I am testing shafts in a machine for material testing. These shafts are rotating with about 3000 rounds/minute. My aim is to measure the temperature of the shafts during the test. Therefor I found the tmp006 infrared thermopile breakout board at Adafruit and Sparkfun. The tmp006 userguide says on page 5 (below figure 2) that the sensor should be placed underneath the target.
Can I also place the sensor above or beside the rotating shaft? Mounting it underneath would destroy the sensor when the shaft breakes after a few days. And is it generally possible to measure the temperature of the rotating shaft or are there better opportunities?
rookiebalboa:
The tmp006 userguide says on page 5 (below figure 2) that the sensor should be placed underneath the target.
Can I also place the sensor above or beside the rotating shaft? Mounting it underneath would destroy the sensor when the shaft breakes after a few days. And is it generally possible to measure the temperature of the rotating shaft or are there better opportunities?
I would assume the reason for placing the unit below the target is because heat rises and may effect a sensor being placed beside or especially above the target. Maybe do some test with a pair of sensors, one below and one to the side and see if the readings differ by more that you can tolerate.
How about placing a standard temp sensor just above the shaft to take advantage of the rising heat.
You might have to do some calibration first and use a second sensor to measure ambient temperature in the general area.
If you need +/- 2 degrees C or better then a more expensive IR temperature sensor may be required or the TMP006 if you can get the location sorted. (It seems to use ambient as part of its calculation, which is why you wouldnt place above a hot object)
Shiny metal (or shiny oil) would be just about the worst case for an IR temperature probe. It may be more accurate if you painted part of the shaft matt black, and kept oil off it. IMHO an air temperature thermocouple close to the shaft sounds more practical, and is bound to be related to shaft temperature - especially if there was a fan.
Shiny metal (or shiny oil) would be just about the worst case for an IR temperature probe.
Please specify why this should be the case. If there's not another heat source hotter than the shaft the readings should be quite accurate, at least much more accurate than a thermocouple with insulating air in between. And if there is a fan do you really expect the blown air to be the same temperature as the shaft?
Shiny metal (or shiny oil) would be just about the worst case for an IR temperature probe.
Please specify why this should be the case. If there's not another heat source hotter than the shaft the readings should be quite accurate, at least much more accurate than a thermocouple with insulating air in between. And if there is a fan do you really expect the blown air to be the same temperature as the shaft?
I am not familiar with the sensor you you mentioned, but I can tell you a few things about the Melexis MXL90614. This sensor (and I have no reason to believe that it's just this particular one) will give substantially different results depending on the emissivity of the surface you are attempting to measure. Different materials at the same temperature will have different amounts of IR radiating from them.
The other consideration is the temperature of the sensoor. It's not the absolute temperature that is the problem, but the difference in temperature across the sensor element itself. When there is uneven heating of the sensor, significant errors occur, and it pays to take steps to ensure that the sensor is evenly heated or cooled.
As well, the Melexis comes in a variety of "view angles". I believe the narrowest angle readily available is 15 degrees. This is important because the sensor will give a result that is the average temperature sensed on all parts of the view, so if you include a lot of view that does not cntain the object to be measured, the temperature change of the part of interest will not have a linear response, and might require a lot of effort to properly calibrate it.
It's an interesting project. Please keep us informed of your progress. I am building a cloud sensor with the MLX 90614.
Shiny metal (or shiny oil) would be just about the worst case for an IR temperature probe.
Please specify why this should be the case.
Basically, the IR sensor is only accurate when measuring pure black body radiation. Reflected radiation messes up the reading. So the closer the object is to a mirror (e.g. shiny metal) the worse it gets. A matt black object is close to optimal.
It's easy to test. Take water at a known temperature and put it in a shiny metal pot. Wait for the pot to heat up then measure the outside temperature using the IR sensor. You should find a big error. Take a matt black cloth or similar and lay it on the pot. Allow to reach equilibrium then measure temperature of cloth. You should find the result is closer to reality.
No, I do not expect blown air temperature to be the same as the shaft, but I would expect the error to be consistent and predictable.