"Doing it Right" or "How to Build a Functional Café Racer"


OK... admittedly this is a very broad topic and I wasn't quite sure where to put it, so I stuck it in the Engines section because it's my intention to make this more of a technical type of article and also because I'm going to be talking mainly about engines.

I was reading a blog post from a friend of mine and it really struck a chord with me. It seems that the café racer genre has really taken off these past few years, but a lot of the newcomers seem to be missing some of the basics. They're immediately drawn to the looks of the café racer, but are perhaps not understanding the function (or even that the form follows the function). Now don't get me wrong... I'm not hating on the guys. Fresh faces are what will keep this knowledge alive and well in the years to come, but only if the knowledge is taught and understood.

Please allow me to repost the blog before we go too much further:
[quote author=kopcicle]What is a "Cafe Racer"?
Make it go ...
We all know that there are gains to be had just in the fine details without resorting to pistons , compression , valves , cams , extensive porting , carburetor and exhaust . Well , no , wait . Within reason and budget that is the idea !
Make it stop ...
Disc upgrade or pursue a lost art in tuning up drum brakes .
Make it turn ...
Make what you have the best it can be or bin it and adapt a modern front end entire.
Make it look ...
Like it's more at home carving corners than sitting outside the favorite pub .
Take pride ...
In the fact that each sub assembly is the very best that you can do with the tools , talent , time and budget available .Only then will the bike reflect that it is more than the sum total of it's parts .
Teach ...
What you learn . Without this not only the lesson but the spirit of the lesson dies with you .
Enjoy ...
What you do , what you build , what you ride .

Our bikes are not only a form of personal expression but a loosely defined art form that stems from the enthusiasts of previous decades . Without growth and change we stagnate and die . In a world of posing and posturing for the benefit of who knows who what is the point of modifying a bike to be just like whatever unless it's meant to be a replica . Be original . Be different . Experiment . Let function be your guide and form will follow .

Our legacy to the next generation of riders and builders is our collective and individual vision . Our passion for that something extra defines our enthusiasm . Our ability to communicate and teach how to learn is our obligation . The definition of a "cafe racer" isn't rooted in our collective or individual past it will be defined by what we choose to do in the future . I've never known a brighter future for the genre in all my years turning a wrench . I can't wait to see what happens next .[/quote]

Now one other thing... there's a been more than a little drama on forums recently and I think one lesson we could all take away from it is that "you catch more flies with honey than you will with vinegar". That is, maintain a positive attitude. Instead of telling someone what they're doing wrong, tell them what they can do right. TEACH them right and wrong so that they can identify it for themselves. Be aware that sometimes there is more than one way to be right, but also be aware that there is almost always one BEST way.

With these things in mind, I'd like to get back to the main topic of this post. "Make it go". My own personal philosophy about café racers can be boiled down to two concepts.

1.) Form follow function - It has to work well first, look good second. Generally speaking, if it works well, it will look good.
2.) Built not bought - It's easy to take this maxim to an extreme, so please use this in the more moderate sense. It's your bike, you need to understand what's going on with it. If something breaks or there's something you want to improve, try it yourself. It may cost you some extra parts when you screw it up, but you've learned something in the process and knowledge is priceless.

Now that the preamble is over. Lets get down to it. This article is an attempt to pass on a bit of the knowledge I've gathered over the past few years. Undoubtedly, there are guys here with a lot more experience and a lot more knowledge than I and I welcome them to please correct me where I'm wrong and chime in with additions where applicable. I readily admit that most of my knowledge comes from reading and studying rather than doing and so keep that in mind. As always, empirical data trumps rhetoric. Something tested, measured, and replicated is something proven. Something written is just words on a screen. If you disagree with something I've said, please post why so we can all learn from it.

A café racer without performance enhancements is just a tractor with a body kit. Do you really want to be one of those kids in a 1.6L Honda Civic with glowing lights under the body panels racing his gutless wonder from stoplight to stoplight? If the answer is, "yes", you can probably stop reading now. ;)

Seriously though, the engine is the heart and soul of your bike and I'm seeing fewer and fewer builds that attempt to improve this key component. I'm not sure why this is, but I suspect that our culture has come to favor looks over performance or perhaps people are a little wary of cracking open something that has so many parts? Cost is also a consideration, but if you can afford to drop $500 on fiberglass seats and tanks and another couple of hundred on paint and upholstery, a bit of money for the engine doesn't seem out of line, right?

Building an engine can most certainly be done in stages, but the most important thing to keep in mind is to take a holistic approach. There are few things you can change that won't also have an effect on something else. Understand the consequences (both good and bad) of each action before you take it. Not all parts will work in all circumstances and the final goal of your engine build SHOULD be the deciding factor of which parts go into it. Building a comfortable long distance cruiser versus building a café racer is more than just adjusting the seat and control locations. The engine characteristics are the soul of a bike.

So what are engine characteristics and how does the design of the engine affect them? Well... put simply, the characteristics of the engine can be categorized by the throttle response, revs, acceleration, torque and a host of even more subjective items. The engine from a semi truck can put out more than 600 horsepower, but you're never going to find one in a sports car (size issues aside). Sports car drivers want high revs. The very successful Honda S2000 can redline at 9,000 RPM. Much higher than most passenger vehicles. That redline is necessary to create an appropriate feel and power for the vehicle's purpose and equal thought should be given to your own engine.

There is something intangible to engine characteristics, though. How much of a grin is on your face after your bike pushes you through a tight corner? Is your bike still egging you on for more throttle even when you're skirting the ton? Or is your bike telling you it's had enough when you start pushing 80? The engine on a café bike should pull strong through the mid range and get even better as the rpms climb. The engine on a café bike should bounce off of the redline after a gear change and actually feel a bit sad that you didn't take it further. The engine on a café bike wants nothing more than to rev itself to pieces. It would love the opportunity to see how fast it can spin before parts start flying out. Is your bike screaming for more or is it screaming "enough"? Your engine should be willing and your bike should be braver than you are. Remember, café racers started out at street legal race replicas, mimicking the race bikes of their time. Would your bike be at home on a race track or did you just build another tractor with a body kit?

So... enough rhetoric already.

Time for some theory. I'll get into more details in subsequent posts, but for the remainder of this post I'm going to talk about the four general ways in which engine performance can be improved. Before that, though, lets talk about a few engine basics just to make sure we're all on the same page.

The Workings of a Four Cycle Internal Combustion Engine (ICE)
This may seem a bit basic to many of you, but I'm of the mind that a decent house needs a decent foundation and so I've included the info here.

An ICE is basically a self-powered air pump. Fuel and air is drawn into each cylinder, compressed, ignited/combusted, and then expelled. Repeat as frequently as possible. Those four steps are as follows:
Intake Stroke - Piston starts at Top Dead Center (TDC) and descends toward the crankshaft centerline. During the entirety of this stroke, the intake valve(s) remain open and the descending piston can be thought of in a similar manner to the plunger on a syringe, drawing fluid into it as it opens.
Compression Stroke - The piston ascends from Bottom Dead Center (BDC) back toward the head as the intake valve begins to close. The fuel/air mixture that was drawn into the cylinder during the intake stroke is now squished into a smaller and smaller volume.
Power Stroke - The compressed mixture is ignited with a spark from the plug and the ignition releases a great quantity of heat. This heat is what causes the gases in the cylinder to expand and produce an increase in pressure. This pressure pushes the piston back toward BDC. Partway down, the exhaust valve opens and begins bleeding excess pressure out the exhaust headers.
Exhaust Stroke - The piston passes BDC and heads back toward TDC. Going back to the syringe metaphor, this is the plunger being depressed and expelling all of the fluid back out of the syringe. During the entirety of this stroke, the exhaust valve remains open. The intake valve will open during this stroke as well.

An astute reader will have noticed that the piston descends and ascends twice for an entire period of the four cycles. This means the crankshaft is rotating 720° for each complete period.

How efficiently each of these four strokes can be accomplished is what determines the performance of your engine. These performance modifications will always go to serve one (or more) of four goals. These four goals are the same four goals for everyone, everywhere, and your modifications need to answer to these objectives.

The Four Goals of Engine Performance
1.) Increase Displacement - All things being equal, a bigger engine will outperform a smaller one.
2.) Increase Revs - Horsepower is a unit derived from torque. Torque is what is measured, horsepower is what is calculated. HP = (T x RPM) / 5252. If you can keep torque the same and increase your revs, you've just "created" horsepower.
3.) Parasitic Losses - The power generated by your engine goes into a lot more than just turning your rear wheel. It takes a lot of energy to spin metal as fast as your bike does and that's even without having to contend with friction of all the components necessary to make it happen. Reduce this friction and inertial losses and your engine will spin faster, sooner, and more of that power will make it to the ground.
4.) Brake Mean Effective Pressure (BMEP) - This is the average pressure within the cylinder, generated by the power stroke. More BMEP directly translates to more torque, which, of course, means more horsepower.

I'm right around my mental limit of typing for the day, so I'll go into more detail on each of the above four items in a later posts.


Slightly Obnoxious
Fantastic thread! Being new to bikes I really appreciate all the effort this is going to take. Subscribed.


New Member
Nice opening post and thread! I must admit I am guilty of building a tractor with a body kit, but my reasoning is sound and once my wife is done with it, it'll be time to tear it open. I'll be following with great interest. ;D


Damn man , did I strike a chord or what ? I'm in this one for the long haul ...



Power Goal #1 - Increase Displacement

As mentioned previously, there are four main methods to employ in the quest for more power. One of the simpler methods is to increase displacement. As with my previous post, I'll start with simple and build from there.

Displacement is the volume of the area occupied by the pistons (at any time) within the cylinders. If you were to measure all of the area occupied by a piston as it moves from BDC to TDC (called the "sweep" or "swept volume" of the piston), this would be the displacement. As the calculation of a cylinder is fairly easy, calculations of displacement are also easy.

The formula for calculating displacement is to first divide the bore (diameter of the cylinder, though diameter of the piston can also used for this calculation) in half. This will give you the radius. The radius is then squared and multiplied by PI (3.1514). The final answer is then multiplied by the stroke (the difference, in distance, between TDC and BDC). You'll now have your displacement in whatever units you used during the calculation process. For most of us, this will be cubic millimeters and so you may wish to divide by 1000 to convert to cubic centimeters. Divide again by 16.387 if you want cubic inches. Finally, multiply by the number of cylinders in your engine for a final answer.

For example, this is the displacement calculation for my own CJ360. Bore is 69mm (non-standard) and stroke is 50.6mm.

R = 69mm / 2 = 34.5mm
R² = 34.5mm * 34.5mm = 1190.25mm
A₁ = 1190.25mm * 3.1415 = 3739.17mm²
A₂ = 3739.17mm² * 50.6mm = 189202.00mm³ = 189.2cc
Final Displacement = 189.2cc * 2 = 378.4cc

This equation can also be "simplified" to one line in the following manner:

Displacement = PI/4 * bore² * stroke * #cylinders

The displacement of your engine directly affects how much fuel and air can enter the cylinder and so more displacement will almost always translate into more torque and more power. All things being equal, a big engine is more powerful than a small engine.

Not all displacement is created equal, however. Many engine designs opt for a longer stroke or a larger bore. Engines with a stroke longer than the bore diameter are called, "undersquare". Engines with a bore diameter wider than the stroke is long are called, "oversquare". If an engine has the same stroke and bore (or are within 5% of one another) then the engine is "square". Generally speaking, motorcycle engines follow an oversquare design, though the more displacement an engine has the more likely it will be approaching undersquare. All HDs of the modern age run undersquare engines and almost all sport bikes will run oversquare.

So lets cut to the chase. Are oversquare engines better than undersquare engines? Well that's kind like asking are apples better than oranges. It'll depend on who you ask and the purposes of the build. Generally speaking, undersquare engines hit peak torque sooner. This gives them a feel of very strong acceleration, but it also tapers off quickly. Oversquare engines take a little while to get up to pace but then will pull harder through the top end.

Undersquare Traits
An oversquare and an undersquare engine will have very different characteristics for the same displacement. Because undersquare engines have a longer stroke, this means their pistons are moving faster than the oversquare engine for a given RPM. Longer distance over the same time means more speed. Engine components can only handle so many forces and as speeds double, forces quadruple. The increase in speed of the pistons directly translates into a need to reduce the engine RPMs before the breaking point of the bottom end is reached.

Piston speed is generally measure using "Mean Piston Speed". This is usually listed in feet per minute or meters per second. Mean Piston Speed is precisely the reason why the Triton was born. The Triumph engine of the day was a better alternative to the Norton because of its ability to rev. The maximum MPS for an engine is usually right around the 4000ft/min or 20m/s mark. Some engines can go as high as 4900ft/min or 25ms but unless you've made modifications, I don't recommend it. To calculate mean piston speed (in ft/min) multiply the stroke times two times rpm and then divide by 60. For my own CJ360 it becomes:

2 * 50.6 * 11,000 / 60 = 18,533mm/sec or 18.53 meters per second at 11,000 RPM

If I were to rev to a mean piston speed of 20m/s, then I'd be hitting nearly 12,000 RPM. Probably doable for short periods of time, but I don't think I'd want to live up there. For argument's sake, lets say I increased the stroke out to 60mm. 20m/s of MPS now occurs at 10,000 RPM. I've had to drop my hypothetical redline by 2,000 RPM to accommodate the new stroke.

Furthermore, cylinder filling can become an issue at higher RPMs because the increase in displacement of the cylinder (but not it's diameter) prevents the use of bigger valves. Next, the increased travel of the pistons creates more friction. A majority of the friction in your engine comes from the piston rings against the cylinder walls and increasing the distance the pistons needs to travel increases parasitic losses. Finally, the longer crankshaft arms (or pin offset) necessary to create a longer stroke also causes an increase in sidewall pressures of the cylinder and on the piston skirts. This is illustrated below this paragraph.

Lets assume this simple image is a side view of your crankshaft with the center of the crankshaft being the point, "C". The black circle is the path which the big ends of your connecting rod follow during the rotation of the crankshaft and the blue line, "c", represents the conn rod, itself. The graphic here represents the most extreme scenario; when the piston is halfway between TDC and BDC and the angle between the cylinder centerline ("b") and the conn rod are the greatest. Let's assume that crankshaft is rotating clockwise and we're halfway through the power stroke. The natural tendency will be for the expanding gases to push straight down onto the piston which will then push down through the conn rod and to the crankshaft. The third law of motion tells us that the crankshaft and conn rod must be pushing back through the piston as well. As this force is not applied directly upward at 0°, then some of the force is directed to the left (in the graphic). In real world terms, the piston is forced against the side of the cylinder wall and how hard it is pushed against the side is related to the angle A, which is determined by the length of the conn rod, c, and the distance from the crankshaft centerline, a.

But, it's not all doom and gloom on the undersquare front. Undersquare engines do provide some very distinct advantages. By increasing the length of the crankshaft arms or by offsetting the pins, we create a greater mechanical advantage. This mechanical advantage is easily thought of as the handle on a wrench. It's much easier to turn a bolt with a long-handled wrench than with a short one. The same holds true for a piston applying its forces onto the crankshaft; things turn easier with long handles. So with a longer stroke, you're not only getting the torque advantage of more fuel and air, you're also a getting a mechanical advantage.

A further advantage of undersquare engines is the surface area of the cylinder, combustion chamber, and piston that is exposed to the ignition of the mixture. As an engine gets more and more undersquare, the area exposed to the initial ignition shrinks in relation to the displacement. This creates thermal efficiencies within combustion that directly translate into a greater BMEP for equal displacement engines. Because of this greater thermal efficiency, higher compression can be employed without the need to use higher octane fuels (which is also beneficial because higher octane fuels burn more slowly).

Also, the longer stroke of the piston aids in port velocities. Generally speaking, the greater the port velocities, the better the volumetric efficiencies (we'll cover that in a later post).

Finally, the flame front within an undersquare engine travels faster and the rate of increase in combustion chamber volume as the piston descends more closely matches the natural characteristics of the expansion of gases created by the combustion of gasoline. This leads to smoother operations and another increase in BMEP.

Oversquare Traits
As you'd expect, the opposite of an undersquare engine is an oversquare engine and so many of their traits are opposite as well. With a shorter stroke you are not only able to rev an engine higher to get more power, you're actually required to do to. Oversquare engines will produce their peak torque at a higher RPM than that of its undersquare cousin. The gives the engine a feeling of wanting to run. Taken to an extreme, however, many oversquare engines will be feel high strung.

With a larger diameters than an oversquare engine, an undersquare engine can pack in larger valves or even more numerous valves. This has the effect of increasing engine breathing at higher RPMs, but it does lower intake velocities and lower RPMs. This is a significant reason why undersquare and oversquare engines achieve peak torque at different spots in the RPM band. Also, the shorter stroke makes those higher RPM forces more tolerable for the engine components and results in lower frictional losses as well.

On the downside, oversquare engines will have a greater area exposed to the flame front and so will generally run hotter while getting less torque per cubic centimeter. These engines need to be revved to get their full potential because they rely on speed for power rather than force.

Oversquare engines are most common in applications that require higher levels of power at the cost of efficiencies. Race cars (yes, even NASCAR) use oversquare engines. Semi trucks, marine diesels, trains, Toyata Prius, etc, all use undersquare engines when efficient operations are more of a concern.

Square Traits
Square engines, obviously, sit between the two extremes. A well designed square engine can have the best of both worlds while a poorly designed square engine will have the worst. Much of this will have to do the selection of peripheral components.

Which to Choose
Well... it's not that simple. Even after laying out the traits of each motor, you can't just pick one and run with it. If your starting platform is an 883 EVO engine, you're going to have a hell of a time reaching oversquare. Likewise, you may encounter some difficulties if you want to stroke out your Ducati 1198.

An increase in bore will require larger diameter pistons. Unless you spend some dough, these pistons will almost definitely be heavier and this has the adverse effect of adding more tension onto the conn rods and crank. More tension means lower RPMs before parts start trying to occupy the same space at the same time. A larger bore will also require the use of a specially made head gasket. You can't have edges of the gasket poking into the combustion chamber or it won't last very long. Copper is a common material for custom head gaskets but it's a bitch to seal properly (especially with iron-sleeved aluminum chambers) and is NEVER as easy as it looks, although I'm not sure it ever looks easy. Going very overlarge on the bore may require your combustion chambers to be remachined or even your entire head to be replaced. In some cases, increase in compression or timing will be required to ensure the flame front continues to propagate at a reasonable speed.

An increase in stroke is accomplished with a new crankshaft and/or pin offsets between the crank and the conn rod. This has the effect of causing the piston to rise higher and drop lower at TDC and BDC, effectively creating a larger circle in which the crankshaft spins. Obviously, there needs to be room within the crankcase for this new length, but there also needs to be room for the pistons. Because the pistons descend lower, the skirts are more likely to make contact with the cases or the crank, itself. On the top end, the piston now might be making contact with the head and so that issue will need to be dealt with as well. Pistons with the wrist pins situated closer to the ring lands (this lowers the piston at both TDC and BDC) and longer cylinders with higher decks are some examples of solutions, though the latter is usually reserved for big spenders or auto enthusiasts. It's also not uncommon to use shorter rods in combination with a stroke increase. This will help to reduce side-loading and usually reduces the weight and hence the forces at play.

So... which to choose? That is going to depend on many factors. Generally speaking, stroking an engine is going to give you more torque than an equal increase in displacement coming from bore alone, but the investment in time and money will be a lot greater. HD engines LOVE a good stroking (don't we all?) and their popularity means there are a lot of aftermarket options available to pursue that avenue. Aftermarket options for other engines may be more limited and so you'll have to undertake a lot of work yourself or follow the path of those that have come before.

My advice is to shoot for a modest increase in bore and then call it good, unless you're running with an HD engine. Any increase in displacement is going to be a good thing and so, for this topic, the focus should be more on what's cost effective for your build rather than what provides the exact traits desired. That said, if anyone feels like stroking out a CB350, please let me know. That's a thread I want to be following.


New Member
keep it coming! and thank you in advance. I wont get to my engine for a long time but when I'm ready to tackle the beast I will be back here taking notes! SUBBED


Born to lose, destined to fail.
"Doing it Right" or "How to Build a Functional Café Racer"

Wow. Just wow. Not only have you given me good advice in the recent past but now you continue to put your $$ where your mouth is and provide this community with great knowledge.

It's a slippery slope when people just assume someone else (a newbie like me) has a ready grasp of knowledge required and then judges them for not completing a build as in depth or technically as others think they should. From someone (me) who has, not a build thread for their bike but instead a "bolt on" thread I cannot tell you how important all this information is for some.


Jump in with both feet.
Right on. Thanks. Based on this thread I did it right when I dropped a bunch of money into the engine before anything else but still lack the knowledge to bring it all together. Performance before looks though. lol.


Kickstart, shift on right, drum brakes and spokes
Sonreir and Kopcicle you are right, cafe racers are about performance and handling, not fashion. Unfortunately, this primary characteristic is lost on many people recently drawn to cafe racers because apparently they are suddenly "cool" or the next 'big thing". Cafe racers have been around since the 1930's and have never gone away. The recent popularity will eventually wane as people realize they actually need to maintain their new fashion accessory.

While I encourage anyone to get on two wheels of any style, but I get really fucking annoyed reading member introductions and cafe build threads prattling on about looks and plans cosmetic modifications ( flat black, root beer powder coating, pinstriping, hole drilling etc) while ignoring basic improvements to the motor, crank balancing, fuel intake, braking, suspension etc. I will take a fast, well handling ugly or stock bike over a beautiful paint job and chrome any day.

We were all noobs once and although I have been riding motorcycles for more than 30 years and building cafe racers for more than 10 years, I strive to learn new build information and skills everyday. DTT and other forums have made information readily available and am thankful for them and I enjoy posting and sharing information if it helps another rider. Tuning information and instruction has never been easier to find and there is no excuse to not use it. With no formal training and being completely self-taught, I have built more than a dozen cafe racers including a highly modified and race tuned Triton. If I can do it, so can anyone else.

"Built not bought" though I appreciate the thought, is a slippery slope. Not everyone has a full shop, lathe, mills, welding equipment, CNC, casting equipment etc so some parts need to be bought. Even with access to a complete shop would you make you own frame? Cast your own engine casings, mill your con rods, design and turn your own hubs, or roll and punch your own rims?? Some things need to bought, but I encourage everyone to DIY as much as possible.

For me, the feeling of pushing a bike I built myself to the top end of its speed, cornering and handling and coming out alive is better than any compliment or praise you may get for how it looks parked in the street.


Active Member
The key for me is making the most of what you have. Not all of us have the resources to mine iron ore,forge con rods etc or even to weld or do simple machining. We have different experience and knowledge but we can all aspire to do the best job we can on our projects.

Fancy paint is nice, but running without fenders or a fork brace is just silly. Flow tested heads with big valves and cams that look like bricks on sticks look great in a thread but are rarely what we need. What we do need are frames that are straight, wheels that run true and brakes that don't drag, levers that pull smoothly and cables the right length, motors that are clean and well adjusted and electrical connections that are clean.

The list goes on, but it really is about maximizing what we already have. Make it work properly. get it so the levers are the right angle for your wrists and the bars and pegs are comfortable. If we don't take care of the basics, all we have a shiny turd. Sorry to be so blunt, but it's true.

Who needs a shiny bike with all that cafe Racer style if it is hard to operate and doesn't do anything well? no one with any self esteem. It's all in the details and if you get those right, the bike will be fun to ride and will look good too.

Sure , we all like positive strokes and affirmation, but they don't last long if my bike is harder to ride than stock and is slower and handles badly at anything above in town speeds. That's not a cafe racer - it's a fashion statement.

Bike can be a ton of fun to ride, and to work on and first and foremost they have to work well and be safe.

That's why so many posts exhort newbies to get it running first and learn to ride it and have fun riding. "building" is a poor substitute for riding IMHO especially for people with only 1 bike and limited experience and cash.

That's just my opinion. Your mileage may vary.

stroker crazy

crazy as a fox
A great thread; sometimes we need to step back to focus on the full picture!

I will be reading this primer on the café racer with great interest. Are there any plans for strokers to get a look-in?




Active Member
I don't disagree with Sonrier's treatise in general, but I'd add that the way to improve performance is first and foremost to reduce lost energy. That's free performance and it's there to be taken. Second is to improve volumetric efficiency and means making it breathe easier. It can include cams and porting, but that's not always necessary or desirable - it all depends on your goals.

You can't beat cubes was an old saying that holds true to this day, a bigger engine will usually make more power than a smaller one, but the costs increase even faster.

let's take some examples. I have say a CB360 and I decide I want more power and less weight. How much more power do I want and what is the path to that power? Do I need bigger pistons custom made to my spec to go into new carbide coated aluminum sleeves and a lightened crank and Titanium rods etc. And at the end of the day if I created the ultimate CB360 how much power would it make and would I have been better off just buying a more modern bike for less cash?

For me, the starting point was always to reduce weight because that's like free power - as long as I don't take that to extremes and have things break.

Then I can make sure that the wheels spin freely and brakes don't drag and there's more free performance. After that I can carefully fit thinner wheels for less rolling resistance or fatter tires for more side grip and I end up with stock sized tires in a soft sticky compound for the best of both worlds.

Then I upgrade suspension at both ends to improve handling so it goes around bends faster or at least more securely and safely.

At that point I can think about modifying pistons, lightening cranks, porting the head, increasing compression and after that I'll think about how much more I want to spend of this sweet riding motorcycle. I'm an engine guy at heart but most of my performance improvement comes from attention to detail and a logical approach and critical thinking.

In the race world, as in the custom world, most modified bikes are slower than stock until the rider starts to put the pieces together and gets the details right. How many threads are about jetting and oil leaks and ignition timing and how many are about cast versus forged pistons or ways to reduce pumping losses? Make the most of what you have with the resources you have available. It's an optimization exercise and not a maximization trip.

Motorcycles are systems and the components don't exist on their own. Everything in life is a balance, and soit is with bikes.


Mechanical Engineering student at Texas Tech
Re: "Doing it Right" or "How to Build a Functional Café Racer"

Awesome post. Were actually learning about auto cycles right now in my thermodynamics class and I find it fascinating. Even more so now that I've seen a literal application of these theories and math now in the forum. Reading to reality makes all the difference similar to what was said in the original post. I was wondering if someone could calculate out the increase of a small improvement such as a velocity stack. I would love to see how much of a difference they make mathematically if someone is able? Thanks for the posts and keep em coming!

Sent from my PC36100 using Tapatalk 2

stroker crazy

crazy as a fox
808shadow - if you're interested in some of the mathematics for intake bellmouth design I can PM to you the pdf file of a research paper by Gordon P. Blair and W. Melvin Cahoon,



Just a slight caveat to my post about displacement. I tend to get caught up quite a bit in the minutiae of things and perhaps didn't make my conclusion strong enough in that post.

A friend read over what I wrote and had this to say:
The longstroke and shortstroke stuff is largely irrevelent in the modern era. A CX 500 has a very short stroke and produces good power up and down in the rev range. Rod length/stroke ratio is a very important determiner of power production and output.
And I do agree. There's usually not much of a bad way to go about getting more displacement. For you newer guys, any way you can get more CCs without adversely affecting your engine, that's what you want to do. While not all displacement is created equal, stroke and bore are not leagues apart, either.

As for rod length:stroke, that is something I neglected to mention too much about. A topic for a future post, perhaps...

Also, I totally agree with what swan and teazer have added. Especially the portion about getting it running and learning your bike before you tear into it. You MUST, MUST, MUST, have a baseline for your modifications. If you modify a bike you've never ridden or don't know well, how are you to know whether your changes are improvements or detriments?

Finally, I also agree with teazer about tackling losses and volumetric efficiency, first. VE, especially, is a topic that will take pages to cover and so I'm leaving it for a bit later. The order in which I choose to address these topics has more to do with getting the "simple" stuff done first rather than implying that this is the order in which things should be done. Trimming weight and, to a lesser extent, increases in displacement can be done with a minimum of investment. When you start talking about significant gains in volumetric efficiency is when the money really starts to matter. A future post, to be sure.


Active Member
808, I don't have the theoretical data but this empirical data was published by David Vizard and is posted on Michael Moore's Eurospares site


Scroll down until you get to VizardVstack001a.tif and download that image.

You will see that taper helps but the key to good flow is a rolled entry to allow air to enter smoothly from all around the stack.


Power Goal #2 - Increased Revs

To first understand why revving an engine produces more power, you must understand that there is a difference between Force and Power. Force is something that causes an object to undergo a change in direction or movement. Power is how quickly that force is applied. Zero force applied over any period of time is always zero power and infinite force applied over any period of time (except zero) is always infinite power.

Since I'm going to mainly talk about horsepower, I'll stick (mostly) with imperial units for this one. This means we'll be dealing with foot-pounds for our unit of torque and horsepower for our unit of power. For true correctness, this force unit is actually called a pound-foot, but the two terms are generally interchangeable, so I'll use with ft/lb as it's a bit more common. A foot-pound is defined as the amount of force, measured at a central pivot point, generated by applying one pound of force at a distance of one foot from that central pivot point. Because ft/lb is dependent on both direction and magnitude, this makes it a vector unit. It can also be thought of as a force that "twists" something. Because torque is a function of both the initial force (one pound) and the initial distance (one foot), it can be increased by either utilizing a larger distance or a larger initial force. Doubling the distance or doubling the initial force will also double the torque.

Hopefully, that gives a basic idea about torque, because you need to understand torque to understand power. Power is how much force is applied over time. The definition of mechanical horsepower (the kind we equate with engines) is 33,000 ft/lb per minute.

If you recall from my first post, horsepower is derived from torque using the following equation:
HP = (T * rpm) / 5252.

So how does that relate back to 33,000 ft/lb/min? Well, all that has to do with our inclusion of Revolutions (it's the 'r' part of rpm) into the HP equation. One revolution is equal to 2pi radians in the math world. Dividing 33,000 by 2pi gets us 5252 and this is the reason for the use of that constant number in the equation.

Enough math for the moment; the important thing to remember is that without torque, you have no power. Applying your torque faster gets you more power. Furthermore, unless you're applying your torque with an appropriate speed, you also have no power. To get power from an engine it must not only produce force, but apply that force quickly. While the human body can produce great amounts of torque (especially with mechanical advantages), our inability to apply that torque quickly is what prevents bicycles from breaking the ton.

So... nitty-gritty time. How does one go about revving an engine faster? Well, it's not quite as easy as changing gears later in the rev range, though this is a basic gist of it. There are several main factors in preventing your engine from operating well at high RPMs and each of these issues must be dealt with in order to create an engine lives up to its potential as it approaches five figures.

Mechanical Considerations
Though most everything on an engine can be qualified as a mechanical consideration, for this section I mean mechanical failure considerations. If you hold the throttle open on your bike, in neutral, you will probably destroy your engine because of a mechanical failure. I say probably, because some engines will fail past that point at which they stop making power and they don't make enough power to get to the immediate point of failure (I'll get to that in a second). On race bikes, over revving an engine is a common cause of failure. When you live on the edge, it's easy to fall off.

When a four stroke engine suffers a mechanical failure it almost always occurs at the end of the exhaust stroke / beginning of the intake stroke. Why this occurs is quite simple and can be reduced to a few key points:

1.) Almost all metals (including those used in the construction of your conn rods, crank, etc) have a higher strength rating when comparing compressive (pushing) to tensile (pulling) forces.
2.) The tensile forces on the crank, piston, and conn rods are the greatest at the end of the exhaust stroke for two reasons: First, the piston is the furthest from the crankshaft centerline at TDC than it is at any other point in the rotation and if rotational speed is held constant, the most force is exerted at the furthest point from the center of rotation. Second, at the end of the compression stroke (the other time the piston is at TDC), there is less stress because the compression of the fuel/air mix combined with the increasing pressure of the ignition of the mixture reduces the tensile loading of the components and so they are less likely to fail. In short, the compressed/ignited mixture is acting like a pillow for the piston and helping to cushion it.
3.) For top end failures (due to valve float, etc), the exhaust stroke remains the most common because the speed at which the piston chases the exhaust valve is greater than the speed at which the intake valve chases the piston. During the compression stroke, the exhaust valve is closed, and so the end of the exhaust stroke remains the critical time. If you float a valve, it's going to be the exhaust valve that eats it first.

So how do you prevent mechanical failures? Think of two words: Stronger and Lighter. Chromoly steel is the weapon of choice for crankshafts, connecting rods, and rocker arms. Titanium makes for excellent valves, though stainless steel is also a good option. For pistons, forged aluminum is generally the best option, though hypereutectic aluminum will usually work for most applications where weight is more important than strength.

Remember, as speeds double, forces quadruple. Pistons weighing in at 200 grams will produce 10 joules of energy at 10 m/s but 40 joules at 20 m/s. Dropping the weight of your pistons by five grams will save you a joule based on mean piston speeds (it'll actually save you more on peak stresses, though). That may not sound like a lot, but on a Honda 360, that's an extra 150 RPM you gain on the redline before a potential bottom end failure. Again, 150 RPM may not sound like a lot, but lets say you can hold a steady 15ft/lb of torque from 11,850 RPM to 12,000 RPM. For the only 150 RPMs, you've gained almost half of a horsepower. Tacking on a thousand RPM instead of just 150 would net you closer to three horsepower, or an 8% increase (assuming that torque holds steady, which it won't).

When it comes to building a maximum effort engine, the little things don't mean a lot. They mean EVERYTHING. A gram here, a gram there frees up a joule here or a joule there, which adds a few more revs to your redline, which adds a few more ponies to your bike.

Pumping Losses and Parasitic Forces
Most people (it seems to me, at least) tend to think only of the mechanical failure portion of chasing RPMs, an equal consideration would be pumping losses and frictional losses. While not sending rods through your crankcase is important, it's also useful to realize that the work your engine must do increases exponentially with the speed at which it operates.

For instance, as the rotational speed of the crankshaft increases, this causes more side-loading on the pistons and the friction between pistons/rings against the cylinder walls increases because of this. Furthermore, it requires more energy to keep your pistons in motion because they are not only traveling faster through the cylinder, they're stopping and accelerating more often. This change in direction, called reciprocation, requires more and more energy as rotational speeds increase. Rocker arms, too, reciprocate and energy is required to move them.

Engines built to withstand high RPMs must also be able to deal with a phenomenon called, "valve float". Valve float occurs when the valves springs are unable to provide enough force to close the valves in the time required. To combat valve float, stronger springs are used in the valve train. Unfortunately, strong springs means the engine has to use more force in opening the springs than previously required. Force over time is power and so by strengthening the springs for more revs, you now require more power to open them.

Pumping losses are a big concern as well. At high speeds, up to 20% of the power your engine creates goes directly into overcoming pumping losses (both above and below piston, though losses above the piston tend to accumulate more quickly in the very high RPM range as compared to below piston losses). And that's even before you factor in friction, decreases in BMEP, or other concerns.

I won't spend too much time on pumping losses as it will be covered on more detail once I start attacking the BMEP section, but I will at least give a definition and a few examples.

Because ICEs are basically air/heat pumps, they all suffer from pumping losses of some sort. And since air is a fluid, frictional fluid laws apply (Newton's third law of motion, again). Think of this in the same way you would think of the the wind while you're riding. There is a bigger difference in wind resistance between 60mph and 90mph than there is between 30mph and 60mph.

Air has inertia (and hence, momentum) and so accelerating it faster to fill a cylinder in less time requires more energy. As your piston is descending and the intake valve remains open, a low pressure zone is created within the cylinder. This low pressure is not only pulling air into the cylinder, it's also pulling up on the piston as it's trying to descend. Likewise, as your piston starts the compression stroke, the mixture within the cylinder must be compressed at a more rapid rate. The relationship between force and acceleration means that as greater acceleration is required, the greater the force must be. These losses occur on the exhaust stroke, too, but in reverse. The spent exhaust gases will initially rush to the lower pressure areas of the headers, but as the pressures within the cylinder and the exhaust system begin to equalize, the force of the piston is required to help expel the gases from the cylinder. As with other fluid dynamics, the force required from the piston to expel the exhaust gases increases as engine speed increases.

Combatting above piston pumping losses can be achieved in a number of ways, but on older engines our usual plan is the same as increasing BMEP, which I will address in a future post. It's also important to note that pumping losses are a byproduct of all ICEs. You will never fully eliminate them.

Below piston pumping losses must be considered as well (especially with two strokes, because the crank cases are not open to the atmosphere). No matter how well your cylinder is sealed, there will always be some blow by. Some of the exhaust gases make it past the piston rings and into the crankcase. This represents an increase in pressure within the crankcase and it is usually vented out through the crank case breather. Though usually not a major concern for 180° twins or inline four-cylinder bikes, below piston pumping losses start becoming a real problem for other configurations. This is because the crankcase volume varies depending on the angle of the crankshaft. For 360° twins, both pistons are at TDC or BDC at the same time. At TDC, the crankcase volume is the greatest and at BDC it is the smallest. As the pistons approach TDC, a lower crankcase pressure requires more force to overcome it. Likewise, as the pistons approach BDC, higher crankcase pressures require more force to overcome it. Regardless of the engine configuration, however, the variations between the cylinder pressure and the crankcase pressure mean that some level of force is being exerted on the pistons as they travel between BDC and TDC. Below piston pumping losses would only disappear if the crankcase pressure and the cylinder pressures were the same, always, which isn't possible.

For limiting below piston pumping losses, your best friend is going to be decent crankcase ventilation.

Combustion Speeds
Finally, there is one more major consideration that prevents high RPMs from being achievable, and that is combustion speed. The gasoline being burned must create pressure at a rate which exceeds the increase in volume of the cylinder as the piston descends. The reason for this is (again) inertia. The momentum of the crank will want to keep the pistons reciprocating long after the the engine stops receiving fuel, air, and spark. In a high RPM situation, the reciprocating nature of the pistons "steals" some of the energy from the combustion event because the piston was going back down whether or not that mixture was ignited. It's the difference in trying to punch someone standing still or punch something that's running away from you. You're going to get a much better hit on the guy who's standing there waiting for it, and the same is true for your pistons. If they're running away from the flame, you're not making power.

There are some tried-and-true methods for increasing the combustion speed of your mixture, however. One common way is ignition advance. By lighting the mixture sooner, you can actually begin building pressure while the piston is still accelerating toward TDC. By the time it reaches TDC, you also have a ton (literally) of PSI built up and so it pushes the piston back down with greater force.

The next, is good atomization of the fuel and air mixture. Smaller droplets of fuel burn faster than bigger droplets and so they release their energy faster. This occurs naturally as your piston speeds increase, which is why additional ignition advance past 3,000 or 4,000 usually isn't necessary. Using the correct jets in your carbs, decent squish bands in your combustion chambers, and tumble/swirl from your valves also helps (more on these things in the BMEP section).

Also, increased compression will help combustion speeds. Due to Brownian motion, an increase of density in the air molecules hammers the fuel droplets into smaller portions and smaller portions means faster burn times. Increased compression also causes a spike in heat which helps to vaporize the fuel right before ignition.

Finally, it should be noted that low octane fuels burn faster than high octane. It's a common misconception that octane is some measurement of power. It's not. Octane is technically a molecule, but it's commonly used as a word to describe a fuel's resistance to ignition. That's right; higher octane fuels don't catch fire as well as low octane fuels. Honda's famous RC166 bike required the use of 87 octane fuel in order to reach its redline of 18,500 RPM. For more "streetable" motors, it's better to opt with higher compression and higher octane fuel as the increase in torque in the lower RPMs is usually more desirable than a few extra turns of the crankshaft at the redline.

In closing, chasing RPMs is actually a losing game (in the long run). You have almost everything working against you. As your RPMs increase, the cost in doing it, while still producing torque, is exponential. It's OK to add a few extra revs once you've tackled some of the other performance areas, but merely adding components for the sole purpose of allowing more revs is not good value for money (though it can be done). One thing I didn't mention in detail is that it's very possible to rev an engine past the point of being useful. As more and more energy is required, that's less and less energy available toward moving you forward. To add insult to injury, the redline of an engine is usually pretty far past peak torque and so while your power may be increasing (slowly), your torque is quickly dropping off. The faster your torque drops, the slower your horsepower rises. A large enough drop in torque and your horsepower begins to fall as well. Building a high revving motor usually requires a peak torque to occur fairly late in the RPM band. This may build winning race bikes, but they won't be much use on the street.


Jump in with both feet.
:eek: I didn't like hearing that last bit. lol. Looks like I'll be building another motor. lol. Thanks for all the input Matt.


Active Member
To Matt's last point, a full race motor is a complete PIA on the street. Nothing happens until it's revved hard and getting away from the lights is fun for the first time and gets more miserable as time goes on until the clutch fails from abuse and then it's game over.

Even for the track, full race cams are rarely the way to go except at Daytona or RA. What you need is torque and you need it at reasonable revs. The two ways to get that at low revs are cubes and compression. Big ports, valves or carbs all hurt low to mid range pwoer and torque.

Thnink mild and think efficient.
Top Bottom