Experiment to determine the effect of pressure on temperature in a thermally dynamic system – phase 1: design and equipment acquisition

Posted: June 11, 2012 by tallbloke in atmosphere, data, Energy, Measurement, methodology
Tags: , ,

A watched pot does boil!

It is remarkable that in the climate science debate, the ideal gas law and its consequences in dynamic systems has been variously forgotten, misinterpreted, denied and ignored. In order to clear up the misconceptions, obfuscations , ignorance, error, and denial, it is time to do some practical science in order to lay the various misapprehensions and mis-statements to rest.

It’s very encouraging to see that ‘Lucy Skywalker’ is intending to replicate the experimental work of Roderich Graeff. This is a serious undertaking and a difficult task, due to the very accurate measurement of small differences required. I have decided the Talkshop is going to enter the fray with some empirical experimental work too.  The aim is somewhat simpler. We are going to measure the effect of Pressure on a contained volume of air which has energy passing through it, as per Ned Nikolov and Karl Zeller’s outline of the situation in Earth’s atmosphere, which is a volume of air  contained by gravity, with sunlight passing through Earth’s day side. This is so we can determine whether there is merit in their hypothesis that the atmospheric temperature profile is underpinned by the effect of gravity on atmospheric mass: warm near the surface where the air pressure is around 14 psi, and cold at high altitude, where the air pressure drops nearly to zero.

Talkshop regulars will remember that a few months back, contributor Konrad Hartmann performed an experiment using pet bottles in direct sunlight. There was some constructive criticism of his experiment, and he made some design improvements which we are awaiting results from. Konrad tested the effect of increased pressure. Initially, we will go the other way, and see what happens when we reduce pressure towards vacuum. After that we will test positive pressures too.

I have been scouring ebay for the equipment we will need to accomplish the task. So far I have a pack of 20 5W ceramic resistors to make a controllable and accurately measurable heat source. And I have also scored a very nice old vacuum pump. This was a real bargain. It’s a Leybold Heraeus, a veritable piece of German precision engineering.

Although it won’t achieve a really hard vacuum, it will get near enough to find out what sort of relationship exists between pressure and temperature as we change the pressure from near vacuuum to ordinary atmospheric levels. John, the ebay seller, very kindly reduced his price when I went to pick it up (it was too heavy for the postal service), when I told him it was for non-profit experimental work.

I’m also awaiting the arrival of another bargain – a Siemens pressure sensor which is insensitive to temperature changes.

It works on the principle of piezo resistivity, using a hybrid ceramic diaphragm to transmit the pressure from the test medium. Full spec here. These are expensive sensors. The new price is 391 Euros. I just won a brand new in-the-box 0-4bar P4 model for £5 on the ‘bay – my favourite shopping channel.  🙂

For the pressure vessel, I intend to use an old ‘camping gaz’ butane cylinder. This is small enough for vacuum pumping times to be reasonable, and large enough for the required separation between heat source and thermistor.

Which brings me to the measurement side of the experimental set-up.

I have found a neat little four input oscilloscope/datalogging module.

This comes with free software and connects to a USB port. Amplification will be needed for the thermistor signal, and a small power supply for this and the the pressure sensor. I intend to build all this into an old SCSI external hard drive enclosure to keep things neat. More on the construction of this unit once I have all the necessary components in stock.

All this effort is being undertaken in the spirit of Einstein’s famous injunction:

Experimentum Summus Judex – experiment is the final arbiter.

It the climate science arena, we have witnessed how computer model output resting on untested theory has been put forward as  scientific truth. This is a fallacy, because ultimately,  hypothesis is mere conjecture until experiments are designed and carried out. If experimental results support the hypothesis, it may go forward to become a fully fledged theory.

We are going back to basics here and doing some real science. Before we conduct the experiments, we will make predictions from our hypothesis, and test them properly. All ideas and input from Talkshop contributors and the wider scientific community are welcome. We want to do this as well as we can within the budgetary constraints imposed on non-institutional research.

Comments
  1. Konrad says:

    Roger,
    This looks like a very good way to look into the N&Z hypothesis. One of the main disadvantages I had with the PET bottles was that I could only test atmospheric pressure and higher. Higher pressures cause more conductive losses through the test chamber walls, reducing the temperature differential to be observed. Tests at 0.25 bar and 0.5 bar should reduce these effects.

    I would have a few suggestions. Firstly an internal circulation fan to prevent stagnation of the gas tested. This could also be achieved by placing the resistor lower than the thermistor in the test chamber to allow convection to keep things moving. The second suggestion would be to shield the thermistor from stray IR, just as I had to shield the thermometer probes from incoming SW in the pet bottles. Also, just as I had to use a fan to stabilise the temperature of the PET bottles before sun exposure, you will need some way of externally controlling the test chamber temperature before starting a run as pump down will cause cooling.

    The use of the resistors as the heat source also offers an opportunity to provide intermittent heating similar to the diurnal cycle on Earth.

    One poster on a previous N&Z thread made a comment concerning heat sinks in high altitude electronics. Some of the figures in the two links below could help with predicting the energy transfer to gases at differing pressures.

    http://archive.ericsson.net/service/internet/picov/get?DocNo=28701-EN/LZT146231&Lang=EN&HighestFree=Y

    http://dspace.mit.edu/bitstream/handle/1721.1/39011/20404614.pdf?seque.provide

  2. Ray Tomes says:

    They way I see this whole issue is to look at the statistics of gas molecules. If we consider two locations vertically above each other, then gas molecules falling from the upper to the lower are accelerated by gravity, while those rising are slowed down. Faster equals hotter and slower equals cooler. It is very easy to calculate the effect of this given the temperature and velocity of air molecules. According to my rough calculations this does lead to a temperature gradient that is about the same as the the observed one. Anyone would have to be foolish to deny that such a factor is there. The only question is whether it entirely explains the variation of temperature with height.

  3. bwdave says:

    If one follows a gas through a centrifugal compressor, the gas warms continuously as it gets compressed. How is this any different than gravitational compression?

  4. Herkinderkin says:

    Gravity schmavity. Compression is compression – period. A given volume of a gas contains a given amount of energy at a given temperature. If no energy is removed and it is compressed so that its volume is smaller, it follows that the temperature must rise. If it expands so that its volume is greater (and no extra energy is added) then the temperature must fall. Because heat moves from hot bodies to cold bodies – the “cold” side naturally picks up heat from the environment, while the “hot side” adds it to the environment. That’s the basis of heat pumps.

    Tallbloke’s experiment seeks to determine how the behaviour of a gas through which energy is flowing varies with pressure.

  5. I wish I knew in advance about you needing a vacuum pump I have one that is just the ticket almost new only slightly used. I was saving it to evacuate a passive heat exchanger before adding the refrigerant.

    Face view of oil level window

    Details of vacuum achievable

    Side view of pump

  6. tallbloke says:

    Konrad: thanks for the comments, suggestions and links. I’ll have a think about how we might use some insulation to advantage, as the warm ambient temperature will affect the low pressure curve compared to theory.

    Ray: I agree, but if you’ve followed any of the threads, you’ll know that there is all sorts of conflicting opinion on the statistical mechanics of Boltzmann distributions etc. I’m hoping we can cut through that with a clear practical demonstration.

    bwdave: Too many other factors with centrifuges for a clearcut set of inferences to be drawn in my opinion. Feel free to start spinning and measuring. 😉

    Herdinkderkin: Quite so. It’s the dynamic part which was missed by Ira Glickstein which led to a lot of confusion when N&Z’s theory was first discussed on WUWT. You’d have thought a Phd physicist would know better.

    Richard, nice looking unit! I would think that the cost of shipping your vacuum pump over here would far exceed what I paid (£40). Thanks for the thought though. I’ve had to remove your (huuuuge) images for now, because they will cause slow page loads for people on slow connections. I’ll make some smaller versions and re-upload them

  7. Just trying to help, sorry for the clutter.

  8. tallbloke says:

    Richard: no worries. You can either make the images smaller by selecting a lower resolution on your digicam for the pics when you take them (640×480 is ideal) or reduce them with a simple program like irfan view before you upload them. Or edit them in wordpress after you upload them. I’ve reduced the display size to 614 pixels wide, the width of the text column.

  9. I just thought the 5 millitorr min might be handy. Thanks for the sizing lesson.

  10. tallbloke says:

    Richard: I’ll be interested to see what my 40 year old Leybold-Heraeus can manage. What is the power rating of the motor on your unit? Mine is 550W

  11. Richard111 says:

    Inspiring stuff TB. Will be watching closely from the sidelines.

  12. Tallbloke

    niiiiiiice to see some more experimental work. Two caveats however.

    First, the phrase is “Experimentum SummUs Iudex or Judex” – U not A – and please will you correct my MS too?

    Second, more important. I may be just sheer stupid, but isn’t this work just verification of the Gas Laws, you know, PV=RT? I do not see this as the same issue as Nikolov and Zeller are dealing with – which I see as Graeff’s much smaller effect of gravity, difficult to measure at lab scale, but multiplied up to geographic scale becomes highly significant as a prime temperature cause – which is then largely but not completely overcome by convection, to reach the uneasy truce of the adiabatic lapse rate.

    I think it is of high significance that all atmospheric planets have a tropopause at the same pressure level as Earth ie around one-tenth of an atmosphere, 0.1 bar. Also highly significant in this context, that in densely atmospheric planets, the winds are so huge. To me this is no surprise, once one grasps the inevitable existence of an ongoing balancing act between gravity and convection in gases.

  13. I never tire of looking at the Hilsch Vortex Tube.

    It works, but its raison d’etre is still a mystery to physicists. To me this phenomenon is clearly in line with all we are looking at here, that supports Nikolov and Zeller’s thesis.

  14. oldbrew says:

    What about ‘expected results’? Maybe the Standard Atmosphere Calculator can help here.
    http://www.digitaldutch.com/atmoscalc/

    It says psi at zero altitude should be 14.965949. By tinkering with the ‘altitude’ parameter i.e. using it as a proxy for pressure change, it will display what the exact temperature should be for any level of psi. It will accept negative altitude.

  15. 115 volt 60 Hz 3.10 amps 1/10 Hp 3450 rpm General electric model 5K010mgr13T so about 355 watts.

    Lucy; We used a lot of centrifugal dust separators to clean flue gas from powdered coal power plants that looked a lot like that with a vertical orientation and the dust removed along the circumference at the bottom, flue gas sans particles out the center top.

  16. tallbloke says:

    Lucy: ” isn’t this work just verification of the Gas Laws, you know, PV=RT? I do not see this as the same issue as Nikolov and Zeller are dealing with”

    No. The key difference is the heating element inside the pressure vessel. This is to emulate the situation N&Z describe where solar energy passes through the atmosphere’s pressure gradient. To repeat what I said in an earlier comment:

    It’s the dynamic part which was missed by Ira Glickstein which led to a lot of confusion when N&Z’s theory was first discussed on WUWT.

    Oldbrew: Thanks, that looks like a useful reference. It looks like you are using an environmental lapse rate of 6.5C/km, yes?

  17. oldbrew says:

    TB: have a look at ‘Graphs’ and ‘Help’ on the SAC menu.

  18. tallbloke says:

    Oldbrew, thanks. Ned Nikolov commented some time ago that you have to be careful with temperatures above the tropopause because the measurement method is different. Whereas in the troposphere you are measuring bulk gas, at very high altitudes the measurements are of individual molecules.

    I think this will change the curves considerably above the stratosphere if you are considering energy per volume.

  19. Scute says:

    Tallbloke and bwdave: I was going to suggest centrifuging on Lucy Skywalker’s Graeff’s Second Law Seminar. For me, the obvious problem was getting such a setup to spin along with some or all of the sensors and their wiring. However, I wouldn’t have thought that the basic principle of centrifuging is out of the question. I had the notion of a cylinder of gas, like Graeff’s, being spun, either eccentrically like a weight on a string; or concentrically i.e. the axis being at the centre of the tube, with the gas being ‘flung’ to the two ends and a partial vacuum arising at the axis.

    As far as I can see, this is the only way it is possible to replicate Graeff’s experiment in an exaggerated form, with exaggerated G force. This would surely make it easier to tease out any effects, notwithstanding the usual problems with convection which might also be exaggerated.

    While I think it is useful to do the experiment as described above and also in the format that Konrad did it, I do think there is something very neat about Graeff’s pressure gradient in that it replicates the atmosphere’s pressure gradient in microcosm.

    The one difficulty with Graeff is that the pressure gradient is small so the effects are small. But by centrifuging a cylinder those effects would be magnified.

    One small caveat with the pressure gradient in a centrifuged tube is that the G force isn’t constant but increases in a linear fashion with respect to radius (g=rw*2) and that would mean that the pressure gradient would be even more non-linear than it is in the atmosphere. This non-linearity would be less apparent at higher rpm because more gas would be concentrated at the end(s) of the tube, in other words, distributed over a smaller portion of the radius. However, it might then be harder to get readings in such a small space unless the initial tube was quite long and robust enough.

    As I said, the problem with this setup is attaching wires and sensors to a spinning apparatus and still getting meaningful readings. But up to the point where you do that, I feel that it is conceptually and experimentally neat to centrifuge Graeff-like tubes.

  20. oldbrew says:

    TB: Yes, the SAC is only valid for the troposphere. Pressures below about 3-3.5psi aren’t relevant to your experiment I would suggest, because they are too low for the troposphere (on average at least).

  21. Stephen Wilde says:

    Glad to see this issue getting the attention it deserves.

    I haven’t had time to read everything so could someone enlighten me on a point ?

    Does the proposed containment vessel adequately mimic the atmospheric radial increase in atmospheric volume with height and the 3D nature of the radial increase ?

    That would create a logarithmic increase in the cooling effect of increased atmospheric height which gives a very powerful cooling mechanism when the atmosphere expands upwards even by a small amount.

    Ideally the vessel would need to be wider at the top and open to space and even then it would only be a 2D representation of the 3D atmospheric response.

    It is the speed and ease with which an atmosphere expands or contracts when energy is added or subtracted that is at the heart of this issue but that is already fully accounted for in the Gas Laws.

    What does the Graeff experiment or the N & Z proposition add to the Gas Laws and the Standard Atmosphere ?

    The Gas Law already caters for radial expansion and opennness to space.

    I think the idea that one could show a gravitational effect in a container of uniform width from top to bottom and closed so as to prevent expansion at the top may be a little fanciful but I’d be content to be wrong on that point.

    Furthermore I think that the expansion and contraction of the atmosphere only counters the effect of increased or decreased solar incoming on the larger or smaller circumference at the top of the atmosphere exposed to solar irradiation.

    To prevent surface warming it is the circulatory changes rather than the density changes that are important.

    A bigger or faster circulation in an expanded atmosphere is what whisks energy away from the surface faster to prevent the extra energy in the atmosphere warming the surface.

  22. wayne says:

    “As I said, the problem with this setup is attaching wires and sensors to a spinning apparatus and still getting meaningful readings. But up to the point where you do that, I feel that it is conceptually and experimentally neat to centrifuge Graeff-like tubes.”

    I wonder if there is a gas that changes color with different temperature that is in the right range? That would remove the problem of wires and sensors within the centifuge tubes. I do agree that some sort of magnification of the effect sure would help in getting it out of the range of sensor sensitivity and accuracy.

  23. Ray Tomes says: “According to my rough calculations this does lead to a temperature gradient that is about the same as the observed one. Anyone would have to be foolish to deny that such a factor is there. The only question is whether it entirely explains the variation of temperature with height.

    Well, Ray, I am sure we all agree. But surely the whole point of the experiment is to answer your “only question”.

  24. oldbrew says:

    Stephen Wilde says:
    “Does the proposed containment vessel adequately mimic the atmospheric radial increase in atmospheric volume with height and the 3D nature of the radial increase ?”

    The Standard Atmosphere Calculator I mentioned earlier should act as a guide there. If the results don’t match the SAC, questions have to be asked…

  25. bwdave says: “ If one follows a gas through a centrifugal compressor, the gas warms continuously as it gets compressed. How is this any different than gravitational compression?

    Please don’t start that old hare running! Herkinderkin has responded briefly but let me expand on his comment.

    Yes, of course any method of compression will warm gas – just as pumping up a bicycle tyre makes it warm. But that is a transient change, soon dissipated as the gas cools down again due to heat loss to the ambient surroundings.

    TB’s experiment (just like Konrad’s) uses a fixed source of heat to continuously maintain the air inside the cylinder at a higher temperature than its surroundings. So, even though heat is constantly leaking out through the cylinder walls, the elevated temperature inside remains fixed (in other words in a steady state) despite the energy continuously flowing through it.

    TB is an excellent experimentalist. When he changes the pressure inside the cylinder from one value to the next, I am sure he will wait long enough for the temperature to settle down again to a steady state value before recording it. This will eliminate the transient thermal effect that concerns you.

    The 64,000 dollar question is: will the second steady state temperature be different from the first steady state temperature? Konrad found that is was not the same, thus providing empirical evidence supporting the Nikolov & Zeller proposition that the temperature in an atmospheric column subject to a continuously varying pressure profile due to gravity (highest at the surface and zero at the top) has a matching temperature profile that is dictated solely by the Gas Laws and is materially unaffected by atmospheric composition (e.g. the presence or absence GHGs).

    TBs experiment aims to be more accurately controlled than Konrad’s so that it can better quantify the warming effect due to gravity, thus pushing climate science a little further down the experimental path from which it it has so sadly strayed.

  26. tallbloke says:

    Hi Stephen:
    We are going to use a closed container at various pressures to sequentially mimic the atmosphere at various altitudes. My budget doesn’t run to conical containers 10km high. 😉

  27. Lucy Skywalker says: “I may be just sheer stupid, but isn’t this work just verification of the Gas Laws, you know, PV=RT? I do not see this as the same issue as Nikolov and Zeller are dealing with…”

    Lucy, you of all people are not sheer stupid. But on this occasion you are wrong. Graeff’s experiment offers just a different way of tackling the same question: what is the temperature profile in a vertical column of fluid subject to gravity? He has chosen the interesting and demanding test of measuring the effect in a column of fluid that is less than a metre high. Measuring a drop in temperature over such a tiny distance is extremely difficult and all credit to him for his persistence in attempting to do so. His experimental results, like Konrad’s are positive.

    TB is now simply following in the time honoured scientific tradition of replicating earlier experiments.

    P.S. Re. your further comment on the Hilsch Vortex Tube, there is nothing mysterious about this. The energy input that cools the fluid comes from the pump that forces it through the vortex. It is just an example of a very, very, very inefficient method of refrigeration! Yes, it is a fascinating curiosity, but of no relevence to the N&Z atmosphere proposition and does not in any way challenge the Laws of Thermodynamics.

  28. tallbloke says:

    “TB is an excellent experimentalist”

    TB is benefiting greatly from the excellent suggestions and comments both here and in private email from interested parties. This is a group effort.

  29. oldbrew says:

    Wayne says:
    “I wonder if there is a gas that changes color with different temperature that is in the right range?”

    How about nitrogen dioxide? Used here in experiments re the Le Chatelier principle.
    http://www.ausetute.com.au/lechatsp.html

    See (b) Changes in Pressure of Gaseous Equilibrium Systems (in the link)

  30. Stephen Wilde says:

    “My budget doesn’t run to conical containers 10km high.”

    Just as well, you’d have to invert it too !!

  31. tchannon says:

    So far I have no idea what is being proposed other than vaguely a point heat source in a container and something about heat flow from that. This doesn’t seem reasonable as worthy of an experiment, so what else?

    What actually is going to be measured?

    [Reply] Hi Tim. Temperature inside and outside the pressure vessel, the pressure inside the pressure vessel, and the input to the heating element. If it doesn’t seem worthy of an experiment to you, it’s probably because you understand the thermally dynamic application of the ideal gas law already. There are plenty who don’t, apparently. – TB

  32. tallbloke says:

    There’s now a post up from Lucy Skywalker taking a closer look at Roderich Graeff’s experimental results:
    https://tallbloke.wordpress.com/2012/06/11/lucy-skywalker-graeffs-experiments-and-the-second-law-of-thermodynamics/

    Please head over there for further discussion of Graeff, gravity column methodology and centrifuge equivalence so we can keep this thread on track with fleshing out the experimental design of the related but different method we have chosen here.

    Thanks

  33. stephen wilde says: “Does the proposed containment vessel adequately mimic the atmospheric radial increase in atmospheric volume with height and the 3D nature of the radial increase?”

    to which TB replied: “…My budget doesn’t run to conical containers 10km high”.

    Stephen, I assume you were winding TB up and a very good joke it was too. Much appreciated.

    But were you serious about the impact of the 3D radial increase in an atmospheric column? The radius of the Earth is 13,000km and the height of the troposphere is 16km. So, to a reasonably good approximation, a 1 square metre column at the surface becomes (13000/13016)^2 = 1.0025 square metres at the top of the troposphere.

    Are you seriously suggesting this difference is significant in any way under any circumstances? If so, why?

    Or am I just caught in the trap of an endless joke?!

  34. tallbloke says:

    Thanks David for putting this in perspective. It’s a quite frequently mentioned idea which not enough people quantify for themselves. It crops up in discussions of back radiation (bigger surface area of spherical surface at the effective height of emission than planet surface), insolation (expanded atmosphere receives more light from Sun) etc.

    I think we can safely disregard this and concentrate on attempting to adequately emulate the atmospheric situation in our small pressure chamber. What I really want from this thread is some discussion of optimal positioning for sensors and heater, and possible pitfalls for our ‘atmosphere analogy’ which can be overcome with simple additions to the setup.

  35. Brian H says:

    Edit: “a veritable piece of German precision engineering” means actual, not fake. Did you mean “venerable”? = aged and honoured? Oldie but not moldy? 😉

  36. Brian H says:

    Another thought experimental rig: a silvered squat cylinder, with fins/dividers running from the shell partway towards the center, spun up in a vacuum along the central axis on a shaft supported by frictionless bearings, then allowed to rotate freely. Hypothesis is that the gas inside pressed towards the shell is and remains hotter than at the core. Various spin rates and gas concentrations could be used.

  37. Brian H says:

    P.S. to above hypothesis: as the spin rate winds down (imperfect bearings, etc.), the temperature gradient smooths and vanishes.