17th IMEKO TC 10 and EUROLAB Virtual Conference

“Global Trends in Testing, Diagnostics & Inspection for 2030”

October 20-22, 2020.

Some considerations about choosing a method

for TCal
Sasho Andonov1, Prof. Dr. Marija Cundeva-Blajer2
1
M.Sc., Skopje, North Macedonia, [email protected], +389 78 482 560
2
PhD, Skopje, North Macedonia, [email protected], +389 75 603 222

Abstract – The global trends in testing, diagnostics and is a paper which actually addresses the Type B
inspections in manufacturing industry are strongly uncertainties which are strongly depending on choosing a
connected with development and implementation of method to transduce the calibration data into message on
Industry 4.0 framework. Two years ago, in the scope of Transmitter’s side (Calibration laboratory) and recreate
initiative Metrology for Industry 4.0, the general idea later the same calibration data from the message on
for Touchless Calibration (TCal) was presented as a Receiver’s side (manufacturer’s laboratory).
method for remote calibration of instruments in
manufacturing industry. II. CONCEPT OF TOUCHLESS CALIBRATION
By implementing the TCal, instead sending the
The simplified concept of TCal is explained on Fig. 1.
instruments, the companies will send only the
Sensor/Transducer to the laboratory. Before that, in
manufacturing premises they will provide Relative
calibration. It is actually “pairing” (with other words: Transmitter
providing a ”mirror image”) of the Sensor/Transducer CAL STD Sensor Tx CPU
and Sensor/Actuators, without having the particular &
expensive standard (etalon).
The method of transforming calibration data is IND 4.0 Network
actually determining the types and characteristics of
Sensor/Transducer and Sensor/Actuators. Choosing a
proper method, means not to increase the Type B CAL UN Sensor Rx CPU
&
uncertainties to the values which are comparable to the Receiver
classical calibration.
In this paper, authors will try to present some
considerations regarding choosing a method based on
transformation of calibration data into Fig. 1. Simplified flow-chart for TCal
Time/Frequency which can be generated with
extremely small uncertainties. The Sensor/Transducer gathers the calibration data
from the CAL STD (which is NMI/BIPM Laboratory
Keywords – TCal, Touchless Calibration, Sensors, Standard) on Transmitter side and particular transducing
Transducers, Actuators. into another quantity is executed. This quantity is
something which can be embeded into digital message
which will be used on Receiver side (the manufacturing
I. INTRODUCTION company) to recreate the calibration data. The recreated
The method of Touchless Calibration (TCal) was data are used for calibration of the CAL UN (Calibration
introduced on XXII IMEKO Congress in Belfast (UK) in Unit). So, the Laboratory Standard cannot “touch” the
2018 [1]. The uncertainties between TCal and classical CAL UN which is in manufacturing company’s premises.
calibration were presented in paper submitted to the Joint The gathered calibration data from the CAL STD is
IMEKO Symphosium in St. Petrsburg (Russia) in 2019 [2] associated with particular transducing into another
and the cost-benefit analysis between TCal and classical quantity which should be actually voltage or current
calibration was presented on IEEE MetroInd4.0&IoT because it can be easy processed. This quantity must be
conference in Rome in June this year [3]. embeded into digital message which will be used on
From all these papers, the benefit from TCal can be Receiver side (the manufacturing company) to recreate the
clearly noticed, but this is not a flawless method. So, this calibration data. The recreated data are used for calibration

Editors: Dr. Zsolt János Viharos; Prof. Lorenzo Ciani; Prof. Piotr Bilski & Mladen Jakovcic 47
17th
16th IMEKO
IMEKO TCTC1010 Conference
and EUROLAB Virtual Conference
“Global Trends in Testing, Diagnostics
“Testing, Diagnostics & Inspection as a&comprehensive
Inspection for value
2030”chain for Quality & Safety
October 20-22, 2020.
Berlin, Germany, on September 3-4, 2019

of the CAL UN. As it can be seen, the Laboratory Standard in genera, Type B uncertainties for TCal in total, will be
(CAL STD) cannot “touch” the CAL UN which is miles bigger. That is the reason that we must take care for all
away, in the manufacturing company’s premises. novel uncertainties introduced by these three factors to be
The point is that actually, the Sensor/Transducer and kept as low as possible.
Sensor/Actuator do not need to be integrated as Sensor and The novel Type B uncertainties (their variances!) can
Transducer or Sensor and Actuator. In general, Actuator be presented [2] by the equation (3):
does not even need to exist, because this is dependent from
particuar application where TCal is used. If TCal is used 2
𝜎𝑇𝐶𝑎𝑙 2 2 2 2
𝑛𝑜𝑣𝑒𝑙 = 𝜎𝑑𝑠 + 𝜎𝑒𝑛𝑣 + 𝜎𝑠𝑝 + 𝜎𝐴𝐷 (3)
to calibrate dynamical systems (especially in
manufacturing industries), then Actuator could be very The parameters in the equation (3) are:
much important part of calibration process.  σ2ds is uncertainty caused by incapability of the chosen
Again, on Fig. 1 is presented the general simplified method to provide “mirror image” of
diagram how TCal should look, but not necessary all parts Sensor/Transducer and Sensor/Actuator
need to be present. Of course that for the calibration of characteristics. This is actually the bigest contributor
quantities which are significantly dependent on to TCal novel uncertainties and it will be considered
environmental conditions (pressure, temperature, in details in next paragraph.
humidity, etc.), there is need to provide environmental  σ2env is the uncertainty due to possible difference of
conditions gathered from inside the calibration laboratory environmental conditions data on both sides. TCal in
and adjust them into the manufacturer’s laboratory. In such its general configuration provides also the
a case there is need for additional devices [1],[2] which environmental conditions data for the measurements
will provide “same room conditions” calibration data. where temperature, pressure or humidity can affect the
accuracy of calibration. For the purpose of this paper
III. NOVEL UNCERTAINTIES INTRODUCED BY it will not be exlained in details;
TCAL  σ2sp comes from the speed of communication which
The total uncertainty [4] for TCal calibrations in can be supported by Industry 4.0 network. TCal use is
general (similar to the classical calibration) (expressed by for manufacturing industry and as such it can take into
variances) can be presented by the equation (1): consideration the dynamics of themanufacturing
𝑘
processes which are controlled by some
instrumentation. This Type B uncertainty will be valid
2
𝜎𝑇𝐶𝑎𝑙 2
= ∑(𝜎𝑇𝐶𝑎𝑙𝐴𝑘 2
+ 𝜎𝑇𝐶𝑎𝑙𝐵𝑘 ) (1) only for calibrations of instruments needed for
𝑛=1 dynamic process control and it will not be expained in
σ2Ak is Type A uncertainty of k-th step and σ2Bk is Type B deatails here; and
uncertainty of k-th step. Looking form on Fig. 1, it is clear  σ2AD is uncertainty caused by eventual errors which
that for TCal, k = 1 for Type A uncertainties and k for Type analogue to digital convertors (ADC) on
B uncertainties depends on a chosen method. Transmitter’s side and digital to analogue convertors
Our interest are Type B uncertainties because these (DAC) on Receiver’s side, involved in the process
are uncertainties which are connected by the imperfection could introduce to the calibration process. As general
of the measurement system. For the sake of the accuracy concept, TCal could also provide calibration for
of presentation, it must be mentioned that the Type B instrumentation used for measurement of analogue
uncertainties present in classical calibration will exist also quantities (old equipment). It is known that Industry
in TCal, but the values of these uncertainties will be 4.0 is digital platform for communication and as such
considerably smaller. Reason for that is that these it will need digital signal to be transferred. So, the
unceratinties, similar to the Type A uncertaities, will be change of analogue data into digital data by the use of
apllied to less steps [2] and their total value will be smaller. ADC and DAC is necessity. The biggest error which
In general total TCal Type B uncertainties can be can happen due to digitalization of analogue output of
expressesed [2] as: Sensor&Transducers in TCal, is the quantization
error. More bits will gain better accuracy, so, choosing
2
𝜎𝑇𝐶𝑎𝑙 2 2 2 2 a 24-bits converter [5], [6], will provide relative error
𝐵 = 𝜎𝑇𝐶𝐶 𝐵 + 𝜎𝑇𝐶𝑎𝑙 𝑛𝑜𝑣𝑒𝑙 = 𝜎𝑇𝐶𝐶 𝐵 + (𝜎𝑑𝑠 (2)
2 2 2
+ 𝜎𝑒𝑛𝑣 + 𝜎𝑠𝑝 + 𝜎𝐴𝐷 ) equal to 0.06 x 10-6 for 10 volts full scale. As it can be
noticed, the range of this error is same as the range of
Josephson voltage standard uncertainty. Other errors
As it can be noticed from (2), due to introduction of
can show up due to temperature drifts, but there are
the three additional factors (Sensor/Transducer on
A/D (D/A) devices which have temperature sensor
Transmitter’s side, Industry 4.0 Network in between and
inside (as 24-bits Texas Instruments device ADS 1226
Sensor/Actuator on Receiver’s side), the TCal will
[6]), which are used to compensate the temperature
introduce in additiuon, few novel Type B uncertainties. So,
drift. So, the errors triggered by temperature drift are

Editors: Dr. Zsolt János Viharos; Prof. Lorenzo Ciani; Prof. Piotr Bilski & Mladen Jakovcic 48
17th TC 10Conference
16th IMEKO TC10 and EUROLAB Virtual Conference
“Global
“Testing,Trends in Testing,
Diagnostics Diagnostics
& Inspection as a&comprehensive
Inspection forvalue
2030”chain for Quality & Safety
October 20-22, 2020.
Berlin, Germany, on September 3-4, 2019

not significant if such a A/D (D/A) device is used. Receiver’s side (“mirror image”). The possible differences
That is the reason that this uncertainty is assumed to in the “pairing” can be mesured by the Null Detector and
be neglectable. they can be used as stadard deviation of variance σ2ds
(Fig.3).
Having in mind all these notions, it can be stated that
equation (3) could be written as:
“Mirror”
2 2 V1 f1 = f2 V2
𝜎𝑇𝐶𝑎𝑙 𝑛𝑜𝑣𝑒𝑙 ≈ 𝜎𝑑𝑠 (4) DC Voltage In VFC Out In FVC Out
Source

V1 V2
IV. UNCERTAINTY CAUSED BY INCAPABILITY TO Null
PROVIDE “MIRROR IMAGE” Detector

As mentioned in previous paragrpahs, the most σds2

important thing regarding the TCal is that the method
which is chosen must provide “a mirror image” (Fig. 2).
Fig. 3. Relative calibration for TCal

Transmitter The “pairing” during Relative calibration will be

Sensor& achived by producing same input/output characteristics for
CD Tx Processor
Transducer Sensor/Transducer and Sensor/Actuator. There the
diagram of “pairing” the input/output characteristics for
Industry 4.0 Voltage-to-Frequency Convertor (VFC) and Freqency-to-
“Mirror” Network Voltage Convertor (FVC) [2],[3] is presented on Fig.4.

RCD Sensor&
Rx Processor F (V)
Actuator
Receiver 𝝈𝒅𝒔

FVC
Fig. 2. “Mirror Image” for TCal
(VFC)

Whatever the used method for TCal is, the ”mirror

image” [2] means that Sensor/Actuator and Rx Processor
on the Receiver’s side of Fig. 2 must be image of
Sensor/Transducer and Tx Processor on the Transmitter’s V (F)
side of the Fig. 2. With other words: Whatever the
processing of calibration cata (CD) is applied on the Fig. 4. Adjusting input/output characteristics for VFC and FVC
Transmitter side, the inverse processing must be applied to for voltage TCal
the Receiver side to provide accurate recreated calibration
data (RCD) or expressed by formula: As it can be seen, the 𝜎𝑑𝑠 is not constant and as the
range of calibration is exteding, so the 𝜎𝑑𝑠 will increase
RCD = CD + σds (5) also. That is the reason that input/output characteristics of
the devices must be adjusted and it can be achieved only
Providing “a mirror image” must be tested (and by Null Detector used in Realtive calibration.
adjusted!) by Relative calibration (Fig. 3) [2] [7].
Using Fig. 2 and Fig. 3, the uncertainty σ2ds (in this case V. CHOOSING A METHOD FOR TRANSDUCING
expresed as standard deviation [7]) can be presented by CALIBRATION DATA FOR TCAL PURPOSES
following formula:
Principle of working for TCal is: Instead to send the
σds = RCD − CD (6) instruments to the laboratory for calibration, the
calibration laboratory will send to maunfacturer’s
So, the Relative calibration is needed to provide laboratory the calibration data and there, thes data just
“pairing” between Sensor/Transducer and Sensor/Actuator needs to be recreated. But this “recreation of data” on the
which is main condition to recreate the calibration data on Receiver’s side (manufacturer’s laboratory) is the biggest
issue!

Editors: Dr. Zsolt János Viharos; Prof. Lorenzo Ciani; Prof. Piotr Bilski & Mladen Jakovcic 49
17th IMEKO TC10
16th IMEKO TC 10Conference
and EUROLAB Virtual Conference
“Global Trends in Testing, Diagnostics
“Testing, Diagnostics & Inspection as a& Inspection forvalue
comprehensive 2030”chain for Quality & Safety
October 20-22, 2020.
Berlin, Germany, on September 3-4, 2019

Calibration data will be “sensed” by

Sensor/Transducer and there it will be transduced into 2
𝜎𝑇𝐶𝑎𝑙 2 2 2 2
𝑛𝑜𝑣𝑒𝑙 ≈ 𝜎𝑑𝑠 = 𝜎𝑆𝑇 + 𝜎𝐼4.0𝑁 + 𝜎𝑆𝐴 (7)
quantity which must be in such a form to be recreated in
manufacturer’s laboratory to provide calibration. There are where 𝜎𝑆𝑇2
is uncertainty introduced by Sensor/Transducer
two important characteristics when looking the appropriate on Transmitter side, 𝜎𝐼4.0𝑁
2
is uncertainty introduced by
method for TCal: Industry 4.0 Network and 𝜎𝑆𝐴 2
is uncertainty introduced by
A. Chosen method for transformation (transducing) Sensor/Actuator. It is clear that all these three uncertainties
calibration data needs to have considerably smaller need to be small to contribute to the TCal effectivness. If
uncertaity (at least 100 times!) than expected uncertainty they are at least 100 time smaller then the equation (2) will
of the quantity which needs to be calibrated; and be transformed into (8):
B. Chosen method (presented by type of
Sensor/Transducer and Sensor/Acuator!) needs to be easy σ2TCal B = σ2TCC B + σ2TCal novel ≈ σ2TCC B (8)
produced on Transmitter’s side and easy reproduced on the
Receiver’s side. So, let’s try to see from Table 1 what are the
uncertainty in measurements for different quantitities in
A. Smaller uncertainty different industries [8]…
It is wise if we choose to transduce calibration data
into quantity which will have considerably smaller
uncertainty then the calibrated quantity. The uncertainty
σ2ds from equations (2), (3) and (4) can be expressed as:

Table 1. Uncertainties of different standards for quantities in industry

Quantity Uncertainty Best Available Method Method used in Manufacturing Industry

Measurement of length in retail
Laser Measurements (Accurate, but very
manufacturing industry is done by
Length 10 -12
expensive!!! Not applicable in
calipers and uncertainty there is in range
manufacturing industry).
of 10-4.
Use of atomic (cesium and rubidium)
Frequency generators avaialbale on the
clocks (expensive, but if GPS satellites
Time 10-15 market can achieve accuracy of 10-6 and
and Radio Clock Transmitters are used
frequency meters of 10-8.
achieve uncertainty will be 10-12).
In this areas, the measurement
In the case of forces between 30 – 50 MN.
Force 10-3 instruments provide uncertainty of 10-3
For smaler forces (10 – 5 MN) only 10-2.
[9].
Electrical In the best case, this uncertainty applies to Most of the calibration laboratiries for
10-8
Quantities voltage measurements. industry could provide up to 10-6.
For industry and hospitals, the uncertainty
Ionization The best case scenario for gamma ray
10-7 can be 10-2 (best available personal
Radiation radiation.
instrument!)
The best case scenario in different For mercury termometer (industrial
Temperature 10-5
industries. purposes!) the uncertainty is 10-1 oC [10].

The Table 1 presents the best case scenarios for the particular industry, there is need for additional resarch of
standards in primary laboratories, but for industrial the available literature.
measurements, the manufacturer’s companies cannot Let’s say, there is company which is producing
always afford themselves such an uncertainties. It is products where dimensions are important. And let’s say,
clarified in the last column of Table 1. The main point with they are producing casings for electronic or
Table 1 is that these are just aproximately values. Actually, telecommunication equipment for retail.
calculating of accurate values is pretty much complex Could such a company afford to have laser instrument
process for all ranges and all purposes of the for measuring the box dimensions?
measurements, so the values inside Table 1 should be The answer is: Maybe, but it will be too expensive!
accepted as general numbers. For any specific purposes in And what is the benefit of that?

Editors: Dr. Zsolt János Viharos; Prof. Lorenzo Ciani; Prof. Piotr Bilski & Mladen Jakovcic 50
17th
16th IMEKO
IMEKO TCTC1010 Conference
and EUROLAB Virtual Conference
“Global
“Testing,Trends in Testing,
Diagnostics Diagnostics
& Inspection as a&comprehensive
Inspection for value
2030”chain for Quality & Safety
October 20-22, 2020.
Berlin, Germany, on September 3-4, 2019

Bying such a laser measurement equipment is defying or frequency could be very benefitial from the aspect of
the business logic. In the business and manufacturing Type B uncertainties.
industry, there is a simple “rule of the tumb” which is
known to every beginner and expert in economy: When the B. Easy to create and recreate TCal calibration data
costs of production are going up, the profit is going down. Using a time/frequency as method for transducing
The tolerances for these products in manufacturing calibration data is good even from point of view of easy
industry are 10-3 m and why would such a company invest creation and recreation of these quantities. The
into laser measurement equipment which costs thousands time/frequency generation and measurement with
of Euros? Uncertainty of 10-12 is needed only for some available time/frequency signal generators and
aplications in some industries (space, aviation, etc.) and measurement instruments is very good. It can be seen from
even there only particular application requires such an the last column in Table 1. In addition, the pricess of these
uncertainty of the measurements. instruments are not so high as for Vector Analyser or some
So, nevertheless there are measurement systems other high-tech instrument.
which could provide extremely good accuracy and Going further, for implementation of TCal for
precisions, the companies in manufacturing industry do calibration of the quantities which have uncertainties of the
not need to invest in those systems. range of 10-3 or higher, using time/frequency
In addition, looking the Table 1, it can be noticed that Sensor/Transducer and Sensor/Actuator can cancel a need
Time measurements can provide very good for Relative calibration. Simply, having in mind that
uncertaininties. The point is that there are another two Frequency Metters and Signal Generators can provide
benefits of using Time mesaurements: uncertainty of 10-6, the value for 𝜎𝑑𝑠 will be 10-3 times
a) The good (and cheap!) time standards are widely smaller then required uncertainties, so adjusting 𝜎𝑑𝑠 by
available through satellite navigation (GNSS1) and Relative calibration can be neglected.
Radio Clock Transmitters; and In addition, for such a case, there is even no need to
b) Time can be transmitted as signal with particular send a Sensor/Transducer to calibration laboratory.
frequency which is actually convinient for TCal Simply, the calibration laboratories could provide all
purposes. necessary calibration information (calibration data and
environmental data) to their website and the
To be more precise, something very similar to TCal is manufacturer’s company could actualy access these data
used in satellite navigation. Actually, satellite navigation there.
is working on this principle: Of course that calibration laboratories need to be paid
The GNSS satellites are equiped by one or two atomic for these services, but this can be easy achieved, simply by
clocks (cesium and/or rubidium) and the satellites have providing User Name and Passwords for the
embeded into each of its periodically transmiting signal the manufacturerer’s laboratoris which are willing or have
burst of extremly accurate frequency [11]. The GNSS already pre-paid for the calibration services.
receiver on the ground is equipped by cheap and inaccurate
quartz clock, but by extracting this frequency from the VI. USING PIEZOELCTRIC CRYSTAL OR
GNSS signal, the receiver use this frequency to calibrate SEMICONDUCTORS TO BUILD SENSORS
its quartz clock. Due to imperfection of the quartz clock,
the calibrated GNSS receiver can keep the accuracy for Investigation all possibilities to find proper method to
very short time, but this time is enough to process the implement TCal in reality, the authors investigated a lot of
signal and to provide navigational information about possibilities and one which showed considerable
location with uncertainty of 10-8 (for purposes of aviation). advantage is a use of piezoelectric crystal. The
GNSS satellites from different consteekkations have piezoelectric crystal for building Sensor/Transducer on
different period for transmitting GNSS messages. For Transmitter’s side and Sensor (which will provide
example, the GALILEO GNSS signals (messages) are Actuator with necessary data to adjust values on calibrated
transmitted periodically every 12.5 minutes [12], so the the CAL UN, if needed) on Receiver’s side has a lot of
calibrated accuracy of the quartz clock in the GNSS advantages.
receiver must be kept for that time. It is well known that piezoelectric crystal [13] has a
By using a transducers for TCal which could feature to generate a voltage when a force (strain, pressure,
transduce any quantity (subject of calibration!) into time etc.) is applied to it. The generated voltage is proportional
to the change of dimensions of the crystal which are

1
GNSS is acronym for satellite navigation (used in avi- (Russia) and BeiDou (China). There is no interopera-
ation). GNSS is build up by 4 satellite navigation con- bility between these four systems, but GNSS receivers
stellations: GPS (USA), GALILEO (EU) GLONASS used for aviation purposes should be capable to receive
any of these signals and to process them.

Editors: Dr. Zsolt János Viharos; Prof. Lorenzo Ciani; Prof. Piotr Bilski & Mladen Jakovcic 51
17th IMEKO TC10
16th IMEKO TC 10Conference
and EUROLAB Virtual Conference
“Global
“Testing,Trends in Testing,
Diagnostics Diagnostics
& Inspection as a&comprehensive
Inspection forvalue
2030”chain for Quality & Safety
October 20-22, 2020.
Berlin, Germany, on September 3-4, 2019

proportional (linearly in particular range) to the force REFERENCES

applied. This voltage can be measured with considerably [1] Andonov S. Cundeva-Blajer M. 2018, Calibration for
small uncertainties (10-6) and as such can be used as Industry 4.0 Metrology: Touchless Calibration, Journal of
calibration information. In addition, the opposite will Physics: Conference Series, Volume 1065, Measurement
happen: The piezoelctric crystal will change its dimensions Science
when a voltage is applied. [2] Andonov S. Cundeva-Blajer M. 2019, TCal: Contri-
The benefit of using piezoelectric crystal is that bution to Metrology for Industry 4.0, Journal of Phys-
pairing between two crystals for Sensors on Transmitter’s ics: Conference Series, Volume 1379, Number 1,
and Receiver’s side can be achieved with very high Measurement Science
accuracy during manufacturing process when Sensors are [3] Andonov S. Cundeva-Blajer M. 2020, Comparative
produced. Cost and Benefit Analysis of TCal and Classical Cal-
The problem is how to transfer this voltage into ibration, 2020, IEEE International Workshop on Me-
time/frequency calibration data (information). There is trology for Industry 4.0 & IoT, Rome, Italy, IEEE
possibility to use Voltage to Frequency Convertor (VFC) Explore, 2020
on Transmitter side and Frequency to Voltage Convertor [4] JCGM 100:2008, GUM 1995 with minor corrections,
(FVC) on Receiver’s side, but this solution is not Evaluation of measurement data — Guide to the
applicable for accurate measurements. Adjusting the expression of uncertainty in measurement, 1 st Ed.,
“mirror” image between VFC and FVC could be a problem September 2008
due to difference in linearity of both devices. Anyway, it [5] W. Kester, The Importance of Data Converter Static
can be solved by providing more complicated circuits Specifications—Don't Lose Sight of the Basics, MT-
which will be temperature compensated and it could work. 010 Tutorial, Analog Devices Inc, USA, Oct, 2008
In addition, the semiconductors can be used for some [6] Technical Specification for ADS 1225/ADS 1226 (24-
types of Sensors. The semiconductors share the good Bit Analog-to-Digital Converter with One- and Two-
ability of “pairing” between two devices during Channel Differential Inputs and Internal Oscillator),
manufacturing process. For force (strain, pressure, etc.) Texas Instruments, SBAS346C, USA, May 2006–
measurements, the change of resistance of the seicinductor Revised January 2009
material can be used. There are already semicinductor’s [7] S. G. Rabinovich, Measurement Errors and
Uncertainties: Theory and Practice, 3rd Ed. Springer,
sensors for pressure in aviation, where accuracy is very New York, USA, 2005
much important.
[8] Y. V. Tarbeyev, Measurements and Measurement
Standards, 2011, Physical Methods, Instruments and
VII. CONCLUSIONS AND OUTLOOK Measurements – Vol. I, UNESCO, Encyclopedia of
Two most important thing with TCal is to choose Life Support Systems (EOLSS)
effective and efficient method for achieving requested https://www.eolss.net/Sample-Chapters/C05/E6-08-
calibration accuracy and precission and to provide pairing 01.pdf
between Sensor/Transducer and Sensor/Actuator. [9] Force Measurement Instruments, 2017, Ametek (Pro-
The consideration regarding methods and sensors for duction Catalog), 2017
TCal in this paper should be used as guidance for the https://www.ametektest.com/-/me-
companies which would like to enter this business. There dia/ametektest/download_links/catalogue-chatillon-
are already considerable production technology for force-measurement-instruments.pdf?la=en
building stable sensors with good accuracy and precission [10] Cotrell, M.S., Temperature, 2006, Tutorial for Course
and only additional efforts needs to be put to materialize Mechanical Instrumentation (ME 240), Missouri Uni-
idea of TCal in different areas. versity of Science, Rollo, USA, July, 2020
Using piezoelectric crystals or semiconductors, https://web.mst.edu/~cottrell/ME240/Resources/Tem
having in mind advance in these technologies, could give perature/Temperature.pdf
good results to provide sensors which will support TCal. [11] Blewitt, G., Basics of GPS Technique: Observation
In general, TCal will bring more economic benefit for Equations, article in the text book “Geodetic Applica-
manufacturing industry, but there is more research in areas tions of GPS”, Swedish Land Survey, Stockholm, Swe-
of adapting sensor’s technology to needs of TCal. Overall den, 1997
impementation of TCal will need additional efforst from http://www.nbmg.unr.edu/staff/pdfs/blewitt%20basic
s%20of%20gps.pdf
the manufaturing companies as well as from calibration
[12] Navipedia, website of European Space Agency; July,
laboratories to adapt their processes to TCal. 2020
Anyway, if this is realised and implemented, the https://gssc.esa.int/navipedia/index.php/GPS_Naviga
maintenance of the TCal system for calibration tion_Message
laboratories will be very cheap. The benefits for the [13] Prokic, M., Piezoelectric Transducers: Modelling and
manufacturing companies are already explained in [3]. Characterization, 2004, MP Interconsulting, Le Locle,
Switzerland, 2004

Editors: Dr. Zsolt János Viharos; Prof. Lorenzo Ciani; Prof. Piotr Bilski & Mladen Jakovcic 52