Q: The thermocouple has no display / The controller reads 1700 °C

Cause: The thermocouple does not have a closed circuit. The cause is a broken wire or a defective cable. For Type S thermocouples, note that the platinum wire is relatively soft. Excessive bending can cause the wire to break. Many controllers display the maximum temperature of the thermocouple in the event of a wire break. Solution: If the thermocouple does not display a temperature despite being properly commissioned, the thermocouples in use should be tested for continuity. If this is the case, the remainder of the measurement path should be checked for an interruption, particularly a cable break. If there is suspicion that the fault lies with the thermocouple, the thermocouple must be sent in for inspection and a detailed fault analysis must be performed.

Q: The thermocouple always reads only 30°C (±10°C)

Cause: There is a short circuit in the thermowires within the connector head, and the temperature of the connector head is being measured. Solution 1: Check the connection cable inside the connector head. No strands or parts of the shielding should be visible or twisted. Solution 2: If no short circuit is visible in the connector head, the short must be in the thermocouple itself. This unusual error can only occur during the initial installation of the thermocouple. If the inspection confirms the suspicion that this short circuit was “temperature-related,” the repair will, of course, be covered under warranty.

Q: Thermocouple Shows a Decreasing Temperature When Heated

Cause: This behavior is usually caused by incorrect polarity of the compensation wire connected to the thermowires. If no change is visible despite the colors matching (white to the white terminal and orange/red to the orange terminal), the silicone insulation sleeves were attached incorrectly. This unusual, manufacturing-related error can be corrected by simply reconnecting the wiring. Solution: It is recommended to peel off the silicone sleeves with your thumbnail and replace them. Another clue is the different thickness of the temperature sensors, with the thicker one always being the negative wire.

Q: Thermocouple is drifting

Cause: Due to the varying thickness of the thermowire, Type S, R, and B thermocouples do not drift over many years. If drift does occur, check for external damage, whether mechanical or chemical. If no external influences are apparent, the thermocouple should be sent in for inspection. Note: PtRh-Pt thermocouples always drift downward. This means they indicate a lower temperature than the actual temperature. If the new, supposedly drifting thermocouple indicates a higher temperature than the installed thermocouples, it is more likely that the installed thermocouples exhibit some drift. Tip: Order a matching test thermocouple along with the thermocouple. This test thermocouple is made from the same wire batch. Since it is not constantly exposed to temperature, drift is unlikely in this case. Solution: Send the thermocouple in for a free inspection. If drift is detected in the thermocouple without any external damage, the thermocouple will be replaced free of charge, provided it is still within the warranty period.

Q: No metal protective tube provides protection against interference voltages

Appearance: A thermocouple operates with interference voltages inside a ceramic protective tube. The thermocouple feed-through is located between the heating coils. Cause: The interference voltages are induced in the thermocouple wires. It cannot be assumed that this problem can be easily solved by using a metal sheath instead of a ceramic one. In fact, protective tubes made of chromium, nickel, and steel contain magnetizable elements such as iron, chromium, and nickel; however, the alloys used for protective tubes are not magnetic, so they do not shield the induction fields of the thermocouple wires. Grounding a metal protective tube generally does not improve the situation. Here, the feed-through is located in the wrong place. Solution: The correct solution is to place the feed-through in a location where no inductive fields are present. If this solution is not feasible, a filter must be used or more modern measuring instruments must be employed. A tube made of low-alloy steel would protect the thermowires from induced interference voltages only until the Curie point (approx. 770 °C) is exceeded. At higher temperatures, the metal protective tube acts only as an electrical shield, which is ineffective against inductive fields. It shields only against interference voltages of 500 Hz or 500,000 Hz, which occur only in induction furnaces and do not interfere with the devices anyway.

Q: Interference Voltage in Gas-Fired Furnaces

Appearance: Temperature measurement in an indirectly gas-fired oven is inaccurate. A measurement shows 15 V DC at both terminals of the thermocouple. The thermocouple’s protective tube is made of Pythagoras. No fault was detected here. The voltage could not have reached the thermocouple from the heater or via induction, since the oven was gas-fired. Cause: No fault was detected here. The voltage could not have reached the thermocouple from the heating element or via induction, since the oven was gas-fired. Solution: One terminal of the thermocouple in the connector head was routed out using a copper wire and grounded. As a rule of thumb for interference voltages, remember: Grounding rarely helps. In this exceptional case, however, it made sense to ground one terminal of the thermocouple. With the connection between the thermocouple and ground, the unexplained 15 V interference voltage disappeared.

Q: Glass-silk-insulated heating wires show incorrect temperature distribution

Phenomenon: Thermocouples are used in a vacuum to measure the temperature distribution within a charge. The temperature is approximately 1000 °C. Although the built-in thermocouples show a constant temperature, the temperature curves of the batch thermocouples drop sharply. Cause: The thermowires are contaminated by their own electrical insulation. Glass is not a pure material and begins to melt at a temperature of 1000 °C, releasing its sodium, potassium, silicon, and other ions onto the thermowires. The result is a rapid decline in thermoelectric voltage. Solution: Glass-silk-insulated thermowires can only be used up to a temperature of about 400 °C.

Q: Where can I learn more about thermocouples?

Appearance: The thermocouple has not changed significantly since its discovery in 1821. The Type S thermocouple (platinum-rhodium-platinum) has remained unchanged since its introduction in 1885. It is noteworthy that numerous errors and misunderstandings still occur in practice. Solution: The book Thermocouple Practice was published in its 4th edition in 2014. It contains over 50 years of practical experience from the founder, Dr.-Ing. Dipl.-Phys. László Körtvélyessy. It is available from Vulkan-Verlag .

Q: Oxygen probe: C level spikes

After installation, the C-level fluctuates erratically. Cause: The oxygen probe always indicates the true ratio between the furnace atmosphere and the ambient air. A fluctuating signal means that ambient air is entering from an external source. This can have various causes, such as a defective air supply, a broken probe tube, or a leaky furnace door. Solution: Check the area around the probe for defects. Spray nitrogen, argon, or forming gas and observe the probe’s voltage. When you spray the gas onto the leak, less ambient air enters the furnace atmosphere and the probe’s voltage rises.

Q: Oxygen probe: C level remains at ~0 mV

Symptom: After installation, the probe does not display a C level. Cause: The oxygen probe always displays the true ratio between the furnace atmosphere and the ambient air. If the probe’s voltage remains at ~0 mV, the connections for purge air and reference air are likely reversed. The constantly open reference air is sufficient to keep the oxygen ratio nearly identical. Consequently, the probe’s voltage is low. Solution: Swap the two hoses and monitor the voltage. It should soon begin to rise. The rate of increase depends largely on how quickly the oxygen in the probe’s protective tube is consumed.

Question 1

The thermocouple has no display / The controller reads 1700 °C

Accepted Answer

Cause: The thermocouple does not have a closed circuit. The cause is a broken wire or a defective cable. For Type S thermocouples, note that the platinum wire is relatively soft. Excessive bending can cause the wire to break. Many controllers display the maximum temperature of the thermocouple in the event of a wire break.

Solution: If the thermocouple does not display a temperature despite being properly commissioned, the thermocouples in use should be tested for continuity. If this is the case, the remainder of the measurement path should be checked for an interruption, particularly a cable break. If there is suspicion that the fault lies with the thermocouple, the thermocouple must be sent in for inspection and a detailed fault analysis must be performed.

Question 2

The thermocouple always reads only 30°C (±10°C)

Accepted Answer

Cause: There is a short circuit in the thermowires within the connector head, and the temperature of the connector head is being measured.

Solution 1: Check the connection cable inside the connector head. No strands or parts of the shielding should be visible or twisted.

Solution 2: If no short circuit is visible in the connector head, the short must be in the thermocouple itself. This unusual error can only occur during the initial installation of the thermocouple. If the inspection confirms the suspicion that this short circuit was “temperature-related,” the repair will, of course, be covered under warranty.

Question 3

Thermocouple Shows a Decreasing Temperature When Heated

Accepted Answer

Cause: This behavior is usually caused by incorrect polarity of the compensation wire connected to the thermowires. If no change is visible despite the colors matching (white to the white terminal and orange/red to the orange terminal), the silicone insulation sleeves were attached incorrectly. This unusual, manufacturing-related error can be corrected by simply reconnecting the wiring.

Solution: It is recommended to peel off the silicone sleeves with your thumbnail and replace them. Another clue is the different thickness of the temperature sensors, with the thicker one always being the negative wire.

Question 4

Thermocouple is drifting

Accepted Answer

Cause: Due to the varying thickness of the thermowire, Type S, R, and B thermocouples do not drift over many years. If drift does occur, check for external damage, whether mechanical or chemical. If no external influences are apparent, the thermocouple should be sent in for inspection.

Note: PtRh-Pt thermocouples always drift downward. This means they indicate a lower temperature than the actual temperature. If the new, supposedly drifting thermocouple indicates a higher temperature than the installed thermocouples, it is more likely that the installed thermocouples exhibit some drift.

Tip: Order a matching test thermocouple along with the thermocouple. This test thermocouple is made from the same wire batch. Since it is not constantly exposed to temperature, drift is unlikely in this case.

Solution: Send the thermocouple in for a free inspection. If drift is detected in the thermocouple without any external damage, the thermocouple will be replaced free of charge, provided it is still within the warranty period.

Question 5

Some batch thermocouples display unreliable readings

Accepted Answer

Appearance: Several batch thermocouples are used in a furnace; they are all inserted into the holes in the batch to accurately record the temperature. The recorder is connected to the thermocouples via two wires. Some charge temperatures are displayed smoothly and with slow fluctuations, while others are unstable, with the temperature curve being interrupted multiple times and slightly shifted. For some thermocouples, the instrument display adjusts quickly and precisely to the charge temperature and remains stable there; for other charge temperatures, the display is slow and hesitant, showing small jumps after stabilization.

Cause: The temperature distribution measurement is affected by interference voltages. Although nearly the exact same setup and instrument were used in the furnace, the thermocouples are not all connected to the charge in the same way. With some thermocouples, the measuring tip touches the metal charge, while other measuring tips happen not to penetrate as deeply into the bore. As a result, they are isolated from ground. This condition can be easily detected with a test lamp by disconnecting the thermocouples from the device on both terminals.

Anyone who believes that grounding is a good protection against interference voltages will be surprised to learn that it is precisely the thermocouples grounded through contact with the load that cause interference. The explanation for this is as follows: The measuring instrument is connected to the power grid via its transformer. This connection arises from the electrical capacitance between the transformer windings and from additional stray capacitances. These capacitances cause 230 V to enter the instrument’s measurement circuit via the transformer and reach the thermocouples through the input stage. If a thermocouple remains “floating”—that is, not in contact with the load—then this applied measurement voltage cannot cause a significant current in the input stage. This thermocouple therefore remains free of interference. However, if the measurement point is grounded, a strong current flows from the thermocouple to ground, and this current causes significant interference in the input stage.

Solution: Grounding of the measurement point can be prevented by inserting a small ceramic bead in front of the thermocouple. In general, sheathed thermocouples with a measuring junction isolated from the sheath should be used for batch temperature measurements. The time constant of a batch is always several orders of magnitude greater than that of the thermocouple. A fast response is not required here.

Question 6

No metal protective tube provides protection against interference voltages

Accepted Answer

Appearance: A thermocouple operates with interference voltages inside a ceramic protective tube. The thermocouple feed-through is located between the heating coils.

Cause: The interference voltages are induced in the thermocouple wires. It cannot be assumed that this problem can be easily solved by using a metal sheath instead of a ceramic one. In fact, protective tubes made of chromium, nickel, and steel contain magnetizable elements such as iron, chromium, and nickel; however, the alloys used for protective tubes are not magnetic, so they do not shield the induction fields of the thermocouple wires. Grounding a metal protective tube generally does not improve the situation. Here, the feed-through is located in the wrong place.

Solution: The correct solution is to place the feed-through in a location where no inductive fields are present. If this solution is not feasible, a filter must be used or more modern measuring instruments must be employed. A tube made of low-alloy steel would protect the thermowires from induced interference voltages only until the Curie point (approx. 770 °C) is exceeded. At higher temperatures, the metal protective tube acts only as an electrical shield, which is ineffective against inductive fields. It shields only against interference voltages of 500 Hz or 500,000 Hz, which occur only in induction furnaces and do not interfere with the devices anyway.

Question 7

Interference Voltage in Gas-Fired Furnaces

Accepted Answer

Appearance: Temperature measurement in an indirectly gas-fired oven is inaccurate. A measurement shows 15 V DC at both terminals of the thermocouple. The thermocouple’s protective tube is made of Pythagoras. No fault was detected here. The voltage could not have reached the thermocouple from the heater or via induction, since the oven was gas-fired.

Cause: No fault was detected here. The voltage could not have reached the thermocouple from the heating element or via induction, since the oven was gas-fired.

Solution: One terminal of the thermocouple in the connector head was routed out using a copper wire and grounded. As a rule of thumb for interference voltages, remember: Grounding rarely helps. In this exceptional case, however, it made sense to ground one terminal of the thermocouple. With the connection between the thermocouple and ground, the unexplained 15 V interference voltage disappeared.

Question 8

Glass-silk-insulated heating wires show incorrect temperature distribution

Accepted Answer

Phenomenon: Thermocouples are used in a vacuum to measure the temperature distribution within a charge. The temperature is approximately 1000 °C. Although the built-in thermocouples show a constant temperature, the temperature curves of the batch thermocouples drop sharply.

Cause: The thermowires are contaminated by their own electrical insulation. Glass is not a pure material and begins to melt at a temperature of 1000 °C, releasing its sodium, potassium, silicon, and other ions onto the thermowires. The result is a rapid decline in thermoelectric voltage.

Solution: Glass-silk-insulated thermowires can only be used up to a temperature of about 400 °C.

Question 9

Calibrated thermocouples produce an inexplicable temperature distribution

Accepted Answer

Phenomenon: Twelve calibrated thermocouples in an oven, inserted at twelve different locations, result in an inexplicable temperature distribution, even though the correction values for the individual thermocouples were taken into account precisely.

Cause: The calibrated thermocouples may have already been in use, and the insertion depth of the thermowires during calibration was different from that during measurement. Another possible error is that the thermowires were inserted to the same depth during calibration as during measurement. However, the calibration was not performed for all thermocouples at the same time—for example, six thermocouples on one day and the remaining six on another day. Even with the most precise calibration, small errors occur. For example, thermowires from one roll might receive a correction of +0.8 K on one day, while their “twin” thermowires from another roll receive a correction of -0.7 K on a different day. Since the thermowires are practically identical, they should receive a correction of +0.3 K each time. During the measurement, the thermocouples calibrated on the first day show an apparent temperature difference of +1.5 K compared to those calibrated on the second day. In practical terms, this calibration is very accurate in absolute terms. A few years ago, such precision was not found in industrial practice, but only in large research institutes. Nevertheless, these very precisely calibrated thermocouples are of little use for measuring temperature distribution.

Solution: All thermocouples used for temperature distribution measurements should either be calibrated simultaneously or subsequently calibrated against one another using a copper tube, in a high-precision tube furnace, or in a ceramic tube within a larger furnace, and their relative deviation measured. It may sound a bit strange that the plant engineer “calibrates” the very expensive thermocouples—which come with a separate certificate from an expensive institute—inside a copper tube. However, this relative calibration is necessary only for these highly accurate thermocouples to ensure that the temperature distribution measurement is not skewed by apparent temperature differences. If these twelve certified thermocouples are already showing signs of aging, this preliminary or subsequent calibration is essential. The immersion depth must remain constant during both the temperature distribution measurement and the calibration. The temperature controller should display the average temperature of the twelve calibrated thermocouples when the measurement is below 800 °C. At higher temperatures, the sheathed thermocouples used here age. Since they age at the same rate, the measured temperature distribution is relatively accurate; however, the reference thermocouple—in a triple-thermocouple configuration—should be verified using a PtRh-Pt test thermocouple. It serves as a reliable reference point within the furnace and provides the necessary correction to the relative temperature distribution if a sheathed thermocouple was connected to the reference thermocouple during the temperature distribution measurement.

Question 10

Where can I learn more about thermocouples?

Accepted Answer

Appearance: The thermocouple has not changed significantly since its discovery in 1821. The Type S thermocouple (platinum-rhodium-platinum) has remained unchanged since its introduction in 1885. It is noteworthy that numerous errors and misunderstandings still occur in practice.

Solution: The book Thermocouple Practice was published in its 4th edition in 2014. It contains over 50 years of practical experience from the founder, Dr.-Ing. Dipl.-Phys. László Körtvélyessy. It is available from Vulkan-Verlag.

Question 11

Oxygen probe: C level spikes

Accepted Answer

After installation, the C-level fluctuates erratically.

Cause: The oxygen probe always indicates the true ratio between the furnace atmosphere and the ambient air. A fluctuating signal means that ambient air is entering from an external source. This can have various causes, such as a defective air supply, a broken probe tube, or a leaky furnace door.

Solution: Check the area around the probe for defects. Spray nitrogen, argon, or forming gas and observe the probe’s voltage. When you spray the gas onto the leak, less ambient air enters the furnace atmosphere and the probe’s voltage rises.

Question 12

Oxygen probe: C level remains at ~0 mV

Accepted Answer

Symptom: After installation, the probe does not display a C level.

Cause: The oxygen probe always displays the true ratio between the furnace atmosphere and the ambient air. If the probe’s voltage remains at ~0 mV, the connections for purge air and reference air are likely reversed. The constantly open reference air is sufficient to keep the oxygen ratio nearly identical. Consequently, the probe’s voltage is low.

Solution: Swap the two hoses and monitor the voltage. It should soon begin to rise. The rate of increase depends largely on how quickly the oxygen in the probe’s protective tube is consumed.

Question 13

Oxygen probe: The C level responds very slowly

Accepted Answer

Symptom: After some time in operation, the probe repeatedly reacts to changes in the atmosphere.

Cause: The probe’s reaction rate is influenced, among other things, by the quality of the electrical contact between the electrodes and the FZY tube. High-quality contact results in low internal resistance. If the electrode is chemically corroded, this can affect the reaction rate. Another possible explanation is an increased presence of carbon in the vicinity of the probe. Carbon has a very high affinity for oxygen and thus slows down the probe’s reaction.

Solution: Flush the probe with ambient air. If graphite is present, it will be catalytically burned off. As a result, the probe’s voltage should begin to drop. Once it reaches ~0 mV, the probe is clean and should respond quickly again. If purging does not produce any change, the gas electrode is likely damaged. The probe should be sent in for inspection.

Implementation	Outer Diameter	Flange Diameter	Gradient
G3/8	16.6 mm	14.95 mm	1.34
G1/2	20.96 mm	18.63 mm	1.81
G3/4	26.44 mm	24.12 mm	1.81
G1	33.25 mm	30.29 mm	2.31

DN	Maximum Nozzle Diameter Ød1 [mm]	Inner Ring Diameter Ød2 [mm]	Outer Diameter Ød3 [mm]
16	20	17,2	30
25	28	26,2	40
40	44,5	41,2	55

DN	D	C	F1	F2	d1	W	K	L	n
25	115	18	2	4,5	68	43	85	14	4
40	150	18	3	4,5	88	61	110	18	4
80	200	20	3	5	138	106	160	18	8

Support for Our Products

Warranty

Thermocouples

Oxygen probes

(Leakage)

Delivery Times

Serial Number

Process Connections

pipe thread

Small flange per ISO 2861

Compression Fitting

Compression Fitting

Flange in accordance with DIN EN 1092

Sealing Surface, Type B

Sealing Surface, Type G

Frequently Asked Questions

DN	D	C	F1	F2	d1	W	K	L	n
25	115	18	2	4,5	68	43	85	14	4
40	150	18	3	4,5	88	61	110	18	4
80	200	20	3	5	138	106	160	18	8

DN	D	C	F1	F2	d1	W	K	L	n
25	115	18	2	4,5	68	43	85	14	4
40	150	18	3	4,5	88	61	110	18	4
80	200	20	3	5	138	106	160	18	8

DN	D	C	F1	F2	d1	W	K	L	n
25	115	18	2	4,5	68	43	85	14	4
40	150	18	3	4,5	88	61	110	18	4
80	200	20	3	5	138	106	160	18	8