1. Home
  2. General Electronics
  3. Cheap thermometer calibration technique?
  4. Page 2

Cheap thermometer calibration technique?

are

the

The internal shutter is for NUC (Non-Uniform Correction) -- focal plane array sensors in general (even visible light) have gain and offset differences from one pixel to the next. Part of it is periodic with the internal structure of the chip or with polish marks. The rest is just purely random. Visible light sensors can be selected to eliminate this to some extent (pro video cameras have nonuniformity correction, but the camera manufacturers won't admit it). IR detectors can't, and the amount of nonuniformity in the uncooled detectors can be astonishing; as of five years ago it could be 100x as much as your intended signal.

So every once in a while the shutter comes down, the internal logic recalibrates the NUC, and the camera continues on.

There are only a few companies that actually manufacture imagers (FLIR Sweden is one). So many companies that "make" IR imagers are just plopping an OEM module made by someone else into their case.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google?  See http://cfaj.freeshell.org/google/

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html
Reply to
Tim Wescott
Loading thread data ...

Doesn't 63/37 solder melt at 485° F at sea level?

That would calibrate the thermocouple.

The IR is a bit trickier as the surfaces you examine have differing emissivities.

You likely have a calibrated instrument, but forget to shift for emissivity differences.

For the IR to cal correctly, you need a "black body calibration source, which is typically an Aluminum ingot painted with IR paint.

You can learn a lot here:

formatting link

These guys are the tops.

Reply to
The Great Attractor

Freezing point and melting point are also pressure related, and so they too vary.

Also, water has a very low emissivity, and would be a very poor choice for a reading on the IR instrument.

Reply to
The Great Attractor

For his thermocouple. Not good choices for the IR device, however.

The body temp thermometer is VERY accurate usually (mercury type), but you need a good black body source for the IR cal session.

Reply to
The Great Attractor

Very good call, Chuck.

Spot on !

Not if the surface that got anodized was "shiny", as it were. A good black body source is aluminum (thick) as it conducts heat fairly evenly, but the ideal surface is a very matte finish, or even concentric rings cut into the face and then painted or anodized after a grit blast session.

Shiny is bad, which is why water is bad, despite it also being a very good, even conductor of heat. The shiny surface reflects the IR back into the medium, hence reduced emissivity.

You can find a fairly decent emissivity chart here, as well as a very good primer on the subject:

formatting link

Reply to
The Great Attractor

third

difficult to place more

stirred very rapidly,

as I found out

measure.

Not really. It depends on the focal length of the instrument. Shiny surfaces do reflect a lot of their IR emissions back down into the medium however.

A mirror finish of nearly anything yields results based on reflections, yet is still very dependent on the optical system utilized by the instrument.

Not very much though... if at all. Surface quality is the most determinant factor, not "color".

A sprinkling of copier toner would work better, but be much "messier".

Is that a "messier function"? :-]

Yes, but the underlying heat has to push through the tape medium, and there are losses.

Any matte finish brings one closer to ideal. Flat black paint (very thin coat) is best.

Take a look here for some really good facts:

formatting link

Reply to
The Great Attractor

For the thermocouple, sure. Add some salt even. Not good for the IR tho.

Boiling water is nebulous as one has to decide what "boiling" is, and a hearty boil can well be far above the boiling start point was/is.

Top posting is utterly retarded, boy.

You're a TOFU retard.

Reply to
The Great Attractor

Pointing at the finned side would yield better emissivity than the "smooth" side would.

Anodized, extruded Al usually has a fairly shiny surface quality.

The ideal black body source is a MATTE finish, so grit blasted Al with some matte black paint (very thin coat) would be best/better.

Yields about a .97 Emissivity. The anodized Al (regardless of color) yields about ten full points lower, if not worse.

Surface quality is a VERY important factor here.

Have a look here for a bit more info. Even though you seem very knowledgeable about it, this should help.

The ideal source, of course, is a cavity.

formatting link

Reply to
The Great Attractor

Color isn't the issue. The surface of water is VERY reflective, so any IR the medium generates get reflect BACK into the medium. This is why water has such a poor emissivity.

A matte finish Aluminum block/box filled with a known temp water (circulating) would likely be only a couple tenths of a degree off the water temp, and would allow the IR gun to be calibrated very well.

Reply to
The Great Attractor

Miles off the mark, as usual.

Reply to
The Great Attractor

Garner some facts here:

formatting link

Reply to
The Great Attractor

You're nuts.

Reply to
The Great Attractor

The term is matte.

Water is a bad idea for the IR, and the color doesn't matter.

The term for today is:

SURFACE QUALITY

formatting link

Reply to
The Great Attractor

Hard to believe, but it is.

Still, not a good source to cal IR with.

Skin is about 0.96, but skin temp is always several degrees lower than internal body temp.

A GOOD, NIST traceable black body source is only about 0.98

1.0 is not achievable.
Reply to
The Great Attractor

K. Irani invented the "resistor bolometer" back in 1960.

formatting link

They have imagers, but likely get their sensors from an OEM house, as does FLIR most likely. Chip fabs ain't cheap endeavors.

The best thing about a good IR device maker is the electronics behind the instrument.

FLIR buys their calibration sources from these guys.

These guys are the top dogs in non-contact temperature measurement.

We made some that are still in use at the shuttle launch pads that have the longest IR focal length made for a calibrated instrument.

The observation shacks are 1000' from the pad, and the instruments observe a ten foot spot, and detect a 200° C change within a one foot spot in that circle (1° FOV). The area observed is known as "the protected area". It is near the blast chutes on the pad.

When IR imagery was in its infancy, their device was $90,000, and was 4 frames per second, and we had early 386s back then, and there was not even a way to get the images off onto VHS. At least it was 16 colors.:-] We also had one that looked through a microscope.

Now, they are likely 16 bit color, and the image data can be manipulated post capture. These guys are pretty good.

I used to build a 2.5 foot by 4 inch, gold mirrored, rifle stock and scoped analog unit that power companies used to point at insulators and transformers to get status without having to climb poles or towers.

Now, they just point an imager at them. And a fairly cheap one at that.

The world has come a long way since '87.

Here's a good primer:

formatting link

Reply to
The Great Attractor

The probe also has to be calibrated. Ideally, NIST traceable with a correction chart provided.

Reply to
The Great Attractor

third

difficult to place more

stirred very rapidly,

trickey, as I found out

measure.

How can the focal length change the emmisivity or the effective temperature? If the surface is a mirror in the thermal IR, all you ever image is a reflection of *something*, but you don't measure the temperature of the mirror itself. Shiny copper has an emissivity of about 0.05, so 95% of whatever the IR meter is seeing is the reflection of something else.

Having tested a number of ways to increase the emissivity of shiny metallic surfaces, I know that black whiteboard marker brings the emissivity almost all the way up to 1.0. And it's easy to wipe off after you're done. Is 1.0 within your definition of "if at all"?

Do the math. Delta-t across scotch tape, working against air, is usually tiny.

Gosh, didn't somebody recently say that

'Surface quality is the most determinant factor, not "color"'?

Funny, they keep mentioning "black" too.

John

Reply to
John Larkin

What do you think the emissivity of an ice cube is?

John

Reply to
John Larkin

0.98, about as high as anything else on the planet.

John

Reply to
John Larkin

Take a look here for some really good facts:

formatting link

Water = 0.98

Aluminum varies from 0.04 to 0.3 depending on finish, absolutely useless for IR temperature measurement.

AlwaysWrong.

John

Reply to
John Larkin

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.