed. With the +/- keys, the value is changed, and the button is released again. The change is now completed. The operation panel does not have any other connections than the VE bus. There are no relays or other inputs/outputs as in usual controls. Therefore, it cannot be too small. Only the number of objects, which it can handle, sets the limit.

Fig. 13 shows an example of a structure. Above to the left is seen the control of the A 1 line. To the left drive and input for the two data lines. An operating panel is to have access to both data lines since it has to get information from the sensors as well as to transmit messages to the controlling objects. Note the two pull-up resistance of 2200 Ohm. They keep the data line high. Lowermost, at the middle, is seen an SMPS trans- forming + 24V to + 5 V. Here, there is used an SMPS since the current consumption from the display is large. To the right, lowermost, is seen an EEPROM mounted in a socket. By means of this, one may make a copy of the whole setup and put it into the board. If an accident occurs, the setup is to be read into a new operating panel. At the top all the operating buttons are seen. To the right display and display drivers where multiplexing occurs. Since the MCU cannot take care of details (they are moved out into the objects), it has time to perform this task.

Power Supplys

The power supply is an object itself with properties, methods and events. The power supply transforms the mains voltage to a voltage of 24 - 30 V. A back-up battery may be connected to the power supply whereby it becomes charged. In case of power failure, the battery will provide the power needed.

The object will have two fuses for 24V, whereby a short-circuit will not render all ob- jects without power. Among the properties of the object is found normal voltage and final charge voltage, maximum current at output as well as charging the battery. By overload, the voltage is reduced so that the current is kept constant. If the voltage goes too far down, an alarm is given. This is also the case if the mains voltage disappears and the battery continues to supply the objects.

Table 10 shows an example of methods, properties and events for a power supply. Table 1

Regulation of speed of rotation:

Table 2. Example of Documentation for an Object

Conveyor belt 1 function 31 : Moving slag from silo to crusher Object no. 14. A Code table:

^xThe object responds with data =AA, which is the most difficult to transmit. Used for fault finding. Comments:

The belt conveyor is allocated object no. 14. A. The belt conveyor can only run one way as minimum is 0. Greatest speed is 5000mm/s, but limited to 1000 mm/s. At start, 4 s is used for maximum speed, by stop 5 s to stop. The regulation is PI, and the tractive force in the belt is limited to 70% of the allowable. The belt will run according to code 48, if code 32 is above 100. Otherwise, they will run according to code 32 (manual control), which is in % of the allowable range.

Table 3. Example of Documentation for an Object (continued)

Belt conveyor 2 function 3: Moving large stones from crusher to container Object no. 15. A Code table:

²The object responds with data =AA, which is the most difficult to transmit. Used for fault finding Comments:

The belt conveyor is assigned object no. 15. A. The belt conveyor has not speed regulation since maximum speed cannot be set. The belt speed is 630 mm/s. The regulation is proportional in time with an interval of 20 seconds. Min. running time is 2 s, and min. stop time is 1 s. The belt will run according to code 48 if code 32 is above 100. Otherwise, they will run according to code 32 (manual control) which is in % of the allowed range. When code 48 is over 50% of maximum speed, the belt will run constantly, if code 70 is set to 0 (on/off).

Table 4. Example of Documentation for an Object (continued)

Crusher 1 in function 101: Crushing slag supplied by belt conveyor 1 Object no. 8.3 Code table:

fault finding Comments:

The crusher is assigned object no. 8.3. The crusher has not speed regulation since maximum speed cannot be set. The speed for the crusher is 1400 rpm. There is 10 s operation in star before switching to delta when starting. The crusher will run according to code 48 if code 32 is above 100. Otherwise, it will run according to code 32 (manual control) which is in % of the allowed range. When code 48 is over 50% of max. speed, the crusher will run continuously. The crusher does not have interval operation. Table 5

Table 6

*Protected by Set switch "Protected by Set and Cal soldering bridge Table 8: Properties, Methods, and Events for VE132 Exhaust Object

*Protected by Set switch

**Protected by Set and Cal soldering bridge

Table 9

Claims

1. A method for controlling a processing plant which include one or more objects, where each object is a pre-defined parameter depending on the function of the object, in which:

- the object controls the process parameter so that it falls within the pre-defined parameter,

- the predefined parameter is communicated to the object as a value which by data processing on the object is translated to a control parameter, - a control parameter is communicated between a control unit and one or more objects by data communication on a data bus,

- the confrol parameter is structured as a signal containing objectaddress.code.data- type.data.check. wherein:

- objectaddress defines the object concerned, - code defines the object function concerned,

- datatype defines whether they are control data, object data or log data,

- data is the actual confrol parameter, e.g. temperature, moisture or parameter inquiry,

- check is a check signal, - when the control parameter is parameter inquiry, the object communicates the actual parameter value to the control unit on the data bus.

2. A method for controlling a process plant according to claim 1 where the predefined parameter is communicated to the object as a value which by data processing on the object is translated to a control parameter between 1% and 100%).

3. A method for controlling a process plant according to one or more of claims 1-2, where one or more objects communicate the control parameter on the data bus to at least one other object.

4. A method for communicating without any loss of data on a data bus with collision in a control unit for a process plant as indicated in one or more of claims 1-3, wherein: - the data bus is initiated by providing it with a first predefined voltage,

- a transmitting object sends data on the data bus by attempting to give the data bus a desired voltage level by, in the case of a first type .of bit, e.g. 0, attempting to draw the voltage level on the data bus to another predefined voltage level, and in the case of a second type of bit, e.g. 1, to let the data bus keep the first predefined voltage level,

- the transmitting object measures the voltage of the data bus in order to verify whether the voltage level is at the desired level,

- the transmitting object continues transmission if the desired voltage level is present on the data bus, but discontinues the transmission if the desired voltage level is not present on the data bus.

5. A method for handling alarm in controlling a process plant according to one or more of the claims 1-4, including one or more objects wherein:

- an alarm object makes an inquiry to other objects about the real size of a process parameter,

- the alarm object compares the returned value with a predefined alarm limit,

- if the alarm limit is reached, the alarm object issues an emergency order to the object in question.

6. A method for power supply for objects with control lamps in a process plant according to one or more of claims 1-5, wherein:

- the object is supplied a voltage which is considerably greater than necessary for driving a processor in the object,

- a control lamp is placed with the possibility of being short-circuited by an electronic switch, e.g. a transistor, in series with the processor of the object in the part of the circuit where the voltage is substantially greater than necessary for driving the processor of the object,

- the object turns on the control lamp by opening the electronic switch,

- the object turns off the control lamp by closing the electronic switch.

7. An object for a process plant where the object comprises a processor, data communication means, inputs and/or outputs for reading and controlling a process parameter and a predefined object address.

8. An object according to claim 7 where the object includes means for receiving standard codes with associated parameter value via the data communication means and for changing a parameter in the object, e.g. the real value of the process parameter, the goal for the process parameter or an alarm condition.

9. An object according to one or more of claims 7-8 where the object includes means for transmitting standard codes with associated parameter value via the data communication means which is reflecting a parameter in the object, e.g. the real value of the process parameter, the goal for the process parameter or an alarm condition.

10. An object according to one or more of claims 7-9, where the object is supplied with a voltage that is substantially higher than necessary for driving the processor of the object, and where the object includes a control lamp supplied with power in series with the processor of the object.

11. An object according to one or more of claim 7-10, where the object includes an output for power supply to a unit reading or affecting the process parameter.

12. An object according to one or more of claims 7-11, where the object includes an alarm input, means for preventing change of a parameter in the object when the alarm input is in a predefined condition.

13. An operation object for a process plant, the operation object comprising a processor, data communication means and an operation panel.

14. An operation object according to claim 13, where the operation object comprises means for manually selecting and transmitting standard codes with associated parame- ter value via the data communication means, and for receiving and displaying received standard codes with associated parameter values on the operation panel.

15. An operation object according to one or more of the claims 13-14, where the op- eration object is supplied with a voltage which is substantially higher than necessary for driving the processor of the object, and where the object comprises one or more control lamps supplied with power in series with the processor of the object.

16. Use of an operation object according to one or more of claims 13-15 for control- ling one or more objects according to one or more of claims 7-12 according to a method described in one or more of claims 1-6.