1 Topic by driggers (edited by driggers 2013-06-01 20:39:47)

  • driggers
  • Member
  • Offline
  • Registered: 2012-10-07
  • Posts: 162

Topic: [resolved] 20 degree temperature errors? software wrong temperature

IR thermometer reads: nozzle 180, bed 95

Software readings say: nozzle 190, bed 75

Why is my temperature reading so wrong?  I am quite sure my IR thermometer reading is right.  It saves the max temp while I hold the trigger, pointed at the bed or nozzle.

Is this difference going to be constant? I think I remember the room temp software readings to be correct, so I guess that answers the Q with a negative.

Should I do anything specific?

SD2 Sanguinololu 1.3a atmega1284p, wood platform, lawsy's carriages, braided fishing line, pallet wood overhead spool mount, carboard/magnet enclosure, glass bed, E3D v6, bed levelling knobs, extended z-stop, 25A DC-DC SSR for bed heater, everything fixed to the SD2 frame, marlin firmware with some adjustments and extra failsafes enabled.  I'll never give up on you, little printer that could(n't)!

2 Reply by ronsii

  • ronsii
  • Member
  • Offline
  • From: PNW
  • Registered: 2012-10-28
  • Posts: 1,723

Re: [resolved] 20 degree temperature errors? software wrong temperature

Ir on shiny metal = inaccurate

best way to get accurate measurements is to put tape or paint the target area.

3 Reply by adrian

  • adrian
  • Member
  • Offline
  • From: Canberra, Australia
  • Registered: 2013-01-15
  • Posts: 1,623

Re: [resolved] 20 degree temperature errors? software wrong temperature

What Ronsii said.. this is what we meant when we remarked about calibrating for the various emissivity in the thermal pics we've posted...

https://en.wikipedia.org/wiki/Emissivity

4 Reply by driggers

  • driggers
  • Member
  • Offline
  • Registered: 2012-10-07
  • Posts: 162

Re: [resolved] 20 degree temperature errors? software wrong temperature

Emissivity should affect the readings thusly:

Citing values on http://www.infrared-thermography.com/material-1.htm and http://www.testequipmentdepot.com/fluke … _chart.pdf

Glass has emissivity of 0.92, so my reading of 95 is a black material of 103 degrees.

Emissivity of brass is less relevant, because there is black insulator tape over it and a lot of burnt plastic on it.  Those are more emissive, and thus hotter, so the max reading is from something that is closer to 1 than brass.  So perhaps the nozzle is closer to the mark than the bed.

Accounting for emissivity, the bed is up to 28 degrees over temperature.

SD2 Sanguinololu 1.3a atmega1284p, wood platform, lawsy's carriages, braided fishing line, pallet wood overhead spool mount, carboard/magnet enclosure, glass bed, E3D v6, bed levelling knobs, extended z-stop, 25A DC-DC SSR for bed heater, everything fixed to the SD2 frame, marlin firmware with some adjustments and extra failsafes enabled.  I'll never give up on you, little printer that could(n't)!

5 Reply by ronsii

  • ronsii
  • Member
  • Offline
  • From: PNW
  • Registered: 2012-10-28
  • Posts: 1,723

Re: [resolved] 20 degree temperature errors? software wrong temperature

Another thing that affects readings is field of view of the measuring device... some people think it is measured exactly where the laser pointer is if it has one, also the aperture is not 100 percent constrained to the listed field of view especially on the cheaper models form china.

6 Reply by elmoret

  • elmoret
  • Filastruder
  • Offline
  • Registered: 2012-10-11
  • Posts: 6,048

Re: [resolved] 20 degree temperature errors? software wrong temperature

driggers wrote:

Emissivity should affect the readings thusly:

Citing values on http://www.infrared-thermography.com/material-1.htm and http://www.testequipmentdepot.com/fluke … _chart.pdf

Glass has emissivity of 0.92, so my reading of 95 is a black material of 103 degrees.

Emissivity of brass is less relevant, because there is black insulator tape over it and a lot of burnt plastic on it.  Those are more emissive, and thus hotter, so the max reading is from something that is closer to 1 than brass.  So perhaps the nozzle is closer to the mark than the bed.

Accounting for emissivity, the bed is up to 28 degrees over temperature.


Wrong-o, my good man.

Aluminum and glass are reflective, so some of the IR your gun is recording is from anything above it. You happen to have a 190C source just a few inches above...

Filastruder - the robust, inexpensive filament extruder
E3Dv6 All Metal Hotend - the premier 3D printing hotend

7 Reply by driggers

  • driggers
  • Member
  • Offline
  • Registered: 2012-10-07
  • Posts: 162

Re: [resolved] 20 degree temperature errors? software wrong temperature

elmoret wrote:
driggers wrote:

Emissivity should affect the readings thusly:

Citing values on http://www.infrared-thermography.com/material-1.htm and http://www.testequipmentdepot.com/fluke … _chart.pdf

Glass has emissivity of 0.92, so my reading of 95 is a black material of 103 degrees.

Emissivity of brass is less relevant, because there is black insulator tape over it and a lot of burnt plastic on it.  Those are more emissive, and thus hotter, so the max reading is from something that is closer to 1 than brass.  So perhaps the nozzle is closer to the mark than the bed.

Accounting for emissivity, the bed is up to 28 degrees over temperature.


Wrong-o, my good man.

Aluminum and glass are reflective, so some of the IR your gun is recording is from anything above it. You happen to have a 190C source just a few inches above...

My knowledge of IR thermometers is not deep, but I read that glass is IR opaque, and the emissivity of most types of glass is around 0.92.  This is all evidence that my readings off the glass of not more than about 10% off, and that the aluminum and kapton underneath is virtually irrelevant.

Your point seems intuitive, but it goes contrary to evidence that I have no reason to disbelieve.

SD2 Sanguinololu 1.3a atmega1284p, wood platform, lawsy's carriages, braided fishing line, pallet wood overhead spool mount, carboard/magnet enclosure, glass bed, E3D v6, bed levelling knobs, extended z-stop, 25A DC-DC SSR for bed heater, everything fixed to the SD2 frame, marlin firmware with some adjustments and extra failsafes enabled.  I'll never give up on you, little printer that could(n't)!

8 Reply by elmoret

  • elmoret
  • Filastruder
  • Offline
  • Registered: 2012-10-11
  • Posts: 6,048

Re: [resolved] 20 degree temperature errors? software wrong temperature

I'm not sure how else to explain it to you. Glass reflects IR. Maybe this will help:

http://www.moistureview.com/wp-content/uploads/2011/01/IR000132-vl1.jpghttp://www.moistureview.com/wp-content/uploads/2011/02/IR000132-ir.jpg

http://www.moistureview.com/blog-018

Filastruder - the robust, inexpensive filament extruder
E3Dv6 All Metal Hotend - the premier 3D printing hotend

9 Reply by driggers

  • driggers
  • Member
  • Offline
  • Registered: 2012-10-07
  • Posts: 162

Re: [resolved] 20 degree temperature errors? software wrong temperature

ronsii wrote:

Another thing that affects readings is field of view of the measuring device... some people think it is measured exactly where the laser pointer is if it has one, also the aperture is not 100 percent constrained to the listed field of view especially on the cheaper models form china.

Yes, I have been careful to hold the IR opening quite close to the thing I'm measuring.  It's a self correcting problem in this case, because there are huge jumps in temperature between different parts, which makes it obvious what I am reading.  Also, my IR holds a max temp value, so I just wave it around different parts of the glass, or at the nozzle, and it tells me which is the hottest part.

SD2 Sanguinololu 1.3a atmega1284p, wood platform, lawsy's carriages, braided fishing line, pallet wood overhead spool mount, carboard/magnet enclosure, glass bed, E3D v6, bed levelling knobs, extended z-stop, 25A DC-DC SSR for bed heater, everything fixed to the SD2 frame, marlin firmware with some adjustments and extra failsafes enabled.  I'll never give up on you, little printer that could(n't)!

10 Reply by driggers (edited by driggers 2013-06-01 20:24:53)

  • driggers
  • Member
  • Offline
  • Registered: 2012-10-07
  • Posts: 162

Re: [resolved] 20 degree temperature errors? software wrong temperature

elmoret wrote:

I'm not sure how else to explain it to you. Glass reflects IR. Maybe this will help:


Hold the phone!  While you were posting this, I did an experiment.  Try to read the stove top's temp through glass.  The Glass is IR opaque, so the reading did not change.  But then I took a reading by reflecting the laser off the glass and onto the hot burner and lo and behold!  A reading of 100 degrees!

Thanks for your patience big_smile.

So maybe I am getting a partial reading of the nozzle when reading the glass in the printer.  I will measure again with more care (edit: and non-reflective) later.

SD2 Sanguinololu 1.3a atmega1284p, wood platform, lawsy's carriages, braided fishing line, pallet wood overhead spool mount, carboard/magnet enclosure, glass bed, E3D v6, bed levelling knobs, extended z-stop, 25A DC-DC SSR for bed heater, everything fixed to the SD2 frame, marlin firmware with some adjustments and extra failsafes enabled.  I'll never give up on you, little printer that could(n't)!