Author Topic: Talking Thermodynamics  (Read 123840 times)

Offline steam guy willy

  • Full Member
  • ****
  • Posts: 2252
Re: Talking Thermodynamics
« Reply #1230 on: August 23, 2019, 02:59:50 AM »
Hi MJM, thanks for the explanation...and i suppose if the conical piston were pointed the maths would still hold up !! ?  Does the surface texture also have any effect on the outcome of these forces..ie turbulence that i think was mentioned in a previous post ?

Willy

Offline MJM460

  • Full Member
  • ****
  • Posts: 927
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #1231 on: August 23, 2019, 01:30:34 PM »
Hi Captain Jerry, I would have to agree with the one word for Willyís first question, but while correct (for the first question), in my view, it is not at all informative.  Besides the post contained two questions. 

But significantly, a one word answer to a question on a complex matter does not fit with my learning style.  It inevitably leads to additional questions such as how? Or why? Or what are the implications? 

A one word answer also seems to me to reduce the subject matter of the question to a collection random unconnected facts to be learned off by heart, like times tables or spelling.  To me, this type of learning rarely leads to understanding.  I still canít spell, and I really like having a calculator handy.  So while my style might justifiably be described as unnecessarily wordy, I hope it does help those who hang in there for enough of the subject, to connect the answer to things they already know, and preferably lead to a little more understanding.  Or at least prompt a few thoughts which might lead to more conversation and eventually to increased understanding.

Hi Willy, if the conical piston really comes to a point, it might result in the maths failing if it leads to a divide by zero for example, but the principle still holds, and the net force on the piston is straight along the cylinder.  However the maths mostly involves the projection of the piston crown onto a flat surface, which involves a simple regular shape, so I believe the divide by zero issue does not occur.

It does not matter if the piston crown is some irregular shape, rough or even an inverted crown to make a cavity in the top of the piston.  The total projected area in each direction is still the same, even if there are undercuts.  Though the internal stresses in the piston will be more complex.  And of course the side walls of the piston should preferably be smooth unless the rings are dimensioned to keep the piston from touching the cylinder walls.  The analysis of forces on piston rings and o-rings all relies on the same principle of pressure being the same in all directions.

In more advanced engine design, (another subject out of my league, but others are clearly well into it), the inlet to a cylinder is directed to give a swirl or perhaps a well distributed combustion pattern, or to avoid hot spots etc.  Certainly fluid velocities involved in these ideas will involve friction effects on the piston which are certainly not balanced in the way the hydrostatic pressure is balanced.  My answer is based on static forces due to pressure, and does not account for such dynamic effects.

Mind you, I would expect that the hydrostatic pressure forces, and the inertial forces due to the rotation of the engine and the continuously changing angle of the conrod are each much larger than the friction forces due to gas inlet velocities.  But whatever friction forces there are simply add to the hydrostatic pressure force.

I hope that helps,

MJM460

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

Offline Captain Jerry

  • Full Member
  • ****
  • Posts: 1028
  • Summerfield, FL USA
Re: Talking Thermodynamics
« Reply #1232 on: August 23, 2019, 07:42:28 PM »
I mean no disrespect to you or Willy by offering a one word answer.  Of course it brings up a lot of questions and that is how understanding is developed, not by rote memory. However it is very difficult to anticipate all of the questions and so the discussion of a silly question that you haven't anticipated is often worth having.


Resolving the force vectors on a conical piston face makes it easy to see that the horizontal vectors all cancel out at the axis of the cone, but suppose they don't cancel out. Suppose there is no corresponding opposite force?  What happens if the piston is cylindrical but the top face is the result of a diagonal slice...a flat surface at 45 degrees to the axis of the cylinder? It seems that all of the horizontal vectors are in the same direction.


Question #2. How would you go about making the piston rings?


Jerry
NOTARY SOJAK

There are things that you can do and some things you can't do. Don't worry about it. try it anyway.

Offline steam guy willy

  • Full Member
  • ****
  • Posts: 2252
Re: Talking Thermodynamics
« Reply #1233 on: August 24, 2019, 03:15:21 AM »
Hi MJM and Jerry,  The thing about delving in deeper with accepted wisdom and established facts is that there is always an exception to make the rule !!!  I do always have more questions with the answers given  to replies but they tend to sort themselves out with deeper thought and sleeping on it ...!!  Keep up with the good work with this fascinating subject...

Willy :) :)

Offline MJM460

  • Full Member
  • ****
  • Posts: 927
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #1234 on: August 24, 2019, 09:05:35 AM »
Hi Captain Jerry, I know what you mean about it being difficult to anticipate all the questions that arise out of a particular answer.  Indeed, some of my longer replies have been due to that exact problem.  Often the more detail I put into my reply, the more aspects I see that ought to be covered, but in the end I have to draw a line.  On the other hand, there are times when a one word answer will do.  It depends on the context I suppose.  But your input is always welcome.  I much prefer conversation to one question, one answer.  Most things are more complex than that.

The cross section of the piston crown does not have to be symmetrical for all the horizontal components of force to be balanced out.  It is not easy to picture perhaps, but if a section is more parallel to the axis the area on which the pressure acts is smaller, while a section that is nearer flat as we normally understand the top of a piston involves a larger area spread over the same increment of height.  That is where calculus comes in, it basically divides the whole surface into small increments, each small enough that it can be considered a flat surface to which only one angle applies, then adding up all the horizontal components of the force on each small bit of the area.  It does not matter that your piston crown has a 45 degree slice, it will still be balanced by the components on the other side of the piston, and there is no sideways thrust due to the static pressure.  An inlet stream directed at that surface might provide a sideways thrust, but I believe that force would be very small compared with the pressure forces and the conrod forces on the piston.  Whether it is large enough to cause noticeable wear on one side of the cylinder, I donít know.

Machining of piston rings is something I still have to learn.  I see Brian sometimes buys ready made ones, but many other threads describe how they are made by people who know far more about the procedure than I.

Hi Willy, sleeping on the question is a good way of using hidden brain resources to sort something out, it is an amazing and very peculiar part of how our brains work.  But as you know, you are always welcome to come back with follow up questions if I have not addressed the aspect that puzzles you, or if in the morning it is still not clear.  But as to the exception to every rule, it usually means the rule is incompletely stated, or perhaps even wrong.

Thanks everyone for looking in,

MJM460

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

Offline Captain Jerry

  • Full Member
  • ****
  • Posts: 1028
  • Summerfield, FL USA
Re: Talking Thermodynamics
« Reply #1235 on: August 25, 2019, 04:35:05 AM »
Good evening, MJM,


Question #2 about piston rings was mostly facetious but question #1 about a diagonal piston face was very serious and I am afraid that your answer did not resolve it for me.  I have slept on it or to be more precise, I lost sleep over it, and yet it remains a question.  My "Crap-O-Cad" skills are not so good and they don't include color so I have had to resort to full-blown 3D Alibre to help me understand it and to be able to share it here as well.


Alibre' can output PDF files for general sharing so I have included a list of PDF files as attachments.  If you are not able to see these, please let me know and I will see what else I can get.  I can output .STEP files if you have a 3D cad that can read them.


The files show several views of a Green Piston in a transparent Cylinder.  The top of the cylinder is sloped at 45 deg to the axis of the piston and the face of the piston is colored Red.  For the purpose of discussion, the piston is a perfect fit in the cylinder so we can ignore any by-pass pressure acting on the flank of the piston...we are only talking about pressure acting on the Red face of the piston and forces acting on the exposed inner piston wall as well as on the cylinder head are also not part of the question, only those forces acting on the red face of the piston


The second PDF shows a view of the face straight on and of course this is an elliptical shape and the pressure in the cylinder, acting perpendicular to  the face is at a 45 deg angle to the axis of the cylinder.  If this force is resolved into two component forces, there is a vertical force (downward) on each integral location on the face.  There is also a horizontal force acting on each integral location and because it is acting at 45 deg to the surface, it is a horizontal force and is exactly equal to the vertical force.  Unlike an irregular piston surface, all of these resolve force vectors are parallel and in the same direction... there are no off setting (negative) vectors as a result of the pressure.


There is also a view of the piston from the top and the piston appears as a Red circle and the area of this circle that the vertical force vectors act on.  This force is countered by the piston rod and crank and can be measured (or calculated).  There is a bottom view of the piston as well and the bottom face is green which indicates that there is no cylinder pressure acting on this face.


There is a "front view" of the piston and the top surface appears as a Red circle of the same size as the Red circle when viewed from the top and this represents the horizontal forces acting on that face and this force is exactly equal to the vertical force. For clarity, we can say that this force is acting from "front" to "back."


Now, there is a "back view" of the piston and it can be seen that it is all "green", an indication that it is not seeing any cylinder pressure to off set the force from the piston face.  The only thing preventing the piston from being blown out the back wall is contact with the back wall of the cylinder.


There is also a side view showing that there is no left-right or side to side force on the piston.  In this view, the piston is all green.  The view from left and right sides are equivalent.


THEREFOR: It has been shown that for THIS configuration, there is force causing the piston to bear on one side of the cylinder wall and the force is considerable.


Jerry


PS: How's that for wordy?



NOTARY SOJAK

There are things that you can do and some things you can't do. Don't worry about it. try it anyway.

Offline AVTUR

  • Full Member
  • ****
  • Posts: 56
Re: Talking Thermodynamics
« Reply #1236 on: August 25, 2019, 06:03:01 AM »
Jerry (and all)

Good morning

I have been following this but with quite a time delay. Therefore I have not entered the discussion. Also, I know nothing about sugar solutions.

I cannot open your .pdf files. Can you print and then scan them as .jpgs?

AVTUR
There is no such thing as a stupid question.

Offline MJM460

  • Full Member
  • ****
  • Posts: 927
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #1237 on: August 25, 2019, 11:39:58 AM »
Hi Captain Jerry, now I see the issue you are thinking of.  The words did it, I am another having difficulties with the pdfís, they appear for a second or two then become a blank page.  Very puzzling, as the thumbnail and the file size still seem visible and complete.  I have seen this issue before, but not sure of the cause.  I would suggest printing then scanning either as jpg or PDF, as scans saved in PDF format seem more reliable.  I donít know if the cad programme PDF format is not quite standard, at least not to Apple requirements.

For the problem as you have described it, I have to agree, the piston would be pushed against the cylinder wall by the unbalanced force.  Fortunately the forces on the cylinder are still balanced, so the engine will not rotate around its crankshaft, or not due to this cause anyway.

 It then it is my turn to ask the ring question.  Do you think it possible to make rings around the cylindrical walls, and following that 45 degree face, to create this situation in a real engine, and if so, why would you want to?

(I should have attributed that question to its source, but I donít know the name of the person who posed it to Alexander Bell, but we all know the answer!)

However, you are correct, and my answer to Willyís question does not apply to this configuration.

I am also thinking about real engines in this context.  Could this configuration really be created?  If we have very smooth flat surfaces, and we press them together so as to exclude the air between them, they stick pretty well due to atmospheric pressure on the other side.  Gauge blocks can be wrung together in this way for example.  Blocks wrung together in that way do not slide very easily, so the question arises if we could make a piston to such a perfect fit, would it actually slide up and down as required for the engine to run?  I am not sure that it would be possible to make a piston and cylinder with such a fit all around, but I will leave it to others to answer that.

On the other hand if we assume a tiny gap around the piston, there will be leakage past the piston.  There will be a drop in pressure as the fluid flows down the cylinder wall where the piston wall is present, and I think we can see that the profile on the side with the short wall will be different from the profile on the longer side.  (Draw it out to make it easy to see.). This again leads to the conclusion that again, the force is unbalanced.

However in this case we are dealing with a dynamic pressure change as the fluid flows down the wall, so my answer has to be qualified as a static pressure analysis, with negligible change of elevation.

If we have a conventional well fitted ring located at a point where it can extend right around the piston, there will be gas pressure between the cylinder wall even where the gap is very small.  The horizontal forces above the ring will be balanced across every diameter, whether the wall is vertical or sloped at that point on the circumference.  And the forces below the ring will be balanced by symmetry, whether they are classed as static or dynamic pressures.

Does that adequately address your concerns?

But certainly, this is another case where the left field question leads to further understanding.  I hope this clarification actually clarifies, and that you do not loose anymore sleep.  I donít want to cause that!  And I hope that between us, we have not added confusion.  But if things are still not clear, please continue the conversation (after some sleep!)

Hi Avtur.  I also had troubles with those pdfís, but I am not really an expert on sugars.  I did just on spec, try searching for the chemical composition of sugars and found some very good articles among all the usual rubbish.  To open them without any background would have been truly intimidating.  However the similarities with the structure of the hydrocarbons, which I am familiar with, were striking.  It made them relatively easy to read enough to see how it applied to the question at hand, and I hope it allowed me to give a reasonable answer to Willyís question.  But it certainly did not make me an expert on all things sugary.  Most of us have knowledge in our specific field that can help us understand things in other areas if we can find that initial similarity.

Thanks for looking in,

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

Offline Captain Jerry

  • Full Member
  • ****
  • Posts: 1028
  • Summerfield, FL USA
Re: Talking Thermodynamics
« Reply #1238 on: August 25, 2019, 03:48:27 PM »
Good morning MJM,


So you see that the piston ring question was not totally facetious.  It is relevant only as it is used to define the shape of the expansion chamber and the intersection of the face of the piston with the cylinder walls.  As you point out, if the piston ring is square to the cylinder then all horizontal vectors above it resolve to zero.


The piston ring could be an O-ring but it was the problem of cutting the ring groove parallel with the face that disturbed my sleep.  Some people may be able to solve a problem while sleeping, my sleep engine seems to have a way of creating a problem.  The ring problem is not the ring. It is the sealing of the piston (back to the hobby aspect) but since this is a theoretical experiment, I chose to provide the seal by defining it as a perfect fit.  I suppose an oil film could do it well enough and would also solve the friction. 


The engine rotation problem that you mention could be solved by alternating the orientation of the slope in a multi cylinder engine or it could be put to good use in a radial engine.


To those that tried and failed to open the .pdf files above, I apologize.  I think the problem is that my aging version of Alibre' outputs an obsolete version of .pdf so I have attached jpeg files that should open.


I too, hope that this has not muddied the waters by stirring up the bottom but it should be remembered that no laws were broken or even bent.


Jerry
« Last Edit: August 25, 2019, 04:11:30 PM by Captain Jerry »
NOTARY SOJAK

There are things that you can do and some things you can't do. Don't worry about it. try it anyway.

Offline steam guy willy

  • Full Member
  • ****
  • Posts: 2252
Re: Talking Thermodynamics
« Reply #1239 on: August 26, 2019, 01:24:59 AM »
Hi MJM , well ,i wasn't expecting quite as much discussion as this has generated but it is good that this concept has been explored...it is one of those questions that may seem intuitive but then becomes much more complex ..so thanks to everybody for chipping in   :D :D I did have more questions that would have followed on ,but they seem to have been explored by others...

So thanks to everybody for looking in

Willy

Offline MJM460

  • Full Member
  • ****
  • Posts: 927
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #1240 on: August 26, 2019, 12:54:00 PM »
Hi Captain Jerry, the extra conversation is most welcome.  Doing the ďthought experimentĒ to simplify a complex problem to a simpler one that can be analysed is often a useful exercise. 

I had certainly not considered the particular case, as I was only thinking of a piston with a crown that was not flat.  So normal rings, when the crown does not have to be symmetrical.

The analysis assuming a pressure profile through the leak path around the piston does indeed support your thought that there would be a sideways force.  Even with an oil film, which will also carry the static pressure.

But I return to the question of whether this would be useful.  The friction due to the side force would presumably be enormous.  The normal symmetrical force distribution minimises this, and with a flat topped piston, or a symmetrical one, and sufficiently close tolerances, an oil free compressor can be built, that relies on grooves without rings to create a Labyrinth to minimise the leakage flow, and relies on the fact that there are no sideways forces on the piston to cause contact with the cylinder walls.

Similarly we can look at a conventional slide valve steam engine.  In larger sizes and with high pressure steam, the out of balance forces on the valve surfaces create such enormous friction that piston valves are the preferred solution in those cases.

So quite a bit of learning involved due to your out of the box thinking, thank you.

Hi Willy, certainly the unexpected leads to more creative thinking.  It is fascinating, the turns which this thread has taken.  Glad that you are still following along.

MJM460




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

Offline derekwarner

  • Full Member
  • ****
  • Posts: 526
  • Wollongong ...... Australia
Re: Talking Thermodynamics
« Reply #1241 on: August 26, 2019, 01:20:07 PM »
 :old:.............message deleted ..........Derek
« Last Edit: August 26, 2019, 01:23:49 PM by derekwarner »
Derek Warner - Honorary Secretary [Retired]
Illawarra Live Steamers Co-op - Australia
www.ils.org.au

Offline steam guy willy

  • Full Member
  • ****
  • Posts: 2252
Re: Talking Thermodynamics
« Reply #1242 on: August 27, 2019, 01:11:11 AM »
Hi MJM, A more practical question now..we have learnt about labyrinth seals ..so are there tables for best practice with cylinder diameter and pressure.?  Say a 20 mm bore and 50 Lbs  square inch..with a 8 mm thick piston... also will the individual grooves vary ??. so number of grooves ,width and depth ?? Another short question !!!  Also with this weeks bath i switched off the temp gauge between  readings .As last time the gauge read a lower reading after 10 mins  and yes it did, it went from 24  to  18 after a while I was wondering if the meter would also read a lower reading from ambient when first switched off after 10 mins ?? !! Really Hot today 33 celsius !! Also with labyrinth grooves ...if you had a really long cylinder with a very long piston could you have enough grooves to actually stop the piston moving ??

willy
« Last Edit: August 27, 2019, 01:57:24 AM by steam guy willy »

Offline MJM460

  • Full Member
  • ****
  • Posts: 927
  • Melbourne, Australia
Re: Talking Thermodynamics
« Reply #1243 on: August 27, 2019, 12:26:31 PM »
Hi Willy, while conventional piston rings attempt to block off the leakage path between the high pressure above the piston and the lower pressure underneath.  When they are made in one piece, with a small gap, that gap is the major leakage path for a well fitting ring in a smooth cylinder.

Labyrinth pistons do not attempt to block the flow path, but simply to reduce the leakage flow to an acceptable low level.  There is a complete, though small gap between the piston and the cylinder through which gas can flow.  The labyrinth is constructed by grooves in the piston.  When the gas flowing down the wall reaches a groove, it expands into the greater flow area and because it is a sudden expansion, there is little or no pressure recovery, Bernoulli does not apply to sudden changes in the flow passage, there is loss of the kinetic energy of the gas into turbulence.  It is then accelerated into the small gap separating the grooves, which again requires energy which comes from the static pressure.  So at each groove there is more pressure loss for a given flow than if the groove was not there.

The total pressure difference between the piston top and bottom is determined by the operation so all the pressure losses result in a smaller flow. 

To be really useful in a high pressure situation, you need a long piston with many grooves.  However, in my small steam engines, I take the approach that grooves result in less blow by than a smooth piston wall, so canít do any harm.  In reality my piston is probably too short and too few grooves to make much difference.  But having machined the grooves when the piston was on centre in the lathe, I can come back and add some cotton or graphite packing, which will almost certainly be more useful.

In full size, the principle is used for oxygen compressors by one company that I know of, you can search for labyrinth piston oxygen compressors.  They are expensive beasts so really only used where there is no other viable alternative.  They have to be very accurately made with very small clearances.  I am sure that they would have guides or calculation ,ethos to give the required number of grooves, but they would not be likely to publish that.

I have described the action as though it is all static, but of course the piston is moving in the normal way, and this complicates the flow analysis, but you can see the piston is surrounded by a flowing gas, so I feel that the actual drag on the piston will be quite low.  The drag is mostly due to the viscosity of the gas which is, of course, much lower than the viscosity of any oil film.  I canít see that it would ever actually stop the piston, more likely it lowers the friction drag a little, but I donít want to have to try and prove it.  I hope that answers the question.

With regard to your meter, I would be suspicious of the batteries.  I donít know what sort of power supply it uses, but in my experience, cheap measuring instruments rely on a stable power supply as Roger stated.  If the battery voltage tails off as the battery is used, that will affect the calibration.  New batteries of the best possible quality may help you get more consistent readings.  With button cells, you need the Silver or Lithium chemistry, depending on the voltage your instrument needs, not alkaline batteries.  If it uses a 9V or normal AA or AAA batteries, you need reasonably fresh ones of the best quality you can get.

Hi Captain Jerry, one further thought on your piston with the angled top, when we were talking about potentially being useful for a radial engine, we missed the point that the forces on the cylinder are the same whether from the piston or directly from the gas and always balanced for every case I can think of.  So I think that leaves us with the friction as the most obvious effect of that arrangement.

Hi Derek, good to see that you are still looking in.

Thanks to everyone looking in,

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

Offline Captain Jerry

  • Full Member
  • ****
  • Posts: 1028
  • Summerfield, FL USA
Re: Talking Thermodynamics
« Reply #1244 on: August 30, 2019, 01:52:23 AM »
Model engine pistons don't have to fit all that well, at least mine don't.  They don't have to do anything but run smoothly so I can appreciate the symphony of motion. Loose fits make smoother movements with less friction and they will run slowly, almost without effort so I can run them without the hissing sound of air bypass and compressor noise.  That is one way to enjoy your time in the shop.


But it leads to a question.  Why don't we run our model engines on higher pressure, say in the range of 50-60 psi?  Or even 80-100 psi? Because with no work to do they would overspeed and tear themselves apart or jump off the bench.  They would need a governor to limit their by reducing avail air pressure or volume and/or both and if we load them up with some kind of resistance like a friction brake, their air leaks and resulting condensation running every where would be embarrassingly obvious. 


To specify a volume of air without also specifying the pressure and temperature has no real meaning. If an engine will run nicely on a given volume of air at 10 psi, why can it not be made to run well on 1/5 of that volume at 50 psi, assuming a constant temperature?


I seem to be spending most of my shop time these days devising methods of eliminating air leakage and bypass and ways to measure the results even if only on a comparative basis.


Jerry





NOTARY SOJAK

There are things that you can do and some things you can't do. Don't worry about it. try it anyway.