Author Topic: Talking Thermodynamics  (Read 194610 times)

Online derekwarner

  • Full Member
  • ****
  • Posts: 833
  • Wollongong ...... Australia
Re: Talking Thermodynamics
« Reply #270 on: September 13, 2017, 01:30:49 AM »
Morning Paul......without digressing too far from the thread, I have found my $15.00 digital laser pyrometer absolutely invaluable in understanding temperatures in my steam plant...[with simple understanding of actuals]...[without these, all I knew was that things were hot :Mad:]

My boiler steam discharge valve presents as 135 degrees C, however this is an external reading spotted on the body of the valve - from the steam tables, the boiler at 3 Bar  is producing steam at ~~142 degrees C....so everything is relative in comparison]

I can trace the temperature to the lubricator body, the steam regulator, the steam inlet fitting to the engine....the exhaust and all the way back to the de-oiler] - monitoring gas temperature is interesting......including lagging of the gas line

One simple point I can confirm [before & after test] is that insulating the steam tube from the boiler to the engine, provides steam that is ~~3.?? degrees C degrees hotter that in the uninsulated state....[and this is over approx. 220 mm long run of 1/8" OD copper tube]

My interest in this area was that I was producing too much water condensate in the de-oiler and needed to understand what was happening. To this end I have increased the exhaust tubing from a combination of 1/8" and 5/32" to 1/4" x 0.014 full flow K&S brass

[The insulated tubes with the X's were the 1/8" & 5/32" exhaust tubes to and from the de-oiler which are now redundant]


Whilst we are thinking seriously about efficiencies, a $15.00 set of digital scales [the type available were 7kg max]  are essential for confirming the volume/weight of gas burnt......[my gas tank has a published volume of 105gm......I have managed to get 103gm inserted at 26 degrees C]

Apologies MJM for deflecting out of line.... :happyreader: ....Derek
« Last Edit: September 13, 2017, 05:38:54 AM by derekwarner_decoy »
Derek L Warner - Honorary Secretary [Retired]
Illawarra Live Steamers Co-op - Australia
www.ils.org.au

Offline MJM460

  • Full Member
  • ****
  • Posts: 1647
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #271 on: September 13, 2017, 06:01:38 AM »
Hi Paul, thanks for your encouraging words, and for being kind enough not to mention that it was not one of my best efforts.  Sometimes a day does not go quite right, even when building a knowledge base.  I am determined to do better on the engine performance.  But first your thermocouples.  Obtaining these is much easier these days.  The go-to combination in industry is Copper-Constantin for all except quite special temperature ranges.  I believe the junction is electrically welded, but no need to worry.  Nearly all reasonable digital multimeters have one supplied.  The temperature- voltage characteristic is well known, even laid down in international standards.  It is a bit non linear, but that is all built into the multimeter.  Probably less expensive for the whole combination than trying to buy a length of the wire in hobby quantities plus plugs to fit your meter.  I included a picture of one of mine earlier, and Willy included a picture of his which had a nice stainless steel sheathed probe.  I have to experiment to see how the probe goes with my thermowells, but it would be great for stack gas, as well as for Willy's coffee.  I even checked mine at ice point and boiling point to see if it was really correctly calibrated.  That is probably over the top, but all the ones I have tested have come up spot on, give or take the unavoidable experimental error involved.  (Actual atmospheric pressure for boiling point, plus the difficulties of getting a true equilibrium ice/ water mixture for the ice point.  It takes lots of ice, only enough water to let you put the thermocouple in.)  Tempted to include a "war story" about purchasing small quantities of thermocouple wire, but I believe this is strongly discouraged.  If you don't have s temperature scale on your meter, it is worth considering upgrading your multimeter, Christmas is not far away.  Avoid the cheapest.  If you look at the wire into the plug, some are very flimsy.  However if your multimeter has the temperature scale but no probe, Jaycar and Altronics both sell both types separately at a reasonable price.

Back to the ideal adiabatic engine performance - see if I can make it clearer this time!

Refer back to my little sketch.  You will remember that I measure the boiler inside temperature where steam is saturated and both liquid and vapour exist together, so the steam tables gave me the pressure of 0.175 MPa (absolute).  I assumed that pressure still at the engine inlet, as the pipe is short, and no throttle valve.  I measured the engine inlet temperature, 138 deg C.  It is above saturation temperature, so steam there is obviously superheated, so temperature and pressure are independent, and two independent properties are necessary and sufficient to determine all the other properties.  I interpolated the steam tables to get the remaining steam properties, particularly specific volume, enthalpy and entropy.

Paul, you will not be the only one not used to spreadsheets, so try following the procedure from post #267, one step at a time, to put in that first formula for interpolation, and let me know what point you get into trouble.  Spreadsheet formulae are so useful for so many purposes that I don't want to leave anyone behind on that one, even if you don't want to go much further.  If you have a computer with Office on it, try Excel, or you might download Open Office, (it's free).  On an iPad, or Mac, worth buying Numbers, but there are many others.  Lotus and Multiplan were two very early ones, they are all sufficiently similar until you try much more advanced functions.  I am sure that you will not regret the effort.

The engine exhaust is more tricky.  We know the pressure is atmospheric, I assumed 100 kPa, but the steam could still be either wet or superheated.  For my real engine, I measured the exhaust temperature as 104 deg C.  Now the steam tables tell me that for 100 kPa, or 0.1 MPa, the equilibrium temperature is 99.6 deg, so our exhaust is superheated.  Hence pressure and temperature are independent and hence sufficient.  Another row of interpolation of the superheat tables and I have my exhaust steam properties.

Real engine exhaust steam by interpolation -
T = 104, P = 0.1, v= 1.72, h = 2684.22, s = 7.380.


It's worth noting that if you don't have a superheater, or only an ineffective one, your exhaust steam will probably be wet steam, and we are stuck, but superheated exhaust is more informative.

Before we look at what that means, let's look at the ideal reversible adiabatic engine performance.  Again, the exhaust pressure is 0.1 MPa, but where are we on the curve?  This is where that property, entropy comes in.  The second law of thermodynamics says for an ideal adiabatic engine operating between points 3 and 5 (on my sketch), s3 = s5.  We already have s3 = 3.018 from interpolation of the steam table for the engine inlet, so s5 = 3.018.  Two independent properties, P and s, so the steam is completely defined, but how do we find those other properties?

First, look at the saturated steam table row for 0.1 MPa, and check the entropy columns.  You will see that our ideal engine exhaust entropy of 3.018 lies between the dry vapour value, sg, and the saturated liquid value, sf, which means that the steam is wet.  For we steam we can calculate a quality factor, or dryness factor.  Quality is another of the properties, but it only exists in the region between saturated liquid and dry saturated vapour.  My text book gives it the symbol x, and you can think of it as the mass fraction of the steam in the boiler which is vapour, the rest being liquid.  For our ideal adiabatic engine exhaust the dryness fraction, x is calculated as follows _

x = (3.018 - sf)/(sg - sf). We substitute the values from the steam table and then

x = (3.018 - 1.3026)/(7.3594 - 1.3026) = 0.99 or quite close to saturation.

Now we use the dryness fraction to calculate the other properties by simple linear proportion.  We really only need h5 so let's calculate it.  We look up hf= 417.46 and hg= 2675.5, then

 h5 = hf + 0.99 x (hg - hf) = 417.46 + 0.99 x (2675.5 - 417.46) = 2652.92

If you are using paper and calculator, you will notice that between hf and hg columns there is one headed hfg.  The definition is  hfg = hg - hf to save you one subtraction.  At one time, none of us had computers, or even calculators.

Now, still talking about our ideal adiabatic engine, we apply the first law of thermodynamics which tell us that the work done on the piston, W, by each kg of steam is W = h3 - h5 = 2745.91 - 2652.92 = 92.99, say 93.  The units used by the tables are kJ/kg, or J/g.

We multiply this by our steam flow rate, 0.296 g/s to get 27.53 J/s or Watts.

That is the work done by an ideal adiabatic engine with my steam conditions and flow rate.  No engine can exceed that, and any real engine will produce less!  Not very impressive, only 2.2% of the heat released by burning the fuel, but it cannot be exceeded, or even matched by any real engine.  No point beating myself about the head because I can't get 30%.

Now to the really interesting one, the exhaust of my real engine.  What can that tell me?

That will be our topic for next time.  This one has been a bit long, but where to split it?

I hope that everyone is still with me,

Thanks for following along.

MJM460

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

Offline MJM460

  • Full Member
  • ****
  • Posts: 1647
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #272 on: September 13, 2017, 06:29:55 AM »
Thanks Derek for those excellent photos of your beautiful engine and boiler.  Your post appeared while I was inserting mine.

No apologies needed, you are right on topic.  I believe I have at last a clue as to where your condensate comes from.  I am attending my granddaughters choir performance this evening, hence my early post.  I gather from the pictures that you might not have a superheater, so after the next post, I will have a go at calculating the exhaust steam properties, starting from saturated steam instead of superheated.  It might reveal something interesting.

I also have an infrared thermometer. It is very useful, but as you are aware, it has inherent errors so is best for comparative measurements.  The outside of a tube is always cooler than the steam inside unless the tube is well insulated, but then you can't use infra red.  However, the difference between ends of a tube is probably more than close enough to the difference in temperatures inside.  Interesting that the insulation makes 3 degrees difference.  I suspect that 1/8 is very small and a significant restriction in that engine, close to being a laminar flow orifice, like a domestic fridge.  You don't want to fill your piping with temperature elbows like I have done in your boat, but on the bench for test purposes, they can be very instructive.  However, replacing the plug in your boiler with a thermowell allows use of a thermocouple to check the gauge pressure reading any time, even in the boat, by just poking in the thermocouple, so I would think may always be worthwhile.

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

Offline paul gough

  • Full Member
  • ****
  • Posts: 457
  • Tropical Queensland, Australia
Re: Talking Thermodynamics
« Reply #273 on: September 14, 2017, 12:51:54 AM »
Thankyou Derek and MJM for the comments regarding thermocouples/multimeters and digital thermometers, things have certainly advanced in this area since the early seventies when I was getting preferred temperature ranges of reptiles and I will look around to see what is available.

MJM, I am interested to know what constitutes a good thermowell. I take it these are used where the temperature probe cannot be inserted into the medium for direct measurement through a gland nut arrangement. Obviously one wants to get the closest to the internal temperature as possible, so does one manufacture the orifice of the well to a push fit for the probe dia. and make the profile at the bottom the same profile as the end of the probe to get contact here also? Or alternatively, does one use a clearance around the probe and some proprietary filler that sets to gain the appropriate heat transfers. The design of a good thermo-well would seem to me to be an absolute necessity to avoid introducing errors and ending up with spurious results. Are there any other considerations or do's and don'ts regarding the acquisition of data and trip ups with instrumentation etc.

Derek your temps regarding insulation are interesting, however it would be helpful to know the material used and the thicknesses. From my experience in full size stationary steam facilities pipe lagging is quite thick, as a rough guide it was as large in diameter or often larger than the size of the flanges on the pipes and nearly always had sheet metal cowling around it. Now dragging from the depths of memory, (1960s), regarding steam locos, I think the asbestos lagging rope used around the steam pipes on NSWGR locos, eg. to the air pump, was about 1/2 inch, so the dia would be roughly twice that of the pipe. I wonder if lagging on models is adequate, have you done any tests on a pipe or a vessel with a set thickness then increased it by, x2,x3,x4 thickness to see what the optimal thickness of a particular insulating material is?

Online derekwarner

  • Full Member
  • ****
  • Posts: 833
  • Wollongong ...... Australia
Re: Talking Thermodynamics
« Reply #274 on: September 14, 2017, 01:38:41 AM »
Paul......

The detail of the model lagging for the 1/8" OD copper tube from the boiler steam isolation stop vavle to the engine is as follows

3.175 diameter tube.....add a 1.78 diameter section Viton o-ring [to act as a diameter guide] at each end of the spool, wrap 1.0 diameter cotton string around the spool [superglued at ends & occasionally at bends] , paste a wet premixed cellulose [Bunnings type] Polyfiller material into the string and to exceed the major diameter of the O-Ring...[this may take 3 or 4 coats of Polyfiller & sanding to achieve the oversize diameter]

After sand back to the uniform diameter, two coats of enamel primer, one coat of gloss enamel paint....so the insulation OD for this tube is ~~ 8mm diameter, or a nominal 2 mm wall thickness.

I have bent & set ~~ all tube spools to include 90 degree bends as this lagging material has very little resistance to bending and stress cracking will appear through the enamel top coat at bends if care is not taken...[this also requires the spool fittings have sufficient clearance when being dissembled so as not to require any moment of bending]

On area of concern is the non insulated fittings in the line of components [lubricator, regulator and the entry fittings to the engine itself].....my Scottish steam regulator is a 20mm bronze cube ...[I have lagged external faces of the regulator with timber planks] but it is clearly a source of heat loss...[image below]....these screwed of flanged fittings in the lagged steam line provide an excellent point for the temperature comparisons....being as such, provide totally repeatable set points 

For exhaust lines I have used the similar insulation relationship which is governed by the 1.78 section O-Rings for the correspondingly larger diameter tubes etc

I have not conducted any varying diameter lagging test comparisons, suffice to say I can place thumb & forefinger around the steam line directly below/from the boiler isolation valve [135 degrees C]....and the lagging OD is hot, but not sufficient to burn skin

Sheet metal sheathing over lagging on full sized steam applications is I suggest only to provide resistance to any form of mechanical abrasion, as wet pasted or preformed asbestos and latter synthetic insulation material is very soft

Lagging in full sized applications includes covering all flange & connection points or and inline component parts as these are potentially a source of gross heat loss

Derek



« Last Edit: September 14, 2017, 12:56:45 PM by derekwarner_decoy »
Derek L Warner - Honorary Secretary [Retired]
Illawarra Live Steamers Co-op - Australia
www.ils.org.au

Offline paul gough

  • Full Member
  • ****
  • Posts: 457
  • Tropical Queensland, Australia
Re: Talking Thermodynamics
« Reply #275 on: September 14, 2017, 07:55:13 AM »
 Derek, thanks for the detailed reply. My experiences in four full size plants has seen most of the pipe flanges exposed, i.e. high temp hot water heating systems for a uni. campus, fruit juice processor and food manufacture, only the drug manufacturer lagged most of the flanging. All these were below 200 psi though and accept that these plants were not chasing efficiencies as one of their primary goals. Interestingly the uni underground distribution system of quite a few kilometres, (about eight from memory), had eight and six inch piping with no insulation, the ground was the insulation! This was sometimes helpful in winter, a leak could be found by going for a long walk following the route of the piping, a telltale wisp of water vapour or a relatively warm wet area somewhere near the spot would indicate the leak. My experience with small, other less 'organised' plants has demonstrated insulation is often a very neglected area or of no concern to the operators. Regards Paul Gough.

Offline Zephyrin

  • Full Member
  • ****
  • Posts: 766
  • near Paris, France
Re: Talking Thermodynamics
« Reply #276 on: September 14, 2017, 08:33:31 AM »
some friends at my club use a ribbon of plaster for orthopaedic contention to hold a layer of insulating stuff around pipes, (just like on a broken leg !), a paint layer or a teflon band keeps away the moisture.

Offline Admiral_dk

  • Full Member
  • *****
  • Posts: 3752
  • Søften - Denmark
Re: Talking Thermodynamics
« Reply #277 on: September 14, 2017, 12:03:05 PM »
Here in Denmark we have real central heating - ei. Aarhus where I work has 300,000 people living and the heating is provided from two plants for the whole city + some villages are ; one being the electricity plant and the other the waste incinerator.

"Water" leaving the incinerator plant is between 200 and 250 degree C and around 80-90 degree C when it arrives to the buildings some 10-100Km. down the line. I don't know if there is any heat exchangers along the way.

The pipes are insulated so that the outside diameter is tre times the metal pipe in the middle - local pipes are around 250mm. ~10" and the ones leaving the plants are 2-3 times that diameter.

Pipe integrity is tested continuesly with the help of two copper wires running parallel in the insulation for the whole length of the piping. An electric pulse is transmitted out through the wires and the reflection is measured. If the resulting reflection arrives back at the right time and has the correct shape, all is well and a new pulse is transmitted. Any dammage to the insulation results in an earlier reflection and the time from transmission to reception gives the exact distance to the dammage - so they know where to dig even though there might not be a visual leak at the site.

The pipes where not insulated in my childhood and that was also the reason the rusted and broke - plus no snow or ice on the surface over them - but even back then we never had the heat missing for more than a few hours and after the insulated ones I have newer experienced any breakdowns.

EDIT : I really should make a point about the fact that the heat would have been wasted / is a waste product from the primary function at the two plants + it is a nice in the winter where we are around -10 degree C.
« Last Edit: September 14, 2017, 09:03:06 PM by Admiral_dk »

Offline MJM460

  • Full Member
  • ****
  • Posts: 1647
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #278 on: September 14, 2017, 01:40:57 PM »
Hi Paul, I don't think the thermowell design is over critical, the main thing is to provide a way of inserting the element so it is surrounded by steam without compromising the pressure containment.  No problems with liquids creeping through the wires and their insulation etc.  The main issue with contact resistance is that it slows the response of the thermowell which adds to all the other issues that make temperature measurement inherently slow response.  Special quick response designs are available, but not really needed for our application.  The first approach is usually to put a few drops of glycerine in the thermowell to improve the heat transfer, but very messy in a model where we probably do not really intend a permanent installation.  Slow response alone does not affect accuracy, as in principle, there should be no continuous heat loss once the thermocouple is heated.  So the next issue as you imply, is conduction away from the fitting where it is in the atmosphere and conduction along the wires.  Some insulation around the insertion point in the piping is probably more important on our models as the insertion fitting is closer and more bulky relative to the thermowell depth.  In industry, vortex shedding induced vibration leading to breakage is more of a problem than minor heat losses.  I make my models along the lines of the industrial ones which are basically machined from a bar, and internally drilled for the thermocouple element.  Small ones usually screwed into the appropriate pipe fitting, but larger ones are flanged.  The sheathed thermocouples could be inserted through a gland, but I think the simplicity and reliability of a thermowell is a more practical solution.

You have prompted a few ideas with your insulation comment.  Very interesting to see the different approaches in different environments.  In hydrocarbon industries, insulation falls into two basic categories, hot and cold.  Cold insulation has to be right, and cover complete, otherwise ice breaks the insulation off.  Hot insulation is applied for safety reasons, to reduce the surface temperature so people do not get burned.  And of course also for heat conservation.  It is generally a compromise between what can be maintained with normal practice, and cost.  The cheapest part, and the most result in terms of reduced heat loss is the straight pipe sections, usually with special formed sections for elbows.  Much more fiddling required to cover flanges and valves etc. which are really a relatively small part of the total heat loss.  However there are generally also higher quality specifications which are applied when heat conservation is more important.  Generally we used to use magnesia, but Calcium silicate, foam glass and fibre glass are increasingly used, unfortunately.  I say unfortunately because they wick up any oil spills then become flammable.  Installations are almost universally outdoor, so metal sheeting is applied for weather protection, and also for preventing mechanical damage, which in plant sizes includes people walking on the piping.  Obviously some different considerations from those of well protected indoor installations.  Some experimentation on different thicknesses would be interesting, but might be best carried out on a test rig with a steam pipe say 1 metre long, so there is some chance of measuring the reasonably small differences.  The first layer makes the most difference, then diminishing returns.

Thanks Derek, for describing your methods in detail.  You certainly achieve a good looking result.  As you say, avoiding damaging the insulation after installation can be a challenge.  Looks very realistic for what I have seen in ships engine rooms.  My silicone tape is not so elegant, but it is flexible, and I can always add another layer.

Hi Zephyrin, I have also heard of plaster of Paris being used in this way, I guess the source of the materials depends on who you know.  But that plaster soaked gauze is a very convenient way of applying the paste and then reinforcing the set material.

Admiral _dk, thanks for that description of your district heating plants.  Don't have them here, 37 C today, and Spring has barely started.  Also, a very interesting method for detecting insulation faults.  It is a real problem for such extensive systems, and these days, even that low temperature heat is valuable.

I wanted to get back to the calculations on that exhaust steam, but it is already late, so next time.  I am back in the long paddock, having turned my head towards home the short way.  The Buildup has started, should have left for home a week ago.  Today's temperature was accompanied by quite high humidity, which persisted through the night.  Seven thousand two hundred kilometres so far, but only 4400 home from here now.  And should be able to post most days.  Please understand if I miss a day, as there are some gaps in the service.

Thanks for looking in,

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

Offline MJM460

  • Full Member
  • ****
  • Posts: 1647
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #279 on: September 16, 2017, 10:35:54 AM »
Real Engine output -

Hi everyone, Internet gaps appeared quicker than I anticipated, so glad I had mentioned it.  Usually available each night, but the place we found last night only had the wrong carrier.  Along the long paddock, there are occasional reflecting antenna dishes with a little pad at the focus.  If you put your phone there, you get Internet and or phone reception, but not even a meter away.  Ok for email and messages, and stuff that can be read off line, but not much help for reading an active forum.  I will try and stop next time I see one with enough warning, and get a photo.  Possibly three or fours days time, as we should be good for the next two nights.

Last time went into acknowledging very interesting and relevant comments that people had posted, so tonight back to the test run data.  You might remember that I had calculated the power output of  an ideal adiabatic engine, using the second law to determine the exhaust entropy, which defined the steam condition so allowed calculation of the enthalpy.  The first law then allowed us to calculate the output of that ideal engine as 27.5 watts.  Next, we note that we have a measured exhaust condition for the real test engine, it was superheated to 104 deg C at atmospheric pressure, so we could find the exhaust enthalpy directly by interpolating the superheated steam tables as 2684.2 kJ/kg or J/g.

Now the first law says the work produced by our real engine is h3 - h4 = 2745.91 - 2684.22 = 61.7 kg/kg (J/g).  We multiply this by our steam rate, 0.296 g/s also W = 61.7 x 0.296 = 18.3 J/s.

Not much compared with 668 watts in our steam, only 2.7 % efficiency, based on heat in the steam, even less based on fuel energy.  But it is 66% of the output of that ideal engine, which was only 27.5 watts, and we know that cannot be exceeded or even equaled by any real engine.  So, not so shabby after all.  If I was an engine manufacturer, I would try and use an efficiency defined by comparison with an ideal engine in my sales pitch!  I think both the actual numbers, and the fact that they can be calculated at all is the most interesting thing about the whole exercise of seeing what can be found out by applying basic instrumentation and the laws of thermodynamics to a simple test run. 

Of course you are wondering how we can calculate a power output when the engine was running uncoupled, surely not producing any output at all.  It is quite correct that the engine was not producing any shaft output power, but the observation helps our understanding of what these calculations really mean.  And another illustration of different definitions of power.

The power output calculated from the first law, (h3 - h4) x steam flow rate, gives the work done by the steam on the piston in the power strokes.  Remember way back, how heat is converted to work?  Portions of this work are then used inside the engine, before we get anything to the output shaft.  Some goes into pushing the exhaust on the other side of the piston out through the ports, even before there is a net differential pressure and hence net force on the piston.  Some of the force on the piston goes into overcoming the friction of the piston in the cylinder and the remainder produces the torque necessary to overcome friction in the bearings, and pins.  Some steam, it may be quite a bit in my case, may even bypass the piston by flowing between the piston and cylinder, thus producing no work.  I only have one or two labyrinth grooves, and this will not be nearly enough, especially if the piston clearance is too great.  However, wherever the work is used, it is not available at the output shaft, so I have to get more power out of the engine to have any output at all.  I suspect it would be a bit optimistic to hope to double the power produced by the steam without increasing the losses so that we had another18 Watts available to drive something, but so many of the engines on this forum do produce enough useful work to overcome more than engine friction, that I am sure I can get something out of it.  Obviously the next project has to be a brake of some kind to provide a predictable load for the engine for more tests.

Before I do more testing, as that must be on hold for the moment, is there anything else we can learn about where this power is consumed?  Now I am not a politician, I have posed the question without being sure how I will get an answer, let alone what troubles I will get myself into in the process.  But there are a couple of formulae that relate mean effective pressure and torque to the power of an engine.  I will use those to try a few calculations and see if they yield anything interesting for next time.

Thanks for looking in

MJM460
« Last Edit: September 16, 2017, 10:43:43 AM by MJM460 »
The more I learn, the more I find that I still have to learn!

Offline steam guy willy

  • Full Member
  • *****
  • Posts: 3238
Re: Talking Thermodynamics
« Reply #280 on: September 16, 2017, 03:34:16 PM »
Hi, I have just bought the two volume set of the John Farey  "A treatise on the steam engine"  from 1827 and what a treat that is !! there are 90 pages on MR Woolf alone !!!!! A quick question on his compound engines .........................as the HP cylinder is  supplied with full steam with no cut off, how much work is done by the LP cylinder that is moving anyway as it is coupled directly with the HP side ????...Answers on a very large postcard please !!!!!Also does the water level in the boiler have any significant detrimental effects on efficiency ?? Thanks for all this info from all our subscribers .
« Last Edit: September 16, 2017, 11:11:13 PM by steam guy willy »

Offline MJM460

  • Full Member
  • ****
  • Posts: 1647
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #281 on: September 17, 2017, 02:17:21 PM »
Hi Willy, the second question first.  Heat transfer is determined by three resistances in series, the film resistance for transfer from the hot gases to the copper, the conductivity of the copper, then the film resistance from the copper to the fluid on the inside of the boiler.  Obviously we would the first two should be unaffected by the water level, however the internal convection coefficient is very different for steam compared with boiling water.  Boiling water has a very low film resistance, or high conductivity, partly due to the specific heat of water and partly due to the agitation by steam bubbles which continually carry away the heated fluid so it is replaced by cooler water.  This effectively increases the temperature gradient, and hence gives much better heat transfer.  Transfer to steam is much less effective, partly the lower specific heat of the steam and partly because there is no vigorous mixing due to the bubbling, just normal convection.  Usually when calculating the surface area of a small boiler, only the area below the water level is counted, and heat transfer to the steam ignored.  When the water level is low, the heat transfer will be less, so losses to the stack more, and efficiency lower.  However, all of this is for a fired boiler.  For your electric boiler, this does not apply, as the element should always be fully submerged in liquid, while the boiler shell should be well insulated so there is minimal heat loss.

The second question is a little more difficult.  I believe that what happens in the compound engine is that the volume of the exhaust side of the hp cylinder is getting smaller during the exhaust stroke, while the volume of the inlet side of the lp cylinder is getting larger, but at a greater rate due to the larger diameter.  So the total volume contained in the exhaust side of the hp plus the inlet side of the lp is getting larger or expanding.  At the hp piston face, work is being done on the gas, the energy coming from the steam on the inlet side, while at the lp piston face, the steam is doing work on the lp piston.  As always, the work is the pressure times the change in volume, but calculating the actual amount of work is complex as the pressure is changing as well as the volume.  The work input by the hp piston means the process has heat input at the same time as work is being done so it cannot be compared with an ideal adiabatic process.  With the geometry and timing all known, I assume the process could be analysed, but it is beyond me.  Is that about the right sized postcard?

Regarding the additional calculations on my little engine test, I have tried to calculate an equivalent mean effective pressure, and an equivalent torque from the usual formula, without getting any sensible answers.  I think I am making a mathematical mistake somewhere, and just need some time to check it all.  Just not sure what at the moment, really puzzling.  May need to look at some other topics for and come back to that one, unless someone has noticed the problem or had more success with the numbers.  At least I have had some success with comparing a model engine with the thermodynamic model even with the simplest of test set ups.

Crossed the Tropic of Capricorn late today, back in the temperate zone, cooler days and cool nights (making sleeping easier) only four days from the tropical heat and humidity on the coast.

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: 457
  • Tropical Queensland, Australia
Re: Talking Thermodynamics
« Reply #282 on: September 17, 2017, 04:36:05 PM »
Thank you Willy for asking these questions and a very great thank you to MJM for the lucid explication of heat transfer in the boiler, it caused me to have an "Of course, what a fool am I" moment. Despite knowing of the various boundary resistances and even more if you add a soot layer on the fire side and a scale layer on the water side I had completely missed grasping the differential in heat transfer between steam and water when considering the circumstances pertaining to the poor steaming of a particular design Gauge 1 loco, a 12inch gauge loco that often defied the efforts of those who fired her and the behaviour of a vertical boiler when it had lower water levels. This one element left out of my thinking has finally explained a conundrum of many decades for the vertical and the 12" loco and illuminates perfectly part of the poor performance of the G1 loco. It is one thing to 'know' a number of facts but missing one critical one or not putting them all together can leave one sorely bereft of satisfying explanations for the behaviour of things! I owe you a beer, VB I presume. Regards, Paul Gough.

Offline steam guy willy

  • Full Member
  • *****
  • Posts: 3238
Re: Talking Thermodynamics
« Reply #283 on: September 18, 2017, 02:43:50 AM »
>Hi, MJM, I will be firing up my boiler later on today BST and could you give me all the measurements and volumes and temperatures that i need to record. Will it matter how full the boiler is ,or do you just need the volume of water. The pressure gauge is not very accurate but could i slide my temp gauge probe under the lagging. any suggestions would be welcome. It will be coupled up to my engine so that will be fun . !!

Offline 10KPete

  • Full Member
  • ****
  • Posts: 1606
  • Nordland, WA, USA
Re: Talking Thermodynamics
« Reply #284 on: September 18, 2017, 04:24:00 AM »
Geez, Willy, that's a beautiful engine.....

Pete
Craftsman, Tinkerer, Curious Person.
Retired, finally!
SB 10K lathe, Benchmaster mill. And stuff.

 

SimplePortal 2.3.5 © 2008-2012, SimplePortal