Metal temperature measurement


Thread Starter

S. Jeyaprakash

Dear Members
We request you to suggest the best methods for our following

We use to monitor the metal temperature of superheater tubes of our
boiler. The location of measurement is in flue gas path( the temp. of FG
path which varies from 500 deg C to 850 deg C). The metal temperature
will be around 500 deg C. Now we use mineral insulated thermocouples and
we weld a metal pad at the measuring location and insert the
thermocouple inside the pad so that the junction touches the metal. The
measuring location will be insulated suitably to prevent the gas
touching the pad.But it happens that sometimes the thermocouple comes
out of the pad, insulation also goes off and we are not getting the
measurement. Instead the thermocouple starts reading the flue gas
temperature giving the operator various unwanted limits in boiler
We request the list members/experts to study this problem and suggest
us the best possible method to measure the metal temperature measurement
in boiler applications

Thank you in advance

S. Jeyaprakash
In the first place, it might be wise checking with your boiler supplier what is their current practice, if any of that kind.
In the case it's a modification of your own, obviously it is not a technical one because she
does not work as expected.
Infrared pyrometer for surface temperature may be one alternative.
A second one may be welding two tits of the thermocouple material in place (on the tube) and extending fair size wire outside the gas flue
containment (support must be provided). This alternative requires welding compatibility is included in the design.
As a third alternative and also as alternative to welding compatibility: weld an adapter of same metal as the superheater on the superheater tube. Design the adapter to accept flange type T/C,the flange of the T/C to be spring compressed to the flange of the adapter (something like the exhaust gas collector on automobile car). Springs do not like high temperature (check first). This adapter c/w Thermocouple assembly shall be as short as possible. At this point you may use either grounded T/C or ungrounded. However type T/C, a temperature measurement of the metal tube has been created, somewhat not perfect (never it is anaway).
Remember thermocouples (T/C) subjected to hot/cold cycling of that kind read lower temperature (several deg C) than actual.
In the case the metal temperature (relatively speaking) is of importance to your operation, install new T/C on new start of boiler.
A removable weld pad design has become very popular with a lot of companies that need to monitor this temperature for metal stress calculations. in the past a lot of utilities simply welded TC wire on the tube that had high temp fiberglass insulation and some metal overbraid. that method was cheap initially, but th einsulation would become contaminated and eventually allowed a path for electricity along the length of the wire. waht insued was that the readings were higher than they would have been oterewise. You see, the flames are hotter than the tubes. These misreadings cause the user to rellace good tubes when there was lots of life left.
A later method used was to use 3/16 or 1/8 inch diameter inconel or ss sheathed thermocouples with a pad welded directly to the thermocouple tip. Type K thermocouples are the primary sensor used because of the temperature range. there were and are lots of problems with this method that drove some back to the less expensive wire thermocouples. in a sheath, ther are chemical constituants that attach to the thermocouple wires that actually change the metalurgy of the thermocouple wire. so basically you end up with thirrd materials in the circuit and along with the severe temperature gradients seen in a boiler, the readings couild be all over the place. it was another problem altogether when it came time to replace these sensors because they had to be remopved with the bad tubes. New sensors had to be run. It is a messy and at best an uncomfortable job to pull out he old sensor, measure what length it was, but a new one, run it through a feed through that most of the tiime makes you remove all the sensors. Then you have to weld it to the new pipe and make sure you didn't overheat the thermocouple when you did you welding.
a much better solution is now available through JMS. A permanent weld pad made of the same material as the superheater tube is welded to the tube. (no welding procedure infractions here) A new type N thermocouple with 1/8 inch diameter Stabaloy sheath is installed and locked to the weld pad with a SS cotter pin. the thermocopuple tip is so designed that the thermocouple wires are within a 1/16" of the surface of the tube with exemplerary metal to metal contact. the stabaloy sheath contains few chemical components that will contaminate the thermocouple. the type N is used so that the peculiar problems associated with the use of type K thermocouples is eliminated. here's the beauty of the fix....
When the tube has to be replaced, you just remove the cotter pin and install a new pad. then you just insert the sensor into the new pas and you are finished. No feed throughs to worry about, no damage of the new thermocouple to worry about and you don't have to breath in nearly as much soot because you are out of there in minutes as oppossed to days.
pg 1-14 at
[email protected]

(sorry if i misspelled words or didn't capitalize. I don't know how to type with out my correction program)