Direct vs. indirect measurements - for AgingYoung and others

[Folly]

This is split from the "Danny Klein - Fuel from Water" thread.

A long discussion started when, in response to a comment that the Sun's temperature could not be measured, followed by a comment that it certainly could, AgingYoung said

Upthread there was a mention of measuring the surface of the sun and it was pointed out that we could measure that temp. We can’t. We can derive or calculate a temperature. It was mentioned that we could move instruments further from the sun (maybe at some distance where they wouldn’t melt) and extrapolate the temperature. That’s a technique of calculating a value and isn’t a direct measurement. There are measured quantities like distance. You can actually take a tape and measure distance. Then there are derived quantities like speed (mph). Then there are calculated quantities like the surface temperature of the sun. It’s probably splitting hairs but these are distinctions made in instrumentation.

A number of other comments followed, like

This might help clarify the point. There are some quantities that are based on standards and you can take an instrument that is calibrated to that standard and directly measure an unknown quantity. Temperature is based on a standard. If you were sick and we wanted to know what your temperature was, I could take a thermometer (calibrated to that standard) and directly measure your temperature.

Frequency is a quantity that is considered derived. A method to calculate frequency is to measure the time between two consecutive occurrences of the event (the period) and then compute the frequency f as the reciprocal of this time..

F=1/t

When you measure temperature using derived measurements you aren't directly measuring; you are calculating it. You can't take a thermometer and directly measure the temperature of the sun. As you know it's too hot.

To the point of what is being sensed and temperature readings: if you are using a thermocouple it's sensing temperature change. You're directly reading temperature. If you're sensing light firstly you're sensing a derived unit (freq) then inferring temperature. You are not directly sensing temperature.

There is indeed a difference. That difference is, 'what is being sensed.' This point is fundamental to the field of instrumentation and process control. Basics in any discipline are established so that people within it can be on the same page when discussing the details. Usually someone not interested or versed in a particular discipline looks at things differently. That's a common occurrence.

and

It might be hard to follow but I'll give it one last shot. In a thermister the change in resistance is a function of temperature. The instrument varies as a function of the variable you are sensing. Mercury in a thermometer expands and contracts as a direct function of temperature.

When you're measuring a variable that is a function (i.e. freq) of the actual variable you want to measure (i.e. temp) you aren't directly measuring it. That's the case when first measuring frequency then inferring what temperature is. Temperature isn't the first variable that's causing the change in your instrument; it's not a direct measurement.

A number of people noted that in this sense, all temperature measurements are indirect, and any claim that spectrographic measurements of the temperature are less direct than a meat thermometer are, at best, misguided. After a number of claims that perhaps this is just too complicated for everyone, and this is just the way it is in industry, I posted a small quote from http://www.isa.org/InTechTemplate.c...tManagement/ContentDisplay.cfm&ContentID=9464

All temperature measurements are indirect. That is to say, the measurement is the measurement of volumetric expansion (liquid-filled thermometer), dimensional change (bimetallic thermometer), electromotive force (thermocouple), resistance (resistance temperature detector, or RTD), radiated energy (radiation thermometer), or some other characteristic of a material that varies predictably and reproducibly with temperature.

At this point, as expected, AgingYoung leaps on the next paragraph on this page

However, in industrial process measurement and control, the concept of direct and indirect temperature measurement has a different meaning. A direct measurement is a measurement of the temperature of the product itself. An indirect measurement is a measurement of some other temperature from which one can infer the product temperature.

saying it confirms what he has been saying all along, and

I might add that example clarifies direct and indirect measuring of temperature in a process but still the variable being measured is temperature. When we are attempting to know the temperature of the sun we aren't measuring temperature; we're measuring frequency (among other things) and inferring temperature from that. There is a slight distinction between 'indirect' and 'inferred' temperature measurements.

AgingYoung, did you even read the two paragraphs that you and I quoted between the two of us? In the first sense of direct measurement, all temperature measurements are indirect. A mercury thermometer is indirect, as it measuring the thermal expansion of a column of mercury, which is itself influenced by the subject temperature. A spectrographic temperature measurement is indirect because it measures the emitted black body radiation.

In the second sense (the "industry sense") described in the paragraph that you quoted, both instruments could be doing direct or indirect measurements. In fact, in the third and fourth paragraphs of the linked site, they even have examples:

An example of direct temperature measurement occurs when, as in roasting meat or making candy, it is possible to insert a thermometer directly into the product; insertion or immersion thermometers are often used.

One uses an indirect measurement in baking bread because one controls the oven air temperature. It is not practical to insert a thermometer into the bread because the action adversely affects the quality of the bread.

AgingYoung: you first tried blaming SI units, and then you insisted it's due to the differing sense of direct measurement used by industry. Do you still insist that we can not directly measure the temperature of the sun, while we can directly measure the temperature of a cup of coffee?

 

[stormer]

I got it, haven't you?

I read the other thread entirely, and this is what I understand AgingYoung to mean.

....Do you still insist that we can not directly measure the temperature of the sun, while we can directly measure the temperature of a cup of coffee?

If you stick this in your coffee to get the temperature, it is a direct measurement.

If you use this to measure the temperature of your coffee, it is not a direct measurement, but a derived one.

Hahaha!

(Now, I got to ask, why are you all trying to convince AgingYoung otherwise? Don't you all get frustrated by it?)

 

[Dilb]

(Now, I got to ask, why are you all trying to convince AgingYoung otherwise? Don't you all get frustrated by it?)

It's important to try and answer things properly so that people who genuinely want to learn have the right idea presented.

 

[stormer]

I have to say I really do admire all of our patience with dealing with something / someone like this.

 

[Paul C. Anagnostopoulos]

If you stick this in your coffee to get the temperature, it is a direct measurement.

It is? Seems to me it involves the heating of the glass, which then indirectly heats the mercury.

~~ Paul

 

[ponderingturtle]

It is? Seems to me it involves the heating of the glass, which then indirectly heats the mercury.

~~ Paul

Which expands and you measure the expansion. Not as dirrect as it seems

 

[stormer]

It is? Seems to me it involves the heating of the glass, which then indirectly heats the mercury.

~~ Paul

Oops, I don't think you should take that one sentence in the post to reply to, but rather read the post as a whole? Maybe even read my other post after that to see the context?

 

[Paul C. Anagnostopoulos]

Oops, I don't think you should take that one sentence in the post to reply to, but rather read the post as a whole?

I read the entire post. Perhaps you mean that the temperature gun is more indirect than the thermometer?

Maybe even read my other post after that to see the context?

This one?

I have to say I really do admire all of our patience with dealing with something / someone like this.

~~ Paul

 

[stormer]

Hey Paul. The direct measurement or indirect measurement is not a distinction I am trying to make, but rather one AgingYoung was in the Denny Klein - Fuel from Water thread.

You should know, in that thread, you asked that discussion about these things be continued here. I did not want to comment on temperature measurement in that thread.

I was just quite (I can't find the right word, so I'll just stick with...) surprised that that discussion resulted in this thread due only to one poster insisting on his view about the direct and derived measurement.

By AgingYoung's opinion, the thermometer would be a direct measurement of temperature, while the InfraRed non-contact would be a derived measurement of the temperature. I don't agree, and I was amused at that poster's opinion.

Hahaha!

I go on to say that (in not so many words) "I admire the patience of the people who were trying to get the message across to the OP even though the OP was being stubborn and contradictory to their own words about direct temperature measurement and derived temperature measurement". I made a typo there in saying "our" when I wanted "your". My mistake.

Hope this helps clear things up

 

[joobz]

Why is this an issue? Why does it matter that the calculation of the sun's surface temp be a indirect measure?

AgingYoung mentioned that this is an important issue with process control, which is true. there you need to know the response time of the probe (indirect having slower responses typically) and the offsets they may have (surface jacket temperature as a function of internal tank temperature). These things need to be programmed in and modeled, otherwise you get wrong readings.

BUT, I can only assume that those who did calculations on the sun surface temp took into account enough of the issues to provide a reasonably accurate value. If you have an issue with the calculation or assumptions made in the calculation, than those should be addressed. but to claim that the values can't be (or ever be) trusted because we only use indirect measures is just silly.

 

[ponderingturtle]

Why is this an issue? Why does it matter that the calculation of the sun's surface temp be a indirect measure?

AgingYoung mentioned that this is an important issue with process control, which is true. there you need to know the response time of the probe (indirect having slower responses typically) and the offsets they may have (surface jacket temperature as a function of internal tank temperature). These things need to be programmed in and modeled, otherwise you get wrong readings.

BUT, I can only assume that those who did calculations on the sun surface temp took into account enough of the issues to provide a reasonably accurate value. If you have an issue with the calculation or assumptions made in the calculation, than those should be addressed. but to claim that the values can't be (or ever be) trusted because we only use indirect measures is just silly.

He was asserting initialy that you can't know the temp of the sun because it was an indirrect measurement. Now we never touched on the accuracy of the measurement, but only about how you are measureing it.

 

[Paul C. Anagnostopoulos]

Hope this helps clear things up.

It does, thanks. So where is AgingYoung?

~~ Paul

 

[joobz]

He was asserting initialy that you can't know the temp of the sun because it was an indirrect measurement. Now we never touched on the accuracy of the measurement, but only about how you are measureing it.

I've never looked into how they measured/calculated the sun temperature, but that seems like a splitting hairs.
It's like saying, "We don't know the exact value of Pi"

What would the value be used for? I'm guessing that the indirectly measured temperature is useful enough for most calculations. (as is using pi=3.14) If you need better accuracy, you can use pi=3.14159, but is that the case here?


Like I said, if he feels that the value is wrong for some fundemental reason, then it is a valid point. If he feels the value is wrong simply because it is a "indirect measure", then he's whole point is moot.

 

[ponderingturtle]

I've never looked into how they measured/calculated the sun temperature, but that seems like a splitting hairs.
It's like saying, "We don't know the exact value of Pi"

What would the value be used for? I'm guessing that the indirectly measured temperature is useful enough for most calculations. (as is using pi=3.14) If you need better accuracy, you can use pi=3.14159, but is that the case here?


Like I said, if he feels that the value is wrong for some fundemental reason, then it is a valid point. If he feels the value is wrong simply because it is a "indirect measure", then he's whole point is moot.

His whole point has always been moot, he just managed to troll it out longer.

 

[Myriad]

This raises the question, is there any such thing as a truly "direct" measurement? Don't the very concepts of "measurement" and "directness" contradict one another?

Use a ruler to measure the distance between two marks on a wooden board. The scale on the ruler doesn't show the distance between the marks; it shows the distance between the adjacent points on the ruler. Close, but not direct.

Of course, you could use the scale on the ruler to show how far apart two scale marks on the ruler are. But there's no unknown value in that scenario. You're not measuring anything. Similarly, you can define the distance between your two marks on your board as a new unit of displacement measurement, and then measure other things (indirectly) using that unit. But defining the unit, in itself, is not measuring anything.

"Direct" as applied to measurement only works as a relative concept. It's more direct, say, to measure the temperature of the sun from the frequencies of the light of the sun's visible disc than to measure it from the frequencies of the light reflected by the moon.

Respectfully,
Myriad

 

[Folly]

I was just quite (I can't find the right word, so I'll just stick with...) surprised that that discussion resulted in this thread due only to one poster insisting on his view about the direct and derived measurement.

Well, I'm surprised too: I have no idea why I care. For some reason, this really irritated me, and I didn't want to just let it go, even if it was just a tiny, silly claim mentioned in a larger, sillier claim.

This raises the question, is there any such thing as a truly "direct" measurement? Don't the very concepts of "measurement" and "directness" contradict one another?

Use a ruler to measure the distance between two marks on a wooden board. The scale on the ruler doesn't show the distance between the marks; it shows the distance between the adjacent points on the ruler. Close, but not direct.

I had understood comparison to be all measurement was, and because you're comparing two lengths without intermediaries, this is a direct measurement. That seems slightly vague to me, and if there's a better definition, I'd love to hear it.

Of course, you could use the scale on the ruler to show how far apart two scale marks on the ruler are. But there's no unknown value in that scenario. You're not measuring anything. Similarly, you can define the distance between your two marks on your board as a new unit of displacement measurement, and then measure other things (indirectly) using that unit. But defining the unit, in itself, is not measuring anything.

"Direct" as applied to measurement only works as a relative concept. It's more direct, say, to measure the temperature of the sun from the frequencies of the light of the sun's visible disc than to measure it from the frequencies of the light reflected by the moon.

I had also understood there to also be an actual distinction to be made based on intermediary variables. A ruler makes a direct measurement because it compares a length to a length. A thermometer makes an indirect measurement because it compares a length which changes with temperature. I don't ever deal with this in my job, so I don't know whether this is just the sort of thing you hear in school that is completely ignored in the real world, or whether this distinction is actually considered to be of interest. My guess is that it's only the relative sense that is considered, and only so far as it affects the accuracy of the measurement and the conditions under which it applies. I could certainly be wrong: while I was looking for information on this, a German site with English text popped up describing some temperature measurement device that was explicitly stated to be indirect.

 

[davefoc]

I think AgingYoung made two errors and for whatever reason decided to push on after he realized that he had made them.

The first was the direct/indirect thing. He initially thought that a measurement technique that involved direct contact was fundamentally different than a measurement technique that involved measurement of electromagnetic radiation because one method measured temperature directly and the other method involved a calculation and assumptions about the relationship between the characteristic actually being measured and temperature.

Many posts were generated that explained that this wasn't the case and that in fact there is no method of directly measuring temperature using the definition that he proposed.

He then went on to his second mistake which was that the distinction he was making was recognized by industry.

At least in the links he posted the recognized meaning of direct and indirect with regard to temperature measurement was not what he had in mind. An indirect temperature measurement is one where the temperature of one thing is inferred by measuring the temperature of something else.

So AgingYoung appears to have been wrong. Having been completely wrong myself a few times I can say that it isn't always the easiest thing to just fess up and move on. Since AgingYoung hasn't shown up in this thread, perhaps he is tacitly acknowledging that he was wrong right now.

 

[joobz]

I think AgingYoung made two errors and for whatever reason decided to push on after he realized that he had made them.

The first was the direct/indirect thing. He initially thought that a measurement technique that involved direct contact was fundamentally different than a measurement technique that involved measurement of electromagnetic radiation because one method measured temperature directly and the other method involved a calculation and assumptions about the relationship between the characteristic actually being measured and temperature.

Many posts were generated that explained that this wasn't the case and that in fact there is no method of directly measuring temperature using the definition that he proposed.

He then went on to his second mistake which was that the distinction he was making was recognized by industry.

At least in the links he posted the recognized meaning of direct and indirect with regard to temperature measurement was not what he had in mind. An indirect temperature measurement is one where the temperature of one thing is inferred by measuring the temperature of something else.

So AgingYoung appears to have been wrong. Having been completely wrong myself a few times I can say that it isn't always the easiest thing to just fess up and move on. Since AgingYoung hasn't shown up in this thread, perhaps he is tacitly acknowledging that he was wrong right now.

that is most likely a fair assessment.

I will state though that people generally gain much more of my respect when they acknowledge fault or being wrong.

I had tried to start a thread where people would post hypotheses they've made in research that turned out to be strikingly wrong, but that failed completely.

It is hard to admit mistakes.

 

[Iamme]

It is hard to admit mistakes.

Yes. But especially so if you hardly ever make one.

 

[Schneibster]

I feel comfortable maintaining that ALL measurements of any kind whatsoever are indirect. My rationale is that we cannot experience reality itself directly, but do so only through the signals delivered to our brains by our nervous system. Thus, no matter what we think we are measuring, what we are actually measuring is the frequency of nerve impulses from some input or other. And that's only if you admit that "you" and "I" actually EXIST.

In the presence of all this philosophical uncertainty, I have to decide at some point that I will accept certain facts as manifestations of some "reality" "out there." Making what I consider to be minimal assumptions, I find that the "accuracy" of "indirect" measurements is as good as anything I can come up with as a means of quantifying phenomena in my environment. Thus, I see in this light little difference between very, very closely estimating the temperature of a cup of coffee by sticking a thermometer in it and very, very closely estimating the temperature of the surface of the Sun using blackbody theory and measurement of its spectral peak.

Sounds like someone trying to obfuscate the issues to me.