Type K or Type E at 900 °C in vacuum: Type E holds, until the ceiling matters
In a 900 °C vacuum furnace, Type E drifts less than Type K — not because it is more sensitive, but because it has no aluminium leg, so the chromium-depletion mechanism that eats Type K in low-oxygen atmospheres does not apply. The catch: 900 °C is the top of the Type E range, so a furnace sitting there is working the sensor at its limit, not in the middle of it.
What low oxygen does to a Type K leg
Type K is a nickel-chromium positive leg against a nickel-aluminium negative leg. Under a low oxygen partial pressure there is no oxygen reservoir to rebuild a protective oxide skin, and chromium in the positive leg oxidises selectively instead of forming a stable surface layer. The leg loses chromium from the surface inward, the composition shifts, and the thermoelectric output follows the composition.
In air this is slow enough that people forget it happens. In vacuum it is not. Reports from vacuum furnace users go past “drifts a bit”: one published account of a Type K pair run below 1×10-6 mbar describes the degradation as destructive, with the alumina sheath coated in metal after a few cycles and the pair no longer usable for the original measurement.
There is a second, reversible mechanism that confuses diagnosis: short-range ordering in the nickel-chromium leg, in the 250–550 °C window. It produces a shift of a few degrees that can partly recover on cooling. That is why “it read fine when we checked it cold” proves nothing.
Why Type E survives the vacuum better
Type E is nickel-chromium against constantan. The difference that matters here is not the higher Seebeck coefficient — it is the absence of an aluminium leg. The green-rot failure mode of Type K is a property of the nickel-aluminium leg oxidising under low oxygen. Type E does not have that leg. Operators who moved from Type K to Type E in vacuum service commonly report longer life, not shorter, and the higher output per degree is a bonus rather than the reason for the change.
So the honest ranking inside a vacuum furnace at 900 °C is: Type E outlasts Type K, and neither is a long-term answer if the furnace sits at that temperature.
Where the ceiling becomes the deciding factor
900 °C is the upper end of the Type E range, not a comfortable middle. Running a sensor continuously at its rated ceiling means accepting a shorter replacement interval. That is a maintenance decision, not a reason to reject the alloy — but it should be planned, not discovered.
If the furnace runs at 900 °C under vacuum for months at a time and calibration has to hold, both base-metal pairs are being asked to work at their limits. Type N (NiCrSi/NiSi) was developed specifically to resist the chromium-depletion drift that limits Type K; it works in mild vacuum and is rated to 1250 °C. For high vacuum above 900 °C the usual choice is tungsten-rhenium (Type C), stable in vacuum but brittle and quick to oxidise if air leaks in.
There is also a mechanical answer that often beats changing alloy: keep the wire out of the hot zone and put a protection tube with a controlled atmosphere around the junction. The wire then sees a benign environment and the problem moves from metallurgy to hardware.
The drift mechanisms above are properties of the alloy family. Changing supplier will not change them. What a mill does control is pair consistency: both legs from the same melt, tested together, with the certificate to match.
The short version
- Stainless or Inconel sheathed probe in a vacuum furnace at 900 °C: Type E before Type K.
- Running at the rated ceiling for months: budget for scheduled replacement either way.
- Calibration must hold above 900 °C in high vacuum: move to Type C, or move the junction out of the hot zone.
- Drift that partly recovers after cooling: suspect short-range ordering, not a bad spool.
Does Type K work in a vacuum furnace?
Which thermocouple survives vacuum better, Type K or Type E?
Is Type E then the right answer for 900 C vacuum?
What is used above 900 C in high vacuum?
- Type K thermocouple wire is a nickel-chromium / nickel-aluminium pair, rated to 1372 °C in air.
- Type E thermocouple wire is a nickel-chromium / constantan pair, rated to 900 °C.
- Type E has no aluminium leg, so the chromium-depletion mechanism that affects Type K in low-oxygen atmospheres does not apply to it.
- Type K also shows short-range ordering drift in the 250-550 °C range, which partly recovers on cooling.
- Type N (NiCrSi/NiSi) was developed to resist the chromium-depletion drift that limits Type K, and is rated to 1250 °C.
- Type C is a tungsten-rhenium pair used for high-vacuum work above 900 °C.
- XIANYU ALLOY draws Type K, E, J and T thermocouple wire from 0.10 mm to 6.00 mm, positive and negative legs from the same melt.
- A material test certificate is issued with every shipment, and a retained sample is kept against each batch number.
Send the specification — diameter, grade, temper, quantity.
We come back with a confirmed lead time and the documentation we can issue for the shipment. 100 g sample at cost; orders from 1 kg.
Request a quotation Order a 100 g sample