Introduction: Design Criteria for Stirling Cycle Engine

Picture of Design Criteria for Stirling Cycle Engine

This Instructable  will describe a model Stirling cycle engine I built. More importantly, it will list design criteria, materials of construction, and tips so that you can build one of your own design.

This type engine is called a low temperature difference (LTD) Stirling engine, and there are several ways to build one, some of which are described in other Instructables on this site.

Step 1: What Is a Stirling Cycle Engine?

Picture of What Is a Stirling Cycle Engine?

The Stirling cycle engine was invented by Robert Stirling in 1816, so it has been around for a while. It is a heat engine, and is based upon a cycle of heating, then cooling, of a gas (usually air) contained within the engine.  Since a Stirling engine is air tight, during the heating phase the air pressure inside increases, and during the cooling phase the pressure decreases.   A displacer connected to the crankshaft moves the internal air from hot side  to cold side of a cylinder. The change in pressure drives a power piston, which is also connected to the crankshaft. Since there are two stages, hot and cold, it is a two cycle engine

This diagram shows the concept of a traditional LTD Stirling cycle engine.  The components of the engine described in this Instructable  are arranged differently, but the concept is the same.




Step 2: Components

Picture of Components

Here are the components of the engine I built, which are similar to those shown in the previous diagram. I used copper for many of the components due to its excellent thermal conductivity. I used aluminum also; its thermal conductivity is good but not as good as copper. Both materials are very easy to work using common shop tools.

The steps that follow will describe  design details of the flywheel and  crankshaft,  displacer cylinder and displacer piston,  and the power piston.


Step 3: Flywheel and Crankshaft

Picture of Flywheel and Crankshaft

I built the flywheel from a sheet of aluminum; cutting it roughly on a band saw then finishing it on a belt sander. The “hub” is made from two strips of brass bar stock, attached with machine screws and nuts. This helps align the flywheel and prevents wobble. I drilled the hubs and flywheel with a 5/32" drill and pressed in the 5/32" shaft; it was tight enough.

You can see in the photos how the crankshaft and crank pin are designed.  I drilled and tapped a brass bar to thread onto the threaded 5/32" flywheel shaft. The crank pin is just a machine screw that threads onto a drilled and tapped hole in the brass bar. This way it is easy to change the crank throw distance; just drill and tap a new hole at the desired distance. Also, the phase angle can easily be changed by loosening the 5/32" nut on the crankshaft, rotating the crank throw, and re-tightening the nut.

Design details:

  • Flywheel:   Aluminum, 0 .100” sheet, 5” diameter
  • Crank throws:  Brass bar stock; 3/32” x ¼”; drilled and tapped for shaft and crank pin
  • Crank pins: Machine screws and nuts, #2 screws  48 tpi
  • Shaft:   Brass rod, 5/32” diameter.
  • Bearings:   Ball bearings, 5/32” ID,  5/16” OD (from Boca Bearings)
  • Hub: Brass bar stock;  3/32” x ¼”
  • Crank throw distance, piston: 5/16” (so piston travel is 5/8”)
  • Crank throw distance, displacer:  1.0”  (so the displacer travels 2.0”)
  • Phase angle between power piston and displacer piston: 90º
Tip: When you make any round items (like the flywheel) be sure to start by marking it out with a compass and mark the center with a punch.  Knowing the exact center will be important later.

Tip: Brass and aluminum are easy to drill and tap. I don’t always use  a proper tap, I just drill a slightly undersize hole and then use the screw as a tap. The screws are stainless steel so harder than brass or aluminum. I did use a die for threading the shaft. Use 8-32 threads for 5/32” diameter shaft.







Step 4: Displacer Cylinder and Piston

Picture of Displacer Cylinder and Piston

The displacer cylinder and piston can be considered a heat exchanger, because heat from a flame (in this case) is being transferred to the air inside the bottom part of the cylinder. When the displacer piston moves to the bottom of the cylinder, the air moves (is displaced) to the top part of the cylinder, which it cools. The engine makes one heating/cooling cycle each revolution of the flywheel. Since the system is air-tight, it also has an air pressure increase/decrease cycle each revolution.


The displacer cylinder is made from two sections of 2” diameter copper pipe, separated in the middle by a wood spacer. The wood serves as a thermal break between the hot section below and cooler section above. The plate at the bottom is also copper, soldered to the pipe section using 95/5 lead free plumbing solder.  This solder has a melting point of 450° F (232° C), so the heat source has to be controlled. The operating temperature using two candles is 250° F (121° C). Another limitation is the temperature of the epoxy glue used to attach the copper pipe sections to the wood spacer; ordinary epoxy adhesive is OK to about 350°F  (177°C). The cylinder head is aluminum, and is bolted to a mating flange, which is  glued to the upper cylinder. The cylinder head is metal-to-metal contact with the upper copper cylinder, so there can be heat transfer between cylinder and head, therefore both cylinder and head can radiate the heat away and cool the air inside. I used a piece of Teflon for the bearing in the cylinder head.

The displacer piston is made from heavy paper wrapped around a 1 1/2” pipe (actual diameter 1.9”) and glued together, the ends are foam board. Shaft is 0.0625” steel wire. The displacer length is 2.0”.


Design details: 

  • Displacer cylinder: 2” type M copper pipe, ID = 2.0”, OD = 2.125”. Gap between hot side and cold side = 3/8”.

  • Insulating spacer: hardwood, ID = 2.125”, drilled with 2 1/8” hole saw, then used epoxy to glue copper pipe sections to wood spacer.

  • Displacer piston: 1.9” diameter x 2.0” long. As a general rule, displacer piston volume should be ½ total displacer cylinder volume. Bearing in cylinder head is Teflon, with drilled 0.070” hole for 0.0625” displacer piston shaft.

Tip: The present cylinder head is 4” x 4” aluminum, 0.10” plate. For better heat transfer, it could be larger and made of copper.

Tip: Since the displacer piston volume should be 1/2 the volume of the displacer cylinder, I made the  piston length 2", the total length of the cylinder being just over 4". The piston should be a "loose" fit in the cylinder; it is important that the piston not rub on the cylinder wall.



Step 5: Power Piston

Picture of Power Piston

The power piston and cylinder drive the engine, using pressure cycles from the displacer cylinder. They are the only purchased components in this engine. The piston is made from graphite, the cylinder is glass; the parts are manufactured to close tolerance and are practically friction free and air tight.  I glued the glass cylinder into a 3/4" diameter hole bored into a block of wood; opposite the cylinder is a short length of 1/4" copper tubing.

Design details:

  • Power piston and cylinder:  Piston diameter 5/8”; piston stroke 5/16”. Purchased from Airpot Corporation (no, not  the same company that makes coffee pots)
  • Connecting rod: 3/32” brass wire soldered to brass sheet.


Step 6: Power Calculations and Performance

Picture of Power Calculations and Performance

I used a shop built dynamometer to calculate power output of the engine.

Power output of any rotating engine is based upon only two factors: torque and RPM.  I used the device shown in the photo to measure torque, and could count revolutions to determine RPM.  To measure torque, I made a  prony brake out of balsa, then regulated the clamping force around the flywheel shaft by tightening a screw until the RPM decreased, and recorded the force on a digital scale.
P=Power, W
T=Torque, Nm
N=Engine rotational speed, RPM

Here are the equations, data and  calculation:
P =T * π * N/30  or approximately T * N * 0.1047

RPM = 140
Torque arm = 200 mm = 0.2 m
Net scale reading, maximum: =1.7 g = 0.0017 kg
Force = m * g =  0.0017 * 9.8 = 0.017 N
Torque = f * d = 0.017 * 0.2 = 0.0034 Nm
Power =  T * N * 0.1047 = 0.0034 * 100 * 0.1047 = 0.05 W

Not a huge power output, which is why minimum friction is so important.

Other operating data:
Maximum RPM = 175
Minimum temperature difference = 110ºF  (43ºC)


Step 7: Tools

Picture of Tools
I built this engine in a woodworking shop, with no special tools. Both aluminum and brass can be cut with carbide tipped saw blades, but use care. When cutting thin metals on a table or band saw, it is safer to first attach them with double sided tape to a carrier sheet of plywood.

Here are specific tools:

  • Set of small size wrenches (my only splurge)
  • Drill press
  • Table and/or band saw
  • Butane torch for soldering copper
  • Hole saw



blueandmyboys (author)2014-02-01

can you get electricity from this in any way

MenuaC (author)blueandmyboys2016-04-03

yea to power a tiny led

Bill WW (author)blueandmyboys2014-02-01

Excellent question.

This particular engine does not have enough power to drive a generator. But yes, a Stirling engine if large enough, and if it has a great enough temperature difference, can generate electricity.

blueandmyboys (author)Bill WW2014-02-01

Thank you. I found a much cheaper to build stirling out of tin cans on YouTube. Its really nice man, I'll use a concept similar as yours but with wood to make a buggy

Bill WW (author)blueandmyboys2014-02-01

Oh, yes! My engine here is not the cheap or easy option!

blueandmyboys (author)Bill WW2014-02-01

I'm starting to make one that has to do with popsicle sticks. This is a photo of the beginning stage help me out if you can please sir.

Bill WW (author)2013-11-12

Thanks for looking, Tesla.
Yes, the power I calculated is embarrassingly low. I have done that test several times and re-calculated, but still get 0.05 W.
There are other examples of model Stirling engines connected to a small generator, so your goal is not unrealistic. A greater delta T will definitely help; I was limited to around 350F max due having used solder in the construction of the displacer cylinder (duh!). Realize also that a larger power piston diameter and/or stroke will increase power IF you have aequate pressure from the displacer.
Look at the comments from Rimar2000 for additional suggestions.
Let me know if I can help further .
Good luck.

gaieb (author)Bill WW2015-06-16

Back in the late 70s, I was working for a company that produced a sterling engine that ran off a carbonpile reactor, produced 1kw of electricity. It used pressurized helium as the working fluid. I went into some satellite that was needing a source of power without using solar. If I remember right the whole thing was less than on cubic feet in size.

MenuaC (author)gaieb2016-04-03

yea and I built an engine that is perpetual lol ;)

NyczlJ (author)2015-10-16

Ummm... hello. You have said in your article that you used a shop built dyno. May i ask how much did you bought it? Is it available worldwide 'coz i live in Phil. and i am not sure if it is available in here. I really need it as a part of my research to calculate for the engine's speed. Thank you for your appreciation. :)

jaybrahmbhatt22956435 (author)2015-10-10

I want to calculate the power of stirling engine so juss help me out .
Its a very small sterling engine and so the power will also be very small.
Juss guide me how to start

can u tell me how u have maded engine pls............... i really need help just send on this email .......( secured9958465409@gmail)

my email address is above one pls send me images and how did u made

Thank u

shameem_78 (author)2014-04-28

This is great work. Have you tried using steel wool in the displacer instead of foam board and pipe? That way the displacer could also act as the regenerator.

I have been trying to build the 4 cyl swashplate engine described here - - however time and skills are not my greatest friends and haven't made an y progress in a while.

Bill WW (author)shameem_782014-04-29

Thank you Shameem -

I have thought about using a regenerator, the steel wool is a good idea, will try.
I saw the swashplate when I looked thru some of your Youtube postings.
I have the same problem with time and skills - and often trying to do precision mechanical work with carpentry toold.


snoopindaweb (author)2014-04-28

Thanks for the treat.

Bill WW (author)snoopindaweb2014-04-28

Always glad to hear from a corrugator operator again.

Thanks for the comment.

snoopindaweb (author)Bill WW2014-04-29

A big A flute Thank You.

Triangle_Stirling (author)2013-12-19

Very nice engine you have there! is is very well built.

I have a question. I'm from Europe, so we work with Newton/Meter.
How do you get the number of 0.1047?

I calculated this myself, and i'm getting 0.1033
From Wikipedia i get 1 Hp = 33000 ft*lbf /min.
That's 44.741,99235 Nm/ min

Then i get:
P (hp) = T (N/m) x f (rpm) / 7120
P (Watt) = T(N/m) x f (rpm) / 9.68

1/9,68 = 0.1033

am i doing something wrong? Maybe you can tell how you calculated the 0.1047 ?

Thank you!

Sorry for skipping over the derivation and just giving a "magic number".
The basic equation for calculating power from RPM and torque is:
P (Watts) = T*N* π /30 = T*N*0.10472
It can also shown as:
P (Watts) = T*N* π /30 = T*N/9.55

The above equations derive from the basic definition of mechanical power; the time rate that energy is produced: Joules per second, which is same as Newton-meters per second, which is same as Watts.

Bill WW (author)2013-12-02

Thank you.

I will likely have more on Stirling engines later.

heathbar64 (author)2013-11-26

Bill, thanks for sharing this. I've played a little with stirlings, but this is the most lucid and detailed info I have seen.

jott_1 (author)2013-11-15

Nice job.
Do you have an instructable for the dynamometer you used?

Bill WW (author)jott_12013-11-15

Thank you Jott.

Glad you asked about the dynamometer! I have the parts on my workbench right now for a generic dynamometer that can be used for small electric motors. I will be using it for fractional HP 120 volt motors, but it could be scaled up or down for other sizes.

jott_1 (author)Bill WW2013-11-15

I would be interested in seeing it when your done.

iceng (author)2013-11-12

Very Well Done !
Using minimal material components .... Kudos to you !

Years ago a friend in Reno looked at a heat engine
running on the twice daily 50° air temperature change
in relation to an easily accessible constant water table.

The difficulty was the volume of air that needed to be cycled
used more energy then the engine could put out :-(

Bill WW (author)iceng2013-11-13

Thanks for the comment. Your friend's heat engine is interesting. I can't help but think that there is some way to harness solar energy that will be more efficient than photo voltaic cells.

iceng (author)Bill WW2013-11-13

I agree and promote the solar Fresnel mirror steam approach for small works..

action pig (author)2013-11-11

This is very helpful, a friend and I are working through another stirling engine instructable. Thanks!

Bill WW (author)action pig2013-11-13

Appreciate your comment. If I can help with your project let me know.

dddddd (author)2013-11-13

I *liked* that torque measurement scheme, using a brake on the shaft, and measuring its effect as a change in force on a scale at the end of a moment arm.

I liked that graphite piston, too: lubrication and close tolerances at the same time.

Bill WW (author)dddddd2013-11-13

The graphite piston was a bit of a splurge, but it is a nice precision unit.

guitarpicker7 (author)2013-11-13

A thought:

the Stirling is not a 2-cycle engine by traditional definition. It is an EXTERNAL COMBUSTION engine.

I have seen many spinning fans in hot countries with little electric power. Seems odd to start a fire blow warm air, but it works!

Bill WW (author)guitarpicker72013-11-13

Correct, not a 2-cycle engine by traditional definition.

zawy (author)2013-11-13

I can focus an average of12 kilowatts for 6 hours a day (60 kWh per day) while tracking the Sun onto a 3x3 foot area for $200 in parts using aluminized mylar glued to six 4x8 insulation boards. See my heliostat instructable. Can someone build a sterling engine as the receiver? I'll make the Sun concentrator if you get the engine made. The target hot spot needs to be vertical 15 in front of the heliostat.. Non-low-E glass with little iron impurities (not green looking down the side) is needed for the front of the 3x3 black target (use Rustoleum high heat grill paint). I'll be updating my heliostat instructable in about 2 weeks to saw both units in operation with problems fixed and more specific videos on construction.

Kiteman (author)2013-11-11

There are some mechanisms I could watch all day...

Bill WW (author)Kiteman2013-11-12

Thanks Kiteman.
I could send you a much longer video.

Kiteman (author)Bill WW2013-11-13


rimar2000 (author)2013-11-11

Magnific work, Bill!

I have some ideas for the day when I decide to make a Stirling motor (if that event come some day).

1) The shape of the chamber and displacer should be double tapered. This would facilitate almost fully displacement of the air, and consequently the conversion of heat into kinetic energy. If the displacer ends touch the end walls of the chamber, it get cold in the cold end and hot in the hot end. This would be a good thing.

2) The lenght of the chamber and displacer should be as large as possible. This would enhance the heat isolation of hot and cold end. Obviously, the limit is the common sense and the construction convenience.

3) The displacer should have a heat isolator too, at its middle point, to improve the difference of temperatures between its ends.

4) The crank should be modifiable, in order to try different schemas of angles and displacements of displacer and power piston. Maybe 90° is not the optimum angle, and maybe a little delay at one or both ends would improve the efficiency.

5) The fire should have a fan under it, to improve the heat production and transmision.

6) The cold end must have a fan too, to improve the cooling.

Possibly some of these ideas are a foolishness, but I fantasize with their experimentation. I have much more pending tasks than time available, which is a luck...

Bill WW (author)rimar20002013-11-12

Thank you Osvaldo; you comments are very good.
Sometimes I put ice on the cold end.

rimar2000 (author)Bill WW2013-11-13

Another idea I now remember was that the air chamber could have a valve to increase its pressure (more air, more power). This excedent pressure should be compensate by an equivalent on the open end of the power cylinder/piston (a spring?). The construction could be a little more complicated, but the advantage could be positive. In parallel you could use the valve to introduce other fluids into the chamber, i.e. alcohol, acetone, etc, in order to experiment how they affect the performance of the motor.

TeslaBoy (author)2013-11-12

Nice job! By far this is one of the best stirling engines I've seen instructions for. However your power output does seem to be quite low even for an engine of this size and precision. Do you think it's power output could be improved with a greater delta T value? I'm looking to have an output of around 10w.

Rodney7799 (author)2013-11-11

I was wondering if you could tell us the price you paid for the cylinder and piston from Airpot.

Bill WW (author)Rodney77992013-11-11

I wish I could - I bought a couple of sets a few years ago and do not have the price on this particular piston/cylinder set.
Here is the link to the site:

If I come up with more info I will post it.

About This Instructable




Bio: I'm a retired mechanical engineer, woodworker, boater, and inventor. Now I'm getting into wood turning, and have found that all my wood projects ... More »
More by Bill WW:Make a Camera View-sight for Bird PhotographyHow does magnetic field vary with distance?Table saw sled for precise angled cuts
Add instructable to: