IR temp meas of induction heated gears 2

[twinebender]

I am thinking of adding an IR temp sensor to an induction heating process to determine temp of gear teeth. (1650F)
2 things concern me.
1. predictability of emissivity of the steel I am going to use ( SAE 1045 ). Most sensors I have looked at require me to set emissivity. (vendors recommend .85)
2. The part will be spinning and the device pointed at the teeth, which will give non-uniform angle of reflection.

Is this an idea where I can expect accuracy within +-10 degrees day in and day out?

 

[JKEngineer]

I would have concerns about this application, although would not rule it out.

Are you planning on mounting the sensor perpendicular or parallel to the cylindrical axis of the gear? (Sounds like perpendicular.) If you go with parallel and set the sensing area inboard of the root of the tooth, you will get a steady look at the gear. However, there will be an offset to the tooth temperature.

One concern will be the response time of the sensor. If you are perpendicular to the axis then the angle will affect it and the values will be varying. If the rotation is rapid, then the sensor may have trouble keeping up.

Similarly, if the sensor is parallel and outboard of the root diameter, then you will have a varying signal as the gaps go past the sensor. Response time will be a concern (depending on gear size and speed).

You should use an emissivity for the material based on the target temperature. Your safest approach would be to measure it. There are some literature sources for metal emissivities at elevated temperature.

Jack

Jack M. Kleinfeld, P.E. Kleinfeld Technical Services, Inc.
Infrared Thermography, Finite Element Analysis, Process Engineering

http://www.KleinfeldTechnical.com

 

[twinebender]

Jack,

The gears will be rotating at about 30 RPM. Diameters are 5 10 inches. They will sit inside a coil that will completely cover the face of the gear. My present thinking was to aim the sensor down at the part at a 45 degree angle. The center of the target would be at the corner where the root of the tooth and the face meet. So my target would alway be hitting some part of the gear.

Here is a rough diagram.

/ <IR BEAM
/
IIIIIIIII < GEAR
IIIIIIIII

As far a accuracy goes I am only looking for a +/- 20 degrees F at about 1650F.

Rick

 

[IRstuff]

30 rpm = 1/2 cyc/s, assuming 50 teeth, the teeth will be in view for no more than 40 ms.

Unless you're using an imaging pyrometer, it's likely that your device will simply see a blur and therefore will average the reading over a some length of the circumference of the gear.

TTFN

 

[mrpyro]

Hi,

I just joined to respond to a different thread and saw yours.

Perhaps the simplest way to select an IR sensor for this well-known application, is to use a spot thermometer with a spot size that will cover most or all of the area of the region that you wish to see and use a two-color or ratio thermometer. If access is a problem, there are several units made with fiber optics that can get in close. There are a number of them on the market and the ones that would perhaps best suit your use would be ones that work in the near IR , i.e. with wavebands center wavelengths less than 2 microns.

Talk to an sales or application guy at Ircon, or Land Instruments, or Mikron Infrared, or Raytek. They are the big guys in industrial IR sensors and have solid equipment for this use. With a two color device you do not need any emissivity knowledge (or setting) when looking at oxidized steel (That's not true in all cases, but for steels it's accurate enough to more than meet your tolerance limits-I wrote a paper in 2002 on the subject and can get you a copy, if it would help). More than that you don't have to have the entire field of view filled and do not have to worry much about small amounts of smoke or loose scale in the field of view.

They and a few others are listed on the IR Thermometer vendors page of my website at:

http://www.temperatures.com/rtsvendors.html

.

Good luck.