I have been searching for an optical style (probably IR) temperature sensor that can measure up to 3000 °F that doesn't cost hundreds of dollars and can be interfaced to an AVR for automation. (accuracy does not need to be super tight, +-10 °F would suffice. So far I have found many of the "gun" style ones that claim to read above 3500 °F and they go for around $100! The only problem is this kind would need to be "hacked" in order to interface to an AVR, and I would much rather find a product intended for such use. Does anybody know if there's a sensor like this for say, under $300? I'm stumped.

 

How about something like this; it is limited to 1,000 degrees C, but with some IR absorber (such as glass) placed in front of the sensor it may extend the output voltage to cover the temperature range you want.

https://uk.rs-online.com/web/p/infr...VQygYCh3cjApWEAQYAiABEgItrPD_BwE&gclsrc=aw.ds

 

How about something like this; it is limited to 1,000 degrees C, but with some IR absorber (such as glass) placed in front of the sensor it may extend the output voltage to cover the temperature range you want.

https://uk.rs-online.com/web/p/infrared-temperature-sensors/1618103/?cm_mmc=UK-PLA-DS3A-_-google-_-CSS_UK_EN_Automation_&_Control_Gear_Whoop-_-Infrared+Temperature+Sensors_Whoop-_-1618103&matchtype=&aud-772940708119la-299966235540&gclid=EAIaIQobChMI3Map-6nq7wIVQygYCh3cjApWEAQYAiABEgItrPD_BwE&gclsrc=aw.ds

This is a good idea. Is there a way to quantify / calculate the scaling of the signal without doing a calibration. Im wondering if the result would be linear comparing gain (less than 1) to magnitude.

 

You could calibrate the IR sensor (including filter) using some other direct contact or optical method. I think the problem likely to be that the IR adsorbsion filter (such as glass) would absorb too much IR, the trick would be to find a filter material that would half the sensors response, giving a range to 2,000 degrees C.

 

I suppose i could just use the calibration thats already on the device, by simply removing and reinstalling the filter and comparing gain at various points within the device's original range, then extrapolate beyond that. Only problem is extrapolation is pretty much an assumption. I could throw in a type k thermo couple and compare to that instead. I guess ive got some stuff to try.

 

If you check out glass IR absorbsion, it is likely to be much greater than a factor of 2. You might be lucky and find that a thin piece of clear plastic or even transparent tape gives the desired IR cut.

Looking at the instructions for use, the sensor has emissivity settings between 0.2 – 1.0, you might be even luckier in that with a certain emissivity setting, the detector will give useful readings for you.

Otherwise you might be wasting a couple of hundred dollars on the detector.