Engine Design basic questions

[Andrew Pullin]

Hi All,

Being a newly minted engineer (as in Diploma not Engineer with a Degree), obviously I have a head full of knowledge and no real world experience.
I am also learning quite a lot including self learning (I have a book and I am not afraid to use it!). The current book I have is about the design of
four stroke engines, so with pen at the ready I was going to use this book to help design a basic One Cylinder petrol (ok, gas for you Americans) four
stroke engine. Nothing fancy, just something that "should" work and has all of the basic bits like cranks and valves and cam shafts etc. For those
interested, the book is "Engineering Fundamentals of the Internal Combustion Engine" by Willard W Pulkrabek. This guy is apparently a Guru and is
better known for his books about Two Stroke engines but I find it easy to read and understand.

I am at the very beginning of this journey. I am going to read the book and work everything out as I go, then draw it and try to build it. One of the
things I need to decide is how big to make it. I want it to be big enough so that it doesn't cost too much to fabricate, but not so small I need a
watchmaker's lathe to do the machining. I also thought that if it was big enough it might serve some useful purpose in the future. One idea in the
back of my mind was that it might be nice to use the base design for other more complex designs later. My thoughts on this were: a One Cylinder
engine should be easy to turn into something like a V-2 (No. NOT some fancy rocket from the 1940s), that in turn could become a V-4 then a V-8.
By now you are probably thinking that innocence and naivity is so sweet. Well - we all started somewhere.

So, some questions.

What is the smallest practical engine that can be built? I don't want to either make my own spark plugs or source ones only available to Pygmy tribes
of the mythical Himalayas of Antarctica. I want to scale it so that I DON'T have to fabricate EVERYTHING.

What is a good size engine that might be of practical use? My thoughts here were that a 5 Litre V8 is common so what about a 0.5 Litre engine?
This would make a single cylinder engine about 25cc or 1.5cu" - is 1/10th scale too small?

The little I have read so far suggests that a "square" engine, or close to it, is practical. What the book does NOT tell me is that Stroke is easy, but is Bore
the DIAMETER of the cylinder so that 25cc = 5 x 5 or is it the ACTUAL area of the cylinder (Pi x 2.5^2) x 5 = 19.63cc when working out the size? The book
has some great definitions but that particular one is ambiguous.

What are general tolerances used in engines like this? Metric please - I can cope with "thous" but I have 10 fingers not 12 so it is easier for a "youngster"
(over 50) from Australia to work it out. 1 thou = 0.0254 mm - just sitting on the edge of the accuracy of "common" machine tools.

Final question. I have been a member here for a while, mainly lurking and reading other people's posts. I did at one stage download some files from
this site but they seem to have disappeared. I know there was a way to do it but there don't seem to be many files in the downloads section, only
posts about the files. I recall downloading some Steam Engine plans some time ago. I notice I can attach file to this post so is it I just didn't dig deep
enough into the Downloads forum? What I would really like to find is a simple car style engine plan to have a good look at. It doesn't have to be a V8
just one with all of the standard bits - cylinders pistons, crank and cam shafts, valves, etc.

Thanks in advance for all of your help.

Cheers

Andrew

 

[Andrew Pullin]

Hi All,

Disregard my previous about Files. I have found them.

Andrew

 

[Jasonb]

Something around the 20-30mm bore and similar stroke size would be in the right sort of region, if you do stick the same cylinders and heads onto various blocks then probably 20-25mm would keep things managable. I build a lot of engines at 24mm bore which seems to work for me be they petrol, steam or hot air.

You won't find much in the way of tolerancing on any model drawings so make one part to fit the other. but a standard H7 reamer would be used on smaller holes and depending on what if any rings you are running then about 0.005 to 0.01mm clearance for piston in the bore

 

[GreenTwin]

I use a couple of benchmarks when deciding what size engine to design or build.

I have a 12" lathe, so that limits the maximum flywheel size to 12" or less.
A 12" flywheel is not that easy for me to machine, and so I am somewhat standardizing on flywheels with diameters between 10" and 11".

When an engine bore gets small, it makes it much more difficult for me to make cast iron piston rings, and so I prefer a minimum of 1.5" diameter bore, and hope to standardize on a 2" bore, which is larger than many modelers use.

I lean towards making small working-class scale models with poured babbitt bearings, capable of measurable and usable continuous power output, and not just static show-display engines.
My goal is to make a scooter 4-stroke gasoline engine, with a 1.5" or 2" bore, and install it on a scooter of my design, for riding around the neighborhood.

Fasteners is also something I pay close attention to.
It is not easy for me to work with fasteners smaller than a #10, and so that is a minimum for me.
Taps smaller than #10 tend to break off easily, and my eyesight is not what it use to be, so visually, I need something I can see and handle with my fat fingers without difficulty.

I have 0.001" over and under reams, and use those quite a bit.

I can generally machine to 1/2 of 1/1000th of an inch.
Beyond that is not possible with my machines or my patience.
I have found that machining to 1/2 a thousanth is sufficiently accurate for the engines I build.
For model airplane engines, the tolerance/fit would have to be far better, but I don't build model airplane engines.

I am scheming to build some IC engines, and have found a small spark plug, which will be the smallest one I use on any IC engine.
I will have to dig out photos of that plug.

Edit:

Here is the commercial spark plug I have standardized on.
Large enough to be handled by my figers, and robust enough to last a long time.

I plan to use an electronic ignition system.

I will literally design the cylinder head around this spark plug.

 

[Jasonb]

Pat how will you get on with "standardizing" with 2" bore and 10-11" flywheels as going upto that bore will in a lot of cases put your flywheel over size if making to scale?

2" bore is going to be quite big for the type of engines Andrew is talking about, put 4 of those in a line and you have a sizable engine..

The CM-6 plugs can be fitted down to about 30mm bore and leave room for two valves, smaller if it is a 2-stroke. If Andrew is making smaller than that then the most common size is the 1/4" x 32UNEF plugs. There is also hot tube and ignitor ignition if you want something different to plugs.

 

[lohring]

The smallest commercially available engine was the Cox with .010 cubic inch (.16 cc) displacement. My favorite books on engine design are Sir Harry Ricardo's The High Speed Internal Combustion Engine. Paperback reprints should be available at a reasonable cost. The others are Gordon Blair's Design and Simulation of Four Stroke Engines as well as Design and Simulation of Two Stroke Engines. The classic is Taylor's two volume Internal Combustion Engine in Theory and Practice. I have written a series of articles on high power two stroke design if you are interested.

Lohring Miller

​

 

[Roger B]

I started out with a self designed 3cc 4 stroke (16mm bore and stroke). This design was expanded to a 12 cc twin (20mm bore and stroke). Both these use 1/4-32 plugs. The others have been 25mm bore, but I think that Jason's 24mm bore may be better as the pitons and rings can be made from 25mm stock rather than 30mm stock. The petrol versions use CM-6 plugs.

 

[bluejets]

so with pen at the ready I was going to use this book to help design a basic One Cylinder petrol (ok, gas for you Americans) four
stroke engine

Save yourself some headaches and use a proven design.
You'll have you work cut out getting one to run first up as it is.
Practice machining methods, much more important.

 

[peterl95124]

all my current IC engines are about 1" bore,
a Hansen Diesel single cylinder test engine that runs on gas + spark (runs)
a Hansen Diesel single cylinder, the diesel version (work-in-progress)
a De Havilland Cirrus I4 (work-in-progress)
a Duesenberg J4 (work-in-progress)
an I4 test engine for 4-valves-per-cylinder (runs)
a Merlin V12 (runs)
and a Franken-Radial R9 (work-in-progress)
1" bore is a convenient size as valves in the 5/16" to 7/16" range are fairly easy to make
and lap if needed to get to seat and seal
and 1" rings are fairly easy to make, and lap if needed to get to seat and seal
I make my own spark plugs, 1/4-32 for 2-valves-per-cylinder, #10 or #8-40 for 4-valves-per-cylinder
the body is stainless steel and the insulator is either Corian or Macor, the electrode is 1/32" SS wire

there are a couple concepts related to measurement accuracy and when and how much is needed, the first is clearance, here are some "rules of thumb" for clearances from Liston's 1942 "Aircraft Engine Design"
piston head to cylinder bore clearance: .004" per inch of bore
piston skirt to cylinder bore clearance: .0015" per inch of bore
crankshaft journal to bearing clearance: (can't find the reference, but I aim for .001" for my
5/8" main bearings)

and the main reason I stick with 1" bore and no smaller is that these rules-of-thumb still apply to 1" bore but not to smaller, below 1" bore everything gets exponentially harder to make and precision and accuracy get exponentially more critical, read up on this and you'll find people have to go to a lot of extra work to get a small bore IC engine to run at all. and in addition to clearance and tolerance issues there are scaling issues, another fascinating topic to read up on.

so if you are trying to maintain a .0015" clearance between your cylinder bore and your piston diameter you should have a micrometer that reads .0001" and a bore gauge that reads .0001", as well as a 1" gauge block to verify your micrometer at 1", and a 1" ring gauge to zero your bore gauge at 1". I personally don't trust digital calipers that read .0001" and stick with a micrometer for this sort of measurement, but use dial calipers for most other measurements. Rings also have to be machined to the .0001" accuracy, but in my experience also need to be lapped before they actually seal all the way around.

then there's he related concept of tolerance, which is if you want .001" clearance between journal and bearing is .0009" or .0011" good enough,

and so a large part of "engineering" is knowing what clearances and tolerances are needed and acceptable for each measurement. if you buy plans for a model engine these will generally be left out, so you build up your own "rules-of-thumb" based on experience and that of others.

Luckily .001" accuracy is fine for just about everything else in a 1" bore model IC engine, and I'm usually lucky to even get that accurate. For example my multi-cylinder engines usually have cylinder bores that were drilled on a mill with DRO so they are perfectly spaced, but the corresponding crankshaft is hell to turn because they are interrupted cuts and there's always shaft deflection, and always tool bit chatter, and things usually aren't perfectly spaced for a variety of reasons, so all the journals are designed a bit wider than necessary to accommodate such "errors".

and so a large part of "model engineering" where we're always making one-off parts is to know how to design things to be at least a little bit error tolerant, unlike mass production where you just throw away the first N parts until you've figured out how to repeatably make sufficiently accurate parts, another part is know when you will need to throw away your first N parts working things out ! (if you think my engines look good you should see my throw-away parts boxes !)

one thing you didn't ask about is what size machinery is required to make a model engine. I have a 200-lb bench top lathe, and a similar bench top mill, but the mill is not really stiff/rigid enough so I have a 1500-lb(?) bridgeport clone in the garage which is overkill but it was used/worn-out and practically free, and I still use the indoor bench top mill for some drilling + tapping and very light milling. I measure machinery by the pound because accuracy comes from rigidity and rigidity comes from mass and there's just no way around the physics. don't expect to make a 1" bore IC engine on a table top lathe and mill, on the other hand you don't need a bridgeport mill or a monarch 10EE lathe to make model engines either.

anyway, hope that's enough for now, welcome to the journey !!!

 

[minh-thanh]

Save yourself some headaches and use a proven design.
You'll have you work cut out getting one to run first up as it is.
Practice machining methods, much more important.

Simple but that's all it takes for newbies,!!!
Or base on a proven plan and change it up to your liking...

 

[Dr Jo]

Hi Andrew. You say you have no experience. Could I suggest you start by gaining some of the skills necessary by making a proven model engine before you try designing one from scratch. This will also tell you what sized engine your machine tools will enable you to build.

The good news is that model engines are much easier to design than full sized engines, While engine design books will tell you lots about bore stroke, cam lift, angles etc, they probably won't tell you about the "obvious stuff" like clearances, offsets etc that are taken into account by old hands without realising they did it.

Jo

 

[Charles Lamont]

I lean towards making small working-class scale models ...

I know what you mean - the aristocratic ones are so much more temperamental!

 

[bluejets]

then draw it and try to build it.

There is a simple design in here somewhere (plans etc) called "simple simon", a single cylinder engine designed and built by our resident ex-school tech teacher and Duke of Edinburgh award winner, George Punter.
It was initally designed for high school students so , as the name implies.
There were a couple of versions, mainly it started as a poppet valve with single cam exhaust which one could modify by adding the extra parts if time permitted, something that does come about with any group of builders.
Carburation is a simple venturi inlet with mixture control for speed.
Video here...........


And another 2nd take........

 

[Jasonb]

Hi All,


The little I have read so far suggests that a "square" engine, or close to it, is practical. What the book does NOT tell me is that Stroke is easy, but is Bore
the DIAMETER of the cylinder so that 25cc = 5 x 5 or is it the ACTUAL area of the cylinder (Pi x 2.5^2) x 5 = 19.63cc when working out the size? The book
has some great definitions but that particular one is ambiguous.

A "square" engine will have equal stroke and bore diameter so you can either work from a chosen bore or work backwards from a chosen capacity.

If you base it on a bore or stroke then 25 bore x 25 stroke = (1.25 x 1.25 x 3.142) x 2.5 = 12.3cc

If you decide on say a 30cc capacity then you would need to make the bore and stroke just under 34mm each but in practice you would tend to choose a round number of 33 or 34 for the bore and adjust the stroke to either give the exact 30cc or more likely go a bit over or under square and use the nearest whole or half mm for stroke.

On other design worth looking at would be Malcom Stride's Bobcat and Jaguar engines, one is a single the other a twin cylinder. It was his intension to use the cylinder and head parts to make other variants but he never got to do that, there have been variants made, I did an opposed twin and there is also a 4cyl boxer that has been done. It's metric and does not use castings.

 

[Steamchick]

Hi Andrew,
I have watched many threads by Brian Rupnow building engines that he designs. He goes through all the machining - and de-bugging of the finished model - step by step as he does it. Warts and all! With his occasional mistakes and changes of design there is a host of learning opportunities there. He sells plans for a very modest cost, to cover some of his material costs and work. So look him up on this website and but a set of plans for one of his single engines for your first model, then you can have not just the model, but the tutor's text and videos alongside when you make the engine.
There are many others on this site as well, although many make complex multi-cylinder engines as well! - Not really suited to a first engine!
Brian, perhaps you can suggest "the best" of your engines for Andrew's first engine?
K2

 

[Steamchick]

Hi Andrew, I may have missed another reply to this one:
"The little I have read so far suggests that a "square" engine, or close to it, is practical. What the book does NOT tell me is that Stroke is easy, but is Bore
the DIAMETER of the cylinder so that 25cc = 5 x 5 or is it the ACTUAL area of the cylinder (Pi x 2.5^2) x 5 = 19.63cc when working out the size? The book
has some great definitions but that particular one is ambiguous."

The displacement (capacity) of an engine is normally described by the displaced volume of the piston. But there are exceptions (e.g. the extreme is a Wankel engine, with a rotor and chamber totally unlike a conventional bore and stroke).
So you want a DISPLACEMENT = Area of piston x stroke of piston.

But, technically it can get confusing with something like a piston ported 2-stroke, Atkinson engine, etc. and other alternative designs.
A 2-stroke is defined as having displacement of bore (piston) area x stroke of piston. But in a ported 2-stroke engine, the actual working part of the stroke (compression and combustion to start of exhaust) is less than the full stroke. But no-one ever talks about that! Similarly, on conventional 4-stroke engines, displacement is always bore area x stroke, although valve tuning can have exhausts opening before the piston reaches bottom dead centre, and inlet valves open after the piston has started up the compression stroke.
So really the "working" (compression and combustion) displacements are never the same as the quoted displacements.
Engines run slower, and are more "manageable" (TBC?) when the stroke is a bit longer than Bore diameter, Somewhere between 1 x and 1.5 x bore diameter? So most "home designed" models are made that way.
I hope this helps, but also warns of some of the pitfalls that can confuse a bit at first? (I am often confused while learning! = Always).
K2

 

[Mike Ginn]

Andrew
Why not buy a casting kit with drawings and often detailed instructions. There will still be plenty of challenges in the machining and you will be able to decide your self on the build standard. For instance, I make all my own nuts and bolts but that isn't necessary (some say I am daft!) but it does help with the scale. No idea what is available in Aus but there are plenty of castings available in the UK and US - have a look at Hemingway Kits - Projects for the Workshop - hemingwaykits.com
Best of luck
Mike

 

[minh-thanh]

Hi @Andrew Pullin
"Engine Design basic questions"

My story.
A long time ago, before I joined the forum, I wanted to make an engine, not one but many types of engines, Can I design? Yes I can
I started designing a stirling engine, I designed the crankshaft, I had to find a way and make the crankshaft, I designed the cylinder, I had to find a way and make the cylinder, and then the piston and next.... At that point I stopped designing and just focused on figuring out how to make the other engine parts
I have to find the tolerance and fit.....and how to get the engine running
Up to this point, most of my stirling, flame eater, 2- and 4-stroke engine plans were incomplete.
My conclusion is
Designing an engine, creating a 3D model It's not too difficult
Designing an engine, building it and making it run depends a lot on skill, experience and patience to overcome difficulties and failures.
That's my short story
For you , you have knowledge , you have books , you have great information from the forum , you have the internet , what you really need is experience and skills , so pick a plan and build and make engine run
Once you get the engine running, you'll know what to do in your future designs

 

[Jasonb]

The downside of the majority of casting kits available is that they are going to be in imperial and the OP has said he does not want to work in bananas. As it is your first engine adding the conversion to metric including stock sizes, metric threads & tooling etc is not going to make life any easier. It will also push up costs if you have to buy from overseas.

So look for something that will allow the use of metric taps & dies, MOD gears*, metric bearings, metric fixings, etc.

* avoid a design with 2:1 helical gears until you can cut your own.

 

[Mike Ginn]

We live in a world of metric with some countries remaining imperial. Most of my drawings are imperial but all my AutoCad drawings are metric. Its not really an issue and I do find the conversion chart below of great use. Easy to use in Excel. If anyone wants a copy I can post it as a .xlsx or pdf file
I mainly use BA fixings because I have the taps/dies but for non-model use its always metric - the world is held together with M6 bolts!

For measurements I usually revert to a metric/imperial Vernier or for greater accuracy, a digital mic.

I would say that you shouldn't have a hang-up about imperial/metric!!
Mike