A very rough overview of the hardware part of DCS

The DCS is an extremely important piece of equipment that can be called the brain of a chemical plant.

Operators also tend to specialize in DCS, and the era is truly DCS! That was a long time ago (lol)

This technology is almost commonplace now.

Even though DCS does not involve a machine shop, it is very important for a factory.

It is also a problem for machine shops who want to understand the equipment comprehensively that they have been avoiding it for a long time.

If you understand just the outline, it may be easier to understand when talking to an instrumentation engineer.

In particular, the hardware aspect is easy to understand in the sense that it is visible.

The outline of DCS is related to control, so even machine shops may not feel any resistance to it.

DCS system diagram

The DCS system diagram can be expressed simply as shown below.

As a system, first of all there is the field .

In this industry, we call it a field.

Although the field part looks small in this system diagram, we must remember that the field is actually very large.

Both in terms of size and the importance of driving.

There is a station that collects on-site signals .

The I/O rack also fits here.

There are stations operated by humans .

Also called HIS (Human Interface System).

This is the part for humans to obtain and manipulate information.

The control room is the heart of the factory, and this is the part you see in photos of the control room.

It’s a place full of desks and monitors.

control part

The control part, including the I/O rack, has the following functions.

• CPU for process calculation
• Convert field signals to digital signals
• Returns the control-converted digital signal as a field signal

You can imagine performing feedback control as shown below.

To be extreme, the control has been completed up to this point.

What if driving could be done without humans?

Operation monitoring part

A part that is operated and monitored by humans is absolutely necessary.

The elements that make up that part are as shown in the diagram below.

HIS of DCS has the following functions.

• graphic
• Operation guide
• alarm
• trend
• tuning
• group

Among these, the most commonly seen screens are graphic

A flow diagram and information on the required operating values ​​are displayed.

This is the part that comprehensively monitors the driving status and is the heart of the HIS.

I also use trends very often.

It graphs the time response when multiple instructions are given.

Required for driving analysis.

This trend should also be used by machine shops.

Operator ‘s Guide displays special driving situations in words.

An alarm is just as it sounds. He will tell you about any driving abnormalities.

This is an important point when driving.

Alarms will alert you to areas that may be neglected when processing a large amount of information at the same time.

There is also a problem of too many alarms causing operator stress, and recently alarm settings have become an important issue.

It can also be used to record and verify what happened and when.

Tuning and groups are often used in ad hoc driving situations.

Used for PID parameter adjustment at the start of operation, manual intervention during operation, etc.

If you change this arbitrarily, the operating conditions will change easily.

In that sense, it is a very important and dangerous place.

cubicle nest card

The station is probably shaped like a rectangular parallelepiped.

However, station is actually a functional name and does not refer to the actual name.

The panels you see in electrical rooms are called cubicles . It works well on the board as well.

The cubicle has several tiers.

Each of these shelves has a box called a nest .

A printed circuit board called a card is placed inside this nest .

Card type

Let’s look at cards as the smallest unit of stations.

The image is as below.

There are a lot of cards in a box called a nest.

The cards are assigned, for example, one for the thermometer indication signal and one for the thermometer control signal.

There are three main types of cards:

• for signal distribution
• for input/output
• for control

An indication value is input from the field, the signal is converted for control , and the converted signal is distributed to the control unit .

Functional classification

The card has a function to perform various conversions of signals.

The purpose of setting cards for each function is to unify the functions of the cards.

Considering only one point of temperature control, there is no need to assemble this complicatedly.

To efficiently control multiple things in one place, we take the approach of distributing functions.

Examples of functional classifications include:

• analog/digital
• Input Output
• Temperature control information/other process fluids
• alarm
• interlock

Create stations for each function , place nests in the cubicles , and insert cards .

The idea is to make it easier to organize.

This is the same as a company.

• Head office branch
• Manufacturing Department/Office
• Sales/Research/Sales
• public relations
• accounting
• human resources

Create offices and departments for each function , create islands for departments, assign desks , and work .

This may lead to misunderstandings or counterarguments, but I would like to point out that they are very similar in that cards are desks and nests are desk islands .

Signal input/output

The reason it took me so long to understand is simply because no one could explain the diagram below.

This diagram shows the hardware system from detecting signals from process instruments to returning control values ​​to the process.

analog instrument

As an example of an analog instrument, consider temperature control.

To control the temperature, perform the following operations.

1. Detects the reading of the thermometer
2. I/O rack converts temperature electrical information into signal information
3. Control values ​​are determined by controller feedback control.
4. I/O rack converts signal information into control information
5. Transmits control information value to regulating valve and operates

I/O racks and I/O cards are technical terms, so in extreme cases you don’t need to know them.

digital instrument

Digital instruments can be explained in the same way as analog instruments.

As a digital instrument, overflow prevention with a liquid level gauge is easy to understand.

The raw material is shared in a tank, and when the tank liquid level reaches a certain level, the raw material supply valve is closed.

1. Detects the indicated value of the liquid level gauge.
2. I/O rack converts liquid surface current information into signal information
3. Control value determined by controller sequence control
4. I/O rack converts signal information into control information
5. Transmits the control information value to the regulating valve and operates it.

The target of analog control is feedback control using PID, but it can also be used for sequence control.

Digital control is usually not used for feedback control, but for sequence control.

Signal input

I mentioned that the signal input is via the I/O card.

Let’s look at that a little more.

See the diagram below.

Transmits the reading of the thermometer as electrical information in mV units.

As a thermometer, imagine a resistance thermometer.

This is converted into a control signal using an I/O card.

DC 4-20mA is common.

There is an electrical circuit that converts voltage information related to temperature in mV units into transmission information of 4 to 20 mA , and 24V DC is supplied to it.

That level of understanding should be enough.

From this point forward, we will talk about electronic circuits…

Use this signal for feedback control or for recording.

signal output

The signal output is basically OK if you think of it as the opposite of the input.

The final output will be the instrument air pressure.

The operating position of the regulating valve is determined by the pressure of the instrument air.

An electro-pneumatic converter is required to convert DC4 to 20mA electrical signals to air pressure .

I think it’s enough if you understand it as a signal conversion place, just like I/O cards.

electricity and air

Signal transmission methods connecting the DCS and the site include electricity and air .

Air can be thought of as an old technology.

The nozzle flapper is a fairly primitive technology.

Nowadays, almost all electricity is consumed.

Let’s make a comparison in terms of air and electric signals.

electricity and air

The only advantage of air is that it is easier to predict failures.

You may not notice when a blockage occurs in the air piping, and pressure fluctuations are a bigger issue.

As the number of automatic valves increases and the amount of compressed air consumed increases, using air for instrument signal transmission may be risky.

Instructions/records/alarms

Assuming that the main flow of control is as described above, sub-functions include factors such as instructions, recording, and alarms .

Indicator

Indicators are often set on-site.

It is possible to give instructions using DCS, but if the information is not that important, it is set on-site.

This demand is decreasing year by year.

I don’t know the details, but I believe it’s because the capabilities of DCS are increasing.

Or perhaps it is due to the fact that on-site work is being reduced rapidly and communication between the on-site and DCS is becoming more sophisticated.

In any case, the practice of erecting stanchions just to indicate undemanded readings goes against the trend of saving space in plants , and will likely disappear in the future.

recorder

Nowadays, with the advancement of DCS, there are fewer people who are aware of these separately.

Those who are familiar with the old days of recording paper are probably aware of the existence of recorders.

In order to record on paper rather than electronic data, an analog device is required to output the information onto paper.

Conceptually, this still exists as a sub-factor.

If the recorder disappears and you try to do everything with DCS, from the perspective of people other than those involved,

• With DCS, data can be stored indefinitely.
• Unlimited server capacity allows unlimited data collection

There will always be people who misunderstand things like this.

alarm

Demand is also declining year by year.

Alarms within the DCS are distinguished from soft alarms, and alarms displayed on an alarm panel separate from the DCS are distinguished from hard alarms.

The alarm presented as a sub-factor here refers to a hard alarm .

DCS can also send alarms, and when controlling with DCS, soft alarms are also considered part of the control.

On the other hand, hard alarms use DCS signals, but they are often highlighted separately from DCS.

Or perhaps you want to display alarm information for devices that are not imported into DCS ?

A common example is specialized equipment such as refrigerators .

Refrigerators usually have a control panel located locally, and the operation is completed within that area.

In many cases, refrigerator information is not imported into DCS.

Even if it is imported, it is often only the current value.

If an abnormality occurs, an alarm will sound at the site, but it is difficult to notice.

When transferring that alarm information to the DCS room, it is easier to display it with a hard alarm rather than firing with a DCS soft alarm.

This is because there is no need to configure DCS.

power supply

When you think of voltage, what comes to mind in Japan is probably a 100V household power supply.

Few countries use 100V power supply.

440V power supplies are mainstream in chemical plants.

There is also a 220V power source in an old plant.

Instrumentation power supply is 100V or less

By the way, most of the power supplies for instrumentation, which is the target this time, are 100V or less.

I consider 220V and 440V power supplies to be high voltage, but that’s because I’m used to 100V household power supplies.

The reason why the power supply for instrumentation is 100V or less is because the power supply for instrumentation does not require a high voltage .

can also be read as

Most instrumentation power supplies are DC24V.

Most power supplies for instrumentation are 24V, even among 100V or less.

Has anyone heard the word 4-20mA for current?

Machine shops rarely hear about it, so if you’re not conscious of it, you might not hear it.

4-20mA is used as analog instrument signal.

Specifically, there are the following.

• Continuous signals from liquid level gauges, thermometers, flow meters, etc.
• Continuous instructions such as adjusting the opening of the regulating valve

Since it is an analog signal, it is called AI/AO .

The analog signal is 4-20mA current.

20mA is a very dangerous value if you get an electric shock .

Most automatic valves open and close at 100V.

Automatic valves can be said to be control equipment for instrumentation equipment.

This automatic valve often uses a 100V power supply.

General catalogs have a lineup of 100V, 200V, and 24V, but it seems that 100V is often the base.

It is only used for solenoid valves and limit switches, and it seems that both equipment can be used at the above voltage.

Is it because 100V is common in Japan?

In any case, it is correct to choose a general-purpose specification, so I think you should assume that 100V is most likely.

This is used to exchange signals determined by 0-1, such as on-off.

An on-off automatic valve is a typical example.

Most of the instrumentation equipment used in chemical plants is automatic on-off valves.

I want this to be as general as possible.

On the other hand, a special one is a 4-20mA analog signal.

Analog signals are called AI/AO, while digital signals such as on-off automatic valves are called DI/DO .

Duplex

What does duplication mean?

As the name suggests, it means having two pieces of the same equipment .

You only have 1 normal device.

So why do we have two pieces of the same equipment?

The impact of failure is too great

That’s the answer.

DCS collects the signals from each instrument into several devices and processes them.

If an important part breaks, everything may become unusable in an instant .

If it is an analog instrument, even if it breaks down, the instrument cannot be used.

That is the weakness of DCS .

In DCS, the word duplication comes up many times.

Let’s check each part individually.

Control station duplication

The control station is the heart of the control device.

The heart of DCS, which is the heart of a chemical plant.

If this breaks, you’re definitely out.

Therefore, the control station will be duplicated.

Duplication of CPU and memory

Important parts such as the CPU and scale in the control station are further duplicated.

The worst thing you can do is break the control station.

Try to stop one step before the worst

This is the basic philosophy for safe operation of chemical plants.

Redundant power supply

The power supply may seem obvious, but it will be duplicated.

If the power supply stops, all equipment will stop working even if there is no power outage.

Duplicate card

The I/O cards will also be duplicated.

In particular, the control input/output cards are duplicated.

Self-diagnosis function

It is not rational for humans to judge whether or not these duplicated parts are broken.

In the distant past, patrols may have periodically checked.

However, electronic components with self-diagnosis functions are now common.

Naturally, DCS also has this self-diagnosis function.

It performs self-diagnosis, issues a warning if there is an abnormality, and switches to standby mode.

This is normal.

switching operation

Even if you have duplication, if one side is always running and the other is stopped, it may not work even if you suddenly move the equipment that is not working.

I don’t know if it’s broken because it’s not running.

Problems like this can occur.

Therefore, the equipment for duplication is basically

Switch equipment that is operated regularly

That’s what I’ll do.

Mechanical equipment in a factory must be switched manually,

If it’s a part in the DCS, it’s normal that it switches automatically.

DCS and sequencer

When the topic of DCS comes up, of course the topic of sequencers also comes up.

As a machine shop, it would be easier to understand this from a hardware perspective.

DCS

The DCS is typically installed in the user’s factory control room.

The program is created by a factory control person or an external control engineer.

The program controls factory equipment.

Incorporate all equipment that requires automation to operate the factory into the program.

This is the self-created range.

You could call it DIY (Do It yourself).

Most batch-type chemical plants create this in-house.

This is because there is not much control-like control.

The number of instrument signal points that are imported into the DCS is approximately 1000 to 2000 points in total for AI/AO/DI/DO.

I don’t think there are more than 100 points using control-like controls such as ratio control and cascade control.

Even if individual PID control is included, it will be within the range of 300 to 500 points.

In other words, more than 50% is simply on-off digital switching and fixed value control.

The easy control method is already established, and it is a combination game where you just match symbols.

You have to concentrate your efforts on special control.

Creating your own software means standardizing it as much as possible and reducing unique features.

Sequencer

Now, it is said that control in a batch chemical plant is extremely simple.

Up until now, almost all the control targets have been valves and instruments in piping.

As a machine shop, don’t you wonder how the equipment is controlled?

In fact, most equipment only has extremely simple controls.

The equipment that needs to be controlled is a rotating machine.

The control of that rotator moves . Only this.

• to move or not to move
• How fast to move it (inverter)

There are very few complex machines.

For example a centrifuge. For example, a refrigerator.

We do not create our own control programs for these types of equipment.

I usually ask the manufacturer.

Here, the manufacturer creates a program using a sequencer.

Install it on the control panel and connect the manufacturer’s control panel and the user’s control panel.

In this way, signals are exchanged and controlled between the manufacturer’s control panel and the user’s control panel.

Leave the complicated things out!

lastly

We have provided an overview of DCS that mechanical engineers should also know.

DCS system, cubicle nest card, signal, instruction record alarm, power supply, duplication, sequencer

I think you can become familiar with it just by understanding the vocabulary and outline.

Even if you’re a machine shop, if you keep avoiding DCS, you might have trouble eventually.

Please feel free to post your worries, questions, and questions about the design, maintenance, and operation of chemical plants in the comments section. (Comments are at the bottom of this article.)

*We will read all comments received and respond seriously.