Metal sheet temperature measuring

Have someone a solution to measure temperature of a metal sheet what's moving approx 15 m/min ?

Due to colors and variations in the sheet, the emission factor for IR-measuring is changing too much.

I want to measure within 0.2 degrees celcius.

Regards, René

you can get a roller type thermocouple probe

http://www.omega.co.uk/ppt/pptsc.asp?ref=88000_roller&Nav=tema11

bear in mind that no thermocouple is more accurate than +/- 2.0 deg C. and a Pt 100 is only accurate to 0.5 deg C.

Those accuracy specs are for the full temperature range of the device.
You can get much tighter by calibrating within the range of interest.
Response time may be an important consideration when comparing different probe types.

keithkyll said:

Those accuracy specs are for the full temperature range of the device.
You can get much tighter by calibrating within the range of interest.
Response time may be an important consideration when comparing different probe types.

Click to expand...

For thermocouples calibration is only valid until the next time it is heated up. you cannot guarantee calibration on a thermocouple tighter than the limits i have quoted. 0.2 deg C accuracy I do not believe is achievable in an industral environment.

RTDs are a lot slower response than thermocouples unless special designs are used.

I agree that .2 C won't happen. I strongly disagree about themocouples.
Yes, there will be a shift if it's cooled and heated, but not a 4 degree C window. The major shift occurs at the range extremes, such as -200C for a type K. If you stay within a typical range of 50C or so, and calibrate in that range, your results will be much better.