Author Topic: Talking Thermodynamics  (Read 154538 times)

Offline MJM460

  • Full Member
  • ****
  • Posts: 1434
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #75 on: June 17, 2017, 12:05:21 PM »
Symmetry in valve events?

Thanks for the comments steam guy, sorry you need the dictionary, I was hoping to keep this understandable, so that the thermodynamics is accessible to all.  You will not be alone, so if you tell me some of the words you needed to look up, I will try and explain what I mean a bit more clearly.  As with all dictionaries, they tell you the word definition, but often do not help much with understanding the sentence.  And for some forum members, English is not the language of choice, and I want it to be accessible to everyone.

I know what you mean about the plumbing nightmare under the bonnet these days, I used to be able to sit on the mudguard with both feet comfortably down beside the engine while I worked on my old Holden, but not any more.  So much extra equipment crammed in, mostly more about pollution control, equipment such as air conditioning and modern engine control systems than pure thermodynamics, though I notice my Subaru keeps the engine air separate from the cooling air through the radiator, which is pure thermodynamic reasoning.

Yes, some of the energy is converted into noise, but a very small part of it.  It does not require much energy to make a lot of noise.  Most losses are directly or indirectly turned into heat.  I can talk a little about noise if you like, but if I could make your Morris sound like a Rolls, I would be a lot wealthier, and would be able to afford more machines (and castings!). It's on the list.

Derek, I am not quite with you on your aim with the "water retained to water discharged as steam".  Do you want to condense and collect more?  Or do you want collect minimum, just enough to collect the oil and send more water up the stack as steam after removing the oil?  These are very different issues to minimising the back pressure for power output.

Back to the valves.  Some of you may be feeling there must be more resistance in our inlet and outlet piping, especially as we are so often advised to have a free flowing exhaust.  The issue comes down to two areas.  For the inlet piping, it is velocity just as I have described, and then valve opening .  Especially for larger engines, it is important to have sufficiently large piping to keep the velocity low.  The valve opening requirements however are very different for inlet and exhaust.  Surprisingly so.  There is no symmetry there at all.

Most of our valve linkages whether a simple eccentric, Stevensons reversing hear, Hackworth, Joy or one of the many others (the W one used commonly on steam locomotives is too hard to spell), all produce a simple harmonic, or sinusoidal motion to the valves.  They move slowly at the extremes, full open, at each end, and fastest at middle of their stroke, which is of course about 90 degrees different from the piston.  This does mean they move fastest when the piston is moving slowly at the top and bottom dead centre, but they still take a finite time and number of degrees of rotation to open or close fully.

For the inlet valve things are nowhere near as bad as you might expect.   When the valve is opening, the piston is at top dead centre, minimum cylinder volume, and starts to move slowly down.  It turns out that the valve opening roughly matches the rate at which the cylinder volume is increasing (due to the piston moving down) so that the velocity in the valve port stays reasonable.  By reasonable, I mean the velocity required to admit enough steam to maintain the pressure in the cylinder stays within the maximum velocity I have suggested, and the valve opening does not cause any undue restriction.

This is especially true if the exhaust closing at the end of the preceding exhaust stroke provides a good compression before the inlet valve opens.  But even if it does not, the cylinder is at minimum volume and not much steam is required to get up to full pressure.  So it all works as expected.

The exhaust valve is very different!  Can you see why?  But this post is long enough already, so next time the exhaust valve operation.

Thanks for following along.

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

Offline Admiral_dk

  • Full Member
  • ****
  • Posts: 2270
  • S°ften - Denmark
Re: Talking Thermodynamics
« Reply #76 on: June 17, 2017, 03:15:38 PM »
As I see it - the exhaust valve opens slowly when the biggest volume of used steam must escape ..... (closing the exhaust works nicely).

Offline steam guy willy

  • Full Member
  • *****
  • Posts: 2772
Re: Talking Thermodynamics
« Reply #77 on: June 17, 2017, 05:37:30 PM »
Talking about exhaust's  I have always wondered why my exhaust valves are smaller than the inlet valves ??....is it the same with diesel engines ? and is it why they burn out every few years ........

Offline Zephyrin

  • Full Member
  • ****
  • Posts: 559
  • near Paris, France
Re: Talking Thermodynamics
« Reply #78 on: June 17, 2017, 06:26:20 PM »
IMHO, the best way to visualise the position of the steam valve during the cycle is the oval or elliptical diagram:
The oval diagram is in the CW rotation.
The steam valve is plotted (ordinate) in function of piston position in abscissa, normalised here in % of the stroke.
The green line corresponds to the opening or closing of steam exhaust.
Yellow lines are the positions of the external edge of the steam port, the crankshaft side is the upper line, and back side is the lower yellow line.
When the steam valve crosses a yellow line, steam port opens or closes. I.e. on upper yellow line, which corresponds to the crankshaft side of the cylinder, the admission opens near BDC, 0%, cut off being near 60% of the piston stroke.
One can easily appreciate the fast opening of the steam admission near dead center.
One also sees that the steam distribution on both sides of the cylinder is not symmetrical, owing to rod length.


« Last Edit: June 18, 2017, 03:07:21 PM by Zephyrin »

Offline MJM460

  • Full Member
  • ****
  • Posts: 1434
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #79 on: June 18, 2017, 01:49:41 PM »
Exhaust Valve issues

Thank you for the oval diagram Zephyrin, such diagrams can be really useful in visualising valve event timing.  I will come back to that in a near future post.  Can I assume that what you refer to as the steam chest might be what I usually call the valve?  I usually think of the steam chest as the stationary part held to the cylinder block with studs.

Steam guy willy,  thanks for your post and pictures.  At first it might appear to contradict what I have said about exhaust valve opening, but of course, engine design also involves compromise.  Clearly the valves are about the largest that could fit the space.  The designer has to consider whether both should be the same size, or make one larger.  In internal combustion engines, the power output from a given engine is limited by the amount of air that can be delivered into the cylinder, despite losses in the air cleaner and carburettor.  Fuel is much easier to get enough.  So the designer has made the inlet valve a bit bigger to reduce the resistance to air entry.  For the exhaust, there is extra pressure available to get the exhaust gas out , so a little more resistance is perhaps less of a problem, even though bigger exhaust valves would be desirable.  They burn out because they pass very hot gases from (possibly) incomplete combustion at high velocity and including any unburnt carbon.  So very hot and abrasive.  Surprising they last as long as they do.  In contrast, a steam engine has the boiler pressure available to help get the steam into the cylinder.  The exhaust is the steam engine problem.

Thanks for your reply Admiral_dk, you have hit the nail squarely on the head.  At the end of the power stroke, the cylinder is full of steam that is at significant pressure, particularly if there was late cutoff, and the volume of the cylinder at this point is at maximum.  Of course, we can help a bit by early release, the torque due to the force on the piston is quite low within 10 to 15 degrees of top and bottom dead centre.  Can we make some estimate of how much of a problem we have?

I do not want to try and make an exact calculation, but if we assume early cutoff at 15 deg before the centre, and that we want the pressure to be essentially dissipated by 15 deg after, we have say, 30 degrees of rotation to get the pressure down.  At 2000 rpm, which is 33.3 revs per second or 0.03 seconds per revolution, or 0.0025 seconds to reduce the steam pressure to that of the exhaust system.  We then have to make an estimate of how much steam is to be discharged.  If we assume the pressure in the cylinder is 300 kPa(abs) and it expands to atmospheric pressure On release to exhaust, then 1922 mm^3 (including clearance volume) of my little oscillator would expand to 4545 mm^3.  Thus 2623mm^3 of "extra volume" has to be discharged in 0.0025 seconds.

Next we assume the valve is open to the full area of a 3/16" tube in the exhaust, 8.76mm^2, which requires a velocity of 120 m/s.  Applying v^2/2000 with an average density about 1 kg/m^3 gives a back pressure of 7 kPa.

While a pretty rough calculation, this does not seem too bad a result.  However as the valve is opening from closed for the first 15 degree, the average area might be considerably less than I have assumed.  My slide valve engines have an exhaust passage 7 mm wide.  So this must be open by 1.25 mm to give the assumed 8.76 mm^2.  As the total valve travel is only 5 mm, 1.25 mm within 15 degrees is probably optimistic.   If the area is only 4.4 mm^2, the pack pressure would be 4 times, so 28 kPa, and starting to impose serious negative torque on our engine output.  These calculations are very rough, but do serve to give an idea of the range of back pressure imposed by inadequate exhaust passage sizes.  I think enough to show that 5/32" is too small for this engine exhaust, 3/16" dubious and 1/4" better.  This compares with possibly 5/32" for steam supply, and 3/16" not really too large.

Next time, I will summarise the key points relating to valve events and the P-V diagram, and start to look at how our valve behaviour compares with the ideal.

MJM460
« Last Edit: June 18, 2017, 01:56:05 PM by MJM460 »
The more I learn, the more I find that I still have to learn!

Offline Zephyrin

  • Full Member
  • ****
  • Posts: 559
  • near Paris, France
Re: Talking Thermodynamics
« Reply #80 on: June 18, 2017, 03:04:44 PM »
Of course, sorry, I did a confusion between valve and chest...I've corrected the post !

Offline MJM460

  • Full Member
  • ****
  • Posts: 1434
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #81 on: June 19, 2017, 02:33:28 PM »
Thanks Zephyrin,  No problem, I have it almost worked out, but I wanted to check that we were talking about the same direction.

I have spent my time today cutting and colouring, Cutout Aided Design, otherwise known as CAD, trying to work out how best to describe the sequence.  I must admit to a severe case of eyes glazing over when trying to work it out, and have previously given up after setting the valve to just start opening as the crankshaft turns over top dead centre, and, on finding the engine ran ok, letting the exhaust look after itself.  The valve dimensions must have been good enough.

However for this thread, I want to match all the valve events to the P-V diagram to see how well it matches the ideal.  The oval diagram seems to be helping, I hope to have it sorted by tomorrow or perhaps Wednesday.  Have some babysitting commitments for the grandchildren tomorrow.

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

Offline steam guy willy

  • Full Member
  • *****
  • Posts: 2772
Re: Talking Thermodynamics
« Reply #82 on: June 19, 2017, 03:41:52 PM »
Thanks for the info ...Actually the exhaust valve failed during a long journey and i was able to get there and and back on 3 cylinders..about 120 miles !! after this happened...A brief summarisation of thermodynamics might be " Everything is a compromise " when it comes down to building engines !!!! ;D ;D   :popcorn:

Offline MJM460

  • Full Member
  • ****
  • Posts: 1434
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #83 on: June 20, 2017, 02:43:23 PM »
Hi steam guy willy,

I once lost a valve in of a V-8 while on a much longer trip, I was in the Smokey Mountains, and home at the time was in Canada.  No question of driving on.   Getting it fixed became a highlight of our holiday due to meeting some great people.  But I can really feel for you losing one cylinder out of a small four.

Got diverted onto thinking about controlling propane on Brian's thread today, so not quite ready to continue those slide valves, but am coming to terms with Zephyrins oval diagram, so a little progress, and l hope to be ready tomorrow.

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

Offline Dan Rowe

  • Full Member
  • ****
  • Posts: 874
  • Dripping Springs TX USA
Re: Talking Thermodynamics
« Reply #84 on: June 20, 2017, 04:55:42 PM »
Have you considered Dockstader's valve gear programs? One of the outputs is the sine diagram.
http://www.billp.org/Dockstader/ValveGear.html

It would be nice to see some of the formulas you are using, a lot of pipe formulas are on the web in calculator form try Engineers Edge or Engineers Toolbox.

http://www.engineersedge.com/
http://www.engineeringtoolbox.com/

Dan
ShaylocoDan

Offline MJM460

  • Full Member
  • ****
  • Posts: 1434
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #85 on: June 21, 2017, 11:36:19 AM »

Real valves

Thanks for those suggestions Dan, I am glad to see that you are still following and your input always welcome.  I will put those topics on my list.  I am planning on a post on units of measurements and you might be surprised to see how much can be done with little more than the definitions.  But also a little Physics for the three conservation laws I have used and also the laws of thermodynamics. It is on the list, so I will get to it.

Casting our minds back to the P-V diagram, the key events in each cycle are the opening and closing of the inlet valve, followed by the opening and closing of the exhaust valve.  Remember that these events occur on the top of the piston, and on a double acting engine, they also occur on the lower side, but half a turn later so that the piston is first pushed down, then pushed up. 

When we set the valves on a new engine, we take off the steam chest cover, and we can see when the steam port is about to open, set this to happen at piston top dead centre, then we can see where on the crank rotation it closes.  We can also see the maximum opening at each end and so adjust the valve position on the rod so both ends open about equally.  But what of the exhaust?

Most books draw a cylinder cross section through the ports with the valve at mid travel, and show lap, and the relation of the exhaust cavity to the edges of the exhaust ports.  But unless more drawings are provided, I for one, find it difficult to keep track of the exhaust events as the valve is shifted only in my imagination.  So, does the exhaust open and close at the right points as Required for the valve events in the P-V diagram?  How much cut off do we get? And do release and compression occur as we want?

Of course with modern computer programs, we can model the valve movement and watch it on the screen as the crank shaft rotates.  But such programs are relatively recent, and quite expensive if we don't have access through our workplace.  For those of us a bit technology challenged is there a simple solution?

To solve this problem, I have resorted to Cutout Aided Design (CAD) shown in the first attachment.  The valve is the blue part that is a separate cut out that I can move back and forward over the outline of the cylinder ports to see when the ports open and close.  It is actually also useful to check the design for features such as how valve sealing surfaces and the bars between ports relate to provide clear port opening.

Many geometric constructions were developed when people only had hand calculations, and these were converted to allow computer calculations even before modern models appeared.  My text books have four variations and Zephyrin has shown us a fifth which is not in any of my books.  No doubt there are others.

Now that I have studied it carefully, I am with Zephyrin, in feeling that his oval diagram is probably the easiest to use.  Certainly the easiest I have seen so far.  It shows quite clearly the effect of lead and lap for both the inlet and exhaust events in a simple construction, at least simple with the benefit of a spreadsheet to do the repetitive calculations.

The oval diagram is produced by making a graph of valve position against piston position, which gives the oval outline.  The points around this oval correspond to the edge of the steam valve that determines valve opening, proceeding clockwise.  The same points can also be considered the edge of the exhaust cavity that determines exhaust port opening.  The steam lap is shown by the red line, and the valve starts to open when the inlet edge of the valve crosses this line at the piston dead centre.  The port finally  closes again when the oval crosses this red line on the way back to the centre position.  Similarly in the lower half of the diagram for the other end of the cylinder.

My first attempt is shown in the second attachment.  I have saved a little time by assuming a scotch crank design, but with a few more columns in my spreadsheet, and a little more trigonometry, the effect of the conrod can easily be included.

I think this item turned out way too long, so I have split it here and will continue to work through the diagram next time. 

Thanks for reading,

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

Offline Dan Rowe

  • Full Member
  • ****
  • Posts: 874
  • Dripping Springs TX USA
Re: Talking Thermodynamics
« Reply #86 on: June 21, 2017, 02:43:18 PM »
Many geometric constructions were developed when people only had hand calculations, and these were converted to allow computer calculations even before modern models appeared.  My text books have four variations and Zephyrin has shown us a fifth which is not in any of my books.  No doubt there are others.

MJ,
I have about 50 textbooks on steam valve gears, could you say which ones you are using? You said four valve gear graphical constructions are in the books, off the top of my head I can think of Zeuner, Reuleaux, and Bilgram and the oval diagram.

Zeuner was the first one and I have his book but it is really a lot of Greek in the form of complex algebra. The way Zeuner made his first diagrams was very interesting. He made a cutaway steam engine and attached a drafting board spinning at crank speed. A pencil attached to the valve to gave the trace lines that we now call a Zeuner diagram.

I know how to construct a Reuleaux diagram and it would be my choice if the valve is not symmetrical at the center of the exhaust port.

My favorite valve diagram is the Bilgram, for me, it is much simpler to see what happens when the variables of the valve are changed.

If you post the valve dimensions needed, I can construct the diagrams I just mentioned.

The historical designers of steam engines used the drafting board like we use calculators and computers. I have studied the graphical methods used to set out Stephenson's valve gear and my graphical solution matched exactly the Excel program in Don Ashton's book. This was very nice because the answers are not in the back of the book and it was nice to know I got it correct.

Dan
ShaylocoDan

Offline Dan Rowe

  • Full Member
  • ****
  • Posts: 874
  • Dripping Springs TX USA
Re: Talking Thermodynamics
« Reply #87 on: June 21, 2017, 07:30:10 PM »
MJ,
I failed to mention that one of Dockstader's programs is a Zeuner diagram. If you capture a screen shot of the Zeuner valve diagram of the valve you are discussing, I can show a step by step how to draw a Reuleaux and a Bilgram diagram.

Dan
ShaylocoDan

Offline Zephyrin

  • Full Member
  • ****
  • Posts: 559
  • near Paris, France
Re: Talking Thermodynamics
« Reply #88 on: June 21, 2017, 11:36:21 PM »
Hi
Each of the Charlie Dockstader's programs also contains the oval diagram, which can be modified dynamically, amazing, you move a cursor to change a dimension and see the effect on the diagram. In these programs, PV diagram are simply a model, without thermodynamics in them, hence of limited use.
I spend hours (and night too) with it while drawing the plans of my little locos...
But the Dockstader's package is not fully compatible with the recent windows version.

Z.
« Last Edit: June 22, 2017, 07:33:14 AM by Zephyrin »

Offline MJM460

  • Full Member
  • ****
  • Posts: 1434
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #89 on: June 22, 2017, 12:15:20 AM »
Hi Dan and Zephyrin,

You have three out of the four in my text book (by Thomas Bevan), the fourth was called a rectangular diagram.  I also have some of the well known hobby books collected over many years.  But I defer to your expertise on these matters.

My motivation in writing this thread was the realisation over many years of hobby reading that the thermodynamics, that was basic to my every day work, gave me clear answers to many of the questions raised, but never answered in any of the hobby magazines or books I have read.   Or worse, answered wrongly.  The same questions keep coming.  And even simple maths is avoided like poison. 

I hoped that I could build a knowledge base of theory that would help us all understand our models better and remove some of the mystery.  I am not an academic and having people like you looking over my shoulder and joining the conversation is really helpful, just as in building an engine.  And I feel that any forum is best seen as a conversation.

My purpose in looking at valve diagrams was simply to show that the valve events on the P-V diagram actually relate to real valve events in our models, so informing what we are trying to achieve with valve setting.  Most of my machines were electric motor driven so I never had much call to go beyond the eye glazing stage of valve diagrams.  I will be more comfortable when I get back to thermodynamics and other stuff that I can understand.

You both clearly have the expertise in this area.  May I suggest that you start a separate thread (or threads) on valve gear and valve linkages plus any other areas that you think would help.  I would really like to understand those valves in the A and G beam engines that have two valve rods and presumably a two part valve.  If you could eventually get to that .......we would be really cooking.  (I will talk about the gas part.)

If we can get three or four such threads going it might even justify a separate sub board to keep the theory topics together.  After all, aero modellers would not think of trying to advance their hobby without discussing aerodynamics.  We could clearly do with some internal combustion engine topics as well.  A little theory helps in most endeavours.

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