Author Topic: Talking Thermodynamics  (Read 108824 times)

Offline paul gough

  • Full Member
  • ****
  • Posts: 392
  • Tropical Queensland, Australia
Re: Talking Thermodynamics
« Reply #735 on: February 24, 2018, 12:05:42 PM »
Hi MJM, It's been excellent weather for frogs, so been out in the field for a few days. Very good to read the review of the work so far, and the projected exploration going forward. Helps to keep the objective in mind. Was interested to read your comment on 'high temp.' insulation being less effective, so will have to carry out a few comparative tests between the ceramic fibre/kaowool sheet and cork of similar thickness. Can you inform me what the max temp. thin cork sheet, say 2 or 3mm might take before becoming damaged. Found one reference giving 140 C. as the high end but don't know if this is the maximum. Regards, Paul.

Offline MJM460

  • Full Member
  • ****
  • Posts: 819
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #736 on: February 25, 2018, 10:36:46 AM »
Hi Paul, I hope you were able to find some interesting frogs.  Do you count them, or try to find new varieties?

I find with a spreadsheet, that once you get out to around 200 rows or columns it is easy to loose your way.  Perhaps a bit before that!  It is often necessary to use a few columns or rows to just copy key earlier ones, so you keep the ones still being used in new calculations close to the viewing screen.  In addition, a little paragraph to summarise where I am up to helps keep me on direction, a bit like a road map.

I don't have any data for the limits for cork, but basically it is a wood product, it is flammable and so I would not put it where it is exposed to the flames.  In fact, that manifold insulation material that I used is barely keeping the entire outside surface below 140C, though certainly most of the surface is below.  Similarly fibreglass has a melting point that makes me cautious about using it in flame contact.  Possibly there are binders that don't help.  There is a big variation in the conductivity of various materials, but generally structurally strong materials have higher conductivity than typical insulation materials.  For example, cork 0.04 W/m^2.K,  similar to other good insulation materials, compared with magnesite (50% MgO) 2.68,  brick, about 0.4, concrete about 0.13.

When I find a material that I am unsure of, I generally test its flammability with a Meths burner, then with the propane torch I use for silver soldering.  If it catches alight, or just starts smouldering, I generally class it as outside/low temperature insulation only.  Probably needs a longer exposure to be very sure, but there is no need to persist with something doubtful in the quantities we need.

Ceramic fibre is used internally in industrial furnaces, especially where the furnace is to be factory fabricated, as a whole or in pieces for field assembly, it is light weight for transport and lifting on site, and does not lead to big distortions in transport.  However, the fibres suffer from fatigue in high gas velocity areas, so tend to need periodic replacement if subject to high gas velocities.  I would think this would take a long time in a model, so with a suitable fixing system it could be quite suitable.  With the insulation on the inside, the outside temperature can possibly be kept low enough for a suitable paint system, as well as retaining a scale profile.

Forced convection turns out to be a bit simpler than natural convection as the velocity is known, and there are exact solutions to the calculation of the velocity profile, and from this, there are methods to calculate a temperature profile.  Much of my text book is devoted to examination of the difference between exact solutions and assumed polynomial velocity distributions.  That and the difference between constant wall temperature an constant heat flux at the wall.  Reality is probably somewhere between these, yet the four solutions differ by less than 20%, so of the real solution is between the values it is probably within 10% of the average of the four.  Not only does that seem adequately close for our purposes, but at the end of the day, most of the examples around heat loss from surfaces at moderate temperature seem to give a heat transfer coefficient of around 3 to 5 W/m^2.K, so I might try using four, and see how important the heat loss is compared with heat absorbed by the steam and in the stack gas.  This will indicate whether I more insulation would be worthwhile, or perhaps the wall loss could be ignored.

With a stationary boiler, the heat transfer from the walls is by natural convection.  More complicated,as the heating of the air near the wall gives rise to the density changes which cause buoyancy effects to drive the air flow, and to viscosity changes which change the flow behaviour of the air.  Different answers for vertical, sloping or horizontal plates.  I am still trying to understand the implication of the equations, and I am looking closely at the range of answers to comparable worked problems, such as the cooling of an insulated oven.  Again, I may be able to find a suitable order of magnitude answer, to compare with the heat flows in the boiler, and thus its importance.  I think the cooling test with the stack plugged might also yield a reasonable estimate of the losses at operating temperatures.  Might be more useful than extensive complex calculations. 

Well, carpets tomorrow.  House is obviously chaos, so daughter and family dropped in for dinner.  Just the day we needed visitors.  But my wife said she had watched my mother cope with such things many times so knew what to do, and did quite well.  What a wonderful woman.  I did pitch in and peel extra vegetables and lift the heavy dishes in and out of the oven, but mostly followed instructions.  I am not much of a cook.

Might be quiet tomorrow, but I am sure I will check in to keep up with progress on all the wonderful builds.

Thanks for looking in,

MJM460

The more I learn, the more I find that I still have to learn!

Offline steam guy willy

  • Full Member
  • ****
  • Posts: 2090
Re: Talking Thermodynamics
« Reply #737 on: February 27, 2018, 01:44:04 AM »
Hi MJM  still following along saw some interesting things in my books and on the web  Farienhiet scales and  stem conservation and heat properties of combustible materials ....Also what is 'wood alcohol'....
« Last Edit: February 27, 2018, 02:14:27 AM by steam guy willy »

Offline MJM460

  • Full Member
  • ****
  • Posts: 819
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #738 on: February 27, 2018, 11:30:58 AM »
Hi Willy, more interesting historical information from your archives.  I suppose it could have been decided that the human body temperature was 100 and adjust the water boiling point accordingly (at about 213,) but water boiling at atmospheric pressure is probably a better reference point.  Did the article about that locomotive explain the purpose of the water tank above the boiler, other than holding water, of course?

Hydrocarbon chains with an -OH group are called alcohols, but not rings containing the same group, which are called phenols.  Methyl alcohol, also called methanol has 1 carbon plus 3 hydrogens plus that -OH.  Compare this with methane with one carbon plus 4 hydrogens.  Unfortunately methyl alcohol is quite poisonous, and is added to ethyl alcohol to make it undrinkable in methylated spirits, bit good antifreeze.

Ethyl alcohol is the one in drinks, it has two carbons, 5 hydrogens plus that OH.  Some would say it is also poisonous!  Compare this with ethane.

Three carbons gives propanol in the similar pattern.

Methanol was called wood alcohol in the early days of humanity, as it was generally obtained from a destructive distillation of wood.  These days it is synthesised in a methanol plant which uses natural gas, methane, which is burned with insufficient pure oxygen so carbon monoxide production is maximised, then a catalyst is used to produce the methanol, used as a chemical feedstock.  I can't clearly remember all the reactions, but I believe one of the New Zealand forum members worked in methanol plants at one stage.  He may be able to amplify the details.  I always pull up short at the concept of burning natural gas in pure oxygen, no matter how much short of the theoretical oxygen required; intuitively, that seems unwise.  Definitely something not to try at home!  But it illustrates the point that even with an accelerant, for which pure oxygen is about as good as you can get, if you don't have enough oxygen, you will get incomplete combustion.

Well, I decided I had done enough reading about convection and that it was time to try the calculations.  Convection calculations involve a lot of empirical correlations, and the amount of work to do the experimental work is reduced by correlating using some of those dimensionless groups of variables that I mentioned early in this thread.  Most of us have heard of Reynolds number, if you are interested in boats you might have heard of Froude number (or at least used it in terms of a critical speed in knots equal to square root of the waterline length in feet, not exactly dimensionless, but...).  Well convection involves many more.  Numbers like Prandtl number, Grashof number, Rayleigh number, and even a few more obscure properties of air.  Then correlations are found by plotting experimental results in these terms and determining the equations that best fit.  Even the so called "exact solution" that I found, turned out to be an exact solution to equations that were approximations. 

The resulting equations are quite intimidating, I don't think any of you want me to post them.  I don't think I would have tried them without a spreadsheet.  Truly horrible things to input to a calculator.  However, I was quite surprised to not only get an answer to the calculation of the heat loss, but it even seemed quite reasonable.

I used a rough average of the wall temperatures I measured with the infrared thermometer of about  100 C and an ambient of 20 C with the 50 mm burner.  I measured up the outside surface dimensions of my boiler casing, and looked up all the air properties at the appropriate temperatures, and got an answer of 90 Watts.  This is about 15% of the heat input of that small burner, so it's a significant loss, but intuitively, (that dangerous word again) it seems to be about the right magnitude.  Most loss was from the vertical walls and ends, and the inclined sections between the walls and the roof, and only 10 of the 90 Watts from the flat roof.

The heat transfer coefficients came out as part of the calculations, showing some interesting aspects.  The ends are taller than the side walls and have a lower average heat transfer coefficient.  I think this is because the air heats as it rises, so the temperature difference, and hence contribution to heat transfer of the area located at the top of the walls is less.  6.5 for the walls compared with 6.15 for the taller ends.  The inclined surfaces had a higher coefficient, 8.31 which reflects the much smaller height, but I think the calculation assumes cool air at the bottom, whereas those inclined surfaces will have had warm air from the top of the sides.  As the formula are the same for both, I possibly should have just treated them as one taller surface.  This would give a coefficient similar or a bit lower than the ends, perhaps 6.0.   The flat top had an even higher coefficient, 10.2, but again the cooler incoming air is actually already warmed by the sides and inclined walls.  The small contribution to the heat loss despite the higher coefficient is due to the smaller area of the roof.

Two important comments in the the book, first, the accuracy is not likely to be better than +/-25%, secondly, there will also likely be radiation which may be a significant magnitude.  It just said most analytical approaches assume only one mode.  I suppose I could try to estimate a radiation transfer rate and add the two.  But not right now.  It is clear that the heat loss is probably enough to make additional insulation worthwhile.  If I can get a lower surface temperature, I can change that figure, and also the ambient in the spreadsheet, and even an iPad will redo the whole calculation in a flash.  The difference will be more instructive than the actual figure.

I think the next step is to try that cooling experiment, with the stack plugged, and see if that gives anything similar.  Have been putting the house back together.  The carpet installation looks good, and we are pleased with the colour choice, but putting everything back on its place will take priority over more testing.  Then there is a wooden boat festival next week.  What did Chris say about too many hobbies?

I will also try that forced convection calculation for Pauls locomotive.  A similar problem to the inclined walls on my furnace in that the sides of the firebox of the locomotive are in the air stream already warmed by the boiler in front.  But the boiler has a different wall temperature to that of the firebox.  I will think about that for a bit longer but feeling encouraged.

I hope that is all of interest,

Thanks for looking in,

MJM460

The more I learn, the more I find that I still have to learn!

Offline paul gough

  • Full Member
  • ****
  • Posts: 392
  • Tropical Queensland, Australia
Re: Talking Thermodynamics
« Reply #739 on: February 27, 2018, 01:16:47 PM »
Hi MJM, Very interested to read most of the loss was from the verticals , sides and ends, and little from the top. This is pretty much describes my little loco firebox as the top is the underside of the boiler barrel and not a loss. Your figure of 15% loss is also noted, notwithstanding the caveat of +/- 25%. 15% seems to be a reasonable proportion and is probably worth chasing any means to reduce it. I find these explorations fascinating and the findings instructive, thank you for the mental toil. I look forward to more revelations!

I'm just about to depart for some more 'frogging', locally. 10pm to 2am has proved to be a pretty good time range around here, but after a week of it my old carcass is starting to flag. Mostly we just note the species found and where but anything pertinent is noted also. Rarely anything out of the usual, but even just raw numbers help in indicating trends in depopulation or disappearance, all too frequently the only circumstances that prevail. We did have a little highlight the other night when we came across a scrub python with a belly bulge, (probably someones chook), the snake was approximately 4 1/2 metres long. Regards, Paul Gough.

Offline steam guy willy

  • Full Member
  • ****
  • Posts: 2090
Re: Talking Thermodynamics
« Reply #740 on: February 27, 2018, 10:58:50 PM »
Hi MJM, The loco is using the spare steam as a feed water heater during the stationary and less demanding cycles  to heat the water that is held in the large tank. this is then pumped into the boiler using axle pumps rather than using the injectors to push in cold water !!..Interesting info on different fuels ...Peat having a really high flame temp .... but i don't understand the other parameters ! Does radiated heat cause its own 'draft' pushing the convection heat away from the boiler ? or how do these interact ?
Willy

Offline MJM460

  • Full Member
  • ****
  • Posts: 819
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #741 on: February 28, 2018, 12:17:19 PM »
Hi Paul, the greater heat loss from the walls of my furnace is mostly due to the greater area.  Remember it encloses the whole boiler.  The horizontal roof is a small portion of the area.  The heat transfer coefficient is actually higher than for the horizontal surfaces, though I suspect it is a high estimate, as the formula assumes it is ambient air coming in at the sides.  In my boiler configuration, it is the warmed air from the sides doing the cooling of the top.  This reduces the effective temperature difference and so reduces the heat transfer rate.

As always, it is important to consider the meaning of a percentage.  The calculated heat loss was 90 watts, that is 15 % of the burner output of 620 watts, more accurately 14.5%.  The +/- 25% is +/- 25% of 90, so +/-22.5W.  So 90 + 22.5 = 112.5 which is 18% of the burner output of 620 watts. Then 90 - 22.5 = 67.5 which is 11%.  So plus or minus 25% of the heat loss means 11 to 18% of the burner output or  +/- 3.5% of the heat output.  So while 25% error looks bad, it is only 3.5% of the heat output.

You boiler is a quite different configuration.  With the boiler on top, or even forming the top of the fire box, obviously no loss there, the heat goes where it is wanted.  Depending on whether the sides of the firebox are inside or outside the frames, they may or may not be subject to that air velocity caused by the motion.  The bottom is the air inlet, with air flowing in, there is no convection out, however you can probably see reflections of the flame on the tracks, so there is a radiant loss.

The front of the firebox is facing the air flow, but located between the frames, behind the cylinders axles etc, above the tracks and below the boiler, so probably minimal air flow, and the air is probably heated by the boiler and the cylinder heat losses.  This heated air is the main flow into the burner which thus receives preheated air, so your locomotive actually uses waste heat to preheat the air, the classic waste heat recovery system!  It may not be very efficient, but every bit helps.

Not sure about the back of the firebox.  Any air flow in this area is preheated by all the sources I have mentioned.  In addition it is probably easier to tuck a bit of insulation on the outside of the box/duct taking gas from the flame to the centre flue.

I know I have mentioned placing insulation inside the firebox.  This is based on assuming the sides of the firebox are visible in a side view of the engine.  However I suspect I do not really have a complete picture of the arrangement.  With the firebox under the boiler, possibly silver soldered to the boiler, then the sides would be an important "fin", conducting absorbed heat to the boiler, and insulation on the outside would be better, if it is hidden between the frames.  I think I need a cross section through the boiler and firebox to better understand the arrangement.  And a side view of the locomotive.

Counting frogs gives an early indication of possible problems.  Without the count, that early warning is missed, so counting them is an important contribution to research.  I am afraid that I removed the hours from midnight through to about 6 am from my clock some time ago.  They no longer exist!

Those pythons can give you quite a start.  Especially if you are from the western district where snakes are all black, brown or tigers and all quite poisonous.  My wife stepped out of our van in the dark one night when it was parked on our daughters front lawn in Katherine, NT, and nearly stood on the front end of one, of which the back end was still hidden in the garden bushes over two meters away.  The scream brought the whole neighbourhood running.  Right in the city!

Hi Willy.  An interesting idea.  I guess the high mounting of the tank provided a bit of extra inlet pressure to the axle pumps, so the water could be heated to a higher temperature before it vapour locked the pump.

The thing about those flame temperatures is that they are all quite close to 2000 deg F.  The small differences are due to the different air fuel ratios required by the different carbon to hydrogen ratios of the fuels.  Twelve kg of carbon requires 32 kg of air for complete combustion, while only 4 kg of hydrogen requires that same 32 kg of oxygen.  And hydrogen releases over 4 times as much heat per kg of fuel in burning to make water compared with carbon burning to make CO2. The carbon to hydrogen ratios only vary a small amount, reflected in the very similar flame temperatures and heating values.  But peat is usually water logged, so I presume the figures are based on dried fuel, without penalty for the heat required for the drying.

Those other columns are the calorific values of the fuels.  As you can see, there is a higher value and a lower value.  The difference is whether the water is condensed from the flue gas, thus releasing the latent heat.  The lower value is applicable to boiler work as it is assumed the water is lost up the stack while still vapour, this carrying away that latent heat.  Again, there is a quite small range of differences between the higher and lower values, reflecting that small difference between the carbon to hydrogen ratios, and hence the amount of water vapour in the flue gas.

In this table, the scientist uses calories as the unit of heat.  Each calory is equal to 4.180 joule.  When I was at high school, that factor was called the mechanical equivalent of heat.  It was experimentally determined by the pioneers, before the clear understanding of the first law of thermodynamics, so there are a few definitions, each giving a very slightly different value.  However, the Joule is defined by reference to the unit of mass and length, so is much more precisely defined, and is the preferred unit for heat measurement.  The calorie was defined as the heat required to raise 1 kg of water by 1 degree C (similar in principal to the definition of the BTU.)

Air is transparent, or nearly so, to radiant heat.  It is essentially unheated by radiation.  Radiated heat from the sun warms the earth which in turn warms the air by convection.  There are small differences in the degree of transparency of different gases, and the transparency also varies slightly with frequency or wavelength of the radiation, leading to the greenhouse effect, but that is a different subject.  For the purposes of our analysis, it is close enough to assume air is transparent to radiant heat.  This means we can calculate the radiant heat loss and the convection loss separately, and simply add the results.  There is no "pushing of the air" or separate draft.

That is a long enough post without my introducing anything new, but very interesting questions.  I hope my answers help with understanding them.

Thanks for looking in,

MJM460
« Last Edit: February 28, 2018, 12:22:46 PM by MJM460 »
The more I learn, the more I find that I still have to learn!

Offline steam guy willy

  • Full Member
  • ****
  • Posts: 2090
Re: Talking Thermodynamics
« Reply #742 on: March 01, 2018, 11:54:21 PM »
 Hi MJM, thanks for the info.... and we are experiencing a severe cold weather event and there is a wind chill factor making it extra cold, I have heard that for every mile an hour of wind your body drops one degree ? is this true and if you are on a motorbike at 100 miles an hour do you get quite cold !! also does this have any effect on steam railway locomotives ? When flame fuels burn the oxygen in the air does the fire then burn the hydrogen content as well and create water ? or is this another silly question ? !!!

Offline MJM460

  • Full Member
  • ****
  • Posts: 819
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #743 on: March 02, 2018, 12:08:02 PM »
Hi Willy, wind chill definitely cools the body faster than still air, but it is not as simple  as a direct linear relationship with speed.  Basically the wind just increases the heat transfer coefficient between your body and the air.  It accelerates your heat loss, but the wind cannot cool you to a lower temperature than the air.  Of course, if you are sweating, or wet from rain or mist, (or if you have been swimming), the wind can evaporate that moisture, and cool you a bit further.  If the air is humid, the lower limit is the dew point.  It is a bit more complex to calculate the limit if the moisture is from rain etc, but the more moisture in the air, the less can evaporate, so less cooling effect.  If the wind is off the desert on a sunny day, and the actual air temperature is above 37 degrees, it can only cool you by evaporating sweat, otherwise it actually heats you.

Essentially the question is very relevant to the convection discussion.  The wind just increases the convection transfer coefficient.  The simplest treatment of convection is to assume it is like conduction, using the air thickness over which the air reaches that wind speed.  Natural convection has a relatively thick boundary layer, so the temperature gradient is not very high.  Wind effectively makes the boundary layer much thinner, so the temperature gradient, and the corresponding heat loss is much higher.  This higher heat loss will lower your skin temperature, and eventually cause hypothermia.  Feeling cold is a warning to do something about it.  Layers of clothes modify that temperature gradient, but the wind making the air boundary layer thinner, then makes the outside temperature of the clothing lower, thus increasing the heat loss. So you need an extra layer to keep your skin temperature comfortable, normally about 34, compared with your core temperature of 37.  But as our Scandinavian colleagues will remind you, there is no such thing as bad weather, just bad clothing!  So wear an extra layer or two, preferably a wind proof one if you are intending to travel at that speed.  Generally that wind chill factor the weather bureau will announce is a temperature which is estimated to produce about the same heat loss as you will experience in the wind, hence it is an estimate of the temperature it will feel like.

Under normal metabolism, the average human produces about 90 watts.  So comfort is about balancing the heat loss with that heat production.  To much loss and you feel cold, to little and you feel hot.

When the fuel molecules have carbon and hydrogen, so all petroleum fuels, methyl, ethyl and propyl alcohols, which all seem topical at the moment, and all coal, then combustion produces carbon dioxide and water.  As you say the oxygen in air, which is in the form of molecules each of which contains two atoms of oxygen, combine with carbon to make CO2, and hydrogen to make water.  Each 2 hydrogen atoms combine with one oxygen atom to make H2O, or water.  So you need four carbons to combine with one O2 molecule.  That is why we have to decide whether the water in the flue gas will condense or not, and so have a higher and a lower calorific value.  Not a silly question. 

If there is sulphur in the fuel, so some hydrocarbon fuels and many coals, sulphur is burned to form sulphur dioxide, not good at all, which is why refineries are normally modified these days to remove sulphur and produce low sulphur fuels.  And so we all need to pay the extra fuel costs for this additional processing.  If the combustion is really hot, you can even get a small amount of nitrogen combining with oxygen to give various oxides of nitrogen.  So cars these days have relatively low compression ratios, it reduces the production of oxides of nitrogen.  However the majority of the energy release comes from the carbon and hydrogen reactions, the others are more important as pollutants than in energy release.  So not a silly question, I hope the answer makes sense.

Sorry about being absent yesterday.  We arrived at the "cottage" and had no power.  No power means no refrigeration, or limited to about two days, which our batteries can supply.  Our solar panel is enough for lighting, but we need a bigger panel to cover the fridge consumption.  So lots of testing and investigation.  I stop short of sticking the meter probes into live 240 v though, so had to call an electrician.   Turned out the power cord had failed!  Looked OK on the minimum load of a multimeter but significant resistance when tested under load.  So a new cord, and all is well.

Thanks for looking in,

MJM460

The more I learn, the more I find that I still have to learn!

Offline MJM460

  • Full Member
  • ****
  • Posts: 819
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #744 on: March 03, 2018, 11:36:56 AM »
At the wooden boat festival today.  What a wonderful display.  Nearly two hundred wooden boats of all ages and designs.  From the magnificent Lady Nelson, a historical square rigger, to little rowing boats and wood strip sailing canoes.  A really good mix of power, sail and even steam.  All beautifully built or restored and presented with varnish gleaming and paint perfect.  An amazing display for a small town far from the big cities.  I will try and put up some pictures next week when I am again at my computer.

A few days back, I had a go at calculating  the convection heat loss from my boiler, the larger one with the completely enclosed firebox and boiler.  The outside of the casing has a layer of manifold gasket material from the local Autobarn as heat resistant insulation.  I basically just looked up the necessary properties of air at the estimated average wall temperature on the outside of the insulation of 100 degrees C based on several infra red thermometer readings of the casing temperature.  Then inserted these values into the equations presented in my heat transfer text book for natural convection from vertical and horizontal surfaces.  I put all the necessary equations is a spreadsheet, so the calculations are easily checked and easily modified if necessary.  Using an ambient temperature of 20 deg C the answer was 90 watts.  You will remember that the calculation only claimed to be +/-25%, which as a percentage of the total burner heat release of 620 Watts, does not seem to bad.

However, more important than the actual figure is the impact of small changes in some of those variables.  For example, I have been asked about the impact of a variation in ambient temperature.  Using a spreadsheet makes answering these "what if?" questions very easy.  At least it is easy so long as you set up the spreadsheet with a specific cell for each of the relevant variables.  Then you use that cell location in any formulae that use the values, rather than inserting the figures directly.  Then, by changing the number in the cell for the variable, the computer, or even iPad instantly recalculates the whole spreadsheet to give you the new answer.  Not for the whole boiler performance, but just for the convection loss contribution to performance.

The first item I wanted to explore was the effect of how I calculated those inclined sections at the top of the walls.  My first calculation treated these separately.  In case that description is a bit confusing, I have attached again an earlier picture of the boiler prior to lighting up for the last set of tests.  Now that means the calculation assumes the air temperature at the bottom of the inclined section was at ambient temperature.  When I thought about it, the inclined section starts at the top of the vertical walls where the air is already warmed by the vertical section.  Perhaps it would be better, seeing that the equations for an inclined wall are the same as those for a vertical wall, to just add the relevant inclined length to the height of the walls and treat them as one higher wall.
Originally I had calculated 30 watts for the vertical parts of the walls and 14.4 watts for the inclined sections, a total of 44.4 watts.  The calculation for just higher walls gave 38.1 watts, and then of course no additional for the inclined.  So the total is now 38.1 which is 6.3 watts less in the total of 90.

This makes sense, as the the section at the top of the higher wall sees warmer air, so has a lower temperature difference between air and the wall, so less heat loss from that section.  Essentially the take away learning is that if we double the wall height, the total heat loss will be higher, but less than double.

Second, I tried varying the ambient temperature.  The initial figure of 90 W was derived for 20 deg ambient.  When I recalculated for 15 degrees, the loss became 97, while for 25 deg it became 83 W.  This just reflects the difference in temperature between the wall and ambient.

The third exploration was to see the effect of an increase in the measured wall temperature, more accurately stated as an estimated average of the wall temperature readings.  The tests of the insulated casing so far have been done with the small burner, which I have previously concluded is too small for this boiler.  The next series of tests we be done with the larger burner.  Regular readers might remember that this burner seems almost too big.  So a mid size burner is on the project list.  However, with the bigger burner, I would expect the flue gas temperature inside the casing will be significantly higher than with the small burner.  This will likely mean higher temperatures measured on the outside of the furnace insulation.  If the temperature becomes 125, the spreadsheet quickly tells me that the heat loss will rise from 90 to 125 watts.  This reflects the 105 deg temperature difference between the wall temperature and ambient compared with 80 deg ambient used previously.  The tests will have to wait a week or two, but it will be interesting to see the result.

That is probably enough on natural convection until I have some new results.  So tomorrow I will try and explore further forced convection.

Thanks for looking in,

MJM460

The more I learn, the more I find that I still have to learn!

Offline steam guy willy

  • Full Member
  • ****
  • Posts: 2090
Re: Talking Thermodynamics
« Reply #745 on: March 03, 2018, 10:51:06 PM »
Hi MJM, Thanks for all these posts explaining everything you do you must have so much Knowledge about thermodynamics tucked away in your grey matter....so a question about model boats ....In my old copies of ME they talk about model yachts as 4th rater or 125th rater and everything in between !! so what does this actually mean ? Also do you know anything about the really cold weather and its effects on the body and bones that can cause discomfort etc
Willy.....
« Last Edit: March 03, 2018, 11:08:56 PM by steam guy willy »

Offline MJM460

  • Full Member
  • ****
  • Posts: 819
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #746 on: March 04, 2018, 10:52:35 AM »
Hi Willy, those two questions are real outliers for thermodynamics, but like many of us I have many hobbies.  Not possible to have too many, but.... 

My collection of model boats magazine goes back to 1964.  Dis-continued the subscription unfortunately sometime in the eighties, but have bought newsstand copies intermittently since when opportunity permits.  Back in the early days, the 10 Rater used to feature prominently among the designs that were reviewed each month, along with Marbleheads and A class.  The Marblehead is similar in size to the 50/500 class in the US.  I always wanted to build one, but other pressures prevailed, though eventually I completed a Starlet design that featured as the Christmas plan.  It still sails well on the very occasional outing. 

In yacht racing, the problem is always to develop a handicap system to make racing as fair as possible between all the different sizes and designs of boats.  The aim was to balance encouragement of design progress and limits to design to minimise unequal advantage.  The rating formulae were a calculation based on various specified measurements, length, beam, draft, and many others, which gave a single number (not necessarily dimensionless) as the result.  This number was called the rating.  Yachts with equal rating were supposed to be able to raced fairly.  I don't know if the same formula was used in all classes, or in the models, or if there were variations.    But I assume each one is based on a similar idea.  And of course classes all add extra requirements, generally in the hope of making racing more equal by prohibiting features seen to provide an unfair advantage.  Similarly, the twelve metre class of Americas Cup fame was a formula based on various measurements, and if the result was twelve, the boat was classed as a twelve metre.  But I am not sure that there is anything about those boats that actually measures 12 metres.  Modern yachting uses many more sophisticated rules, some even based on complex fluid dynamics calculation, yet I don't believe there is any of them that is universally accepted as meaning that all designs of the same rating can start on one line, cover the same course in any weather, and none have a design advantage.  That elusive quest for making the race a pure test of skill, and eliminate the check book race or fancy design variations is not yet over.  Most big races consider the handicap result as the main prize, though first across the line tends to get the publicity.  But many skippers that don't win are quite inclined to blame the handicapper!  It seems that one design is about the best that can be achieved, so equal designs can compete fairly, with different class rules have different approaches to how much latitude is allowed.  A never ending topic around the bar after the race.  Essentially, some weather and wave conditions will favour a light weight design, while other conditions will favour a heavier design.

Effect of cold, almost a medical question, and if you mean extreme cold it might be better to ask some of the Canadian or Scandinavian members about frost bite.  Exposed skin in sufficiently cold weather will freeze, and like fruit in the freezer, the cells are damaged, so quite a severe injury.  I hope your two feet of snow were quickly warmed up, and not left cold for long. 

My closest first hand knowledge is when we took our baby daughter all wrapped up in her pram which I pushed in front of me, outdoor skating in Ontario.  We took her to the doctor when she developed a rash on her cheeks, the only part not totally covered.  He thought it was hilarious that two adults could come into his rooms in a small country town, without any knowledge of frost bite.  We got a lesson on how to protect her, and ourselves, in the surgery.  Fortunately not too serious, though if you know about it, you can still see marks on her now middle aged cheeks.  And of course reading about mountain climbers and those embarking on Antarctic treks, whose fingers and toes seem to suffer a heavy toll.

However, if you mean not quite such extreme cold, look up hypothermia.  Basically your body operates at about 37 deg C and can only tolerate a quite small variation in core temperature.  You know about too hot if you have ever had the flue.  You skin is normally nearer about 34, but you cannot stand much drop in core temperature.  You produce about 90 watts from normal metabolism.  If you loose much more than this, the body cannot compensate and you get cold. 

If you have insufficient clothing, we all know about shivering, but it becomes more dangerous when we get a bit colder, and shivering stops.  All the body processes slow with no good results.  Decision making fails early, so you are less likely to make sensible, potentially life saving decisions if you are too cold.  Someone else has to take charge and get you warmed up.  Preferably by wrapping you in a thermal blanket and sleeping bag so you start warming yourself.  Other heating methods are considered dangerous.  Some medical procedures cool you more than this, but only under intensive, skilled medical supervision.  But even in pleasant water for swimming, you will eventually succumb if you are not wearing an adequate wet suit, and the time you can expect to survive diminishes rapidly with lower water temperatures.  Hikers, skiers, canoeists and sailors all tend to come across the information if they keep up with practicing their sport in the outdoors.

The boat show continued today with a race.  I think I have a picture of the full size version of the model in your picture, and under full sail.  The weather was perfect, and a race with all these classic sailing yachts under full sail was something to see.  The quick and dirty boat building race was a pile of laughs, but admiring all those boats so beautifully presented, occupied us all day.  Beautiful, but I don't think I would want to take on the maintenance, especially of the many that are too big to take home on a trailer and have to be stored afloat.  I assume they have extensive covers.  Pictures as soon as I get back to my computer.

Thanks for looking in,

MJM460
The more I learn, the more I find that I still have to learn!

Offline MJM460

  • Full Member
  • ****
  • Posts: 819
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #747 on: March 05, 2018, 10:48:34 AM »
Forced convection-

Forced convection is when the air (or other fluid) flow is imposed on the system, independent of the heat transfer.  This tends to cause a much higher temperature gradient between an object and the air, giving high heat transfer coefficients, much higher than for natural convection.  We are most interested in air cooling in the present discussion, though the same considerations apply with water cooling or other fluids.  To simplify the language, I will assume air cooling in the following.  I can follow through with the changes necessary if the fluid is water, or other fluids if it is of interest.

Think of Paul's question about the cooling of the sides of the firebox of his miniature locomotive.  The speed of the locomotive gives the free stream velocity relative to the plate.  I will come back to this example.

Fluid mechanics assumes that when there is forced convection past a flat plate, located parallel to the flow, the layer of air next the plate is stationary relative to the plate.  Then the next layer out experiences a drag force which makes it move in the direction of the air flow.  This layer is affected by the next and so on until the air, which is still quite close to the plate, is essentially the same as the flow more distant, or the free stream velocity.  Generally when the flow reaches 99% of the free stream velocity, it is considered the edge of the air affected by the plate.  This part is called the boundary layer.  Fluid mechanics provides some equations for the thickness of the boundary layer.  It does not matter whether the plate moves through still air, or the plate is stationary and the air moves, the equations are the same.  Think of a miniature observer standing on the plate, the observer cannot tell whether it is the air moving, or the plate.  More realistically, if you put your hand out the window (not in traffic mind you) the force of the air on your hand is the same whether it is due to wind with the car stationary, or due to the car moving.  We used to be obliged to indicate our intentions to turn or stop by hand signals which required the window to be open.  Now it is illegal to have any body part out of the window.  Unfortunately, this is a necessary change and none of us want too much information on the statistics that drove the change.

When there is also heat transfer, let's assume for simplicity of language, the plate is hot, and being cooled by the air stream.  The temperature of the air next to the plate is essentially the same as the plate temperature.  The heat transfer increases the temperature of the next layer, thus reducing its density and viscosity, which both are important to the fluid mechanics for velocity of the air close to the plate.  At some distance from the plate, the air temperature is essentially the same as the free stream air so the concept of a thermal boundary layer seems sensible.  For most fluids the fluid reaches free stream temperature a little closer to the plate than the velocity reaches the free stream velocity.  The heat transfer does not affect the basic fluid mechanics equations, but it does affect the fluid properties.  Thus the fluid properties are evaluated at a temperature mid way between the plate temperature and the free stream temperature, and the normal fluid mechanics equations, which are based on our old friends, conservation of mass, conservation of energy and conservation of momentum, are still totally applicable.  But just as in the natural convection case, we have to use solutions to approximate equations, which sounds to me much like we only have an approximate solution.  But it's has apparently been demonstrated by the necessary practical experiments, to be close enough to be quite useful.

This is pretty heavy stuff, so that will be enough for tonight.  Tomorrow I will try and include some diagrams of that velocity field, and perhaps take it a little further.

Thanks for looking in,

MJM460
The more I learn, the more I find that I still have to learn!

Offline paul gough

  • Full Member
  • ****
  • Posts: 392
  • Tropical Queensland, Australia
Re: Talking Thermodynamics
« Reply #748 on: March 05, 2018, 12:51:54 PM »
Hi MJM, your step by step explanation of this "...pretty heavy stuff," is much appreciated as it makes things quite clear. I look forward to more analysis and hopefully some general conclusions can be proffered.

I was daydreaming, again, about boilers and materials, or more specifically thicknesses of same. I remember when an apprentice seeing a copper inner firebox for one of the small loco classes in the workshop that had been 'forgotten' (?), and it struck me how thick it was compared to the steel versions then in use. Now I accept steel is stronger, cheaper, thinner etc. etc. but I wonder if there was any heat transfer advantage over the steel. Yes, the copper is a better heat conductor but it is also a lot thicker, I'm guessing from memory, but maybe two or three times thicker. Also, would there be any advantage in the extra thermal mass of a thicker material. Just wondered if there might be any counter-intuitive advantage in making a particular boiler component thicker than the normal general approach of thinnest possible. I'm thinking of course with respect to a steam locomotive, which can't compete with the efficiencies of stationary set ups, and in particular small model all copper loco boilers. Any thoughts on this?? Regards Paul Gough.

Offline Admiral_dk

  • Full Member
  • ****
  • Posts: 1271
  • S°ften - Denmark
Re: Talking Thermodynamics
« Reply #749 on: March 05, 2018, 09:38:02 PM »
Frostbite - well I heard a radio talk about extreme cold vs. what we got now (a few days ago).
The expert has been working in the Central Antarctica with daytime temperature of -55C and night -70C and he often didn't use gloves if he had to do something that required precision ...!!!
Having students in for tutorial in the deep freezer at the university he works at, is often in T-shirts ..... but being outside in a typical Danish winter at -1C with a high air humidity is extremely cold to him and requires very good clothes  :insane:

So his conclusion is that the relative humidity is the most important factor about how low temperature affect us and how bad.

I'm not sure it made you any wiser - but you asked.