[Dwight Ennis]

Concerning cylinders...

Let's say you have a bronze cylinder with a 1" bore and a 1" stroke with a boiler pressure of 100psi. The cylinder is 1.50" outside diameter.

1 - how thick do the walls have to be? Punching it out to 1-1/8" bore would leave 3/16" walls. Punching it out to 1-1/4" bore would leave 1/8" walls? Does reduced wall thickness, and thereby mating surface area with the cylinder end cap, impact gasket sealing?

2 - how does cylinder length affect power and performance? If the stroke remains the same, does lengthening the cylinder (thereby increasing the volume of the steam chamber each side of the piston) impact power and performance? Does the added volume increase the cylinder fill time enough to rob power assuming valve timing remains the same?

 

[Semper Vaporo]

Your questions are quite valid... may I recommend

"La Locomotive a Vapeur"
by Andre Chapelon,
Translated by George W. Carpenter.
ISBN No. 0 9536523 0 0

If it don't answer your questions, you are on your own, forging new ground.

 

[Chris Scott]

I thought they were called cylinders not theory?

You might look at the archives for Dave Hottmann's posts on his Ruby theory enlarging. I think he had a fair amount to say about theory wall thickness.

 

[Semper Vaporo]

Sorry, I am going to withdraw my "recommendation". Chapelon's work was more in the line of comparing past and present (his era) locomotives of many various countries and designs and trying to make sense of disparate data taken on them. Besides, it is apparent that in his era most people were very concerned with maximizing the use of exhaust to improve the draft over the fire... without "fire" there is no steam to run the thing.

I would recommend a work by Porta, but I have none (yet) to actually recommend. His work also centered on the exhaust in the smokebox, but I know he also worked on the sizes of the steam passages, and cylinder and valve designs.


I cannot expound on the cylinder thickness question except to also mention the wall thickness needed to contain the threaded holes for the bolts that hold the end caps to the cylinders.


I don't have data at hand about your second question to discuss it well. You do need to have large enough passages to get steam to the cylinder and short enough that there is no time delay getting the pressure there. Of course the bigger the cylinders the more steam will be removed from the boiler on each stroke, which means you need a bigger fire to replenish it... and we are back to a good exhaust system to draw a good draft.

 

[llynrice]

Hi Dwight,

For the original for a cylinder with a 1" bore and 1" length, the force trying to break the cylinder open at 100 psi would be 100 pounds force. If you section the cylinder lengthwise, the total cross sectional area of the walls would be (1/4" x 1") + (1/4" x 1") = 0.5 square inches. Divide the 100 pound force by the area and the tensile stress is 200 pounds per square inch. I don't know what bronze alloy you have; but, Machinery's Handbook shows a yield strength ranging from 20,000 to 50,000 pounds per square inch. Even if you cut the wall thickness in half - yielding a stress of 400 pounds per square inch, it would appear that you have a really significant factor of safety. My calculations are somewhat simplified; but, give a notion of what you are working with.

Keep in mind that screw holes drilled into the walls to keep the end caps in place will weaken the cylinder some as well.

Llyn

 

[HMeinhold]

Dwight,
here my 2 cents:
#1 If you make the walls too thin, you don't have enough meat for the cylinder cover screws. If you tap the holes after having opened out the cylinders you might even distort the cylinder walls (guess how I know...).
#2 Increasing the cylinder length without increasing the stroke increases steam consumption without increasing power. On full size cylinders these dead volumes are kept as small as possible.
Regards

 

[rbednarik]

Dwight,

Although I am strongly against the hot-rod theory of cylinders bigger than 5/8" (if you want to know why, feel free to email me), I can answer your question as to the stroke of the piston in the cylinder.

-Yes, the amount of space that is at the end of each stroke is crucial to the engine's running characteristics. Too much and you will eat water and fuel to keep up with the excessive steam consumption that is used to make a cushion for the piston. too little and you risk hitting the cylinder head with each stroke. The length of the actual piston itself has a direct effect on the power of each stroke, similar to having higher compression and loower compression pistons (thin and thick respectively) in an automobile.

From the book I am reading by Ralph P. Johnson, who was the chief engineer of the Baldwin Locomotive Works, cylinder thickness should not exceed more than 5/32" and no less than 3/32". Of course I have scaled this down from 1:1 proportions and calculations. Now, personally, I would use no less than 1/8" due to the slight distortion in the bore when drilling and tapping the holes for the heads and valve chest studs.

I would highly suggest the book, as it has a wealth of information and is a great companion to Chapelon's book

The Steam Locomotive
R.P. Johnson

 

[Dwight Ennis]

Thanks for the responses guys. Keep them coming. I should clarify that this will be a 1-1/2" scale loco, not small scale.

Good point about steam/water/fuel consumption. I hadn't even thought of that. Shows what a raw recruit I truly am. hehehe

 

[Bob Starr]

Dwight,
Obviously, your question is in reference to your new toy. The answers you have received have referenced full size steam and that can't compare to our models. Basically, because the coefficient of expansion is going to have an effect on things. Your cast irop cylinders will expand less than the brass ones in our smaller scale, but it still doesn't take much for steam to get past the piston. The piston is not the only place you will have problems, the bolt holes will also expand at a different rate than the bolts.
If you expand the length of your cylinder, it will be detrimental to power and this is why. The idea is to get the piston moving the opposite direction as fast as is possible. Remember is essentially going over a hump at TDC. By increasing the volume at TDC you will lower your pressure and will require more steam to start the piston on his next journey.
Personally, I would keep the .25 walls that you now have, .125 is definetly too thin. .1875 is borderline and general practice in one inch bore has always been .25 or thicker.
Does your engine have piston rings?

 

[rbednarik]

Dwight,

Had no idea this was for 1" locos, so I'm afraid that my input is not much use to you.

1" is not my area, although the full size principles and applications can apply in certain situations. I have used the calculations on Ga1 locos before and they do work.

 

[Kovacjr]

Dwight - As for the Falk I spoke to a guy at Cabin Fever that had built and run one and the major issue is not that the loco cant produce enough power its that the loco is too light. Under 100 LBS I think....He said that adding as much dead weight as he could made the pulling much better but still looking at about just 2 adults as opposed to one. I thought of enlarging the Falk to be 2.5" scale and of course it would no longer be correct but it woule add weight and also increase the boiler by almost double. I still want to build a ride on loco but for now I will have to hope that my friend gets his LE 0-4-0 back together this year so I have the chance to run it again.

Traction seems to always the issue with all the scales.......

 

[zephyra]

On questions of teory, I always turn to "How Steam Locomotives Really Work (Paperback) by P. W. B. Semmens, A. J. Goldfinch". Available from Amazon for about $20.

 

[Dwight Ennis]

Actually, it isn't for the Falk - it's another loco I'm contemplating building.

Yes, it has rings.

 

[Charlie Mynhier]

Wall thickness.

T=RxP/S

T= Wall thickness
R=Radius of cylinder bore
P=Pressure in Pounds per Square Inch
S=Allowable stress of Cylinder material.

 

[toddalin]

Posted By llynrice on 05/22/2008 8:12 PM
Hi Dwight,
For the original for a cylinder with a 1" bore and 1" length, the force trying to break the cylinder open at 100 psi would be 100 pounds force. If you section the cylinder lengthwise, the total cross sectional area of the walls would be (1/4" x 1") + (1/4" x 1") = 0.5 square inches. Divide the 100 pound force by the area and the tensile stress is 200 pounds per square inch. I don't know what bronze alloy you have; but, Machinery's Handbook shows a yield strength ranging from 20,000 to 50,000 pounds per square inch. Even if you cut the wall thickness in half - yielding a stress of 400 pounds per square inch, it would appear that you have a really significant factor of safety. My calculations are somewhat simplified; but, give a notion of what you are working with.
Keep in mind that screw holes drilled into the walls to keep the end caps in place will weaken the cylinder some as well.
Llyn

Sorry, can't agree with your calculations.
Assume a cylinder with 1" bore and 1" stroke. Also assume a force of 100 psi.
Area of the top (and bottom) of cylinder = pi*r*r = 3.1416 * 0.5 * 0.5 = 0.7854 square inches x 2 (for top and bottom) = 1.5708 square inches.
Now add the sides. The area = 2*pi*r x length = 3.1416 square inches.
Total area = 1.5708 square inches + 3.1416 square inches = 4.7124 square inches x 100 pounds/square inch = 471 pounds of pressure pushing on the interior of the cylinder. But you are only concerned with the walls, so 3.1416 x 100 pounds/square inch = 314 pounds of pressure on the walls.
As to the 20,000 - 50,000 PSI for bronze, something is missing here (like wall thickness). "Square inches" has only two dimensions and fails to consider the thickness of the walls. Certainly a sheet of bronze foil (the thickness of Reynolds Wrap aluminum foil), would not hold 20,000-50,000 psi. /DesktopModules/NTForums/themes/mls/emoticons/pinch.gif"

 

[llynrice]

Hi Todd,

You seem to be confusing the mechanical stress in the cylinder walls (measured in pounds per square inch) with the steam pressure in the cylinders. Stress in a metal part is calculated by dividing the force applied to the part divided by the cross sectional area which yields a stress value given in pounds per square inch. If you were to put a bronze bar which is one inch square in a tensile testing machine, the bronze would not begin to permanently deform until you applied a force of 20,000 to 50,000 pounds (depending on the alloy). If you were to put a piece of foil one inch wide by 0.005" thick in the machine (cross section = 0.005 square inches) you would reach stress of 20,000 pounds per square inch by the time the tensile test machine reached a pulling force of 100 pounds.

The forces we caluclated are in the ball park of one another. In my calculation, I chose to take a lengthwise cross section. If cylinder wall were 1/8" thick and the cylinder were 1" long the area of the cross section for the wall on each side would be 1/8 x 1 = 0.125 square inches. If you take the cumulative area of the walls and divide that into the force generated by the steam pressure (let's use the 314 pounds force which you calculated) and divide it by the area of the cylinder walls (314/(0.125 + 0.125) = 1256psi) The 1,256 pounds per square inch of stress in the cylinder walls (caused by the 100psi steam pressure) is very much less than the 20,000 to 50,000 psi that would be required to make the bronze yield. In other words, Dwight would have a substantial margin of safety if he bored out to have a 1/8" thick cylinder wall. If he were to substitute a 0.005 thick foil cylinder wall, the stress calculation would be (314/(0.010 + 0.010) = 31,400psi). Obviously, a foil cylinder would be at high risk of failing if the steam in the cylinder had a pressure of 100 psi.

Llyn

 

[Slipped Eccentric]

Dwight, if I may ask, what are you contemplating building? And how many people are you planning on pulling (you and a friend or two, or everybody and all the rolling stock around)? I'd be more comfortable with the thicker cylinder walls myself, though they my not be absolutley necessary.

 

[Dwight Ennis]

Justin - I've been going back and forth between Kozo's Climax and Heisler... having a hard time making up my mind. When I got the Falk, I bought the book, "Logging with Steam", and in it are plans and instructions for building Mich-Cal #2 in 1-1/2" scale. I think I've finally settled upon building that.

The book calls for using a set of cylinder castings from Stuart, and though these castings are still available, they would cost me $383.00 if I bought them today under current exchange rates. I'm therefore considering making my own based upon Kozo's cylinder designs. Kozo uses a 1-1/2" bore on his scaled up cylinders, while Harris uses 1" on his Mich-Cal. Just seemed kinda small to me, but what do I know? Hence my questions.

 

[HMeinhold]

Dwight,
David has a 4 3/4" tram engine which he originally inherited from me. This contraption is powered by a Stuart D10 with only 3/4" diameter and stroke 3/4". It pulls him happily up a gradient on his layout. With 1" cylinders and 1" stroke you get at least twice the power. And then a Shay is not a race horse either... With big cylinders you might tax the boiler capacity.
Regards

 

[llynrice]

Hi Dwight,

I think Henner has a good point. The Harris and Kozo designs are both proven and mixing and matching might be problematic. With the advanced metal working skills you've shown, I'm sure that you can come up with machined fabrications which are a close replacement for the Stuart castings. Sounds like you are embarking on an exciting new project. Hope you'll post upadates on your progress once you get going.

Llyn