Kodak Type 3 Thermometer Accuracy

John Gnagy · Oct 23, 2016

Hey folks,

I recently grabbed a Kodak type 3 on ebay in very well cared for condition, but it reads about a degree higher than the other 3 thermometers I have here. I was previously using a Patterson 12" color thermometer, which was bought new and agrees with the other thermometers I have. While I realize that it's possible that the other 3 thermometers are off, it seems more likely that the Kodak is off - even with its claimed accuracy of +-0.25F. Can these thermometers go out of calibration over time? I don't see any gaps in the mercury and it seems like a closed system like this wouldn't have many ways other than gaps to go out of calibration. I'm not an expert though, which is why I'm asking here!

Thanks,
John

Leigh B · Oct 23, 2016

If the Kodak is the type with a glass tube in a metal holder, the glass tube may have shifted.

I suggest moving the tube so this one agrees with the others.

From a process control standpoint, uniformity among different instruments is far more important than absolute accuracy.

- Leigh

MattKing · Oct 23, 2016

I am assuming you are referring to is a Type 3 Process Thermometer, here is a link with the instructions: https://www.flickr.com/photos/etr420/9280657696/in/photostream/lightbox/

silveror0 · Oct 23, 2016

Don't mean to hijack this thread, but can anyone explain the differences between Types 1, 2 and 3?

RalphLambrecht · Oct 23, 2016

John Gnagy said:
Hey folks,

I recently grabbed a Kodak type 3 on ebay in very well cared for condition, but it reads about a degree higher than the other 3 thermometers I have here. I was previously using a Patterson 12" color thermometer, which was bought new and agrees with the other thermometers I have. While I realize that it's possible that the other 3 thermometers are off, it seems more likely that the Kodak is off - even with its claimed accuracy of +-0.25F. Can these thermometers go out of calibration over time? I don't see any gaps in the mercury and it seems like a closed system like this wouldn't have many ways other than gaps to go out of calibration. I'm not an expert though, which is why I'm asking here!

Thanks,
John

John.do you know the difference between accuracy and precision?With thermometers,the latter is more important than the former.If you have three thermometers reading closely to each other,I would trust them most.Pick one and only use that one.a man with a watch knows exactly what time it is;a man with two watches can never be sure.

Leigh B · Oct 23, 2016

RalphLambrecht said:
John.do you know the difference between accuracy and precision?
With thermometers,the latter is more important than the former.

Precision is a function of the design. It cannot be adjusted by the user.
The only way you can change the precision of a thermometer is to buy a different thermometer

Accuracy is adjustable by the user.
However, that can only be done at one point.
To what extent that is repeatable at other points is a function of the manufacturing tolerances.

- Leigh

John Gnagy · Oct 23, 2016

Leigh B said:
If the Kodak is the type with a glass tube in a metal holder, the glass tube may have shifted.

The Type 3 has the scales painted on the glass, so it can't be adjusted that way.

silveror0 said:
Don't mean to hijack this thread, but can anyone explain the differences between Types 1, 2 and 3

Edward Zimmermann at photo.net once explained the difference between the Type 2 and Type 3. Hopefully he's OK with me quoting what he said here:

"Type-2 is for B&W and designed for 68oF (20C) processing to good accuracy. It is a nice thermometer but not NIST calibrated.
Type-3 is for colour and NIST reference calibrated--- the scales are handpainted on the glass.
-Celsius and Fahrenheit scales (individually calibrated!)
-12-60oC (54-140oF)
-Celsius calibrated (accurate to ±0.125) at 15, 20, 25, 40 and 50oC. That's 5 calibration points!
-Accurate to ±¼oC over the rest of the scale.
-Fahrenheit calibrated (accurate to ±0.25) at 59, 68, 77, 86, 104 and 122oF. Accurate to ±½oF over the rest of the scale.

If its a large Mercury glass thermometer set in a stainless steel holder and its handpainted and calibrated at the above points in BOTH Celsius and Fahrenheit scales and it covers the above range.. then its a Type-3."

(Quoted from http://photo.net/black-and-white-photo-printing-finishing-forum/0094n5)

RalphLambrecht said:
John.do you know the difference between accuracy and precision?

I believe so, and I think the Kodak Type 3 is believed to be very precise (repeatable). Based on its specs, it should also be very accurate, but I'm wondering if that accuracy can be lost over the life of the thermometer.

Thanks,
John

John Gnagy · Oct 23, 2016

Leigh B said:
Accuracy is adjustable by the user.

True, although in the case of this thermometer, I would need to repaint the scales at the readings I care about. I would also need a reference thermometer that I know is accurate, which I don't have.

MattKing · Oct 23, 2016

My Kodak Process Type 3 thermometer serves as my reference. As best as I can determine, while they can be damaged, I don't think they can go out of calibration.

If I was in your shoes, I would use it as a reference, and consider my other thermometers to be off a degree. That doesn't make them unusable. It just makes it necessary to translate their results.

AgX · Oct 23, 2016

John Gnagy said:
I believe so, and I think the Kodak Type 3 is believed to be very precise (repeatable). Based on its specs, it should also be very accurate, but I'm wondering if that accuracy can be lost over the life of the thermometer.

-) Fluid stem thermometers are precice by principle of design.
-) Their accuracy depends on the quality of manufacture, it will not change over their life.

-) There are standards on their accuracy depending on their scale division.

Leigh B · Oct 24, 2016

It's pretty unusual for a mercury thermometer to read high, but not impossible.

This could happen if the mercury column separated quite low in the tube, perhaps not obviously.

The solution is to cool the thermometer below the lowest calibration, so all the mercury goes into the bulb.

- Leigh

AgX · Oct 24, 2016

Well, I agree, that would be an exeption to its inherit precision.
Especially in a case where that seperation would be so small that it would not be obvious.
I never learned about such case, but maybe we should consider it.

RalphLambrecht · Oct 24, 2016

AgX said:
-) Fluid stem thermometers are precice by principle of design.
-) Their accuracy depends on the quality of manufacture, it will not change over their life.

-) There are standards on their accuracy depending on their scale division.

what does scale division have to do with accuracy?

msage · Oct 24, 2016

John Gnagy said:
Hey folks,

I recently grabbed a Kodak type 3 on ebay in very well cared for condition, but it reads about a degree higher than the other 3 thermometers I have here. I was previously using a Patterson 12" color thermometer, which was bought new and agrees with the other thermometers I have. While I realize that it's possible that the other 3 thermometers are off, it seems more likely that the Kodak is off - even with its claimed accuracy of +-0.25F. Can these thermometers go out of calibration over time? I don't see any gaps in the mercury and it seems like a closed system like this wouldn't have many ways other than gaps to go out of calibration. I'm not an expert though, which is why I'm asking here!

Thanks,
John

Definitely use the type 3 as your standard. I have 3 of them and they agree with each other.

AgX · Oct 24, 2016

RalphLambrecht said:
what does scale division have to do with accuracy?

Typically the acuracy of a fluid stem thermometer is related to its scale division (=smallest indicated part of°): the finer the division, the more accurate the capillary is manufactured too.
This also is reflected in the calibration standards.

John Gnagy · Oct 24, 2016

Leigh B said:
It's pretty unusual for a mercury thermometer to read high, but not impossible.

This could happen if the mercury column separated quite low in the tube, perhaps not obviously.

The solution is to cool the thermometer below the lowest calibration, so all the mercury goes into the bulb.

- Leigh

I stuck the thermometer in the freezer for a couple of hours and then let it come to room temp while standing vertically. It now only reads a half degree higher than the other thermometers. That must have been what happened. Thanks!

RalphLambrecht · Oct 24, 2016

AgX said:
Typically the acuracy of a fluid stem thermometer is related to its scale division (=smallest indicated part of°): the finer the division, the more accurate the capillary is manufactured too.
This also is reflected in the calibration standards.

I disagree,Itsensitises your read-out but it doesn't improve accuracy.A scale with a 0.001g read out isn't more accurate or precise than a scale with a 1g read-out.

AgX · Oct 24, 2016

You still mix up read-out and manufacturing precision.

Leigh B · Oct 24, 2016

Sorry, Ralph, but I must agree with AgX on this one.

It's all about manufacturing tolerances.

A precision thermometer is manufactured to tighter tolerances (= higher precision) than a loosy-goosy one.

No manufactured product exists with zero tolerance accuracy.
Even national standards have "uncertanties".

Standards with zero tolerance exist only by statute, not by physical characteristics.
An example is the "meter bar" in Paris that was the established standard of length.

- Leigh

RalphLambrecht · Oct 25, 2016

AgX said:
You still mix up read-out and manufacturing precision.

funny;I thought You did.

MattKing · Oct 25, 2016

Ralph:
If you are speaking about digital readouts, you are completely correct.
But when it comes to the sort of analogue "readouts" you find on an engraved and calibrated thermometer, there is a real connection between the precision of the engraving and the accuracy of the thermometer.

darkroommike · Oct 25, 2016

I think it's more important for a reference thermometer to be consistent than absolutely accurate in a darkroom environment. We are not the Bureau of Standards.

Leigh B · Oct 25, 2016

darkroommike said:
I think it's more important for a reference thermometer to be consistent than absolutely accurate in a darkroom environment.

That is very true.

Since everybody does (or should) calibrate the entire process...
any absolute error in the thermometers will be calibrated out.

This requires that the readings be very repeatable.

- Leigh

RalphLambrecht · Oct 26, 2016

darkroommike said:
I think it's more important for a reference thermometer to be consistent than absolutely accurate in a darkroom environment. We are not the Bureau of Standards.

that is exactly right;repeatability is more important than absolute accuracy

RPC · Oct 26, 2016

darkroommike said:
I think it's more important for a reference thermometer to be consistent than absolutely accurate in a darkroom environment.

Unless you are developing color film, which is very sensitive to temperature. Then errors can lead to crossover or altered contrast in the film.

Kodak Type 3 Thermometer Accuracy

John Gnagy

Leigh B

MattKing

Moderator

silveror0

RalphLambrecht

Leigh B

John Gnagy

John Gnagy

MattKing

Moderator

AgX

Leigh B

AgX

RalphLambrecht

msage

AgX

John Gnagy

RalphLambrecht

AgX

Leigh B

RalphLambrecht

MattKing

Moderator

darkroommike

Leigh B

RalphLambrecht

RPC

Kodak Type 3 Thermometer Accuracy

Moderator

Moderator

Moderator

Privacy & Transparency

Privacy & Transparency