Talking Thermodynamics

[MJM460]

Hi Willy, I have not seen one actually catch fire, but I don't have the long term history that any farmer would have, so I would accept that it could happen.  What I have seen is when you dig out the pile to feed the cows, sometimes there is a little patch of ash deep inside.  Obviously heat generation had proceeded far enough to dry out the grass and then burn it, but probably more of a slow smoulder deep inside the pile (rather than actual flames) when the temperature is high enough and there is enough oxygen in the air that fills the spaces between the blades of grass in the pile.  Any heat generated is trapped in the pile by the insulating properties of the grass around it, so the temperature tends to rise quite high.  In my observation, nearer the surface, there is enough cooling to the atmospheric air so that it does not burst into flame.  However, if the pile is not well controlled and that heating continued without being controlled by further compaction with the tractor, I presume the pile could burst into flames.  At the very least, the patch of ash might be much more extensive, and the feed value of the whole exercise lost.  I would be interested to hear if anyone knows more about these silage piles.

Well, today I managed to get my boiler test analysis to a point where I have something to report.

As I mentioned the other day, testing a gas fired boiler on a cold day (14 C is considered cold around here anyway), really demonstrates the drop in gas pressure as the the temperature drops in response to the heat absorbed by the evaporating gas.  This means the heat input to the boiler is nowhere near steady, so any calculations involving the heat input from the burner are dealing with a continually changing value.  Another feature of this test is that it is the first one where I have had a stop valve so I can keep the boiler closed in until I am ready to start steam production.  My other steam plants simply have the boiler connected to the engine, and rely on finding the point at which the valve ports are closed, assuming they seal tight, which they generally do not.

At this stage, I have some results on the heat input to the boiler, but I have to do a bit more thinking on whether I can draw any useful conclusions beyond the obvious, that the gas tank needs a little more heat input in theses conditions.  But most readers will know that anyway.  I am wondering whether a copper or brass tray under the boiler and gas tank would transfer enough heat to the tank.  Perhaps it will need to extend under the engine and exhaust separator as well.  Or do I need to add fins or a heating coil?

The first attachment shows the data for heating the boiler from 15 degrees to 110 degrees, which was the temperature at which I chose to start steam production.    The steam tables tell us that the equilibrium pressure at 110 degrees is 143 kPa absolute, about 6 psi gauge.  Not high pressure, but more than enough to run an unloaded engine.

The X axis is the time in minutes from light up.  The Y axis on the left hand side applies to the blue and mauve curves which slope up to the right.  The blue curve is the temperature, rising as you might expect.  The mauve curve is the total heat stored in the copper of the boiler and the water it contains, in kJ, again rising with time as you might expect. 

You might remember that the previous heating curves for my Meths fired burners showed the heating rate increasing a little with time, until it became clear that we were producing steam at some rate, so we were no longer accounting for all the heat.  At the same time, when I calculated heat rate per second, it was essentially constant with time.  However, you can see that this time the heating rate decreases with time.

The second Y axis, the scale on the right hand axis, applies to the orange curve.  The units are J/s.  The rate of heat transfer is decreasing with time, consistent with the heat up curve.  This is what you would expect from that decreasing gas pressure.  The thick line is a smooth curve through the data points.  The thin line is a regression line calculated by the computer.  It is of exponential form as you can see from the equation for the line, printed above the left hand side of the chart.  I am not sure if that is the best form of the ones available, but it seems a pretty good fit. 

In case you have not come across a regression line before, it is basically calculated in a way that minimises the sum of the squares of the difference between the line and the data, as opposed to the average deviation.  It is a common definition of best fit for the data.  Despite the irregularities of the curve through the data, it is likely that accurate data would give a smooth curve.  Having an equation allows calculations to use the equation which removes some of the inconsistencies in the data.

For the cooling part of the test, I plugged the funnel with a tissue to minimise the convection cooling through the centre flue, leaving the outside of the boiler as the only cooling surface.    The advantage of this is that I can use the heat loss at each temperature determined from the cooling test to account for the external losses during heat up, and perhaps add this to the heat lost up the stack.  The second attachment is the cooling curve calculated in Joules/sec.  Now I think about it, I should have presented a time temperature and cooling rate temperature as well.   You can see it shows the irregularities that come up each time I calculate these curves.  At this stage, I put it down to the 'per second' calculation exaggerating the inaccuracies of the data.  I am not totally convinced that the temperature readings are entirely consistent.

On the cooling curve, I have also included the regression line.  This time, the exponential curve is the correct form, as cooling follows Newton's law of cooling.  This means I can use the regression line formula to account for the heat loss from the outside surface area, but I am still working on how far I can go with that.

Next step is to calculate how much steam was being generated.  However, this is also dependant on the burner heat release rate, so I may have to do some more experiments with more even gas temperature to draw any useful conclusions.

So that is where I am up to at present, I will try and do a bit more tomorrow.

Thanks for looking in.

MJM460

[Steamer5]

Hi MJM,
 Its nothing unusual for hay stacks to catch fire by self generated heat. The usual reason is that the hay was bailed & stacked still partially green & hence with to much moisture content.
Silage on the other hand isn't likely to burn....well I haven't heard of it, but then given that I'm a townie that's not surprising!....silage is cut green & stacked, sometimes allowed to dry a bit & then compressed to get as much air out as possible, then covered with plastic & sealed up to keep the air out until required

Cheers Kerrin
 

[MJM460]

Hi Kerrin, I think that defines the difference between silage and hay.  We need a little chemistry, to understand what is happening in silage making.  I suspect there is some biological action/fermentation of green grass involved.  Such reactions result in some heat release, and with the insulating effect of the surrounding pile, the temperature at the middle can get quite high.  Probably just encourages the fermentation if the temperature rise is kept in control by compacting the pile to minimise the available oxygen.  However, if the temperature is allowed to rise too high, there is combustion, and the silage is spoiled.  With limited available oxygen, the amount burned is limited, but it is all loss as far as the farmer is controlled.  So an interesting diversion into what thermodynamics tells us about Willy's observation of his pile of grass clippings.  Chemical engineering thermodynamics even involves explaining the amount of heat produced by that fermentation reaction, but too far out of my field for me to present a complete understanding.

I tried different ways of presenting that cooling data, and it produces pretty graphs, but they don't really tell us anything more that is useful.  They do allow calculating and including the heat loss during heat up, so all except the stack loss is accounted for.  However to proceed to the next step and calculate the boiler steam production capacity or efficiency and air fuel ratio, we need to know the burner heat release.  In this test, the decreasing gas pressure means that the heat release varies over a large range, and information based on averages is too far from reality to tell us very much.  I need to light up the boiler again, and shut off the gas much earlier so that the pressure change is less significant.  I might even see if I can slip a bit of copper sheet under the boiler and gas tank to conduct a bit more heat into the gas bottle to help evaporate that fuel. 

A short post tonight as not too much progress to report.  I will look in each day to continue the discussion on any questions, but will be a bit quiet on the boiler testing until I get some more results.

Thanks for looking in,

MJM460

[steam guy willy]

Hi MJM ,  so ,a new word for my vocabulary ..Regression !! ... the curve for the cooling is quite dromederious !! I think you have your own species down under or were they all imported ??  I like the new word but it sounds a bit akward like when Scotland decided to become devolutionied !! I was looking at my water butts on the allotment and they were different colours ...i tested the temperatures with my hand and they seemed slightly different ,so, do the different colours of the rainbow absorb different amounts of heat and are these  equal to the order in the rainbow ??? perhaps i could to an experiment and discover infra red !!!!
Willy

[MJM460]

Hi Willy, I had to read that three times to work out what you were talking about.  You are quite right about the curve of the data now you mention it.  At this stage I assume it is due to some inaccuracies in my measurements as I would expect it to be a smoother curve.  But perhaps there is something about steam bubbles collapsing, I just don't know.  However, it will be interesting to see if it is repeated in further testing, or varies, and what the data looks like if I combine all the data. 

By the way, we don't have any native dromedaries, they were all imported as transport a long time ago, though many are now wild, whether they escaped or were turned loose.  I understand that due to the isolation, they have remained disease free and may even be an ancient bloodline so I have heard that we now export them back to the original source.

Regression analysis is a statistical technique used to identify relationships between data, or trends.  You might learn about linear regression in a basic maths course, well not really basic of course, which attempts to fit a straight line to data, but most spreadsheets offer more advanced analysis and can determine an equation of a line of best fit using not only linear, but polynomial, logarithmic, exponential, power or moving average.  Some call it a regression line, some call it a trend line.  I selected exponential because that is the form of the equation for Newton's law of cooling, so likely to give a good fit.  But the spreadsheet does all the work, I just took the option to show the line and the equation.  I would not like to have to do it manually.

The water butt question gets to the basics of heat transfer, involving both convection and radiation.  In fact, if the bottom is in contact with the ground, it involves convection as well.  I assume the water butt cools overnight.  In the morning when the sun gets up, the air starts warming and the water butt will receive heat from the air by convection.  It will also receive some heat by radiation, especially if it has an easterly aspect and receives direct solar radiation.

We have looked before at the issues of touching different surfaces to judge temperature, but with different water butts, presumably all the same material, just different paint colours, you might expect to tell the difference with some reliability.  So let's assume the differences you observed are real.  Was there any condensation on the outside?  If so, wind would tend to cause evaporative cooling so one exposed to wind might feel cooler than one more sheltered.

But I think you asked about the different colours, and whether that changes the radiant heat absorbed.  Radiation is a more complex issue than conduction and convection.  Appearance does not necessarily tell you much about absorptivity.  However, radiation falling on a surface is either absorbed, reflected or transmitted.  In this case we can ignore transmission, unless you have a glass water butt.  So partly reflected and partly absorbed.

The colour we see is dependent on the reflected light so the red one reflects more red, while the blue one reflects more blue.  In each case, the rest of the energy, not reflected, is absorbed, so contributes to heating.  The surface then emits energy by radiation to everything around.  The temperature is then the result of all the gains compared with all the heat losses.

The distribution of energy depends on the temperature of the body.  If we look at the Suns energy, peak of energy distribution is in the visible region, whereas for your water butt at less than 300 degrees K the peak is about the middle of the infrared region, which we feel as heat, but can't see.  So the paint colours perhaps alter the amount of reflection in each colour or wavelength, the energy from the sun in the red end of the rainbow does tend to be higher than the amount in the blue end, so the red butt might get hotter than the blue one, (in the same order as the colours of the rainbow) where as at similar temperatures they would both (all) radiate a similar amount.  Or does the red one feel cooler, as it reflects more red which contains more of the Suns energy? It all depends on how much energy is left after the reflection.  Very confusing.  Sticking my neck out there, I don't know which one you found felt warmer.  Black might get hottest, as it reflects least and white reflects all colours so possibly coolest.

However, there is absorption in the atmosphere which affects the energy distribution of the light that reaches the surface.  So a complete study quickly gets complex. 

When you mention infra red, do you mean an infrared non-contact thermometer?  It might give you readings to compare with your hand sensor method.  It would make an interesting comparison.  And you could hold a thermocouple against each surface with a pad of insulating material, or just measure the water temperature inside the container.

In summary, the colours probably affect the reflected proportion of the energy, hence changing the remaining energy to be absorbed.  However, variations in cloud cover, which I omitted to mention, wind exposure and the direction the container is facing are also important, making the outcome difficult to predict.  I would be interested to know what you actually absorbed, and to see whether those observations indicate which mechanism is most important in this case.

Thanks for following along,

MJM460

[steam guy willy]

Hi MJM , Thanks for more interesting info and there is so much more to this than is at first imagined. The reason for the question was to find out if different colours are used in industry to make thermal plants more efficient/stable etc... To get any meaningful data with different colours would need so many different variables to be standardised. The comment about infra red was a sort of joke as to how it was discovered/invented !!!!!

[MJM460]

Hi Willy, I don't know whether there have been internet problems or whether everyone is out enjoying the sunshine.  I have not before seen so many hours with no posts on this forum.  Just arrived home from babysitting, to find your post, nearly 11 hours after the previous post.

You can see why most introductions to heat transfer start by mentioning conduction, convection and radiation, but then start talking about conduction.  Some get to basic considerations in convection, but they rarely get to radiation.

I want to scan the graph of radiated energy energy in each wavelength for bodies of various temperature to illustrate some of the points.  No chance to do this while I was out tonight, so I will do it tomorrow as I think it will help convey some of what happens and make it easier to understand.

We certainly feel here that a black car heats up more than a white one, and I must admit to never having bought a black one.  But I am not sure whether there is much difference between the other colours that I have had.  But what is clear is that the transmitted radiation, that which comes in through the glass does very effectively heat up the inside.  And now we have tinted windows, good tinting really keeps a car from getting nearly so hot.  Before they were allowed, I have burned my hands on the steering wheel, and an uncle came back to his car which was parked in the sun to find the upholstery smouldering and the car full of smoke.  So there is a lot of energy in the Suns short wavelength visible radiation which passes through the clear glass.  The longer wavelength radiation from the car interior does not pass back through the glass, so is trapped inside.

Solar water heater collectors are usually painted black, and I have often wondered if it is just standard paint or something special, but I have never seen a white one.

I will scan that picture tomorrow and the reason for some of these observations will become clear.

Thanks for following along,

MJM460

[MJM460]

Radiation Heat Transfer

A little post to add some detail to what I was saying about radiation heat transfer.  A good summary of the subject is provided by the attached data.  I have copied a few pictures from my text book for illustrative and study purposes, but the words are my own.  So please let me know if I have something wrong or just not made it sufficiently clear.

The first picture illustrates the spectrum of electromagnetic radiation.  You can see from the notes on the right hand side of the line that visible light, infra red, X-Ray's and so on are all just variations in wavelength of the same basic phenomena.  The scale is micrometers or thousandths of a millimetre or 10^-6 meters.  Keep this in mind when you look at the second picture.  The notes on the left side of the scale say something about the origins of that radiation if you are interested.

The second picture shows the total energy emitted by bodies at various temperatures.  The temperatures are absolute temperatures, so in K.  Zero degrees C is 283 K.  The horizontal scale is in metres, starting at 10^-9.  So 10^-6 is one micrometer for comparison with the first picture.  The vertical scale is the amount of energy emitted at the particular wavelength, and you can see that both the amount of energy and the wavelength with the most energy vary with the temperature of the body.  And the area under the curve is the total amount of energy from the body at the particular temperature.  The sun's radiation curve is off the top of the drawing, effectively about 5700 K, but of course, what we receive is modified by reflection and absorption by the atmosphere.

All the data in this graph is for a black body, defined as a perfect emitter.  Real surfaces are not perfect emitters so the third picture shows the emissivity of some real surfaces when quite hot.  Unfortunately they did not include brass or brass which we might be silver soldering.  But the emissivity, a number between 0 and 1, is the fraction of the energy emitted by a black body at that temperature and wavelength.  You can see the actual radiant energy from a surface varies in quite an irregular manner with wavelength.  But the biggest factor is still the absolute temperature.

When that radiant heat arrives at a surface, (like the sun on Willy's water butts, or on the earths atmosphere, clouds and all) some is absorbed, some is reflected, and for a transparent or translucent body, some is transmitted through.  Emissivity, absorptivity, and transmissivity are all expressed as fractions between 0 and 1, and the three must add up to 1.  Obviously for timber and metals, in fact anything that is not the slightest bit translucent, the transmissivity is zero, and all the incident energy is either absorbed or reflected.  This brings us to the fourth picture.

The fourth picture shows what happens at some opaque surfaces.  All the incident energy is either absorbed or reflected, so both can be shown on the one graph by simply reversing the scale direction.  So one scale on the left, and the other on the right hand side.  You can see it includes both white and black paint.  Black paint, as you might expect, absorbs all wavelengths equally, and nearly completely.  White paint is the more interesting one.  You can see it reflects most of the light in the visible range, then a second peak at 5 microns, right in the range of those molecular motions indicated on the first picture.  The sloping top  of the main peak with wavelength is probably showing the different reflectivity for the different colours that make up white light.

Now those are the basic characteristics or radiation from a surface, however to understand what happens when an object receives radiant heat from the sun we need a little more data from atmospheric temperature, and what happens when radiant heat falls on a surface.

  The next picture (fifth) is a table of emissivity at ambient temperature and absorptivity to solar radiation.  If we look at the difference between a light coloured surface and a dark coloured surface, you see that light, dark and even black surfaces at atmospheric temperature all emit radiation about equally if they are at the same temperature.  Remember these figures are the fraction of energy emitted compared with an ideal black body.  But the three surfaces are significantly different in their absorptivity to solar energy.  Assuming they are all in the sun at the same angle, and receiving about the same amount of sunlight, the light surface absorbs only 10 - 35% of the energy it receives, the remainder is reflected, so it appears bright from all that reflected light we see with our eyes, while the black surface absorbs most of the incident energy.  It appears dark, as so little light is reflected.  And other dark surfaces are in between.  So the white surface, which absorbs so little of the available sun's energy, would probably feel coolest, while the black surface, which absorbs most of the available energy, should be warmest.  Whew!  That is the same conclusion as I suggested two days ago.  I hope that is also supported by your observations.

The final attachment is for me the most problematic.  It is a table of the transmissivity of glass, with columns for clear glass and different tints.  The table should have an extra line for wavelengths greater than 2000 nanometers, (2 micrometers) listing the transmissivity at some very small number, close enough to zero for these wavelengths.  This seems to me to be critical to the conclusion reached in the problem used to illustrate the use of this table, but only acknowledged in the assumptions made in solving the problem. 

When applying this data to the greenhouse, or car parked in sunlight, we note first that radiation from the surfaces in the greenhouse or the car at a temperature of perhaps 25 - 60 degrees C, 298 to 333 K, is just about entirely in the range greater than 5 micrometers wavelength, while the energy in sunlight has a very significant fraction with wavelength less than 2 micrometers.  So glass lets a high proportion of the sun's energy through, around 88% of visible light and still  67% in the infrared region.  However, as the internal surfaces radiate back, the radiation is long wavelength, so almost none of this energy is transmitted so it is all retained in the car or greenhouse by the glass.

The effect of tinting the glass is that the transparency to the sun's energy is reduced, so less heating overall.  But look at that blue-green tint.  It still lets through 75% of the visible light, but only 22% of the infra red.  So with minimal effect on visibility, only 22% of the infrared is allowed in.  Even allowing for a smaller proportion of the Suns energy being in the infrared region, it must significantly reduce the heat accumulation.  However, whether it is that, or just the overall lower transmissibility of the tinted glass, it certainly works in the high sunlight intensities we experience in this country.  I don't know if the tint on my car is similar to one of those listed or not, but it works.

Unless you want to try analysing the composition or distance of distant stars, or construct an infrared thermometer, you probably don't need to take the maths of radiant heat much further to understand the radiation heat transfer you experience.

I think that all I have to add now is those attachments in the correct order.  I think they are small enough to include all, but if it does not work, I will delete some, and include them in a follow up post.

Thanks for following along,

MJM460

[steam guy willy]

HI MJM, Thanks for all that info...quite strange that black paint and water share the same values...quite a lot to read, learn and inwardly digest !! I am very busy at the moment so not much time to ruminate about thermodynamics !!! However questions are still coming out of the grey matter !!!

[MJM460]

Hi Willy, sorry to be AWOL yesterday.  I am in a house full of visitors so not much time for posting for the next couple of days.  Think of it as time to digest that information on radiation heat transfer, or to think up more questions.

There are many strange, meaning non-intuitive things about radiation heat transfer.  No doubt they make more sense to molecular physisists, but we are only trying to understand model engines.

There is one additional point that I meant to include.  Heat transfer by radiation is still from hot to cold, but unlike conduction and convection, where the relevant temperature difference factor is just the difference in temperature, so T2 - T1, in radiation transfer it is the difference in T⁴ where T is the absolute temperature.  As 0C is 273K, for most of our practical purposes, T⁴ is quite a large number.

Thanks for looking in,

MJM460

[steam guy willy]

Hi MJM , no worries about missing days !! I am also quite busy.... I have got some acrylic paints to do a simple experiment with the suns heating effects, and just need some suitable containers that this paint will adhere to so we will see what happens !!
Willy

[derekwarner]

"or car parked in sunlight"

MJM & Willy....I quite liked the 2017 VW Golf Alltrack.......in deep navy blue, and the maroon so compared the internal temperature of both vehicles as compared to a white vehicle of identical build

The 3 vehicles were parked in the same alignment at the local VW distributor

I used my digital lazer pyrometer & checked the dash temperature, the front seat material and the back seat temperature at 12:30 PM......lunch time

According to the B of M, the external ambient in Wollongong was listed as 32 degrees C

The VW sales person was most objectionable in his manner in me querying the identical vehicle temperatures

In round figures, the navy blue  and maroon painted vehicles were 5.5 degrees C warmer than the white vehicle......a black VW Tigwan parked next to the trio of Golfs was 7 degrees C warmer  than the white Golf..........

Living in  Australia, I have without exception driven white vehicles............

Derek

 

[MJM460]

Hi Willy, I think we all understand busy.  But your questions are always welcome when you get back to thinking about those things.

Hi Derek, I didn't have the infra red temperature instrument last time I went to a car yard (I try and avoid them), so I find your experiment particularly interesting.  Thank you for posting that.  You mentioned the ambient temperature of 32 (I presume a clear sky) but did not mention the internal temperatures you measured. 

I have had two white cars and a couple of darker ones over the years.  White cars have always been said to be cooler, but you don't often get a chance to compare them in really comparable conditions.   Yours are the first really comparable measurements I have seen.   All colours heat up considerably in the sun, so I would also be interested in how much they heated above ambient as well as the differences between different colours.

The relatively small differences you recorded are interesting, in that the colour is making a difference in the expected direction, but not as much as common impressions would suggest. 

However, if you look back at that graph of the distribution of radiation energy with wavelength, you can see that, while there seems to be a difference in reflectivity with wavelength, (or colour in the visible range), most of the radiant energy is outside the visible range, so the small but significant colour difference makes sense for the amount of heat in the visible range reflected from the panels as a proportion of all the radiant heat arriving at the car.

The internal heat accumulation however is more influenced by the combination of the transmissibility of glass with different frequencies, and even more importantly, the difference in the level of energy radiated from the upholstery at perhaps 330 K (57C) and that radiated by the sun at  around 6000 K, even allowing for reflection and absorption in the atmosphere.  Remember, the net heat transfer is due to the difference in T⁴.   And of course the temperature reached is moderated by convection cooling by the atmospheric air.  Window tinting, and those shiny reflector blankets we prop up inside the windscreen do make a real difference, much more than 5 degrees.  So I think I would buy the colour I liked, along with some reflective blankets for the windscreen.

Thanks for following along,

MJM460

[steam guy willy]

Hi MJM and Derek....  so ,i have done the coffee shop ...next stop  car park !!!! I was eating a square heated piece of quiche in a cafe and  was wondering if The cooling of a fixed mass of metal with 3,4,5,6,7,8,corners would cool down quicker than a round slab ?? thinking about squarish cylinder heads on Motorbikes ???
Willy...........

[derekwarner]

MJM.....I saw no real point in confirming the individual internal temperatures....only advising that it was a warm summer day in Wollongong and so the individual differences in temperature relative to the external body paint colour

The important points were same vehicle alignment to the road & sun....same vehicle build = same front windscreen angle to the morning sunlight and lunch time overhead sunlight the internals & identical seat material [black synthetic weave material]...also with the Golf build = same floor mats, same door trim, all vehicles were with their doors & windows closed for the morningsame ....same ...same etc

Derek