Multipoint Temperature Probe vs. Thermowell with Individual Thermocouples

Many bench and pilot reactors are large enough to take temperature readings inside the bed. On fixed-bed plug flow systems in particular, one decision you will face is whether to use a multipoint probe or a thermowell with multiple thermocouples. Both options work, but each has its pros and cons.

The two options

Both put several temperature readings along the axis of a tubular reactor, usually through the catalyst bed. They differ in how the measurement points are held in place.

• Thermowell with individual thermocouples. A closed-end tube runs axially through the reactor. Several thin sheathed thermocouples slide inside it, each pushed to its own depth. For a typical 3/4 in. OD reactor tube, the thermowell is usually 1/8 in. OD and the thermocouples 0.020 in. to 1/32 in. OD, with three to four fitting in one well. At 1/64 in., the thermocouples are too frail and tend to crimp and break; at 1/16 in., they are too large for the well. Each thermocouple is a standard, off-the-shelf part.
• Multipoint probe. A single sheath holds several thermocouple junctions at fixed, specified distances from the tip. It is built to order for your reactor and bed location. It is typically 1/8 in. OD and goes directly into the reactor in place of the thermowell.

Thermowell with individual thermocouples

This setup wins on flexibility, availability, and repair cost.

• Points can move. If the bed has to shift up or down the reactor, you just slide each thermocouple to its new depth. Nothing has to be re-ordered.
• No lead time. Thin sheathed thermocouples are usually available right off the shelf. A custom probe has to be built to specification.
• Easy furnace profiling. You can pull one thermocouple through the thermowell in steps and read the temperature at each position. That gives you the full axial profile of the furnace and shows you where the isothermal zone sits.
• Cheap repairs. If one thermocouple fails, you replace that one thermocouple.
• The process stays closed. Because the thermocouples sit inside a sealed well, you can add, move, or replace them without opening the reactor.

The weakness is the flip side of the flexibility. Loose thermocouples can creep, get bumped, or go back in at a slightly different depth after maintenance. Over a long campaign, you can end up comparing readings that were not taken from the same spot.

Custom multipoint probe

This is my preference, and the reason is repeatability. Every junction is fixed at a set distance from the tip, so every reading comes from exactly the same point every run. A hot spot at the top of the bed today is measured at the same place as last month’s hot spot. That matters when you compare catalysts, run long campaigns, or try to tie a temperature change to something happening in the bed.

A single probe is also simpler to handle than a bundle of loose wires. There is one fitting to seal and one assembly to install, and nobody can put a point back at the wrong depth.

The costs are real:

• Lead time. It is built to order, so plan for it.
• Fixed positions. If the bed moves, the points no longer line up with it.
• Repair. If the probe is damaged, you usually replace the whole assembly rather than a single thermocouple. One failed junction may still leave the other points readable, but you have lost that position until the probe is replaced. Because the probe sits directly in the process, replacing it also means opening the reactor.

Recommended approach: profile first, then lock it in

You do not have to pick one. Use each where it is strongest.

1. Start with a thermowell and individual thermocouples. Install the thermowell and a set of off-the-shelf thin thermocouples. Nothing custom is on the critical path, so the unit can start up sooner.
2. Profile the furnace. Traverse a thermocouple through the thermowell at your operating setpoints to map the axial temperature profile.
3. Place the bed. Locate the catalyst bed in the zone you want, usually the flattest part of the profile, and adjust it until you are happy with it.
4. Fix the measurement points. Decide exactly where you want readings: typically the bed inlet, middle, and outlet, plus any point where you expect a hot spot.
5. Order the multipoint probe to match. Specify the junction positions from your profiling work.
6. Swap it in when it arrives. Keep running on the thermowell while the probe is being built, then pull the thermowell and install the probe in the same fitting.

Practical tips

• Match the probe OD to the thermowell. If both are 1/8 in. OD, the probe goes into the same fitting and sits on the same centerline as the thermowell did. Positions you mapped in the well carry straight over to the probe, and the reactor needs no other changes.
• Measure from one fixed reference. Record every position from the same datum, such as the thermowell tip or the face of the compression fitting. Give the probe vendor positions from that same datum.
• Profile under real conditions. An empty-furnace profile is a good start, but flow, pressure, and reaction heat all change it. Profile again with gas flowing and, if you can, under reaction.
• Allow for the thermowell. A reading inside a well is the temperature of the well wall, which conducts heat along its length. It smooths out sharp gradients, so a narrow hot spot will read lower and broader than it really is.
• Check the probe on arrival. Measure the junction positions against your spec before installing it. A response check with a heat gun, one point at a time, confirms the channel wiring matches the labels.
• Repeat a reference run after the swap. Run a condition you already ran on the loose thermocouples and compare. If the readings agree, you know later shifts are real and not a change in hardware.
• Keep the loose thermocouples. They are your spares, and your profiling kit for the next time the bed moves.