Talking Thermodynamics

[steam guy willy]

Hi Zephrin, thanks for the update...I suppose one would have to open the tap from the methsnburner and blow into the top to get the siphoning procedure to start and then turn it on just a small amount to feed the burner. It is a single acting engine and i cannot see where the exhaust comes out ?? Also the steam outlet is very low down in the boiler ??

Hi MJM, thanks for more info When i changed the engine in my BMW for a similar CC but higher compression  The HP went up from 28 to 35  also the engine revs increased from about 5000 to 6000  !! this i was a bit confused about to begin with but now it makes sense...also it would get up to 120 mph !! With a flash boiler as it is just a tube you don't kneed at boiler certificate  ,however a loco boiler is a collection of tubes joined together, albeit different shapes !!
Willy

[MJM460]

Hi Willy, when you changed the engine on your bike, you almost certainly changed more than just the compression ratio.  The increase from 5000 to 6000 rpm would not quite achieve the specified increase in maximum power unless the torque at 6000 rpm is higher than the torque from the old engine at 5000 rpm.  But to get the new engine to be able to spin at 6000 rpm, I would guess that it is probably balanced to tighter tolerances, may have modified cam grind and timing and possibly improved porting and even air cleaner. 

Of course, it does not run continuously at 6000 rpm any more than the original engine ran continuously at 5000 rpm, and neither operates continuously at full throttle.  But in principal, and in practice, power output can be increased either by increasing the operating rpm, or by increasing the torque or some combination of the two.  In fact, increasing torque is probably preferable, gives better acceleration without jeopardising your license so you probably get more opportunity to try it out.  But the main thing is the enjoyment you get from having made the upgrade, and the exhilaration you get when you have an open road to try it out.

Boilers seem to be the topic of the day.  While the flash boiler may just be a tube, the capacity of a long tube might approach the level where the boiler code starts to apply, but we better keep that one quiet.  More importantly the small diameter means the stress is pretty low, and the piping codes are actually more about stresses from supports and thermal expansion considerations.  Even pulsating forces when reciprocating compressors are involved. 

The thing is that the flash boiler is a cylinder under internal pressure, even if it is not a straight cylinder.  In fact, it probably tends to straighten a little under pressure, like a bourdon tube in a pressure gauge.  The locomotive boiler might be a collection of tubes, but the issue of internal vs. external pressure is critical.  The small diameter  of the flash boiler means that pipe ends can be adequate even if flat.  The ends of the locomotive boiler are effectively very large flat plates which need comprehensive staying to withstand pressure.  Then corners around fireboxes and other localised departures from plain cylindrical shapes cause stress concentrations which all require special attention to design.

In case anyone is wondering, the issue with external pressure is that a cylinder will squash flat under external pressure at far lower pressure than the pressure that would burst it when applied on the inside.  The pressure vessel codes include the procedure for checking external pressure capacity.  It is reasonably straightforward, but takes up several pages of the code.  More than just a simple formula.

However, in that flash boiler I was talking about yesterday, while the outer casing looked like a conventional locomotive boiler, inside was the tightly wound long coil of a flash boiler.   Don't know how he got on when stopping at a station!

Thanks everyone for looking in,

MJM460

[steam guy willy]

Hi MJM, Yes i changed the engine from a  R60  To a R69  engine so yes there are quite a few differences !! I should have made that a bit clearer !!

Willy

[steam guy willy]

Hi MJM, I was idly sitting in the cafe today in the rain  and looking at Snickers chocolate bars i noticed that the calorie content was 210.... so i was thinking about how many chocolate bars it would take to get a steam engine from  ambient temp up to Scotland from London.? this would of course i include the snacking by driver and fireman !!! We have all the figures available so if one had to make the journey one would have enough time to work it out !!! ...any way on a more series note how do the weather fore casters know exactly when it will rain as one can have lots of clouds with no rain as well as some ??

Willy

[crueby]

We need a gauge calibrated in Snickers per mile...

[Jo]

We need a gauge calibrated in Snickers per mile...

I thought there are 26.2 miles in a Marathon or as they call them the other side of the pond a "Snicker" 

Jo

[MJM460]

Hi Willy, I have always wondered if the food calorie matched the thermal calorie.  Remember that in heat units, one calorie is the heat required to realise one gram of water through one degree C at the standard temperature.  This results in one calorie equal to approximately 4.182 Joule.  Similarly, one BTU is about 1055 J.  I have only quoted approximate conversions, as the exact definitions require too many decimal places.

However, I read recently that there is a factor of two between food calories and thermal calories.  Possibly to allow for the energy involved in digestion of the food, but I don't know how the food calorie is defined, so don't quote me on that.

So if we know the draw bar pull of the locomotive and the distance, we know how much work was done, and work is a form of energy.  But we would need to do some tests to determine the efficiency of energy conversion from the food bars to the draw bar work.  The second law of thermodynamics says some of the energy in the fuel must be lost in the conversion.

Weather forecasting is a bit out of my league, but I understand that the availability of large supercomputers and large quantities of real time data, has allowed meteorologists to develop some amazingly good models to solve the equations of mass and heat transfer for the atmosphere.  But there is some variability, and racing ocean sailors compare the output of several different models and additionally, look out the window in order to decide what route to sail through the weather.  The model we use to check the weather for outdoor activities tells us the wind for every hour over a week, and it is usually close enough for the first day and usually only subject to relatively minor changes for the ones after that.  But the models predict not only wind but also temperature and moisture content, and even the point at which rain occurs.

Hi Chris, your meter should be possible with inputs of drawbar pull and distance, but it would require lots of sampling the Snickers to confirm the conversion efficiency for accurate calibration.  You would not want to delegate that responsibility to the elves!

Hi Jo, yet another definition.  I think it would take more than a snicker to get me through a Marathon, but you may have explained why I never finished a marathon, I thought I could stop after 26!  I obviously never looked at the fine print.

Thanks everyone for following along,

MJM460

[MJM460]

While diverting to talk about Snickers, I have not been entirely idle, though you may need a stop motion film to see the progress.  Apart from travelling 6000 km across four states, including towing the van along a sandy track over a sand dune, then leaving the van parked and driving through the deeper soft sand over to the beach  (that's what I call a mobility vehicle), I have been insulating my gas fired centreflue boiler with two layers of thin cork.

Initially the boiler was insulated only with the timber slats that were supplied with it.  To do any meaningful testing I had to make a thermowell to replace the fill plug, so I could measure the temperature in the boiler with a thermocouple.  The first picture shows the thermowell and the original plug.  And of course I did not have the required ME 40 tpi die, so that had to be borrowed first.

I was able to buy a two sheets of 3mm thick cork sheet from the local $2 store (would be great for your clutch plate, Brian, though it did cost a little more than $2).  Some careful arranging of the shape resulted in suitable pieces for two layers on the shell, and enough left for the ends.

The shell was easy, though 6 mm of cork increases the diameter of the shell by 12 mm, and the circumference by Pi times 12 or about 37 mm.  So extra timber strips were required, and of course the brass bands were then not long enough, so timber and brass strips had to be sourced.  However the cylindrical part of the shell was eventually completed.  Definitely not up to exhibition standard, but well enough for a test.  The next pictures show the progress so far.

Paper templates helped prove the layout of disks for the ends.  The actual flat end is a little proud of the ends of the cylinder so an extra ring was cut to fill the gap this created.  And the plaque on the front end of the flue tube required a cut out in the first layer.  The back head was not so easy as I did not want to disturb a well fitted and sealed gauge glass, but some judiciously placed cuts allowed the pieces to be fitted behind the glass.  I looked up the data sheet for the loctite glue I had available, and it still had some strength at something over 100 C so I glued the layers together to help hold them for placement.  Even if the glue lets go it won't matter once the cork is fixed in place. 

But how to hold the circular cork layers on the ends?  I suppose I will have to flange some wide strips or even a disk of thin brass, and catch tabs under the end rings, but I am open to any  suggestions for a better or easier method.  In the mean time, I used some stainless steel picture wire to tie them in place.  Not very tidy, but it will work for a test to see if I gained a worthwhile improvement in performance.  You can see it in the pictures, but I hope to come up with something neater for the long run.

Then finally the big test.  I had run a test with only the original timber, but would have been better to do a couple more to get an idea of consistency.  For all the previous runs of this boiler, it only had the original solid plug, so no way to measure temperature. The ambient temperature for the most recent test was 15 degrees which is a bit warmer than 13 C for the earlier test.  I feel I can probably calculate the effect of this if necessary, but in the event, it did not seem to have caused a noticeable difference.

The end result is interesting.  With the cork insulation, the boiler generated 30% more steam than before.  Allowing for the difference in gas consumption, it would have been 40% more.  This was clearly evident in the engine performance, where it ran at 1900 - 2000 rpm compared with only 700 - 800 rpm for the earlier test.  So a quite pleasing result.  Certainly showed up the deficiency in my engine balancing.

When I calculated the heat loss and produced a graph of heat loss with temperature, the difference is easily seen.  But the importance of this difference is highlighted by the difference losses at steam temperature, compared with the burner output, and so result in the extra steam production.  This graph is in the last attachment.

Apart from the value of the extra insulation on steam production, the other big lesson that comes from this exercise is with regard to boiler design.  It is important to design all the boiler fittings with allowance for the intended thickness of insulation.  Making a neat job around the fittings, and making allowance for removal of the safety valve, filler plug and gauge glass, as well as any extra bushes for feed water or draining was the hardest part of the job.  It is also worth giving some though as to how the end insulation will be held in place.  I still have to work on that.

I hope this is all of some interest,

MJM460

[paul gough]

Hi MJM, Absolutely Fabulous! Some real numbers to go by and a demonstration of what adequate insulation can do for performance. Is this little boiler one of the 'Miniature Steam' models, (4"), ?  Could you report on how noisy these centre flue ceramic burners are as I have not seen/heard one in operation.  I am wondering if you would be interested in another experiment, ( a bit more work too), and getting some more numbers for us. I have a small quantity of 5mm Kayowool ceramic fibre insulation sheet. Too thick for most of my G1 uses but would be great to see a comparison between a natural product, cork, and the boiler re-done and an identical test done with a synthetic one. If interested please P.M. me with dimensions required for barrel and end plates and a delivery address so I can send it down, add a centimetre or two or so extra for trimming etc.  Regards, Paul Gough.

[paul gough]

 Hi again MJM, I have found the white exhaust gasket cement in a tube or the high temperature black or red tubed gasket material, both from 'Auto One' shops, to be successful in securing fibre sheet to outside boilers, inside fireboxes of metho burners and on the inside of smokeboxes. Covers for the end of the boiler and its insulation could follow full size as fitted on most package style boilers, a sheet metal dish with a circumferential lip in one or two pieces that fits neatly onto the barrel end. Paul Gough.

[MJM460]

Hi Paul, I am glad you find the test results interesting.  It seems that not many make the effort to do the measurements, but it's not too hard for just the the boiler.  The boiler is the 3 inch one from Miniature Steam Models. 

You might also be interested in the effect on the outside temperature which I measured with the infrared instrument.  With just the wooden slats, the temperature was typically over 100 C when the boiler was steaming at 110-115 C, which is not very useful.  With the cork under the timber, the temperature at the same location was 55-60.  This at least would not be too dangerous to touch, but suggests that even another layer might be useful.  However, it is already difficult to fit around the fittings, so I think it would be better to just keep that in mind for a future project.  In any case, the losses are getting low enough that it is probably not worth chasing a bit more.  It's getting late now, but I have been meaning to calculate the volume of that steam quantity, and compare it with the engine displacement.  But it is obviously enough for the 12 mm bore by 16 mm stroke double acting engine I have been running.

The burner tends to sing a bit when I first light it, though not loud by any description.  Then it seems to go quiet after a minute or so.  Certainly not like the propane burners of similar diameter that I use for silver soldering.  I have to admit that after that initial light up I am not conscious of hearing it.  But I will take more notice next time I light it up.

My text book gives a thermal conductivity for cork at 20 C of 0.035 W/m.K.  I looked out a few figures for Kaowool, and while they are mostly given for higher temperatures, the trend with temperature suggests it would be about the same at lower temperatures.  So not much to be gained from swapping it over for comparison.  Any difference would be mostly due to any difference in the thickness.  The real difference is in the allowable service temperature, which seems to be around 1300 C for Kaowool.  Cork at that temperature would almost certainly char if not burn.  We used Kaowool on the inside walls of industrial furnaces in the plants I was involved with.  And it is a bit flexible, so we could pre insulate large sections of furnaces and transport them without damage.

It would be easier to apply than cork even, or perhaps especially, on a small model.  However I think I will have to decline your offer.  However, if you want to get rid of some, I would gladly make a contribution for it.  I would try it on the furnace enclosures of my externally fired pot boiler, where the resistance to the higher temperature would be useful.  I don't think cork is really suitable for the higher temperature.  Can you remember where you got it?

I would prefer not to glue the insulation to the ends, it tends to be messy if it ever has to be removed.  Though I will probably reconsider if the boiler gets installed in a boat.  On full size we normally used welded nuts on edge to provide anchors for tie wires to tie the insulation in place until the metal jacket was fitted.  I have tied it on with linen thread then the stainless steel wire for an ugly but serviceable solution for the test, but now I will eventually make something from sheet brass for better looks.  However, I did glue the cork layers together so they did not shift around while I was fixing them.

Thanks for looking in,

MJM460

[paul gough]

Hi MJM, Interesting to hear your view that the kaowool would be approximately equal to cork of comparable thickness. Thanks for the additional information regarding the external spot temperature from the infra-red device. One would intuit this result, but always nice to have some numbers to prove it is actually the case.

The 5mm Kaowool was in fact supplied incorrectly by 'Camden Miniature Steam Services' in the UK. They normally only stock 1mm and 2mm, but their supplier wrongly sent them 5mm, (mislabelled as 2mm),  and it appears no-one checked its thickness before fulfilling my order. Camden was very apologetic and said they will forward 2mm when they receive it. I understand Kaowool is available in Oz but have been told not in these very thin sizes, however, I have not pursued the issue as it was convenient just to order my couple of metres from Camden.  I passed on most of the 5mm to a friend in N.S.W. only keeping a small amount, but again, if you have a use for a small quantity of the 5mm sheet I am happy to send some to you. Regards, Paul Gough. 

[MJM460]

Hi Paul,

Intuitively, more insulation would lower the temperature on the outside, but the extent is not easy to guess.  Heat transfer through the insulation is proportional to the temperature difference across the insulation, while heat transfer to the air by convection is proportional to the temperature difference between the surface and the air further away.  Two heat paths in series, so the heat transferred must be the same in each case, and the intermediate temperature results from that heat balance.  But difficult to guess the relative conductivity between the insulation layer and the convection layer.  Experiment tends to be the easiest way, as the convection coefficient is more difficult to estimate as we have seen before.  I know it would be interesting to do the maths and see how it compares, but I tend to be impatient to move on to other projects.  It is a while since I made much swarf, so I will give that more time for a while.  However always willing to try and answer any questions.  I am enjoying the discussion of thermodynamics, and hope it is contributing to all our understanding of our hobby.

Interesting how you came by the 5 mm Kaowool.  I suspect you will eventually use what you have left.  I will see what thickness I can get before I decide the dimensions of the next boiler casing.  I have the 5/16 tubes to use for the water tubes to compare with the current one which has 1/4 inch water tubes.

I am always a little skeptical about temperature readings from the infra red instrument, as they depend on a built in emissivity constant.  I am never sure how closely this applies to the surface I am measuring, especially if I am comparing readings made on different surfaces.  However, for differences between temperature at different times on the same surface, I think they are close enough.  It was certainly pleasing to get that surface temperature below 60 degrees which is around the level usually specified for personnel protection (might be 65?  I haven't checked a specification recently.  Retirement has that effect.)  So now, I am less likely to be burned if I accidentally touch it, or if one of the grand children do.  Though they are pretty safe, as I have not been able to convince any of them to take an interest.  They do bring other joys however.

Have you been running those locomotives lately?

Thanks for looking in,

MJM460

[Zephyrin]

Hi MJM,
nice results you provide in this post !
Even if we all feel that insulating our model boiler is useful, it is important that real experiments have been carried out to see how much it matters...
On my little locos, I'm far from the thickness you mention alas, just 3mm cork under the cladding in thin (0.25mm) steel foil, to keep model close to scale dimensions. No miracle, I can't touch the engine at the end of the run, not the 60°C you mention on your swaddled boiler!
With such increase in the thermal efficiency, I clearly can reduces the size of the boiler tube on my next loco to cope with the layer of insulation.
Regards

[steam guy willy]

Hi MJM, More good info.... I was wondering how the boiler is constructed  as the chimney and gas burner look to be quite close in proximity ? Also a question about boilers  when you fire up a boiler you input heat into it at a constant time. if you don't draw off any steam the temperature will rise. so can the heat in the boiler ever get hotter than the flame temperature  or will it release the extra heat in a proportional fashion via conduction radiation convection ??  Would it be possible to connect a system that reduces the heat input with the rise in temperature with our small gas heated boilers or is that quite complicated and so we do it manually ? On the electrically heated boiler it does it by switching off the currant with a pressure switch.....
Willy