Where Does Platinum Rhodium Wire Sit in a Thermocouple?

The obvious answer to where platinum rhodium wire sits in a thermocouple is that it sits at the measuring end, and that answer is incomplete in a way that costs money. The wire occupies the hot segment of the circuit only, from the junction to the connection head. Everything after the head is a different cable, and for the platinum types that cable is not made of platinum at all.

A temperature reading is produced by the whole path between the process and the instrument. Each stage introduces its own error, and each stage is bought and specified separately.

The wire described in the title is stage two. It is the only stage made of platinum, and it is usually the shortest part of the circuit. A furnace sensor may have 300 mm of platinum rhodium wire and 30 metres of something else behind it.

If the whole circuit had to be the thermocouple alloy, the cost of a platinum installation would follow the cable length, and a long run would be unaffordable. The standards solve this by allowing a second kind of cable, and the distinction between the two is marked in the cable code itself.

Types R, S and B have no true extension cable in common use, and the reason is the metal itself. A platinum rhodium compensating cable is not made of platinum rhodium, because nobody would pay to run those alloys through a cable tray. What is used instead is a copper-based pair whose output matches the platinum pair closely enough over a limited band of ambient temperature.

Published cable data shows how the R and S compensating conductors have converged. The older RCA and SCA pairs, copper against a low-value copper nickel, have largely been replaced by a single combination, RCB and SCB, which uses copper against a copper nickel molybdenum alloy and serves both Type R and Type S. One alloy now covers the two types, which is worth knowing when an old drawing names RCA or SCA.

A cable tolerance is not the size of the error the cable is allowed to have. It is the extra error the cable may add to a measurement that was already correct, and it is quoted against a temperature band for the conductors.

Two numbers in that table deserve attention. The first is that the R and S compensating combination carries a Class 2 tolerance of ±5.0 °C, three times the figure for a KX extension cable. The second is that the tolerances are quoted for the conductors, and the published note adds that the usable range may be narrower still because of the insulation, which is a question for the cable maker, not for the standard.

The consequence is uncomfortable for a platinum installation. A special-limits Type S wire is held to ±0.6 °C or 0.1 % of reading, and the compensating cable behind it can add five degrees. Buying the tightest wire and the cheapest cable relocates the error instead of removing it, which is why the cable belongs in the accuracy calculation and not in the accessories column.

How much the cable matters is set by how large the signal is to begin with. Published values for the voltage produced at 100 °C show the gap between the base metal types and the platinum ones.

A Type R or Type S pair generates about one sixth of the voltage a Type K pair does at the same temperature. The instrument has to resolve a smaller signal, and every avoidable junction in the circuit contributes a proportionally larger share of the total. That is the practical reason a platinum installation is more sensitive to how it is wired than a base metal one, and it has nothing to do with the wire itself.

The intrinsically safe exception is worth isolating, because it inverts the usual rule. Under IEC 60584-3 the sheath normally takes the colour of the positive leg, and on an intrinsically safe circuit the sheath is blue regardless of type. A blue sheath is therefore information about the circuit and not about the thermocouple.

In the hot segment, from the measuring junction to the connection head, and nowhere else on a normal installation. The wire is the part that has to survive the process temperature, so it is kept as short as the geometry allows. The run from the head to the instrument is a separate cable, and for Types R, S and B that cable is not platinum.

The conductors. An extension cable uses the same alloys as the thermocouple, so the circuit behaves as though the sensor were longer. A compensating cable uses substitute alloys selected to reproduce the same voltage over a limited temperature band, which makes it far cheaper and limits how hot the route may be. The cable code says which one is inside: X for extension, C for compensating.

Because it would be made of platinum and rhodium, and the cost would follow the cable length. A compensating pair is used instead. Published data shows that the older RCA and SCA combinations have largely been replaced by RCB and SCB, a copper against copper nickel molybdenum pair that serves both types with a Class 2 tolerance of ±5.0 °C over 0 to 200 °C.

It can, and on the platinum types it often does. A Type R or Type S pair produces about 0.65 mV at 100 °C, roughly a sixth of what a Type K pair produces, so the same avoidable error is a larger fraction of the signal. The published Class 2 tolerance for the R and S compensating cable is ±5.0 °C, against ±1.5 °C for a KX extension cable, so the cable can contribute more error than the wire is held to.

No. Every joint between unlike metals produces its own voltage, so copper in the middle of a thermocouple circuit adds junctions that were never part of the measurement, and those voltages move with the ambient temperature. The cable has to match the type from the head to the instrument, and the joints have to be made with the correct polarity.

The cable type, with one exception that catches people out. Under IEC 60584-3 the negative leg is white and the sheath follows the colour of the positive leg. On an intrinsically safe circuit the sheath is blue whatever the type, so a blue sheath is telling you about the circuit, not about the thermocouple inside it.

Platinum rhodium wire sits at the measuring end, and that is the whole of its extent: the junction, and the short run to the connection head. The long part of the circuit is extension or compensating cable, and the noble metal types are served by the compensating kind because a platinum cable would be priced by the metre. The cable is not a neutral accessory. Its tolerance is added to the wire tolerance, the signal it carries is about a sixth the size of a base metal one, and the loop resistance, the screening and the polarity all decide whether the reading arrives intact.

DLX Alloy has made nickel and nickel alloy products at Changzhou, Jiangsu since 2002, on a 12,000 m² site certified to ISO 9001 and audited by SGS. Platinum rhodium thermocouple wire is supplied in the B, R and S types, and in the bare wire form the temperature loop needs: send the type, the diameter, the grade and the loop length for a confirmed quotation.

1. IEC 60584-3, thermocouples: extension and compensating cables, tolerances and identification system. Source for the X and C cable codes, for the cable tolerance table, and for the sheath colour rule including the blue sheath on intrinsically safe circuits.
2. ANSI MC96.1, temperature measurement thermocouples. Source for the ANSI cable tolerance column in the same table.
3. Published cable data on the application of extension and compensating cables, SAB Broeckskes. Source for the definition of the two cable families, the conductor materials, the compensating cable temperature bands and the output values at 100 °C by type.
4. Published comparison of extension and compensating cables, Alutal. Source for the statement that the noble metal types have no traditional extension cable, for the role of copper in creating parasitic junctions, and for the working guide on loop resistance.
5. ASTM E230 and IEC 60584-1. Source for the wire tolerances quoted for comparison against the cable tolerances.
6. Published guide to thermocouple measurement and cold junction compensation, Texas Instruments. Source for the requirement that the reference junction temperature be measured and converted to an equivalent voltage.
7. Product data held by DLX Alloy. Source for the types and forms supplied.

The figures quoted are published values and are not acceptance limits. A cable tolerance is the additional error a cable may introduce under the conditions the standard specifies, and it is not a statement about a complete installation. What the loop achieves depends on the junction, the head, the cable, the reference junction and the instrument together, and that is a commissioning question, not a catalogue one.