If you don’t fully understand a process, then making improvements becomes something of an art instead of a science. The artistry in designing an engine is to get as much of the proper mixture of air and fuel into the cylinder as possible in the face of all the compromises and counter-variables in the whole system. Volumetric efficiency (VE) describes how well air and fuel fill the cylinder during the intake stroke. One might think that at the bottom of the stroke the cylinder would be filled with a volume of atmospheric-pressure air, no more, no less, but fortunately, it’s more complicated than that.

Air has weight and therefore inertia. Once the air is moving it will keep moving and if it is sitting still it will tend to stay still. A fascinating example of air in motion is the behavior of air passing over the Sierra mountains in the western U.S. Air moves over the mountains from the west and falls down to lower elevations and then "bounces" off the ground. Glider pilots use this rebounding column of air to get up to 59,000 feet, almost twice the altitude commercial jets typically fly at.

In an engine, the moving slug of air being pulled into the cylinder tends to keep rushing in even after the piston is on the way back up on the compression stroke. This dynamic allows the intake valve to be left open a little longer than you might expect. As a result, the cylinder can be pressurized above atmospheric pressure and still have an open intake valve. So you might say that it had a volumetric efficiency above 100 percent, and 110 percent VE is money in the bank. Add in ram-air to boost the airbox pressure and it just gets better.

Motionless air wants to stand still so the first challenge is getting it moving. Starting the intake air moving is accomplished by the exhaust gases. The exhaust gases scream out the exhaust system under high pressure and create a moving quantity of air. The intake and exhaust valves are actually open at the same time (the duration of that time being known as "overlap") and the exhaust gases "pull" on the intake air. Old two-valve Ducatis had so much overlap that it’s hard to figure out when both valves are closed on the compression stroke. The timing of the opening and closing of the valves (dictated by the cam lift, duration and timing relative to each other and the crank), the shape of the ports (not to mention the length of the intake and the dimensions of the exhaust system) and the instantaneous speed of the piston all play into making a hellishly complicated system that seemingly defies prediction. Nowadays there is software that attempts to predict what will happen with different cams and cam timing but the good stuff tends to be pretty expensive. Picking the right set of cams and timing is as much an art as a science. We’ve always experienced a certain amount of nervous anticipation in the dyno room, waiting to see if this or that configuration is going to be a dog or not. So far our engines have had the tact to wait until we are at the track to reveal their true natures completely.

The best we’ve been able to do in building an engine is work with a few rules of thumb to get to the right ballpark and so a new engine configuration always has a few surprises. One of our YZF600 engines was absolutely immune to jetting changes with some big bumpy cams and ran pretty flat. On the advice of an acquaintance who had done the legwork, we put in an exhaust cam that actually had less lift than stock but more duration and picked up 10 horsepower. It’s enough to make you want to bash your head against a wall. As an aside, we are extremely grateful to all the talented people (Vic Fasola, Chuck Warren, Keith Perry, Chris Weidl, Dave Short) we have talked to who took pity on us and helped steer us in the right direction.

Most folks don’t have the time nor inclination to wear out an engine on the dyno trying lots of different cam grinds and timing. We take our best guess and work with what we get for the most part. Using the following rules of thumb for assembly and adjustment we can get pretty much in the neck of the woods we want to be. There are always compromises and this is another case where having an experienced engine builder who has already built numerous versions of the configuration you are running is a huge advantage.

Designing The Engine Package

In most cases you are really best off using the stock cams. Typically the horsepower return for swapping cams is very low compared to the cost of the cams. If you still feel compelled to change the bumpsticks:

Pick a set of cams. Usually the stock intake cams on modern bikes are pretty good so you might try leaving that one alone. It’s easier to get intake air in than to get the exhaust out since the piston is moving fastest when the intake valve is at maximum lift. The exhaust valve is on its way back down as the piston reaches max velocity, and with engines revving as fast as they do, there’s just not a lot of time for everything that needs to happen. There can be some residual exhaust "poisoning" of the intake charge. We try to find an exhaust cam with similar lift or a little bigger, but more duration, which gives the gas more time to get out and might have a beneficial effect during overlap. Many Suzukis have the same valve spacing which means that the GSX-R600, GSX-R750 and GSX-R1000 cams are, for the most part, interchangeable between models and often between years. These cams are also very cheap compared to aftermarket cams. Digging through your old copies of Roadracing World you can find the cam specs on many bikes and try to pick and choose what you want. Straying from the OEMs, many of the American aftermarket cams have a lot of lift and not so much extra duration. We’ve had better luck throwing down on so-called "kit" or factory produced cams for Yamahas although we have never had much luck with the Yoshimura cams for the GSX-R line.

Time them: After you’ve guessed at a cam, take a stab at timing. Books have been written discussing which way to rotate the cam sprockets to achieve a variety of powerbands and plenty of those books contradict each other. Consider the following compromises: During overlap, the intake valve is opening and the exhaust valve is closing. The exhausting gases drag intake mixture into the cylinder, boosting VE. If the cam is rotated such that the valve opens earlier and earlier, it closes earlier as well, pinching off the air that is rushing into the cylinder and compromising VE and power. Also, as mentioned above, there’s not a lot of time to get exhaust out of the cylinder so the exhaust valve opens long before the bottom of the power stroke, giving the gases a head start. If the exhaust valve is opened too early, however, less work is done forcing the piston down. If the exhaust valve is opened too late, some exhaust is left in the cylinder to mix with the intake mixture, reducing power. But if the exhaust valve doesn’t open late enough it’s not left open long enough to be effective during overlap. Your head might start to hurt at this point, because thinking will consume the sugar in your brain. Orange juice is very effective at replacing these sugars, alleviating the headache that results from weighing the effects of different cam timing settings and major life decisions in general.

A typical cam-timing spec will read something like Intake 100, Exhaust 105. This means that the maximum lift, or "lobe center" for the intake cam occurs 100 degrees of crank rotation past top dead center of the piston on the intake stroke, and the exhaust cam lobe center occurs 105 degrees before top dead center during the exhaust stroke. These centers are determined as the midpoint between when the valve is, oh say, 0.040-inch off the valve seat, opening and closing. The spec does not describe what is happening during overlap, or when the intake valve closes, or the exhaust valve opening, the major considerations for designing the cam timing. A diagram is helpful:

As an example, a stock fuel-injected GSX-R600 has timing of about 102I/105E. When we tried 100I/100E, it had a lot of overlap, but was difficult to ride as the compression braking was tremendous. Changing it to 102I/107E, pretty much as far the opposite way as we could go without the pistons hitting the valves, made for less overlap, and better "scavenging" of the exhaust gases. It lost a few horsepower on top, but was easier to get into the corner, and was less abrupt on the throttle. Your mileage may vary.

Install the cams in a stock (or if using different pistons/rods/whatever, use them now) engine and figure out piston-to-valve clearance. Temporarily install weak springs in place of an intake and exhaust valve spring so you can push the valve bucket down easily. Use some adjustable sprockets for the cams, and tack-weld the flanges on. We’ve found using stainless filler rod and a TIG welder to be adequate. The method for checking cam timing is too complicated to describe in detail here, but in brief, find top dead center using a piston stop (a purpose-built tool or an old spark plug with a blob of weld on it that stops the piston from rotating past top dead center), install a large degree wheel (with TDC indicated midway between the readings where the piston stopped) and pointer on the engine, and affix a dial indicator to measure the height of a valve bucket in the head as the crank and cams are rotated.

Piston stop installed in spark plug hole for use during setting cam timing. Photo Courtesy: Cooper Performance

Degree wheel installed for cam timing. Photo Courtesy: Cooper Performance

At two points, the piston is precariously close to the valve. Somewhere around 12-14 degrees before TDC the piston is closest to the still-open exhaust valve. The rod is stretching, and the exhaust valve may be bouncing a bit so a good conservative clearance is to leave 0.060-inch clearance, measured with the dial indicator by pushing down on the bucket until the valve touches the piston and observing the dial indicator. Around 9-12 degrees after TDC the intake valve is opening and chasing the descending piston. The valve is opening faster than the piston is descending, and the rod might be a little stretched, trying to drag the piston down so there is the potential for a coincidence of unfavorable tolerances. A conservative clearance is 0.040-inch.

Engine set up for cam timing with degree wheel, piston stop and a dial indicator to determine valve lift. Photo Courtesy: Cooper Performance

Having made this measurement you can figure out how much you can take off the head and cylinders without bending anything. If you’re really not sure, leave some room to change the timing. As a last resort, machine the tops and/or piston pockets of the pistons to allow the valve extra clearance.

Porting

If picking cams and timing weren’t enough of a baffling ordeal, porting can really make your hair turn gray (or in the case of AOD, fall out). In theory, superstock rules don’t allow for any porting. As a model of motorcycle gets a little longer in tooth, the castings and the valve seat machine start to not align with each other quite as well as in the early production. As the production run continues, tiny obstructions in the ports, valve seat offsets, and general ugliness begin to appear. It pays to get an early head where the valve seats line up more or less with the port.

Close up of valve seat. They really should just come from the factory like this. Photo Courtesy: Cooper Performance

There are two approaches to porting. The first is to clean things up and to make the ports the way mother nature intended by improving the surface finish and removing any obvious seams and casting imperfections or filling in spots with a fuel-proof epoxy such as JB Weld. The second is to decide you know better than the engineers and grind around, changing the shape of the ports. A tool people use is a "flow bench" which measures the volume of air that a port can pass at different velocities, at varying valve lifts. The thought is that increasing the flow number will increase horsepower. With no direct experience we cannot speak to that and there are plenty of people who feel that big flow numbers don’t always produce big horsepower. A fabulous book, dangerous in the wrong hands for sure, is "Practical Gas Flow" by John Dalton which describes making your own flow bench out of common household materials. We confined ourselves for many years to just cleaning things up and hoping for the best. The last few years we have been bolder and we have taken to changing things around, especially the exhaust ports.

Anyone who races a 01-03 GSX-R 600 should have a head that has tried to swallow a valve. We cut ours in half to be able to study the ports. This is an exhaust port, red arrow indicates an spot in the port which Tim suspects will disrupt flow through the port by creating an eddy to the upper right of it.

If jumping in with the grinder and JB Weld is too intimidating, one option is to send the head off to have it done. The only experience we’ve had with people who know what they are doing is with Cooper Performance, the work done by Jeff Walker who was a Neighbor of the Beast rider in 2002 and an AOD rider in 2003. He did a fine job, it came back quickly, it looked right and the engine made great power.

Porting a head takes a long, long time, peace with the world and is so involved that you can’t really relate to humans for a while after. Sending it out to people who know what they are doing may be money well spent. That said, porting is an activity everybody should try at least once. Harbor Freight has a 1/8-inch-shank air die-grinder. Solid carbide cutters of various shapes, abrasive wheels for blending steel valve seats into ports, and sanding cones come from www.mcmaster.com.

Close up of intake port. Note how the valve seat matches the port well on the right side but not quite perfectly on the left.

Modification And Assembly

If you decide to port the head, keep in mind the following: Try to picture the path of least resistance for the air and fuel. Grasp what the designer/ manufacturer meant the ports to look like while trying to understand how manufacturing the head changed that ideal (i.e., having valve guides sticking out into the air stream). Be the gas molecule. If you have a demolished head, cut it in half, through the ports, using an abrasive saw or a bandsaw so you can really see what’s happening in the ports. Unspeakable horrors are often revealed this way, particularly in the exhaust ports.

Two principles to think about are that the gases don’t want to change direction so they need to be guided gently by smooth-radiused corners and at high rpm the gases take the shortest path from the throttle body/carburetor to the cylinder as possible. On the intake side, the "short-side radius," or floor of the port, just behind the valve head, is an area to pay attention to as it is a shorter distance through the port than by the roof of the port around past the valve stem and guide. Get rid of everything that is getting in the way of the flow and round off abrupt corners. Shave the valve guides flush with the wall of the port and sharpen the dividing septum between the valves. A satin finish on intake side supposedly prevents condensation of fuel on the port walls. Polishing the exhaust ports reduces the carbon buildup and reduces the amount of heat transferred into the head. Remembering that gases don’t want to change direction, raising the roof of the exhaust port sometimes helps as it is a straighter path for the exhaust to get out and into the exhaust pipe. Try to make changes uniform for the whole head. Do a single small modification in each port rather than changing a port in its entirety before proceeding to the next. This process can easily take 15-to-25 hours.

Look at the combustion chamber. Try to picture how the gas is getting out of the port. Remember that the cylinder wall "masks" or blocks part of the valve so try to smooth off the chamber where the gas has the best chance of getting into the cylinder. Between the intake and exhaust valves is the "fence," a little lip that blocks intake mixture from directly shooting out the exhaust port during valve overlap. Leave that alone or your intake charge will go straight out the exhaust pipe.

Gasket surface on a head being decked to tighten up the combustion chamber. Photo Courtesy: Cooper Performance

Do a valve job after porting to clean up where you might have bounced a cutter off the seat. In theory an intake valve job will move the sealing surface of the valve seat to the outer circumference of the sealing surface of the valve which makes for an effectively larger valve, opening the way, hopefully, for more air and fuel. Radiusing or rounding the seat will allow for mixture to flow into the cylinder more easily at low lift. Narrowing the seat surface, ultimately to a thin line, also helps with low-lift flow. A caution that as the seat is narrowed the spring pressure is born by a smaller area, and the valve or seat can be damaged, requiring "freshening up" at some point. Also, the valve runs hotter as less heat is transferred into the head across a narrower seat.

Someone who knows what they are doing will be able to leave a wide-enough seat and clean things up with a radius or 5-angle valve job, improving on the stock 3-angle valve job. Don’t grind or lap the valves as might have been done in the past. Lapping will grind through the hardened surface of the valve. A proper valve job with modern equipment, such as a Serdi Micro, results in the valve sealing against the seat without further grinding. A valve job will invariably screw up your valve clearances which will require you to get thinner shims and, in some cases, actually make shims if you have bottomed out the range available from the OEM.

Cooper Performance. Stock three angle valve job. Each angle is clearly visible by the slightly different shades of gray on the valve seats.

Cooper Performance. This is a very trick drag racing head with a bronze insert combustion chamber. These sorts of inserts would not normally be used in a road racing application but this picture also illustrates the modified intake port (note cut down valve guides) and the single angle valve seat with radiuses instead of angles blending the seat into the combustion chamber and port. The radius valve job is ideal for maximizing low lift flow.

Safety-wire all the plugs and bolts and inspection ports on the engine. Never use a blob of gasket goo, as it won’t adhere after a while.

Once it is assembled, leave it that way. We try to limit compulsive tear-downs to one rebuild during a season of endurance as there’s more chance of getting something back together wrong than something breaking. Your mileage may vary.

Dyno Testing

After assembly, we install the engine in the chassis and start it to make sure the cams aren’t in backwards, the side stand switch is successfully bypassed, the clutch lock-out is bypassed and the wiring is all connected correctly. We’ll usually warm it up once by gently blipping the throttle and then proceed straight to the dyno. Dead-revving the bike in the driveway is nerve-wracking and unnecessary. In normal operation the piston rings are forced against the cylinder wall by combustion pressure. High combustion pressure is produced by the expanding gases trying to push the ponderous bulk of the rider and machine down the track or by fighting the load produced by a dyno. Revving to redline in the driveway just makes the rings flutter up and down without being pushed against the cylinder wall and allows for fuel/air blowby into the oil.

When we get to the dyno we run the engine at half-to-¾ redline under load, varying the throttle continuously to push the rings against the cylinder walls and to get all the new parts used to each other. If we have built an engine and have to go straight to the track we will usually spend the first practice session running the engine to ¾ of the rev range while heavily loading and unloading the engine. From our perspective, the worst thing you can do when breaking in a bike is to run it continuously at low loads. You want high compression pressure to seat the rings but low load so as to let the bearings get to know each other. After a couple minutes or so the engine will usually stop smoking which indicates the rings are sealing. It is then time to adjust fuel/air ratios.

We will hopefully never race another motorcycle with carburetors. We could never get things just so with carburetors on the dyno and would invariably have to go a lot richer on the main jets at the track to compensate for the ram air, not to mention the liver-rotting-fuel-covered hands from multiple jetting changes. Modern fuel injection is a fabulous thing and once we have a good map in the bike we have found that it largely self-corrects for ram-air, altitude, barometric pressure and temperature.

We long ago determined it was a waste of track time to work on engine tuning and that most jetting or fuel injection work should be done on a dyno well before arriving at the track. We have created maps three different ways for fuel-injected bikes. Make sure that, whatever your method, you are using fresh fuel of the type that you will be using regularly.

The old school method was treating the fuel injection like carburetors. We would strap the bike on the dyno with the computer hooked up the fuel injection modifier (we have always used Dynojet Power Commanders for this because we haven’t found another product to be more useful for this task). We would perform old-school dyno runs and make small changes to the map and repeat until we had arrived at what we felt was the best map we could create. This was a very time and labor-intensive process.

Then Dynojet introduced the Tuning Link, which connected the dyno with an exhaust gas analyzer and the Power Commander. By performing dyno runs at each of the designated throttle positions the Tuning Link modifies the fuel map to make a perfect 13:1 air/fuel mixture at all throttle settings. If you trust the exhaust gas analyzer. We have had some very, very good results with this process. One of our bikes would barely run after everything we had done to it (cams, pistons, etc) and then ran like a scalded dog with the fuel map created by the Tuning Link. Never in a million years could we have come up with such a map on our own.

Most recently we have split the difference. Some purists will sniff their noses at Dynojet dynos and claim that the fuel maps created by the Tuning Link, although very good, are not perfect because the resolution of the Dynojet equipment is insufficient. The last mapping experience we had was using a combination of the above two approaches. We started with a Tuning Link map, then put the bike on a Superflow dyno (with speed compensated ram-air) and a high quality exhaust gas analyzer. A very experienced operator (in this case the aforementioned Jeff Walker) used the information gathered from the dyno and the exhaust gas readings to make small but noticeable improvements to the fuel map. This process took an additional two hours or so.

A supersport 2003 R6 with looser connecting rod bearings, radius valve job, shaved head and degreed cams getting the air/fuel ratio set using a SuperFlow dyno and a Dynojet Power Commander. This power on this bike increased from 6-10% across the power band. That’s an exhaust gas analyzer in the tail pipe and the fat white pipes by the front wheel are speed compensated blower to simulate the ram air effect.

Pressed for time we would have been happy using the Tuning Link map but since this is racing and we were all grateful for the improvements that Jeff achieved.

Melissa blissfully forgets the hours of work in the motor and fuel injection when she gets to pull the silky smooth trigger on a very powerful R6. Photo Courtesy of Louis Gagne.

Correction: We got a picture switched in the last article. The caption is for head skimming, but the picture is of a head in a Serdi Micro valve-seat-cutting machine. Sorry.