Process Control

Field Instrumentation
 Sensors

• We have been
using a variation
of this diagram for
the entire course
• Let’s now focus
on the measuring
device (or the
sensor)
https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcSy_IxXpxwPJZrJbsaUtnMdA_sps4BFoRY3sruSe0ESv57ZRu4i

Slide 2
 Some Issues
• (Some obvious ones)
What is the parameter we want?
• (Some not so obvious ones)
WHERE is the parameter we want?
Is the parameter we want PORTABLE?
Will it be expensive to get the parameter?
Once we have it, what do we do with it?

Slide 3
 And … quick review
• Some control loops
are open loop.
These are ones
that we do in
manual, or where
very low
interaction is
required.

https://www.researchgate.net/profile/G_Khatri/publication/275003974/figure/fig1
4/AS:311308386881549@1451233111654/General-block-diagram-of-control-loops-
Slide aOpen-loop-and-bClosed-loop-or-feedback-loop.png
4
 Different ways to
categorise the data
• WHERE will the data reside, and what will we do with it?
Type of Data Location Issues
Local Field • Cheap
• Perfect for open loop control where operator not needed often
In Control Room Control • Expensive (need cables and programming to move it to control
Display only Room room)
• Perfect for open loop control where operator not needed often
In Control Room Control • Expensive (need cables and programming to move it to control
active in a loop Room room)
• Needed for closed loop control

Slide 5
 Different ways to
categorise the data
• WHERE do you put the sensor?

• Sometimes we put the sensor in the process

(expensive, need to shut down the process
to fix the sensor which is not good $$$)
• Sometimes we bring the process to the
sensor (cheap, can isolate the sensor from
the process, so we can fix the sensor
without shutting down the process).
Slide 6
https://teenalmyers.files.wordpress.com/2014/04/goodidea.jpg
 Different ways to categorise
the data
• Instantaneous or trend? Instantaneous data is used once and
“thrown away” – while trend is used and saved and analysed. Trend
requires data storage

Slide 7
 Different ways to
categorise the data
• Control or Safety
• Remember the layers of protection? When you get to the boundary
condition of your control, the control component shrinks and the
safety component grows

Safety Control Safety

Slide 8
 How good does the data
have to be?
• This is a wastewater treatment process
• How accurate does the temperature data need to be?
• How accurate does the pressure data
need to be?
• How accurate does the composition
data need to be?
• How accurate does the flow data need
to be?
• Data cost money Slide 9
https://www.conserve-energy-future.com/wp-content/uploads/2016/01/Wastewater-Treatment-Plant-e1454500044384.jpg
 Data is not straight forward
• We never measure the parameter.
• We always measure something else, and then use mathematics to
get our parameter.

• A speedometer does not measure speed. What does it measure?

Slide 10
 The 4 (5) basic parameters
Parameter Issues and Comments
Pressure • Measured using Force or Force balance (therefore,
gauge pressure usually measures)
• Cheap
• Portable
• Robust signal
• FAST!!!
• Data can be collected outside the process
• <John likes pressure data>
Slide 11
 The 4 (5) basic parameters
Parameter Issues and Comments
Temperature • Measured using Force for local instruments or
electricity for control. Can use laser for control.
• Cheap (laser not cheap)
• Data not portable at all
• Weak signal
• SLOW!!!
• Device has no strength, so it needs its own
“house” so it can collect data
• <John likes temperature data> Slide 12
 The 4 (5) basic parameters
• This is a thermowell. It is open at one end (so we can install our temperature
device) and sealed at the other.
• The temperature device “lives” inside the thermowell
• The thermowell protects the device from the process http://p.globalsources.com/IMAGES/PDT/B1078749752/Tantalum-Thermowell-and-Sleeves--Tantalum-
Thermowell-Tantalum-Sleeves-Tantalum-tube-Tantalum-Rod.jpg

pressures (anything above trivial pressure will cause our

instrument to be crushed). http://www3.emersonprocess.com/rosemount/TCTP2_External/Calcul
ation.aspx?local=en-us

• Must be inserted into the process. Small

diameter piping will probably require the
thermowell to be https://encrypted-
tbn0.gstatic.com/images?q=tbn:ANd9GcQAhThmi6sV_
RWrVk4lpkeDgNxnTc91paMHpkjcksIYtj61W8MM

inserted in an elbow

Slide 13
 The 4 (5) basic parameters
Parameter Issues and Comments
Level • What do you want, top of the liquid or bottom of the
vapour?
• What do you want, level or inventory?
• Multiple ways, multiple technologies
• Top of liquid data more portable and can be moved outside
of process but not as accurate
• Cheap to expensive
• Semi-fast to fast (depends on technology)
• Weak to strong (depends on technology)
• <John tolerates level measurement> Slide 14
 Before I begin, What level are you
looking for – simple phases?
• Gas / Vapour

• Oil / Grease

Usually we want
the top of the liquid

Slide CV 1-15
 What level are you looking for –
multiple phases?
• Gas / Vapour
• Foam
• Oil / Grease
• Emulsion
• Water
• Solids

• Usually we want
the top of the liquid
• Which level do you want?
Slide CV 1-16
 The 4 (5) basic parameters
Parameter Issues and Comments
Flow • Dozens of ways to measure the flow – all involve getting
the sensor into the process
• Cannot maintain device without shutdown ($$$)
• Most expensive of the four basic parameters – some
flow meters are REALLY expensive
• Most difficult of the four basic parameters to get
accurate
• Medium strength, speed varies generally medium speed
• John tries to minimise flow meters (cost) Slide 17
 The 4 (5) basic parameters
• You can measure the mass flow directly.
• This is a Coriolis meter. Note the size. I expect it would
cost about $10000USD
• Cost grows exponentially with size.
• If you want one for a (say) DN300 pipe, expect to pay at
least 8 digits in cost to buy one.
• Do you really need one?

Slide 18Images provided by www. bronkhorst-cori-tech.com

 The 4 (5) basic parameters
Parameter Issues and Comments
Composition • Because it requires a small “laboratory”, must bring
data out of process to sensor
• Expensive
• Slow (really, really slow)
• Generally not robust

Slide 19
 The 4 (5) basic parameters
https://www.yokogawa.com/au/solutions/products-platforms/process-
analyzers/gas-analyzers/process-gas-chromatographs/process-gas-

• To measure composition, we need an

chromatograph-gc8000/

automated “laboratory” in our facility

• We need to clean the sample so we do
not destroy our little “laboratory”

• Whenever I tried to find an image of

an industrial gas chromatograph, this
image kept coming up, so I guess this
is an industrial on-line GC (not 100%
certain). The bottom is the “lab” and
the top is the required electronics.
Slide 20
 Paradigm shift
• Of the 5 basic parameters, 4 measure accumulation

• You really need to complete the transition from your first

thought being “mass in =“

Slide 21
 Local Data
• Good for spot checks (everything is normal enough)
• Cheap (no cables, no software, no programming)

• Note … there are 3 pumps –

which pump is running?
• Heat exchangers – temp is
good spot check

Slide 22
 Local Data
• Next time you walk
past these firewater
tanks, check the
level in the tanks.

Slide 23
 Data for Trends
• It is possible you will
have so much data,
you will need to
investigate BIG DATA
and IIoT to manage all
of your data.

• Non local data has

lots of extra costs.
Slide 24
https://www.azosensors.com/images/Article_Images/ImageForArticle_964(1).jpg
 Different ways to categorise
the data
• I will talk more about this another day.

Safety Control Safety

Slide 25
 Take home …
• WHERE is the data
• What kind of data do we require
• Data is not free

Slide 26