I will mostly leave this to a specific motor I have and the machine work I have done.
This is a two part story. Once I built a 1951 pan shovel. From the crank up, it was all OEM parts. It was a pretty solid bike, I rode through Mexico and back and the motor was a champ. It was straightforward to build and besides all the tolerances I didn’t give it much thought. I blew a head gasket at the end of the year (stop using those blue head gaskets). If I had torqued my heads properly I would probably still be riding that bike. That motor was later rebuilt into the dual carbed engine for Space Bugs.
After I blew that head gasket I came across an aftermarket set of cases and wanted a big inch motor. This is how I made most of those decisions.
Harleys have a big limiting factor, which is a v twin engine is inherently sort of unbalanced, and for Harley to produce such large motors, they have long stroke lengths. Your stroke is determined by your crank shaft, and is essentially how far your piston travels up and down. The longer the stroke, the bigger your engine displacement due to the piston traveling further up and down. The reason this matters, look at the calculation for horsepower (torque x rpm / 5252). Which basically says the more rpm you can make the higher the horses. Example, a CB750 makes 70 HP at 8,000 rpm, a panhead makes 55 HP at 5,500 rpm. So how do you make a high rpm motor? Japanese motors have short stroke flywheels and spin extremely fast. Every rotation of a Japanese crank shaft, the piston is traveling a fairly short distance up and down. This short distance is important. Pistons can only travel so fast against the cylinder wall before friction destroys them, and we call this piston/cylinder wall speed. Every time the crank shaft spins 360 degrees, the pistons will travel from the very bottom of the cylinder to the top. The longer the stroke, the farther the distance up and down that the piston travels. A 2 inch stroke motor vs a 4 inch stroke motor spinning at the same rpm will have half the piston/wall speed. So even though a Japanese motor can spin at 12,000 rpm, because the crank has a short stroke, and the piston isn’t traveling that far, the piston/cylinder wall speed is obeying the same rate that Harley pistons can travel.
This is a long way of saying the more stroke you have, the farther the piston must travel every rotation and this limits your rpm. If RPM makes power, then I choose to find other ways to increase engine size without losing RPM.
(There are people who make good power out of stroker motors. More than I do. But strokers have an inherently shorter lifespan and I don't like that)
So the other way to make more displacement is to overbore your cases for bigger aftermarket cylinders and bigger pistons. Bigger piston=more displacement=more power. 3.625 is geverally refered as “big bore” in the 74 inch world and requires you to cut open your cases for aftermarket cylinders. After building a moderate 4.25 inch crank and big boring my cases, I have a roughly 88 inch motor.
(Harley crank stroke lengths can vary from 3.5-5 inches. Standard 74 inch motors are 3.9, with 4.5 being the level I consider “big stroke” motors.)
So now I’ve gone from 74 inches of engine displacement to 88. I’ve got a set of cases, cylinders and pistons, and it’s time to realize that it’s not motor size that makes power, it’s how much air and fuel you can stuff in it. This is the second logical step to making more power, after you’ve made the space for fuel and air to compress, we’ve got to cram in as much fuel and air as we can.
Your cam is lifting your intake and exhaust valves. We’ve got two measurements for how the cam operates, lift and duration. Lift is how far the valve opens, and duration is how long it stays open. The cam conversation is an entire book, and there’s a lot I don’t know or I’m not very good at explaining. But short form, this is what I think is important. Stock/EPA era cams don’t pack an optimal amount of fuel into your motor. A cam with bigger lift gives a motor more space to swallow air fuel, but causes more wear to your valve train as it pushes your valve open a little bit farther. Cams with long duration give your motor more time to swallow air fuel, but can begin to interfere with your bike's compression.
This starts the compression ratio discussion and why most compression numbers don’t matter. Compression ratio is essentially how tightly compressed your air fuel mixture is when the piston is at the very top of its stroke and about to ignite. (This is not actually what the number means, but I keep it simple). Anyway, the higher the number, the more compressed, the more explosive the fuel gets and we make more power. But if you take a 10:1 piston, which compresses the fuel a lot, and put a cam with a ton of duration, so much duration that the valves are still open as the piston is moving up, you will push some of the air fuel out of the combustion chamber before the valve closes and when you go to compress there won’t be much left, and you end up with not that much compression after all. Which is why static compression numbers mean nothing, and dynamic compression that depends on cam is so important. Dynamic compression is a true reading as your motor runs, while static is just a theoretical number your piston maker puts on the box. You can take a static compression 10:1 piston and put a massive cam that will bring it all the way down to a below stock dynamic compression of 7:1.
So maybe you're thinking, why would you put a cam with that much duration in your motor if you’re just gonna push all that fuel out? Well it’s sort of an rpm thing, and no cam can fully exploit low and high rpm. The short explanation is that it’s hard to get your motor to fill up with fuel when it’s running fast. Air only wants to move so quickly, so to keep the air fuel mixture flowing into the motor at high rpm we overlap the intake and exhaust valves at the same time for a moment, that way the momentum from the exhaust leaving will help pull in fresh fuel from the intake. This is called cam overlap. This overlap at low rpm when air velocity can easily fill and escape the combustion chamber will cause a loss of power. This is an oversimplified way of looking at it, and there’s a ton of cam profiles that do lots of different things, but the point is, cams have a lot of effect on compression and how well your motor is going to fill. Big, long duration cams have a tendency to blow off a lot of compression at low speeds and then kick in at high rpm making more power. Smaller, short duration cams build lots of compression at low rpms but can struggle to fill cylinders at highway speeds, or raise compression so much you cause pre detonation. If you are building a motor to be powerful around town, short durations, early intake and good lift cam will make power at the bottom of the rpm range and can work well with standard compression pistons. If you are planning on a lot of highway miles, long duration, late exhaust close really packs fuel in at high rpm and works well with high compression pistons. NO cam can do it all.
So the last consideration I lump into “efficiency”. The displacement and cam all operate in a kind of theoretical world until we put the rest of the motor around it. Head shape, piston design, carb velocity all have a huge impact. I will only focus on a couple points here.
People have this idea that porting involves making your intakes bigger. Early motors, knuck shovel and pan have poor flowing heads because their ports are too large to begin with. This is part of the reason these motors like to be bored out, you can start to take advantage of the large intake volume with a 80+ inch motor. I reshape the castings around the valve guides as well as deshrouding around the seats, making sure not to remove any volume. We look to get more linear air flow through the intake and a clean smooth exhaust port to reduce heat absorption. I have mirror polished both combustion chambers and piston with mostly unknown effects, this supposedly can reduce engine temp but I mostly find it helps to reduce carbon buildup because it can’t stick as well.
After shaping the ports, I have decisions to make regarding head and piston shape. I once asked Reid from MACSPEED about a cam and piston choice. He told me instead of inventing a new recipe, to take a look at EVO motors. There’s a huge variety of performance options and info for modern engines, and sometimes you can retrofit these performance recipes back into pans or shovels. A big development Harley made with the evo was the use of flat top pistons and squish bands, both of which I wanted to fit into my shovel. The big advantage you see with a flat top piston vs a hemi/domed piston is spark propagation. Air fuel doesn’t necessarily burn very fast or evenly, and when you have a piston with a large dome in the middle and a spark plug on only one side of the head, your spark needs to burn up and over the top of the piston like a mountain. So evos do away with domed pistons but completely reshape the heads as well. In order to fit a flat top piston in the shovel I begin a long series of cutting a flat surface into my heads, and then decking the cylinders to decrease the distance from my piston and the top of my heads. Every cut I make to my cylinders, the heads drop slightly closer to the top of my pistons. As they begin to touch I go back the heads and cut my flat area farther back until I have my desired compression ratio as well as piston/head clearance. I eventually end up with significantly shorter cylinders and shovel heads that have a big flat ring on the outside.
The distance from piston to this flat spot is important, not only because if I get it wrong my pistons crash into the heads, but most modern engines take advantage of what’s called a squish band. Squish band is a critically close area between the piston and cylinder head, that as fuel air is getting squeezed together before combustion, this squish band compresses and mixes the fuel air straight towards the spark plug for very quick and even combustion. The squish band in my shovel essentially looks like a flat ring all along the outside of the head, and is roughly .050 away from the piston at TDC. These are the final touches to the motor that give it the efficacy and power that a motor can reliably make. I don’t have a ton of experience with hot rod building, but one of my favorite sayings is “compression is fun for a little while”. There’s a lot of ways to make power, but I also like when the motor stays together.
But to wrap it up, how did it all work? The real lesson I am learning from experimental motor work is that power is a recipe. Even with appropriate cam and carb choices you may be leaving half your potential power on the table based on piston shape. It’s the combination of your head shape, piston, valve size, spring weight, etc that will determine how that new cam or exhaust will effect your motor. If you consider the fact that some people are pushing close to double the HP compared to stock in 80 inch evos. The power is not coming from increased displacement, it is perfect harmony of all the internals.
So to make power and find my own recipe it’s been hundreds if not thousands of hours of building and rebuilding the same motor. Sometimes it comes apart for me to cut .020 inches from my heads and reassemble. Two weeks later I do it again. But holding your own flywheels really brings you close to your bike. Blasting past a couple knuckleheads going 100mph, there’s no better feeling.
SB