2. 2
Temperature
What is temperature ?
The temperature of an object can be described
as that which determines the sensation of
warmth or coldness felt from contact with it.
www.AutomationSys.Org
021-88535820-21
021-88535820-21
4. 4
Modern temperature scales
Fahrenheit
Celsius
Kelvin
Rankine
www.AutomationSys.Org
021-88535820-21
021-88535820-21
5. 5
Learning Objectives:
Temperature Sensors
Have a basic understanding of operation
principles for both an RTD and T/C and
understand the benefits of using one vs the other
Know the T-well styles and how to select the most
appropriate for processes
Know the main factors that affect time response
7. 7
RTD Sensors
What is an RTD?
What is an RTD?
Resistance Temperature Detector
Operation depends on inherent characteristic of metal (Platinum
usually): electrical resistance changes when a metal undergoes a
change in temperature.
If we can measure the resistance in the metal, we know the temperature!
Platinum
resistance
changes
with temperature
Rosemount’s
Series 65 (High Temperature),78, 88
Rosemount’s
Series 65, 68, 58
Two common types of RTD elements:
Wire-wound sensing element
Thin-film sensing element
8. 8
RTD Sensors
Resistance Changes are Repeatable
What is a RTD?
– Resistance Temperature Detector
– The resistance changes of the platinum wiring can be approximated by
an ideal curve -- the EN 60751 curve (Pt100)
0
50
100
150
200
250
300
350
-200 0 200 400 600 800
o
C Ohms
0 100.00
10 103.90
20 107.79
30 111.67
Resistance
(Ohms)
Temperature (o
C)
International Resistance
vs. Temperature Chart:
EN 60751
EN 60751
9. 9
EN 60751 Tolerances
Accuracy
Temperature Resistance Grade A Grade A Grade B Grade B
°C Ohms ± °C ± Ohms ± °C ± Ohms
-200 18.52 0.55 0.24 1.3 0.56
-100 60.26 0.35 0.14 0.8 0.32
0 100.00 0.15 0.06 0.3 0.12
100 138.51 0.35 0.13 0.8 0.30
200 175.85 0.55 0.2 1.3 0.48
300 212.05 0.75 0.27 1.8 0.64
400 247.09 0.95 0.33 2.3 0.79
500 280.98 1.15 0.38 2.8 0.93
600 313.71 1.35 0.43 3.3 1.06
10. 10
RTD Sensors
RTDs Do Not Follow the Ideal Curve
TEMPERATURE (o
C)
RESISTANCE
(OHMS)
Ideal EN
60751
Curve
The EN 60751 standard
describes an IDEAL
Resistance vs. Temperature
relationship for Pt100
385 RTDs.
Every Pt100 385 RTD is
slightly different - they’re not ideal!
Real Sensor Curve
Because RTD sensors are not ideal - they do not perfectly match
the EN 60751 - there will be error introduced into the temperature
measurement.
This Error is the Sensor Interchangeability Error!
11. 11
Exercise:
Find the Interchangeability Error
Your customer is operating a process
at 100°C and is using a Platinum RTD...
What is the maximum error that will be
introduced into the temperature measurement
from Sensor Interchangeability?
+/-0.35 deg C for Class A,
+/-0.8 deg C for Class B
Fortunately, Sensor Interchangeability Error
can be reduced or eliminated by calibration!
12. 12
Red
Red
White
White
RTD Sensors
2, 3, and 4-wire RTDs
4-wire RTD
Why use a 2-, 3-, or 4- wire RTD?
2-wire: Lowest cost -- rarely used due to high error from lead
wire resistance
3-wire: Good balance of cost and performance. Good lead
wire compensation.
4-wire: Theoretically the best lead wire compensation method
(fully compensates); the most accurate solution. Highest cost.
Typically use copper wires for
extension from the sensor
Sensing Element
(I.e. wire-wound, thin film)
13. 13
2-wire or 4-wire RTD ?
If the sensing element is at 20°C,
F What would be the temperature measured at the end of the extension
wire using a 2-wire assembly
F What would be the temperature measured at the end of the extension
wire using a 4-wire assembly
Red
White
2-wire RTD
6 metres of copper extension
wire, lead resistance =
0.06 ohms/metre
(1 ohm = 2.5 deg C approx)
Sensing Element
(I.e. wire-wound, thin film)
Temperature questions.
Error for a 2 wire assembly
0.06 x 6 x 2 = 0.72 ohms or 1.8Deg
C
This means that the temperature
measured at the end of the cable
would be 21.8 Deg C
Error for a 4 wire assembly
As the lead resistances can be
accounted for the temperature
measured at the end of the cable would
be 20.0 Deg C
14. 14
Temperature Sensors & Accessories -
Types of RTDs
Dual Element RTDs available
– Rosemount 65 Series RTD
Red
Red
White
Black
Red
Red
Green
Blue
Blue
White
Dual Element:
Two 3-wire RTDs
Supports Hot Backup capability
Dual element adds only $5 over single element RTD!
Reduce the risk of a temperature point failure
18. 18
AGENDA
Temperature Sensors and Accessories
– RTD Sensors, Rosemount Offering
– Thermocouple Sensors, Rosemount Offering
– Thermocouple vs. RTD Sensors
– RTD Calibration (Characterization)
– Accessories
– Time Response
19. 19
Temperature Sensors & Accessories -
Thermocouples
Process
Process
Temperature
Temperature
Hot junction
– Two dissimilar metals joined at a “Hot” junction
+
-
Cold junction
MV
T
What is a Thermocouple?
Seebeck Effect
In 1821, Thomas Seebeck discovered that if two wires of dissimilar metals were
joined together at both ends and one of the ends is heated,a continuous current
flows through this thermoelectric circuit.
If this circuit is broken, the net open circuit voltage is a function of the junction
temperatures and the composition of the two metals.
How do they work
20. 20
Temperature Sensors & Accessories -
Thermocouples
What is a Thermocouple?
– The measured voltage is proportional to the temperature difference
temperature difference
between the hot and cold junction!
+
-
MV
Heat
Hot junction Cold junction
Voltage
(mV)
Temperature (o
C)
o
C Millivolts
0 0.000
200 8.137
400 16.395
600 24.902
800 33.277
1000 41.289
Thermoelectric Voltage
vs. Temperature Chart:
TYPE K THERMOCOUPLE
T
-10
0
10
20
30
40
50
60
-50
200
450
700
950
1200
21. 21
Temperature Sensors & Accessories -
Thermocouples
Temperature range
Cost
Signal level
Why would you use one type of thermocouple over another?
Why would you use one type of thermocouple over another?
0
5
10
15
20
25
30
35
40
45
50
0 200 400 600 800 1000 1200 1400 1600 1800
Type R
21.0 mV at 1760°C
Type J
42.3 mV at 760°C
Millivolts
Temperature (C)
Type J
Type R
22. 22
Correct!
Wrong!
All thermocouple lead wire extensions MUST be with the
same type of wire or the correct compensating cable!
Another Hot
Junction is
created… not
good!
Temperature Sensors & Accessories -
Thermocouples
Cannot use copper wire for extensions! T/C wire is more
expensive to run and much more care must be taken with
installation!
23. 23
Exercise:
Find the Temperature
A type K thermocouple is lying on
your desk.
You hook up a voltmeter to the
free end of the thermocouple.
What voltage would the voltmeter
read
What’s the
What’s the
Temperature?
Temperature?
The thermostat in the room reads
20°C.
o
C Millivolts
0 0.000
10 0.397
20 0.798
30 1.801
Thermoelectric Voltage
vs. Temperature Chart:
TYPE K THERMOCOUPLE
Answer
0.000mV
because both
junctions are at
the same
temperature
24. 24
Temperature Sensors & Accessories -
Rosemount T/C Offering
Type J
Type J
– Iron / Constantan
» Black, White,
» -40 to 750 °C
» Least Expensive
Rosemount Thermocouple Offering: Series 185
Rosemount Thermocouple Offering: Series 185
Type K
Type K
– Chromel / Alumel
» Green, White
» -40 to 1100 °C
» Most Linear
Type T
Type T
– Copper / Constantan
» Brown, White
» -40 to 350 °C
» Suitable for use at
lower temperatures
+ - + -
+ -
J K T N
J K T N
Colour Codes to IEC 60584
Colour Codes to IEC 60584
+ -
Type N
Type N
– Ni-Cr-Si / Ni-Si-Mg
» Pink, White
» -40 to 1200 °C
» Alternative to Type K
but more stable at
higher temperatures
25. 25
Temperature Sensors & Accessories -
Special Thermocouples
Type B
Type B
– Pt, 6% Rh / Pt, 30% Rh
» 38 to 1800 °C
Type S
Type S
– Pt, 10% Rh / Pt
» -50 to 1540o
C
Type R
Type R
– Pt, 13% Rh / Pt
» -50 to 1540 °C
Other T/Cs
High temperature range
Industrial/ laboratory standards
LOW EMF output!
26. 26
AGENDA
Temperature Sensors and Accessories
– RTD Sensors, Rosemount Offering
– Thermocouple Sensors, Rosemount Offering
– Thermocouples vs. RTDs
– RTD Calibration (Characterization)
– Accessories
– Time Response
27. 27
Temperature Sensors & Accessories -
Thermocouples vs. RTDs
Why would you use a RTD over a thermocouple?
Why would you use a RTD over a thermocouple?
28. 28
Temperature Sensors & Accessories -
Thermocouples vs. RTDs
Better Stability
Better Stability
• T/C drift is erratic and unpredictable; RTD’s drift predictably
• T/C’s cannot be recalibrated
Greater Flexibility
Greater Flexibility
• Special extension wires not needed
• Don’t need to be careful with cold junctions
Why would you use a RTD over a thermocouple?
Why would you use a RTD over a thermocouple?
Better Accuracy & Repeatability
Better Accuracy & Repeatability
• RTD signal less susceptible to noise
• Better linearity
• RTD can be “matched” to transmitter (Interchangeability error
eliminated)
• CJC error inherent with T/C’s; RTD’s lead wire resistance errors
can be eliminated
29. 29
Temperature Sensors & Accessories -
Thermocouples vs. RTDs
Why would you use a thermocouple over a RTD?
Why would you use a thermocouple over a RTD?
30. 30
Temperature Sensors & Accessories -
Thermocouples vs. RTDs
Applications for Higher Temperatures
Applications for Higher Temperatures
• Above 550°C
Lower Element Cost
Lower Element Cost
Faster response time
Faster response time
• Especially when the hot junction is grounded to the
sensor sheath
More rugged
More rugged
Why would you use a thermocouple over a RTD?
Why would you use a thermocouple over a RTD?
31. 31
AGENDA
Temperature Sensors and Accessories
– RTD Sensors, Rosemount Offering
– Thermocouple Sensors, Rosemount Offering
– Thermocouples vs. RTDs
– RTD Calibration (Characterization)
– Accessories
– Time Response
32. 32
EN
60751 Curve
The EN 60751 standard (programmed into all our transmitters)
describes an IDEAL Resistance vs. Temperature relationship
for Pt100 385 RTDs.
Temperature Sensors & Accessories -
RTD Calibration
TEMPERATURE (o
C)
RESISTANCE
(OHMS)
Class B
Tolerance
Standard EN 60751 Curve
Class B Tolerance
0.8o
C at -100o
C
0.3o
C at 0o
C
0.8o
C at 100o
C
1.3o
C at 200o
C
1.8o
C at 300o
C
2.3o
C at 400o
C
The goal is to find out what the real RTD
curve looks like, and reprogram the
transmitter to use the “real” curve!
Every RTD is slightly
different - they’re not ideal!
33. 33
1 and 2-point Trim Reduce Sensor
Interchangeability Error
0
50
100
150
200
250
300
350
400
-200 0 200 400 600 800
A 1-point trim shifts the ideal
curve up or down based on
the single characterized point
Temperature (°C)
Resistance
()
• Use one or two points to trim
the sensor to a transmitter
A 2-point trim shifts the ideal curve
up or down AND changes the slope
based on the two characterized
points
Temperature (°C)
Resistance
()
0
50
100
150
200
250
300
350
400
-200 0 200 400 600 800
Calibration option X8: Customer receives actual resistance values at two temperatures
Calibration Option X9: Customer receives actual resistance values at one temperature
One Point Trim
Use with X9
Two Point Trim
Use with X8
34. 34
Callendar-Van Dusen Constants Generate
True RTD Curve
Ideal IEC751 curve is replaced (in
the transmitter) with the REAL
curve of the sensor
Temperature (°C)
Resistance
()
Customer Receives
RTD-specific constants: • The V options also provides the
customer with Callendar-Van Dusen
Constants
• These can be plugged into the
transmitter to create the REAL RTD
curve!
Ro
99.997
0.003845
1.61027
0.172491
IEC
751 Curve
REAL
RTD
Curve
35. 35
V Options Provide Callendar-Van Dusen
Constants
» Sensor characterized over standard temperature range
» Callendar-Van Dusen constants provided on sensor tag
» Resistance vs. temperature table not included
Ro = 99.9717
= 0.00385367
= 0.172491
= 1.61027
The V Options also provide the customer with
Callendar-Van Dusen Constants
When you order a sensor with a ‘V’ option, you receive the C.V.D. constants
on a metal tag! You do NOT receive an R vs T table for that sensor!
36. 36
Callendar-Van Dusen Constants
Map the Real RTD Curve
Rt = Ro + Ro[t-(0.01t-1)(0.01t)-(0.01t-1)(0.01t)3
]
Rt = Resistance at Temperature t (°C)
Ro = Sensor-Specific Constant (Resistance at t = 0°C)
= Sensor-Specific Constant
= Sensor-Specific Constant
= Sensor-Specific Constant
Using the Callendar-Van Dusen constants
ELIMINATES sensor interchangeability error!
Temperature (o
C)
Resistance
()
Class B
Tolerance
• The transmitter does not use the
EN 60751 standard curve.
• Instead, the C.V.D. constants are
used in the equation below to
create the true sensor curve
EN
60751 Curve
37. 37
Accuracy is just as Important as
Repeatability
Model based control algorithm
Instrument failure
– Replacement interchangeability
Process cloning
There are many reasons why a customer
should be concerned about accuracy in
addition to repeatability:
Transmitter/Sensor matching and referencing are easy ways to
increase the accuracy of a temperature point measurement!
38. 38
AGENDA
Temperature Sensors and Accessories
– RTD Sensors, Rosemount Offering
– Thermocouple Sensors, Rosemount Offering
– Thermocouples vs. RTDs
– RTD Calibration (Characterization)
– Accessories
– Time Response
39. 39
Sensor Accessories -
Sensor Connection Heads
– Rosemount/Heraeus Heads are available for
» General purpose sensors
» A range of IP Ratings (See Data Sheets)
» Terminal block sensor, or flying lead sensor
» Single or dual-element sensors
» Epoxy coated aluminium,or stainless steel or plastic options available
» CENELEC/BASEEFA approved version available
Used for head mounted transmitters, or sensors with Term.Blocks
Typical Rosemount Connection
Head (Other styles are available
see Data Sheets)
40. 40
Sensor Accessories -
Extension Fittings
Extension fittings are used for a number of reasons
including:
– Heat dissipation from the process to the transmitter
– Extend sensor through tank
jacket or pipe insulation
– Ease of accessibility through
mounting in hard to reach areas
– Disconnect sensor from process
without full disassembly (Union)
Various types
– to suit customer requirements
0
10
20
30
40
50
60
75
100
125
150
175
200
225
Thermow ell extension mm
Temperature
rise
C
above
ambient
Process
Temp.= 250 °C
Process
Temp.= 540 °C
Process
Temp.= 815 °C
41. 41
Temperature Product Training -
Thermowells
What is a thermowell (T-well) ?
– A unit that protects a sensor from process
flow, pressure, and corrosion
– Allows for sensor removal without process
shutdown
– Slows response time (by typically 5 times)
Why are there different material types?
– To handle different corrosive environments
– To handle different temperature and pressure limits
42. 42
Temperature Product Training -
Thermowells
Thermowell Failure
T-wells can fail under certain conditions
Fluid flowing around the T-well forms a
turbulent wake called the Von Karman trail
The wake alternates from side to side at a
specific frequency dependent on many
variables
If that frequency exceeds 80% of the T-well’s
natural frequency, the T-well can fail!
43. 43
Preventing Thermowell Failure
– Application acceptability can be determined by knowing
the following:
» Thermowell Style
» Thermowell Material
» Thermowell Dimensions
» Fluid Velocity
» Process Pressure
» Fluid Density
» Process Temperature
» T-well failure calculations can be carried out by Rosemount
Temperature Applications Groups
Temperature Product Training -
Thermowells
Look on the back of your Sensor PDS!
44. 44
Thermowell Design Styles:
Comparison Table
*
Tapered
Stepped
Straight
Process
Pressure
Time
Response
Wake
Frequency
1
3
2
1 or 2
3
1 or 2
1
1
1
Drag
Force
2
1
2
Rating:
1= Best
Stepped
Straight
Tapered
45. 45
Thermowell Mounting Styles
Threaded
» Most common
» Easy to remove and install
Weld in (Not available with tubular wells)
» Non-removable
» Used in high velocity, temperature and pressure
fluids
» Used in non-leak applications
Flanged
» Used in corrosive environments
» Used in high velocity, and high temperatures
Additional Services
– various non-destuctive tests are available . . .
46. 46
Fitting the Pieces Together -
Tubular Assembly
Sensor
Head + Thermowell with integral
Extension
47. 47
Fitting the Pieces Together -
Non-Tubular Assembly
Sensor
+ Thermowell
Head + Extension
48. 48
AGENDA
Temperature Sensors and Accessories
– RTD Sensors, Rosemount Offering
– Thermocouple Sensors, Rosemount Offering
– Thermocouples vs. RTDs
– RTD Calibration (Characterization)
– Accessories
– Time Response
49. 49
Temperature Point Response Time
Sensor
Thermowell
Transmitter
Process
Factors Affecting Temperature Point
Response Time
Process
Transmitter
Thermowell
Sensor
75.4 °C
50. 50
Sensor Time Response
Type of element
– Wirewound RTD
» externally or internally wound
– Thin-film RTD
– Thermocouple
Element packaging
– Element coating, potting
– Contact between element package & sheath
Sheath thickness and material
Factors Affecting Sensor Response Time
element
sheath
ceramic
bore
Al2O3
packing
OD
51. 51
Sensor Time Response
Type of element
Thin-film has slightly faster response time than wirewound
– Thermocouples do not vary significantly
Element packaging
– Rosemount RTD’s are packed in magnesium oxide to provide
optimum thermal conduction within the sheath
– Grounded thermocouples are twice as fast as ungrounded
Sheath thickness and material
– Rosemount uses 316SST and Inconel (for high temperatures) for
sheath; both are very good thermal conductors
Factors Affecting Sensor Response Time
52. 52
Sensor Time Response
Typical
Wirewound RTD 10 - 16 s
Thin-film RTD 8.0 - 12 s
Ungrounded thermocouples <4 s
Grounded thermocouples <2s
* All results based on standard conditions: time required to reach
63.2% sensor response for water flowing at 3 ft/sec.
53. 53
Factors Affecting Response Time
of Sensors in Thermowells
Thermowell design style
(thickness at tip)
– Stepped is the fastest
Distance between sensor sheath
and thermowell (y)
– Spring loaded sensor ensures
contact at the tip
– Industry practice suggests using
thermally conductive fill can
significantly reduce time lag
y
Sensor
Assembly
Thermowell
Thermally
Conductive Fill
54. 54
Time Response for Sensors in
Thermowells
Thermowells can slow the time response by
at least a factor of 5 times
55. 55
Factors Affecting Transmitter
Repsonse Time
Time response depends on
element (complexity of calculation)
2-wire RTD 440 - 760 ms
3 & 4-wire RTD 520 - 920 ms
Thermocouples 300 - 750 ms
Transmitter update time (output)
every 1/2 second
Process
Transmitter
75.4 °C
56. 56
Process Factors in Temperature
Response Time
Velocity of the material
Thermal conductivity of the material
Density and viscosity of the material
Process time constants can be from
seconds to hours:
Process
Water @ 3 fps t = 1 min
Air at 50 fps, 40-80o
C = 11 minutes
Oil agitated in a bath: t = 13 minutes
Oil not agitated: t = >45 minutes
75.4 °C
57. 57
Summary: Temperature Point
Response Time
Sensor < 7 to 10 sec
Sensor in Thermowell 60 to 120 sec
Transmitter .5 to .9 sec
Process Seconds to
Hours
• Thermowells and process material/conditions have the
greatest effect on temperature point response time
58. 58
Learning Objectives at the END:
Temperature Sensors
Have a basic understanding of operation
principles for both an RTD and T/C and
understand the benefits of using one vs the other
Know the T-well styles and how to select the most
appropriate for processes
Know the main factors that affect time response
Editor's Notes
#2:We can describe the temperature of an object as that which determines the sensation of warmth or coldness felt from contact with it. (The hotness or coldness of an object.)
or
THE PROPERTY OF A SYSTEM THAT DETERMINES WHETHER IT IS IN THERMAL EQUILIBRIUM WITH OTHER SYSTEMS.
#3:Temperature can be measured in a variety of ways. Some of the more common methods are:-
Expansion of a gas, liquid, or other material such as mercury.
Change in resistance of metals. (Sir Humphrey Davey)
Platinum resistance thermometers - Sir William Siemen
Thermoelectric effect. (Thomas Seebeck)
Pyrometer. -- Measures the amount of emitted radiation (usually
infra-red). (Max Planck)
Bimetallic. -- Unequal expansion of two different metals welded together.
A not very common method is,
The Cricket Thermometer -- Count the number of chirps a cricket makes in 15 seconds, add 40, and the result is the temperature in Deg.F.
#4:Hook (1664) -- Every degree represented an equal volume equivalent to 1/500th part of the volume of the thermometer liquid.(Red dye in alcohol). The first meteorological records used this scale.
Fahrenheit (1724) -- Based on three fixed points, Mixture of ice, pure water, and ammonium chloride or sea salt (0 Deg F), a mixture of ice, and pure water (32Deg.F), and the temperature of the human body (96 Deg.F).
Centigrade (Anders Celsius 1742) --Based on two points. Boiling point of water (0), and the melting point of ice (100). This was later changed to 0 Deg C for the melting point of ice, and 100 Deg C for the boiling point of water.
Celsius -- The centigrade scale has now been dropped in favour of the Celsius scale which is defined as the triple point of water (0.01 Deg) and the boiling point of water at standard atmospheric pressure (99.975 Deg.C.).
Kelvin -- The temperature at which the pressure of an ideal gas is zero is defined as 0 K, and the triple point of water is set at 273.16 K.
Rankine -- The temperature at which the pressure of an ideal gas is zero is defined as 0 K, and 1 Deg R is equivalent to 1 Deg F (0 Deg C = 491.67 Deg. R.
#7:Have one of the experienced sales people answer this.
#8:Have one of the experienced sales people answer this.
#11:ANSWER = +/-0.35°C for a Class A, or +/-0.8°C for a Class B
This means that if a Class B sensor which is at the minimum value is replaced with one of the maximum value, the temperature would appear to have increase by 1.6°C
(EN 60751 tolerances)
#12:
Find the error in a 2-wire:
18 A.W.G. copper wire, Resistance = 6.385 Ohms/ 1000 ft
-Assuming a length of 20 ft (6 meters), this adds 0.33°C error to the measurement. If the gage wire is higher (e.g. 22), the error shoots up to 0.84°C!
#13:
Error for a 2-wire assembly:
0.06 x 6 x 2 = 0.72 ohms which is equivalent to 1.8°C.
This means that the temperature measured at the end of the cable would be 21.8°C.
Error for 4-wire assembl y:
As the lead resistances can be measured, the temperature measured at the end of the cable would be 20.0°C
#21:Ask:
1. Which is the lowest cost? (Types J,and K)
2. Which covers the highest temperature? (Type K)
#23:Answer: 0mV
How?
1. A thermocouple does not measure temperature, it measures temperature difference between the two junctions.
2. Since both junctions are at the same temperature, the output, which is the difference in temperature between the two junctions must be 0mV
#24:What are the most commonly sold types of T/C’s? (Type J and K)
#27:It is also more probable for poisoning to occur in a thermocouple than in an RTD simply due to the amount of metal in the Thermocouple. In an RTD, the sensor metal itself is limited to a very small region. In a T/C, the sensor metal runs from the tip of the T-well, and the entire distance to the transmitter.
#44:Process Pressure: Is dependent on the thickness of the T-well wall only, and for most (“normal”) cases, the thicknesses of the T-wells we offer are identical.
Time Response: Obvious. Stepped are the best: Less material to heat up due to smaller diameter.
Wake Frequency: Stepped are the worst. They Have the lowest natural frequency and cause the highest wake frequency.
** Key equation: Wake frequency/Natural < 0.8
Price: From the V.1 Sensor PDS: Stepped and straight t-well are the same price, tapered are more expensive
Drag Force: Stepped is best due to smallest diameter tip. Less area, less force. With the other two, it’s dependent on other things. Sometimes one will be better, sometime the other will be better.
*****NOTE: (From Khoi’s friend): Top three t-wells (All threaded): Stepped, straight and then tapered (57.1% of t-wells sold). 84% of t-wells sold are threaded