(909) 838-4587 ed [at] le-suspension.com

Tribal Knowlege TZ250 Tuning Secrets

“Tribal Knowledge” is an unfinished work based on what I learned about tuning a TZ250 by racing a few of them in Hawaii and North America. It was my idea but I needed a writer. Jim Hubert wrote it based on a few long phone conversations. Michael Gougis helped with the editing. The knowledge herein is not mine, it is the shared knowledge of a community of racers and it is yours to use. By using this knowledge you are obligated to pay it forward.

Ed Sorbo 3/26/14

The missing part number for the Kawasaki wrist pins is 13002-1066. Ed Sorbo 4/8/14

Tribal Knowledge

(The Care and Feeding of a TZ250)

Copyright 2004, Jim Hubert
This article is the culmination of a series of conversations with former 250cc GP front-runner, Ed Sorbo. For those of you who don’t know Ed…

Best race of all time: Brainerd in ’99 Finished 6th. Started from last on grid. 15 MPH slower than the guys racing with. Announcer said he moved from paradise to live in a motorhome. That is why! WRITE INTO A STORY

From the outset, Ed wanted it clear that although he might be the one willing to sit still long enough for me to write it all down, the information contained herein is not all his own discovery. Ed feels indebted to a number of racers and tuners, especially his racing partner — Bruce Lind. In fact, it was Bruce who coined the phrase “tribal knowledge pool” when referring to the body of information amassed by TZ racers. This article is a start at disseminating that knowledge. Most of the information is hands-on nutsy-boltsy stuff but there is a fair mix of theory for those of you who want to know “why” instead of just “what.” Don’t be scared off by this because, as Ed is fond of pointing out, “If you’re smart enough to ride the motorcycle, you’re smart enough to work on it.”

This article discusses the 50.7mm-stroke vees from model years 1991 to 1999. Those model years comprise the bulk of the machines inexpensively available. If fact, Ed estimates there are probably 1000 such machines in the US today. The AMA’s opinion notwithstanding, 250cc 2-strokes are not going away overnight. Considering that most parts for the ’99 model are backward compatible to the ’91 machine, and that Yamaha is obligated to provide parts for 10 years for a given model year, racers should have ready parts availability until at least 2009. Additionally, Ed believes that if there is sufficient demand, aftermarket suppliers could step in to fill the void.

In many ways, the short-stroke vee engine represents a good compromise for the racer on a tight budget. For one thing, the engine cases — which are based on the TZR250 street bike — are pressure cast. This equates to a longer lifespan than the sand- cast cases used on 2000+ models. Similarly, the short-stroke engine should have a longer service life than the “square” motors which are more heavily stressed.

Ease of Maintenance: Ed says many racers get the wrong impression of how much effort is required to keep a TZ in top condition. “They see us working on our bikes all the time and figure it’s a bitch to keep them running right.” Quite the opposite is true, “We work on the bikes because it is so inviting. Ease of maintenance was high on the list of priorities when the TZ was designed.” For example, an experienced mechanic can remove and replace a TZ crank in an TWO HOURS, BRUCE AND I DID IN 1.5 hours at the racetrack. The crankcases split vertically, so that the bulk of the engine can remain in the frame for the procedure. If you destroy the crankshaft in a 4-stroke, your weekend is over.

Full Disclosure: Racers endorse products and services for a variety of reasons — sponsorship, contingency money, perceived value, future consideration, and friendship immediately come to mind. When we mention a product or service by name, it is entirely possible that an alternative, competing one, will work just as well, perhaps better, for you. We have endeavored to be as objective as possible. In some cases (Rino fender, for example), there is no competing product. In others cases (Kawasaki wrist-pin trick), we hope it is apparent that we don’t have an agenda to forward. However, in the interest of full disclosure, we want you to know that Ed was WAS AT DIFFERENT TIMES (IS?) sponsored by Silkolene, and received contingency money from VP Fuels. OTHERS? BRIDGESTONE, EBC, SILKOLENE, TSUBAKI, RK, BARNETT, SPEED TUNE, VP

Engine Work

In the old days, engine builders used to talk about “blueprinting” an engine. Although this word means different things to different people, racers came to think of it as correcting the undesirable effects (those most responsible for lost horsepower) of accumulated tolerances in the manufacturing process.

I gained an appreciation of just how good Japanese manufacturing processes have become when replacing California cams with 49-state cams in a 4-cylinder 600. Although the new cams came from a different model year, only one of the sixteen valve shims needed to be changed to achieve the proper clearances.

With that understanding, the following recommendation are not so much intended to correct for sloppy manufacturing as to gain performance by moving the engine a little close to the ragged edge (while still maintaining an acceptable safety margin).

Before we start, it is probably best to get a simple matter of nomenclature out of the way. Throughout this article, we are going to refer to the cylinders as “top” and “bottom”. Although Ed says he has heard a variety of names, those are the most commonly used. There are really are no incorrect names, as long as its clear which cylinder you are talking about. It’s pretty well evident that top, left, and rear all refer to the same thing (as do bottom, right, and front). But what about those “F1″ and F2” timing marks on the flywheel? I just try to remember that the number 1 cylinder is the top, or “best” cylinder.


Deck Height: Deck height refers to the difference in height between the top of the cylinder and the piston crown when it is at top dead center (TDC). THE FLAT PART, NOT THE LITTLE STEP IN THE CENTER. Positive deck height means the piston will protrude out of the cylinder. Negative deck height means the piston will come up short of the cylinder. “Zero deck” means there is no height difference. Changes in deck height can be achieved by raising or lowering the entire cylinder (relative to the engine cases) by varying the thicknesses of the base gasket. Ed recommends using 0.3mm of positive deck height.

When buying base gaskets, Ed says to get the new rubberized steel ones intended for the 2002 model. They fit back to ’91 and are reusable. Although they do cost more than the conventional paper gaskets, you’ll save money in the long run and they are far less hassle as you will see in a moment. These new gaskets are available in 0.2mm through 0.8mm thicknesses in increments of 0.1mm. Their part numbers are listed below:

5KE-11351-20, 0.2mm
5KE-11351-30, 0.3mm
5KE-11351-40, 0.4mm
5KE-11351-50, 0.5mm
5KE-11351-60, 0.6mm
4JT-11351-02, 0.8mm (ED, THIS IS PAPER!)

Note, however, that Ed does not really consider the 0.2mm base gasket reusable as the rubber peels off easily. For the really finicky, “kit” base gaskets are available in increments of 0.025mm. Unfortunately, the average racer cannot get access to the YEC kit parts. They are made available to Yamaha Distributors and selected individuals and teams only.

Identifying Base Gaskets: You can recognize the reusable gaskets easily if you know what to MINE HAVE AN ALL. CORE look for: The rubber exterior is black, and a magnet will stick to interior steel portion. They are also stamped with their thickness. In contrast, the older paper gaskets have an aluminum core. They also have a grey sealant (looks a bit like Yamabond) applied to the perimeter on the “case” side. The worst thing about old-style gaskets is that their claimed thickness is just a nominal value. They will compress on the order of 0.1mm. But no two will compress exactly the same, and the amount of compression varies with thickness. Ed said he would typically install and remove the old-style gaskets at least three times before finally FINDING THE RIGHT SIZE leaving them in place.

Squish Clearance: In 1996, Yamaha began making heads with a bronze combustion chamber insert (said to be more resistant to detonation). Then in 1997, TZs were first configured to run unleaded fuel. Although Yamaha was able to get the same power output using unleaded fuel, Ed says, “You have to rev the piss out of them to do so.”

Assuming it is legal for the class, you will get the most bang for the buck by using leaded fuel in all model years (after the heads have been configured to run it). If you are in a position to upgrade, Ed recommends buying ’97+ heads. All TZs work better with the unleaded head shapes, but with the combustion chamber volume and squish clearance optimized for leaded fuel.

To measure the squish clearance, remove the head and lay a piece of 0.032″ diameter solder across the top of piston in the direction of the wrist pin. Temporarily reinstall the head without the o-rings and using only two diametrically opposed head bolts (properly torqued). Then rotate the engine by hand and bring the piston through TDC to “squish” the solder. Remove the now-deformed solder and measure it with a micrometer. Ed says, “0.6mm is a safe squish clearance.” Adding, “As little as 0.4mm ( .5 )could be used on a slow TIGHT track where the engine will not be highly-revved for prolonged periods.” THE PISTON WILL TOUCH THE HEAD AT .4

Erring Tight or Loose?
Assuming no machine work is done to the head and/or cylinder, and that base gaskets are available in 0.1mm increments, shooting for a 0.6mm squish will result in an actual value within the range 0.65mm (too big) to 0.55mm (too small). On a tight track, Ed says to err on the small side (especially if the volume of the head is greater than optimal). On the other hand, at a track like Daytona where the engine will be highly revved for a long time, Ed would err on the cautious side. In any case, it is probably best to share any error between the deck height and the squish clearance. SHARE WHAT???

Alternatively, if machine work is planned, it is possible to hit the 0.6mm figure exactly. Ed says that because Yamaha’s manufacturing tolerances are so good, there is extremely little variation from part to part. This means that once the proper deck height and squish clearance have been achieved, it is likely to stay that way even though pistons and cranks are periodically replaced.

“I need to tell you about cutting the cases down to allow for +0.3 deck height. Cutting the cases costs a lot less than cutting the cylinders. You don’t have to do it to the old bikes when using the new gaskets but I still want to include it.”

Ed recommends Roland Cushway to do heads.
Building on a Budget:
For optimal performance you would, of course, want to follow the procedure outlined above. However, if you are on a tight budget, you’ll probably just want to optimize the parts you already have and forgo any machine work. Given a choice between having the correct deck height or having the optimal squish clearance — choose optimal squish every time. This means that you will select a base gasket thickness that will yield the proper squish clearance and allow the deck height to fall where it may.
Every year Yamaha subtly changes the TZ’s ports, and yet the power feels no different. All manufacturers do this because they have to change something to get you to buy the new model! As far as porting cylinders goes, Ed feels the return on investment is just not there. About the only worthwhile porting modification is to square off the corners of the transfer ports. If you feel compelled to “port” something, you can always do so by fitting a different thickness base gasket (which will raise or lower all the ports without removing any metal).

Case Packing:
Sometimes called case stuffing, Ed says, “… packing only moves the power around, it does not make more power.” Technically speaking, case packing decreases the crankcase volume which in turn yields a higher primary compression ratio (the ratio of crankcase volume when the piston is at TDC versus BDC). Although case packing can lead to improvements in volumetric efficiency, it also results in increased pumping losses (energy required to draw air into the cylinder).

If you are going to try case packing, start with the area around the reed cage — that is going to make the biggest difference. Also, it is also the easiest place to start as the area is big and it won’t easily break off. (The other region for case packing is the area under the pistons, but that is much harder to accomplish.)

For packing putty, use Moroso 35560 which is a 2-part epoxy. It turns into hard “play-doh” quickly, but the mixture can be worked for as much a 2 hours afterwards.

Prior to applying the putty, rough up the surface really well (using punch marks, etc.), and scrub with soap and HOT water. Don’t use contact cleaner because it can leave a residue to which the putty may not adhere.

Start with a small amount of putty (you can add more later and work it together) and do just one of the four sides. Let it harden a bit and check if the reed cage still fits. When you remove the reed cage, you can see where it touched. Simply shape and smooth the high spots with your finger wetted in water. (The putty will not stick to you with water on your finger, and you can achieve a mirror-like finish.) Let the putty harden overnight, then do another side. If you attempt to do all four sides at once, you probably won’t even be able to get the reed block into the hole.

The only down-side to packing the reed cage area is that you have to remove the upper cylinder from the cases in order to install/remove the LOWER cage *if* you pack the area as tightly as possible. This obviously complicates matters if you have to replace the reeds at the track. That is why Ed recommends packing THE LOWER REED AREA WHILE the upper cylinder it is bolted to the case half. This ensures you will not “over-pack” the area.

Pressure Testing:
A big air leak is the only thing that will cause a TZ to seize a piston. The seizure occurs upon closing the throttle and air is sucked in on the intake side — usually around the intake manifold. Ed learned this the hard way at (WHERE?) PIR during his first AMA race. He crashed after closing the throttle at around 100 MPH for turn 1, and as Ed put it, “…zoop, I’m in the air! Now, Ed pressure tests his engines *every single time* they have been apart.

From my own experience, a pressure test is every bit as important as checking that the axle nuts are tight! A pressure test revealed the previous owner of my bike had installed the intake gaskets on the wrong side of the reed cage (they should go between the cases and the cage). The leak was so pronounced, I was unable to build any pressure at all using a hand-held pump. After a lot of head scratching, I finally resorted to my air compressor and a *carefully adjusted* pressure regulator to discover the source of the leak. Pressurizing the cases with anything other than a hand- operated pump is usually not a good idea because of the risk of over-pressurization (using more than about 8 psi can damage seals).


In retrospect, I should have had a clue something drastic like an air leak was wrong when changing a single jet size make the bike go from acceptable to un-ridable. (Ed says changing a single jet size in a properly operating TZ should never have a drastic effect. If it does, you should be looking for the true source of the problem.) What I think may have happened is that installing and removing the carb torqued the intake manifold so as to improve or worsen the leak.

Building a Pressure Tester:
Although you can buy a pressure tester from Motion Pro or Swedetech, building one is an simple task — especially if you have access to a lathe. THAT IS NOT SIMPLE

The basic idea is to plug the intake and exhaust ports, pressurize the cases to about 7 psi and observe the rate of leakdown. Ideally the pressure will stay at 7 psi. In practice, a leakdown rate of less than 0.5 psi per minute is acceptable. To discover the source of a leak, spray soapy water on the suspected area and watch for the appearance of bubbles.

When pressure testing a TZ, it is possible for the center crank seal to allow air to communicate between the cylinders, but only in one direction (for some reason this seems to be more of a problem with the newer bikes than the older ones). Ed says, tongue in cheek, “If your present center crank seal does not leak, your next one will!” Thus, both cylinders should be simultaneously plugged for testing.

Plugging the exhaust side of a TZ is simple because you can buy expansion plugs intended for pressure testing plumbing at many hardware stores. Buy the size that fits inside couplings for 1 1/2″ PVC or ABS pipe. I USE RUBBER FREEZE PLUGS FOR CARS. A wingnut forces the rubber to expand and provide a good seal. Clean the exhaust port well, otherwise the plug will not be able to seal against the lubricated surface. And wear safety glasses when you test — the plugs can and do pop out.

Capping the intake side requires making some solid plugs that fit inside the intake manifolds to replace the carbs. These plugs should incorporate a fitting to which you can attach a hose. (For hose, I used surgical rubber tubing –like you would use to build a slingshot — because it seals well without clamps.) FUEL HOSE A low- pressure gauge, a tee fitting, and a pressure source are the only other items required. I used the pressure port on a Mityvac vacuum pump for the latter. Some sources recommend the squeeze bulb from a blood pressure measuring apparatus. THATS WHAT I HAVE

Preventing Leaks:
Because the most common place for an air leak to develop is around the intake manifolds, some extra care is needed there during assembly. When changing intake gaskets, don’t over-tighten the 6mm bolts securing the intake manifold (96 in-lbs is the proper value) as that will warp the reed cage. 8 FOOT LBS, INCH LBS FREAK PEOPLE OUT

Additionally, Ed says there are three levels of thoroughness in assembling the intake manifolds. The first is to simply apply grease between manifold and reed cage. The second is to use Yamabond on both side of intake gasket. However, the best NOT BEST, LAST RESORT precaution is to permanently attach the manifold to the reed cage with epoxy and screws. This is also affords an opportunity to “port” the intake manifold by applying case packing putty. YOU CAN PACK WITHOUT GLUEING TOGETHER. Ed says that Kit intake manifolds lack the rubber ears that protrude into the reed cage stating that they resonate at certain engine speeds and negatively impact performance. Ed say to replicate the rubber ear shape with putty, and make the fit as tight as practical. NO, I USE PUTTY TO HOLD THE EARS IN PLACE. THE KIT SETUP IS A MANIFOLD WITHOUT EARS AND A INSERT THAT FITS INTO THE CAGE.

The other place for a leak is around the powervalve. Although a leak on the exhaust side will not a cause an engine seizure, it will prevent you from determining if there is a leak elsewhere, and thus must be fixed by replacing the powervalve o-ring. While you have the powervalve apart, be sure to check the little screw inside that holds the guillotine down so it can’t come up too far (they vibrate loose). The screw must be red Loctited. Guys never take the cover off because it is a messy job. Inspect it once or twice a season. Ed says there are a lot of guys who are going to read this and go “oh-oh” because they have never done it. Those are the ones who need to check. Also, the older bikes use a Phillips head screw which should be replaced with an Allen head screw.



There is a saying in mountaineering that goes, “There are many ways to reach the summit, but you must pick just one.” The path Ed advocates you follow starts with using VP Fuel’s C12 fuel. C12 was VP’s first product. It is a non-oxygenated fuel containing 4.23 grams of lead per gallon giving it a motor octane number of 108. Its specific gravity is 0.717. Ed says *everybody* uses VP’s C12 because it is relatively inexpensive and readily available. Furthermore, C12 has a big advantage over more exotic fuels in that it “stays C12 for a long time.” You can carry it (mixed) from race to race and it may be good for as long as 2 months. Expect to get about 20 miles per gallon at racing speeds.

If money is no object, an alternative fuel is VP’s MR8. It is oxygenated (2.55% by weight) and has a greater lead content (6 g/gallon) than C12. Although MR8 will yield more power, it is substantially more expensive and does not age well. Using MR8, Ed gained 500 RPMs at the end of the straight at Portland. Ed advises that if you are going to try MR8, it should be used within a day or two of mixing with oil. The beauty of MR8 is that it defines horsepower in a can. (Just increase the main jets by 2 sizes bigger what you would use with C12.) For this reason, Ed has, on occasion, practiced with C12 and then switched to MR8 for qualifying and racing.

Dangers of Leaded Fuel:
Tetra-ethyl lead (TEL) is the anti-knock additive used in all leaded fuels. This compound is the most cost-effective method of enhancing octane and was added to gasoline in ever-increasing concentrations from the early 1920s to the late 1960s.

Today, the only fuels in which lead can still be used are intended for aircraft and racing. And racers are often negligent with health hazards that aren’t immediately apparent (lead poisoning, like all heavy metal poisoning at lower doses, is systemic and chronic; you won’t even feel your brain cells slipping away).

I searched VP Fuel’s web site in an attempt to find out if they recommend any additional handling precautions over their unleaded racing fuels. Unfortunately, their web site provides different links to the same Material Safety Data Sheets (MSDS) for both leaded and unleaded fuels. I think the Phillips web site has a much better MSDS for its TT111 racing fuel (which has about the same lead content as C12.)

Quoting Chevron Phillips Chemical Company, “This product contains lead. Fuels containing lead anti-knock compounds should be handled in such a way as to minimize contact with the body. Lead can accumulate in the body with overexposure and cause illness due to effects on the blood, nerves, kidneys and the reproductive system.”

A chemist friend who did not want to be quoted for this article told me that TEL can be absorbed through the skin because it’s lipid soluble and our skin is a lipid bi-layer. Solvents (like gasoline) “de-fat” the skin, that’s why your skin can feel cracked and dried after brief exposure to common solvents. When you spill gasoline on your skin, while the fat in your skin is absorbing some components of the gasoline, the gasoline is “rinsing away” some of your fat. The overall transfer of gasoline components to your body is lessened by the rinsing effect. However, the effective transfer of TEL could be much greater than the change in concentration alone would lead you to believe.

What this all means to me is that I always wear nitrile mechanic’s gloves when changing jets, and never eat anything until I have washed my hands thoroughly.

Premix Oil:
Ed says, “Use plenty!”; recommending a 24:1 ratio. Most racers know this, but we’re going to say it anyway — the more oil there is in the pre-mix, the less fuel a given volume of pre-mix will contain. What this means, in theory, is that going richer on the pre-mix may require richer jetting so that the amount of fuel stays the same. In fact, some racers claim they will alter the amount of oil in the pre-mix to achieve half-size changes in jetting.

Let’s look at the reality of this situation. Say you have been running the Yamaha-recommended 30:1 pre-mix ratio and now want to run 24:1. At 30:1, every liter of pre-mix contains 30 parts fuel and 1 part oil (or a total of 31 parts of fluid per liter). The oil fraction is 1/31st of a liter, or 32cc. The fuel fraction is 30/31st of a liter, or 968cc. At 24:1, every liter has a total of 25 parts fluid. The oil fraction being 1/25th of a liter, or 40cc. And the fuel fraction being 24/25th of a liter, or 960cc. Going from 968cc to 960cc is less than a 1% change. Thus, if your former “perfect” jet was a 370, the new pre-mix ratio would require a 373. Not a jet you’re likely to find in your spares. The bottom line is that to get a half-size jet change when running a 370 would require a pre-mix change greater than from 30:1 to 24:1.

Another point to consider is that oil is not just a lubricant, it is also a “fuel”. In fact, the energy content (measured in Btus per unit mass) of oil is typically somewhat higher than that of gasoline. Thus, burning oil adds energy to the combustion process.

However, oil does lower the octane of the fuel in which it is blended. To combat this, certain 2-stroke oils (e.g., Silkolene Pro2) have been formulated to maintain the octane rating of the fuels in which they are blended. Conversely, oils like Silkolene’s Comp2 (which is primarily intended for MX bikes) do not. Comp2 oil is about half the cost of Pro2, and Ed says it seems to work just as well as Pro2. Ed believes this is true because C12 fuel has more than adequate octane for a TZ.

The point to take away from this is that you need to use a relatively large quantity of a quality oil. You won’t be doing your engine or yourself any favors by trying to run an outboard motor oil at 100:1! And speaking of such, many years ago Gordon Jennings reported that an outboard manufacturer published research showing an increase in power as the oil fraction was raised in 2-stroke pre-mix. They concluded that something around 10:1 produced the best power. Because this ratio was not economical, smoked greatly, and tended to foul plugs, they felt that around 20:1 was probably a practical limit and a good compromise. I wonder if that would still be true with today’s oils?

Ed recommends using distilled water with a “wetting” agent. Water Wetter, a product of the Red Line Synthetic Oil Company, is one example of such an agent. At the least, its properties are useful to lubricate the water pump seal. Whether or not it actually increases the thermal conductivity of the coolant has not been verified by the author.

If the claim is true, the product must remove more heat energy from the engine. This means that for the same volume of water flowing per unit time, the temperature must go up on the engine outlet side. When this volume of water comes across the radiator, it sheds this larger energy by having a greater delta T (change in temperature) across the radiator, just as it has a greater delta T across the engine.

One could verify the claim by measuring delta T of the water across the engine, or across the radiator, or by measuring the delta T of the air across the radiator.


Because they lack idle-stop screws, a TZ won’t idle unless you hold the throttle. This is not really a problem since TZs won’t run very smoothly below about 4000 rpm anyway. When warming up, Ed usually holds the throttle to achieve a minimum running-speed of about 4000 rpm and blips the throttle higher from there. He says there is no benefit to warming the engine up at lower rpms — it just takes longer. Conversely, he does not like to “rev the piss out of it” while warming up either. Stay in the middle area, averaging around 6000 rpm.

Ed also likes to be cautious when the engine is dead cold. The temperature gauge on the new bikes just reads “cold” below 20C. Although Ed prefers to bring the water temperature up gradually when the engine is that cold, he doubts anything could really get hurt even winging the throttle. (You *can*, however, damage the engine when it is cold and you rev it too high under load. But in the pits, this is not going to be an issue.)

Ed feels a water temperature of 40C is okay to leave the pits. But he wants it at least 45C before using full throttle. 60C is the engine’s optimal operating temperature. Generally speaking, 2- stroke engines make maximum power at much lower coolant temperatures than do 4-stroke engines.

Keeping the bike at its optimal operating temperature is somewhat of a challenge since TZs lack thermostats. Typically duct tape on the radiator is used to raise the water temperature when racing in cold weather. But duct tape there can disrupt the aerodynamics of the bike. Andy Edwards proved this on his land-speed record holding TZ250. Andy gained 5 MPH by removing 3 pieces of duct tape from his radiator.

However, a more difficult problem to solve is keeping the temperature down to 60C in hot weather. Often, the solution is to fit a larger radiator from a different motorcycle. For example, on years which employed a flat radiator, fitting a radiator from a similar vintage GSX-R600 is a possibility. However, different spigots would have to be welded on along with different mounting brackets. Instead, Ed recommends installing ’96 bodywork on an older bike, and fitting the curved radiator from a newer TZ. On bikes that already have the curved radiator, Ed is not aware of anything better than the TZ kit radiator.

Finally, for those of you who are concerned about such things, Ed thinks the correlation between the older mechanical temperature gage and the new electronic thermometers should be quite close.

Gearbox Oil:
You can’t judge if the gearbox oil is old by how the bike shifts. Since no combustion byproducts are present to contaminate the oil, shifting does not seem to degrade as the oil gets older.

How does Ed know when to change the gearbox oil? He looks at it when changing internal gear ratios. If it looks good, he just reuses it — otherwise he changes it. What does bad oil look like? It may sound crazy, but think about in terms of drinking it. New oil is so clean it looks like honey — almost good enough to eat. As it get dirtier, it looks less and less like something you’d want to ingest.

Ed uses Silkolene SRG75 and changes it “occasionally”. He knows some guys who change gearbox oil at every race. Club racers — running the same track, never changing internal ratios — might want to get on a regular schedule (maybe every 3 races). At the opposite extreme, Rich Oliver changes his oil every session.


You don’t *have* to buy the most expensive fancy oil. Ed has even run regular motorcycle engine oil. A less expensive oil specifically for gearboxes will provide just as much protection so long as it is changed frequently. The expensive oils are typically low-viscosity — offering good protection while giving up less horsepower to friction. In addition, you may be able to go longer between oil changes.

Real horsepower can be lost to friction in a “splash” type gearbox. Although modern TZs’ gearboxes are pressure-fed using an oil pump, this was not always the case. In his __Sportbike Performance Handbook__ Kevin Cameron writes, “… Kawasaki’s KR250 two-stroke road racer had a very tight gearcase, and when given its ‘correct’ fill of 1,100 cc of gear oil in 1975-76, it got so hot it burned the black paint right off the gearboxes. Cutting the fill to 600 cc made the situation tolerable if not desirable. Gearboxes on many modern two-stroke Grand Prix (GP) bikes are therefore pressure lubed….” Those of you with non-pressure-fed gearboxes (vintage racers), need to keep this in mind.



Assuming that your are starting with a sub-optimal configuration, and that repeated changes in the same direction make performance improvements — there will come a point at which further changes begin to make things worse. What you are looking for is the peak. Other factors being equal, a tuner will strive to err on the conservative side of the peak.

Treat your 250 as two separate 125s. You’ve probably heard the term “stagger” used in conjunction with TZ250s. This refers to the asymmetric tuning needed to achieve optimum performance.

Asymmetric tuning is needed because there are mechanical differences between the two cylinders. Each cylinder’s primary compression ratio (which is dictated in part by crankcase volume) is different. This is due to the positioning of the center crank seal. One crankcase has two crank bearings housed within its crankcase volume, whereas the other crankcase houses only a single bearing. There are also differences in the intake and exhaust tracts that require asymmetrical tuning. For example, packaging constraints (how the engine fits in the frame) dictate the exhaust pipes not be identical to one another. The top cylinder is fitted with larger jets, and can tolerate more ignition advance. One way of remember this is that the top cylinder works harder.

Ignition Timing:
Incorrect timing can easily blow up an engine. Set your TZ’s timing by looking at the “burn back” on each spark plug’s ground electrode. Burn back is defined as the length of the clean (without carbon deposits) portion of the ground electrode. Generally, one practice session is not long enough for the burn back to stabilize; you’ll need at least two 20-minute sessions.

Starting at the void (region where the spark will form between the center electrode and the ground electrode), you want the ground electrode to burn back 3/4 of the way to where it attaches to the threaded spark plug body. See Figure x.

The timing must be evaluated and set independently for each cylinder. Ed thinks you will probably find that a stagger of 0.3 – 0.4mm on the ignition timing will be required to achieve equal burn-back.

Unlike a 4-stroke engine, where timing is measured in degrees, you can very accurately measure a 2-stroke’s ignition timing with a dial indicator as a distance from BTDC. You’ll find the procedure for this in your service manual. (This article is not a replacement for a service manual and a parts catalog. You do have a service manual and a parts catalog, don’t you?) THIS IS IMPORTANT AND NEEDS TO BE AT THE START.

Once the correct stagger has been determined, you can make subsequent timing adjustments by moving the entire stator plate. This allows you to set the timing via just the top (more accessible) cylinder. The angular movement will be the same for each cylinder, and the relative difference between the cylinders will be maintained almost perfectly. Based on the TZ’s stroke (50.7mm) and rod length (110mm) the following tables can be computed:

0.1mm steps

Advance BTDC Advance BTDC Delta Advance
millimeters degrees degrees
1.0 14.57 ****
1.1 15.29 0.72
1.2 15.97 0.68
1.3 16.63 0.66
1.4 17.27 0.64
1.5 17.88 0.61
1.6 18.47 0.59
1.7 19.05 0.58
1.8 19.61 0.56
1.9 20.16 0.55
2.0 20.69 0.53
Looking at it from another point of view:

0.5 degree steps

Advance BTDC Advance BTDC Delta Advance
degrees millimeters millimeters
14.5 0.99 ****
15.0 1.06 0.07
15.5 1.13 0.07
16.0 1.20 0.07
16.5 1.28 0.08
17.0 1.36 0.08
17.5 1.44 0.08
18.0 1.52 0.08
18.5 1.60 0.08
19.0 1.69 0.09
19.5 1.78 0.09
20.0 1.87 0.09
20.5 1.96 0.09

For example, with an initial timing of 1.2mm and 1.5mm (exactly 0.3mm of stagger) advancing the timing by 1.0 degree results in a timing of 1.353mm and 1.670mm (now the stagger is 0.317mm). You have to carry the calculation to 3 decimal places to see the error. Setting the timing to an accuracy of better than +/-0.05mm is pretty unlikely — so for timing changes you are likely to make (around 1 degree) the error can be neglected. If, however, you intend to move the timing by, say, 3 degrees you may want to timing the cylinders independently.


Ignition timing and combustion chamber volume are interrelated. Combustion chamber volume determines compression ratio. No amount of rich jetting will save you from detonating an engine that has incorrect timing TOO ADVANCED or too small a combustion chamber, or for that matter inadequate fuel octane.

The stock timing number is a good place to start. Advancing the timing (firing the spark plug when the piston is farther away and approaching TDC) is less conservative and more prone to detonation.

Advancing the timing typically yields a boost in low-rpm performance, but it may inhibit the engine’s ability to rev freely. Conversely, retarding the timing allows more heat into the exhaust system. LETTING THE ENGINE RUN COOLER.

In some regards, the temperature of the exhaust gases themselves may be used as a tuning tool. In a 2-stroke, the acoustic waves produced by the engine are put to use by the exhaust system. REFLECTION

A 2-stroke engine’s exhaust system is designed with a specific length — called the tuned length. In concert with the port timing, the tuned length determines the engine rpm at which peak power will occur.

In explanation: The speed of sound is proportional to the temperature of the medium in which it is traveling (hotter = faster). The tuned length of the exhaust system assumes a certain wave speed which, in turn, assumes a certain gas temperature. Thus, the acoustic wave speed (and peak power rpm) varies with exhaust gas temperature. THIS IS GOOD STUFF!

By altering the exhaust gas temperature via timing and jetting, the wave speed changes. This interaction with the tuned length can be used to move the peak-power rpm. YES, IT HAPPENS, BUT WE DON’T NOTICE.

Jetting is the key to making horsepower! Do not expect to jet the bike correctly by feel — it can’t be done. The main jet would have to be 10 sizes too rich to feel a difference in power. And, luckily, with leanness, a TZ will just get slower, and slower, and slower. It won’t blow up. You could use a number 100 main jet; the bike will be slow; the engine will run hot; but it won’t blow up. The only way you can blow up a TZ is from an air leak (see the section on pressure testing).

Remember our mountaineering saying, “There are many ways to reach the summit, but you must pick just one.” The path Ed advocates you follow to reach your jetting goal is to start with VP’s C12 fuel and a premix ratio of 24:1. This combination will make the following jetting information relevant.

Recommended Settings:
Ed’s recommended settings are to use the standard R-6 nozzles; 17.5 pilots; adjust the air screws for 1.75 turns out; and get 4.0 cutaway throttle valves (slides). Bruce Lind came up with 4.0 slides by trying *all* of them. NOT TRUE, I HEAR.


The higher slide cutaway lets more air when the throttle is closed. Yamaha made its standard settings rich to be safe when the throttle is closed, but these settings do cause some performance issues. One example: Just cracking the throttle after it has been closed and the power delivery comes on too abruptly. This occurs because there is too much fuel in the system, and when you get back on the gas the engine “burps.” The same effect can be noticed when you close the throttle for a moment in something like the kink at Road America (You have the throttle pinned; back off for a second; then pin it again.) If it burbles, that is a sign of richness which can be solved with the higher 4.0 slide cutaway.

Ed says to use stock needles and keep the clip in the middle (#3) position. If you need to adjust the midrange F/A mixture, Ed recommends that you change the nozzle instead of the clip. (It should be a clue to you that Yamaha never intended you to change clip position because it is not easy to do.)

Main Jet:
There are many methods to arrive at the proper main jet, and we will detail many of them later in the article. For now, just remember there is no single “magic” jet. Unfortunately, many racers have preconceived ideas about what the main jet should be and are reluctant to follow the tuning advice of experts. You often hear them say things like, “But then I’d have to install a (pick a number) jet — and I can’t do that!” After hearing a statement like this, Ed likes to quote Australian racer/tuner and general 2-stroke god Kel Carruthers, who used to say, “Never mind the bloody number mate, give it what it wants.”

Carb Sync:
Carb synchronization must be *exact* (a 1mm difference in slide height is intolerable). You can easily tell if the carb sync is correct just by listening. When you rev the bike in neutral it should sound like a nasty four-cylinder, 4-stroke. If the slides are out of sync, it will sound like an EX500 (parallel twin). Use your finger to feel for a protrusion of the slide into the venturi. Adjust both carbs exactly the same, and make sure both slides can be retracted completely out of the airstream.

Power Jets:
The power jets add fuel in the midrange. Generally, leave them stock. You can add a stagger to the power jets if you have 6 or more sizes of stagger on the main jets.

The electrical part of a power jet (the solenoid) sometimes goes bad. A flaky power jet is difficult to diagnose. The engine just gets a bit lean, but only on one cylinder. Therefore, it is a good idea to test them occasionally.

To test a power jet, simply unplug the solenoid from the wiring harness and unscrew the solenoid as if you were going to replace the jet itself and apply 12 volts DC from a battery. (The solenoid is not polarity sensitive.) You should see the pintle retract and extend each time you apply and remove power.

Because it is next to impossible to repair the solenoid if its electrical connection comes off internally, it is prudent to zip tie the cable to the body of the power jet. This preventive maintenance provides strain relief and should keep the cable from becoming detached in a crash or due to rough handling.

Carb Tricks:
Remove the carb overflow hoses (sometimes called vent hoses). It does not matter if they are there or not.

If the weather where you race permits, remove the choke components and plug the carb body with epoxy. This makes the carbs easier to get out of bike.

Cut about 1/4″ off each throttle return spring. This will make the throttle pull a little lighter.

Replace the Phillips head carb clamp with an Allen head screw. You will likely have to use a ball-end wrench for installation and removal.

When changing jets, use a Ratio Rite to catch fuel coming out of the carb. An old, rimmed, cookie tray works well for bottom carb. Ed likes to change jets in the bottom carb with front wheel on a stand so any stray fuel won’t contact the tire. Don’t remove the float bowl completely. You can access the main jet and the emulsion tube by just removing the 17mm nut at the bottom of the float bowl. When re-installing the nut, be careful not to tighten it as that will just make it more difficult to remove next time.

Dial-a-jet is “bullshit.” Can’t imagine it is very accurate.

Speedtune is a product marketed by AIM Racing Products. It is a system for measuring exhaust gas oxygen (EGO) content. Tuning based on EGO is one method of correctly jetting the bike. When using an unpressurized airbox, a reading of 25 on the Speedtune display is optimum. A bike with a pressurized airbox requires a reading of 40 (which is richer).

These readings are only valid at wide open throttle (WOT) at the end of a straightaway. If you say, “But straights vary in length.” Ed says, “Exactly!” That is the point, you are jetting your bike precisely for that track, under the prevailing ambient conditions.

One jet size will change the Speedtune display by about 5. Although you are shooting for a specific value, say 25, you are not going to get it closer than 1 jet size, anything within the range of 25 to 30 would be acceptable.


Testing an EGO Sensor:

“Use a high-impedance DC voltmeter. Use a propane torch set to “high” and the inner blue flame tip to heat the perforated area of the sensor. You should see at DC voltage of at least 0.6V within 20 seconds. If not, the most likely cause is open circuit internally or lead fouling. If okay so far, remove from flame. You should see a drop to under 0.1V within 4 seconds. Is still okay, heat for a full 2 minutes and watch for drops in voltage. Sometimes the internal connections will open up under heat. If the sensor is okay at this point, and will switch from high to low quickly as you remove the flame, the sensor is good. Bear in mind that good or bad is relative, with port fuel injection needing faster information than carbureted systems.” — wps.com/LPG/o2sensor.html

Piston Color:
Piston color is another way to judge jetting. With an unpressurized airbox, the piston crown should be the color of coffee with two creams. Again, a pressurized airbox requires a richer mixture and the piston crown should look like black coffee.

With a brand new piston, if you run a 20-minute session; do a good plug chop and come back in — everything will still look brand new. As with judging burn-back, two sessions are probably the minimum time needed to get usable color. More time is better.

For example, at a club race, you should do a plug chop at the end of Saturday practice. Then in the evening, pull the heads and evaluate where you want the jetting for the Sunday race. The extended running time needed to see color changes (plus the fact that you must pull both heads) is why another system like a Speedtune or exhaust gas temperature (EGT) is so valuable.

Be aware that it is possible to have so much fuel going through the bike that the piston crown will be washed clean. The uninitiated may assume that lack of color means the jetting is lean. The way to know that the opposite is true is to compare your bike against another TZ250. If your buddy’s bike is “a frickin’ rocketship” and he’s using a 350 main jet and you are using a 450, the remedy should be obvious. That is the beauty of having a friend who races a similar bike. It won’t be exactly the same — you just can’t copy his numbers — but it will get you in the ballpark.

Exhaust Gas Temperature:
Install the EGT probes 6″ from the face of the piston. Use 1200F as the maximum peak EGT.

Bicycle Speedometer:
Jetting can be determined by trial and error based on top speed. For this, a bicycle speedometer — with a mode to capture the peak speed attained — is extremely useful. Simply swap main jets until the highest peak speed is found. Of course, there may be mitigating factors such as traffic; drive out of the preceding corner; head wind; tail wind; drafting; and even measurement error. You have to do your best to minimize these other factors when using this method.

Standard Deviation:
Even if you don’t use a bicycle speedometer to jet your bike, the following discussion on repeatability of measurements and standard deviation is worthwhile. It is worth re-reading until it makes sense to you.

Standard deviation is a term from statistics defined as “a measure of dispersion around the mean.” The point to remember is that any time a number of quantitative measurements are made, the values will not necessarily be exactly the same for a variety of reasons. There will likely be some “scatter” in the data.

Let’s say you repeatedly measure the top speed attained at a given track on a given day without making any changes to the bike. Further, let’s say the speeds you recorded were 138.2, 139.0, 137.6, 137.7, 138.1, and 138.4 mph. The average speed is 138.0 mph and its standard deviation is 0.8 mph.

Now let’s say you make a jetting change, go back out on the track and record a top speed of 139.0 mph. Was that change beneficial? Knowing the standard deviation can help answer that question.

In the example above, the jetting change resulted in a speed difference of 1.0 mph, which is greater than one standard deviation (0.8 mph), away from the average. So the answer to the question, “Was the change beneficial?” is, “Probably, yes.”

Why “Probably?” The farther a given change moves you from the average, the greater your confidence can be that the change was effective. Mathematically speaking, one standard deviation (in the above example 0.8 mph) gives a 68 percent confidence level. Two standard deviations (1.5 mph) would give 95 percent confidence. And three standard deviations (2.3 mph) would give 99.7 percent confidence.

The point is not to get bogged down in the math. Simply put, the less consistency there is in a series of measurements, the greater its standard deviation, and the less certain you are whether a small change is helpful or not. Personally, I would not put much faith in anything less than about 1 standard deviation since approaching a 50 percent confidence level is basically a coin toss.

Finally, remember that the concept of standard deviation is useful any time you can make repeated measurements under the same conditions: top speed, lap times, dyno testing, micrometer measurements, and so on.

One Change at a Time:
If you truly want to understand the effects of a change, hopefully, it makes sense to you to change only one thing at a time. The reality is that, in racing, there are never enough hours in the day. One has to make educated compromises and changes in which the effects can be independently evaluated For example, making simultaneous jetting and suspension changes. If the top speed improves but the lap time does not, then the jetting change was probably beneficial and the suspension change was probably not. Of course, the hard data may have to be tempered with subjective evaluations like, “I think the suspension change allowed me to get on the gas earlier permitting a better drive, possibly yielding the top speed increase… but it upset my confidence elsewhere on the track, resulting in the poorer lap time.” You just have to do your best under the circumstances.

Here’s one final thought on standard deviation: Let’s say the top speeds from the example above were actually recorded in the following order: 137.6, 137.7, 138.1, 138.2, 138.4, 139.0. A blind application of standard deviation would not be appropriate because there is a definite trend in the data (that of steadily increasing speeds). The trend is likely not due to random variations and there are other factors at work. The competent tuner looks for trends as well as considering the standard deviation.

Relative Air Density:
Once the bike is set up and running perfectly, your ongoing challenge is to keep it that way. A relative air density (RAD) gauge may provide the information you need to alter the jetting for different atmospheric conditions. RAD increases with an increase in barometric pressure (e.g., going to a lower elevation) and/or a decrease in ambient temperature. Conversely, RAD decreases with a decrease in barometric pressure and/or an increase in temperature.

Standard temperature and pressure (STP) is defined as a temperature of 59 degrees F with a barometric pressure of 29.92 inches of mercury. STP is one point at which 100 percent RAD exists. Although bike tuners typically deal with RADs that are less than 100 percent, a RAD of greater than 100 is possible. (Operating in a cold environment, snowmobile tuners regularly work with RAD numbers greater than 100.)

If your bike is jetted spot-on at a RAD of 100 and makes 100 horsepower, it could make 110 horsepower at a RAD of 110 with the proper jetting. Similarly, at a RAD of 90, the best you can hope for is 90 hp.

Making horsepower is all about providing the appropriate amount of fuel for the amount of combustion air available under the prevailing atmospheric conditions. Neglecting volumetric efficiency, each time a piston descends, an engine pulls in the same VOLUME of air, but the MASS of that volume of air varies with temperature, barometric pressure (elevation) and humidity. Chemically speaking, it is the mass of the air (oxygen portion) that determines the mass of fuel needed for optimal combustion efficiency (horsepower production).

Your job as a tuner is to jet the bike for the air density you encounter from track to track and day to day (and sometimes even from hour to hour). Typically, tuners using a RAD gauge will produce a copious notebook (developed by trial and error) detailing the proper jetting for a given RAD.

Why trial and error? Firstly, a change in RAD does not dictate a directly proportional change in jetting. (For example, a 5-percent change in RAD does not equate to a 5-percent change in jet size.) Secondly, identical RAD numbers can be achieved under vastly different atmospheric conditions. The same RAD number can be seen in a location with higher temperature and higher barometric pressure as one with lower temperature and lower barometric pressure. For example, when it is 100 F at Willow Springs (elevation of 2600 feet) the RAD will be about 84.5. At Pikes Peak (elevation of 5400 feet) the same RAD would occur at a temperature of 45 degrees F. Although the RAD numbers are the same, optimal jetting would be one size different for a nominal 350 main jet. You’ll see why shortly.

A simple mechanical RAD gauge can yield accurate results under the proper conditions. “Proper conditions” are low or constant humidity and, additionally, the elevation must be constant or the temperature must be constant. Although changes in barometric pressure due to weather are taken into account by a RAD gauge, their effects are usually minimal — a much greater factor is a significant change in elevation.

Most of the time, you can achieve satisfactory results with a mechanical RAD gauge and copious notes. On only two occasions does Ed claim to have been screwed by RAD. But, a mechanical RAD gauge is not the panacea it is made out to be, because it only gives a “bottom line” number without allowing you to see the individual contributions of the two factors — temperature and barometric pressure — that comprise that number.

The best way to jet a bike is to treat separately the two or three factors that influence RAD. Why two or three? Humidity sometimes enters the picture. When it is considered, we call the resulting value “corrected” RAD.

Because mechanical RAD gauges do not take humidity into consideration, many racers use spreadsheets or tables to account for it. In some of the country, humidity is a negligible factor. However, in large parts of the U.S., humidity is a factor. For example, at Brainerd 90 degrees F with 90% humidity is not unheard- of. Under these conditions almost 5% of the *available* “air” is displaced by water vapor. This results in a corrected RAD which is 4 smaller than the uncorrected value.

The word approximately was used above because 5 percent of the *available* is displaced. Brainerd is about 500 feet above sea level, at 90 degrees F, the RAD would be about 93, neglecting humidity. Saturation 4.8%, Partial pressure 4.3% Corrected RAD 89.

Jetting must change in direct proportion to absolute temperature. A 10 percent change in absolute temperature necessitates a 10 percent change in jet size. Absolute temperature is a scale referenced to “absolute zero” rather than to the temperature of freezing water. To convert Fahrenheit into absolute temperature, add 460. (This scale is called Rankine for those of you who care. It can be use to confuse your friends, as in “Boy it’s hot out today — 555 degrees R!) For example, the difference in absolute temperature between 70 degrees F and 90 degrees F is not quite 4 percent (90+460) / (70+460).

Jetting does *not* change in direct proportion to barometric pressure. A 10 percent change in barometric pressure necessitates only a 7.7 percent change in jet size. This is because a change in barometric pressure also affects the pressure exerted on the fuel in the float bowl. Therefore, a change in barometric pressure will *automatically* alter the fuel-air ratio somewhat.

How do you go about achieving an X-percent change in jetting? Fortunately, the number stamped on Mikuni hex-head jets represents a nominal flow in cc per minute at a particular test pressure. Therefore an X-percent change in jetting is just an X-percent change in jet number. For example, a 360 jet is about 3 percent richer than a 350 jet.

As a side note, round-head Mikuni jets aren’t as simple to use because their numbering scheme is based on aperture size instead of flow rate. Because flow is proportional to area, it takes an extra step to convert aperture diameter into area. The formula is: area = pi * radius squared. For example, a 107.5 jet has a nominal diameter of 1.075mm. The area of this jet is about 0.91 square mm.

A 3 percent richer jet would be a 109 (which does not exist, so a 110 would have to be used).

Incidentally, the barometric pressure reported on a TV weather forecast is not the actual barometric pressure — unless you happen to be at sea level. Reported barometric pressure is always corrected back to the pressure it would be if you were at sea level. Thus, in Denver for example, when the barometric pressure is reported to be 29.6, the actual air pressure is more like 24.6 (figure about 1 inch of mercury for each 1000-foot change in elevation).

Author’s Note:
To address some of the shortcomings inherent in mechanical RAD gauges, I have developed a prototype electronic RAD gauge that corrects for humidity and displays RAD and/or “percent change” needed in jet size from a base-line corrected RAD. In the latter mode, the unit is “zeroed” when the bike is running perfectly with known jetting. I may make this unit commercially available if sufficient demand exists. Interested parties may reach me via the contact info given at the end of this article.

Ed: 3% change in RAD is 1 jet size. 1 jet size 360 to 370 is about a 3% change

****Use weather report and RAD formulas to predict jetting****

Tuning Summary: As Ed says, “Part of tuning is just keeping perfect pieces in perfect condition *all the time*. It’s all about doing all the stuff that you know how to do…every…single…time.” On the few occasions when Ed’s bike was faster than Rich Oliver’s, Ed had everything perfect — some of it by luck, some of it by plan. Ed goes on to say, “That’s really the big difference between ‘regular’ guys and the guys riding the factory bikes.”

As a professional mechanic, Ed feels he knows all the tricks the factory mechanics do — the difference is that the factory guys have the time to do *all* the stuff *all* the time. It does not matter how tired the mechanic gets — he’s not riding the bike! This leads us into the next topic, that of maintenance.


A top-end, which includes piston, pin, bearing, circlips and ring, should last 600 miles. Note that this is considerable longer than Yamaha’s recommended replacement interval of 500km (310 miles).

Since this is the first place we have mentioned the piston pin, it is probably a good time to bring up the fact that superior pin exists. It is from a Kawasaki (dirt bike REMINDER). This lighter, drop-in replacement has Kawasaki part number xxxxxxxxxxx.

When assembling a top-end, liberally apply pre-mix oil to the piston, pin, and small-end bearing. If you should have difficult removing a used piston pin (this is more of an problem with 2000+ bikes than the older ones), the recommended tool is a piston pin puller. As an alternative, you can drive the pin out with a drift (often just a socket extension) and a medium-size hammer. Start gently, but don’t be afraid to give it a good whack if that’s what’s needed.

It should be obvious that pistons with visible cracks need to be replaced immediately. If you put off piston replacement too long, eventually pistons will break along their skirts. When a piece comes off a piston, here’s what can happen:

Sometimes the piece will simply bounce out the exhaust port without causing damage. However, pieces can hit the spark plug and close the gap. This makes the engine run on one cylinder. If this occurs, it will likely put you at a competitive disadvantage. Sometimes the piece beats up the top of piston and underside of head making a terrible noise and should cause you to pull off. (The head may still usable after polishing out the damage.) Worst case, a piece gets caught and gouges the cylinder, gouges the cases or damages the crank bearings.

Although pistons come in four sizes: A (56.000 – 56.005mm cylinder), B (56.005 – 56.010mm cylinder), C (56.010 – 56.016mm cylinder) and D (56.016 – 56.020mm cylinder). The reality of the situation is that almost every cylinder is going to require a D piston. The reason is that pistons quickly shrink.

New pistons need three heat cycles for break-in. The pistons shrink as they go through the first few heat cycles. A heat cycle is defined as bringing the water temperature up to 60 degrees C, and letting it cool all the way back down to the ambient air temperature. A cold day is best for break-in. Ed says, “A 120- degree day at Willow is not good for this.”

Break-in new pistons in the pits following the procedure outlined below: For the first heat cycle do not exceed 4000 RPM. It does not matter if you hold the throttle steady or wing-it up to 4000. For the second heat cycle, use a maximum RPM of 6000. For the third head cycle, hold it mostly at 6000 RPM, but occasionally blip it to 8000. Then, in a practice session, take it easy for 2 – 3 laps using no more than 3/4 throttle.

Polishing a Damaged Head:
Clean the combustion chamber as best you can with sandpaper, Scotchbrite, or steel wool. Smooth any sharp edges so they do not glow hot and cause pre-ignition.

Ed has heads (on his TZ, that is) that have dings in them. They don’t have any projections — they are smooth with indentations. By definition, some horsepower is lost due to heads that are less than perfect. There is some compression loss as well as other factors which may or may not affect performance. Of course, engine builders would love to re-cut your heads every time there is a minor imperfection — they get paid to do it. It boils down to a budget issue. As Ed says, “Minor indentations are sure as hell not something you can feel!”

Yamaha’s recommended maintenance interval of 1500km correlates well with Ed’s recommendation of 1000 miles. This applies to both old (’91 – ’99) and new (2000+) crankshafts. That said, cranks have been know to run *much* longer. If your track is the complete opposite of Daytona (no big heavy loads for long periods) then the wear and tear on your bike is going to be substantially less. Dayton is the worst; followed by Brainerd; Willow Springs; and Portland. A tight track like Pikes Peak means a light load. The slower the track, the longer your stuff will last. In Hawaii, Ed ran a crank with unknown initial mileage for 4 more years — with 2 events per month for 12 months each year!

Ed’s pretty sure a new crankshaft does not need any break-in, but still takes it easy for a while. He recommends 2 to 3 laps break- in at no more than three-quarters throttle.

During assembly, grease the crankshaft bearings with any quality grease. This is critical for initial start-up. You don’t want to skid a bearing when there is no oil from the pre-mix. Grease is better than oil because it stays put. This permits the crank to sit, unused, for as long as necessary prior to installation. It is possible to apply so much grease you’ll foul the plugs. As Ed puts it, “Don’t goop it on like you’re doing steering head bearings.”

When the crankshaft does fail, the failure occurs at the big-end bearing. The bearing runs flat; the piston rises farther than TDC and hits the head, bending the rod. The worst thing about this is that it destroys the big-end crank pin. The pin is part of one crank wheel. It cannot be pressed out and replaced as Yamaha does not sell crank wheels. Once the crank pin is damaged, you can no longer rebuild the crank. Thus, it is cost effective to rebuild the crank before the pin is destroyed. A crank rebuild (parts + labor) costs about $500. A new crank costs about $1100.

Who do you trust to rebuild a crank:
Roland Cushway $500 includes parts
Bruce Lind does not do it for the general public

When a crank is rebuilt, both big-end bearings and rods are replaced. The rebuild process necessitates pressing apart the crank wheels. Bruce Lind says you can rebuild a crank 3 or 4 times before the big-end pin does not press back into the opposing crank wheel tightly enough. Even then, you can weld the pin to the crank wheel for one final rebuild (And even that is not the end if you are willing to grind the weld off and start over — but that is a lot of work).

A lightened crankshaft will make the engine accelerate a bit faster, and Rich Oliver says the bike handles a bit better (changes direction quicker). But, Ed doesn’t know too many other people who complain about how slowly their TZ changes direction! Ed says, “The stock crank is definitely the way to go.”

Cylinder: While prudence might dictate that you’d take it easy on new cylinders for a couple laps, they require no special break-in. Ed says, “If my bike just got a brand new cylinder, and it’s race time — I’d start the bike up in the pits, go out for a warm-up lap and then hammer it. The cylinder get bigger as it gets hotter (giving more clearance).”

Ed says, “Measurements and procedures found in the service manual for determining cylinder wear are meaningless.” He does not ever re-plate a cylinder unless it is damaged. Furthermore, when having a cylinder replated, do *not* give them a used piston for sizing. Only a new piston will do. As mentioned earlier, pistons shrink quickly.

Who re-plates cylinders? What should it cost?

Clutch: Unlike most motorcycles, the clutch in a TZ is “dry”. This means that it runs in open air instead of in an oil bath. The advantage is that the clutch does not experience drag due to the viscosity of the oil. The result is more usable horsepower. The downside is that heat transfer into air is poorer than into oil.

The friction plates are driven by the clutch housing (some call it the basket) via the primary gear. The steel (aka drive) plates transfer power to the gearbox by being pressed into contact with the friction plates via the clutch springs.

Friction Plates:
INNERS & OUTERS? Friction plates are made primarily of aluminum with a layer of friction material on each side. When a clutch fails, this friction material is fried off the outer (those at the ends of the stack) friction plates. This results in aluminum to aluminum contact in possibly two places: between the pressure plate (on the outside) and its adjacent friction plate; and/or the clutch boss (on the inside) and its adjacent friction plate. Aluminum to aluminum contact is bad, bad, bad!

Clutch maintenance should be done regularly. Ed replaced the stock “outer” friction plates (the ones that touch the pressure plate and clutch boss) at every AMA event. During his racing in Hawaii, Ed was able to use the same clutch for a full year running 4 races per event by properly maintaining the clutch.

There are no alternatives to stock friction plates. Barnett used to manufacture aftermarket plates for the TZ but no longer does so because the market is too small.

Steel Plates: As the name implies, steel plates are made entirely from steel. They require regular inspection. A change in color does not indicate a need for replacement, but a warp does. If the bike gets difficult to shift (because the clutch won’t disengage well), the steel plates are probably warped and need to be replaced.

Check for warped steel plates by laying them on a known-flat surface. If there are no gaps, the plate is not warped.

You can also check for warp by holding any two steel plates together. If you can see light between them, at least one of them is warped. If there is no gap, turn just one over and repeat to make sure the two plates are not warped exactly the same.

When steel plates are really badly warped, they will not come off the hub easily.

Clutch Springs:
Clutch springs press the steel plates and friction plates together. Ed used stock springs “forever” in Hawaii. The light lever pull of stock springs gave him finer control for better starts. If you have concerns about a slipping clutch, you can use the heavier Barnett springs. Some racers use three stock springs with three Barnett springs.

Clutch Tricks:
The clutch housing is riveted to the primary gear. It is possible for the housing to come loose and shake on the gear. To prevent this, weld the rivets to the gear on the backside. Do it a little at a time and let the weld cool so you don’t cook the o-ring. Alternatively, you can tighten the rivets in a press or peen them with an anvil and a drift.

Finally, Ed really likes the following idea shown to him by Jason Dave. Jason modified his clutch housing and hub to permit installing an extra steel plate. This allows the friction material on the outer clutch plate to work against steel rather than aluminum.

Spark Plugs and Caps:
To many of us, TZ spark plugs seem familiar (14mm thread, 3/4″ reach, 13/16″ hex) yet have several unusual features. They have a short overall length due to the external ceramic insulator; employ a tiny ground electrode; and exhibit an unusual heat range (10.5).

Plugs in the early years (’91 – ’93) are different from those in the later years (’94 – ’99). Both types of plugs are available in three heat ranges: 10, 10.5, and 11. (But Ed believes using the stock 10.5 heat range plug is critical.)


The early plugs have NGK part number R5184-105. These are the non- resistor type, and have a recommended gap of 0.024″. They cost $22.34 from monarchproductsinc.com. The later plugs have NGK part number R6179A-105P. These are the resistor type, and have a recommended gap of 0.022″. They cost a whopping $31.60 each.

Many people (myself included) used to think the “R” in the part number specified a resistor-type plug (as it does in most other applications). However, with TZ plugs, the “R” simply means “racing” and has nothing to do with whether the plug does or does not employ an internal resistor. It is a simple matter to determine if a plug is the resistor type with a multimeter. Measure between the inner electrode and the metal connector at the top of the plug. A non-resistor plug will yield a value near 0 ohms (essentially, you will be measuring the resistance of your meter’s test leads). A resistor plug should measure about 5000 ohms (within 20 percent, anyway). If the multimeter reads infinity or “open circuit”, the plug’s internal resistor is broken and the plug should be replaced.

Plug caps can, and should, be tested in this way too. TZ spark plug caps are an often-overlooked maintenance item. Again, the resistance is nominally 5000 ohms, and anything within 20% of that value would be acceptable. When a cap fails, its resistance will be *much* higher than that (typically an open circuit). By the way, you can test the caps without removing them from the plug wire. Simply measure the resistance between the plug cap and the metal laminations of its ignition coil. The resistance should be that of the cap plus the secondary winding resistance of the coil (which is about 6200 ohms). Thus, the combination should measure about 11,200 +/- 20%. Only if you measure something outside that range should you separate the cap from the wire to determine which component is bad. When you do need to replace your plug caps, Ed recommends not buying stock Yamaha ones. Instead, he prefers to replace the caps with NGK caps intended for use on watercraft. They have NGK part number TB05EMA. They are inexpensive, readily available and reliable.

Incidentally, an engine with a “bad” or open cap/plug may not exhibit any symptoms. It just means that some of the ignition system’s available firing voltage will be wasted in jumping the gap inside the plug or cap and not doing any useful work inside the engine.

What’s the big deal about using resistor plugs or caps? As you may know, both are commonplace in OEM automotive applications. This is because ignition systems create large amounts of electromagnetic interference (EMI). EMI is what makes a crackling sound in an AM radio that varies with engine rpm. EMI can also wreak havoc with any electronic system that uses a microcontroller — like a digital ignition, fuel injection computer, or data acquisition system.

Many people claim that a capacitive discharge ignition (CDI) system can be damaged by using non-resistor plugs when the application calls for a resistor plug. In my professional opinion, this is unfounded. That’s not to say there isn’t a valid reason for using resistor plugs/caps in a racing 2-stroke.

In a typical ignition system for a 4-stroke engine, the ignition coil is charged by a 12-volt battery. The ignition coil’s “turns ratio” then steps-up the battery voltage to what’s needed to fire the plug. However, the CDI system used on modern 2-stroke engines works slightly differently. In it, a capacitor is charged to something on the order of 200 – 400 volts. This capacitor then dumps its energy into the ignition coil when commanded to make a spark. This allows a 2-stroke’s ignition coil to have a much lower “turns ratio” than that of a 4-stroke coil (while still producing enough voltage to fire the plug.)

Although we think of electricity as moving at the speed of light, the time it takes for a spark plug to “arc over” from the instant it is commanded to do so is far from instantaneous. One of the reasons CDIs work so well on 2-stroke engines is an inherent part of their ignition coil’s lower turns ratio. Specifically, it allows for an extremely rapid build up (or rise time) of the ignition voltage. However, if the rise time of the spark voltage is too long, some of the available spark energy is dissipated by leaking off to ground via the “somewhat conductive” paths that form on the spark plug’s insulator. These conductive paths are the result of fouling. And, if there is insufficient energy remaining to create a spark — a misfire occurs.

However, this fast rise time can also make the spark itself have a very short duration. (The spark ignition of a combustible mixture is a statistical process and if the spark is present for too little time, the mixture will not be ignited.) This is one reason increasing a plug’s gap can improve performance — it increases the probability that the A/F mixture will come into contact with the spark and cause combustion to occur for that cycle. The limit of increasing the gap is the possibility of mis-firing as it takes more voltage to create a longer spark.

The bottom line is that resistance (in either the plug or the cap) causes the spark to last longer than it would if there were no resistance (increasing the statistical probability) the spark will ignite the mixture, yet with a short enough rise time so as not to allow the spark energy to bleed off sufficiently for a spark not to be produced at all.

There are two mechanism by which TZ spark plugs “wear out”. One is fouling and the other is due to rounding of the sharp edges of the electrodes. Electrons prefer to jump off a sharp point rather than a blunt edge because a sharp point increases the concentration of the electric field). Think of it in terms of number of electrons per square inch. Thus, sharp edges on spark plug electrodes are good because they lower the voltage required to initiate a spark. You often hear about an ignition system being able to produce 40,000 (or whatever) sparks. RESEARCH 40kV / INCH IN OPEN AIR Of course, it takes more voltage to create a spark in the high- pressure fuel/oil/air environment of a TZ’s combustion chamber (typically on the order of 10,000 to 15,000 volts).

In airplanes, the “bottom” cylinders in radial engines are much more susceptible to lead fouling than the top ones because of gravity. This may also be true of the bottom cylinder in a TZ.

Some people suggest that cleaning spark plugs by blasting them with glass beads is no good because of the chance of the beads getting stuck between the insulator and the plug body and the being released into a running engine. Firstly, with the design of a TZ plug, the insulator nose is very easy see. Secondly, the exhaust port is so big, any released beads would probably just bounce out without doing any damage whatsoever. As a group, it is hard to find individuals more concerned with safety than airplane mechanics and pilots. The cost of an engine failure is high. I figure if glass bead blasting spark plugs is good enough for airplane mechanics, it should be good enough for motorcycle racers.

Spark Plug Heat Range:
Spark plugs *do not* create heat, they only remove heat from the cylinder. From one heat range to the next is the ability to remove 70C to 100C from the combustion chamber.

In *any* application (from lawnmower to F1 car) spark plugs must operate in the range 500C to 850C. They must operate above 500C to avoid wet (fuel and oil) fouling, and below 850C to avoid glowing hot enough for pre-ignition to occur.

In his __Motocross and Off-road Performance Handbook__ Eric Gorr claims that one spark plug heat range will alter the EGT by 50F.

Ed runs stock reeds because they are soft and when they break apart, nothing gets damaged. He says, “Replace them when they look like a rat has been chewing on the edges.” In general, reeds will last longer than a piston does.

Of course, there are kit carbon fiber (CF) reeds and aftermarket CF reeds. Because CF reeds are hard, they can scratch engine internals when they come apart. Also, CF reeds cost a lot more. Ed’s not convinced they offer any real advantage. As he puts it, “I’m pretty sure all the top guys are running the stock ones. The claim is that CF reeds pop open more abruptly and therefore make the jetting more responsive. Whether that is true or not — I don’t know. I’ve never put my TZ on the dyno to do a back-to-back comparison. It is not something that is different enough that you can feel it when you ride.”

Silencer Repacking:
How fast the rider is and how high-speed the track is determines how often silencers need to be repacked. Ed says, “If you keep them packed, it makes the engine happy. I probably should have done it a hell of a lot more often! Repacking is one of those things that you should do once a year, and twice a year is better. It is not done often enough because it is such a pain to drill out and replace the rivets.”

Aftermarket packing are available for dirt bikes which may be tougher and longer lasting than the stock stuff. Ed has always felt that someone should make screw-together silencers. Stock ones are riveted for positive retention and light weight, but if a screw comes loose you can tighten it.

Toomey cans — big snap ring at the back.

Buying OEM Parts:
Buy your parts from a Yamaha dealer. Contrary to what grey market sources may tell you, the parts will come from a warehouse in the U.S. Delivery of TZ parts doesn’t take any longer than other parts ordered from Yamaha. Ed says, “Tell them you are a racer. Ask them to sell the parts at cost plus 10%. The dealer may be willing to do this if you offer to pay for the parts up front (and they won’t have to inventory them).” Ed recommends that if you need one of something, order two and start building your own inventory. If the dealer tells you the part number is invalid, it probably just needs a trailing “-00”.

If you order from Bruce Lind, he will get the parts from Japan.
Special Tools:

1) Deck Height Tool. This tool is necessary to establish the deck height as explained in the section on Engine Work. Ed feels Swedetech makes the best. It consists of 2 dial indicators on a nice aluminum bracket. It is better than the Yamaha one because it averages out the side-to-side wobble in the piston. The Yamaha tool is cheaper, but it only measures in the center of piston. Included with the Swedetech tool is a curved piece of aluminum that sits on the rotor and extends the timing marks so you can set ignition timing more precisely.

2) Dial Indicator. Used to set ignition timing. It screws into the spark plug hole.

3) Rotor Puller. Used to get rotor off the end of the crankshaft. It is the same tool as used on many Yamaha dirt bikes.

4) Scrap Aluminum Bolt. Used to jam between the crank and balance shaft gear when tighten the crank nut.

6) Clutch Holding tool. One can be made from two steel clutch plates with a beat-up old screwdriver welded on as a handle.

7) Crankcase Pressure Tester.

Toolbox Chemicals

Contact cleaner
Precision Parts Cleaner (a slow evaporating contact cleaner, costs half as much as contact cleaner)
Silkopen (graphite penetrating spray)
Lacquer thinner (cleans old tech stickers of windscreen)
Simple Green


If jetting is the key to making horsepower, then gearing is the key to lowering lap times. Gearing in a vacuum is impossible, gearing must be chosen for a particular race track, rider ability, and sometimes, weather conditions.

External Gearing:
Therefore, you need a range of final drive ratios. A reasonable number of combinations can be made using three countershaft sprockets and six rear-wheel sprockets. Ed feels that those sprockets are all that’s needed to run the AMA races. He recommends using a 14-, 15-, or 16-tooth countershaft sprocket. And, says to avoid fitting a 13-tooth sprocket because it will cause the chain to bend too sharply, thereby decreasing its life. Common choices for the rear are 36, 37, 38, 39, 40, and 41 teeth. Although they are available, Ed never uses a 34- or 35-tooth rear sprocket.

On a 2000 model, the chain adjusters will accommodate this range of gearing. However, on older bikes, you may need two different chains, one with 2 extra links. Alternatively, you can use a single chain with 2 master links and a short splice. Ed uses only rivet-type master links — never the clip-type master links.

Once they know the proper gearing for their track, club racers may not need any extra gearing. However, having a choice may be beneficial. For example, when fighting a headwind, lower gearing might be appropriate. Similarly, a dramatic change in air temperature could alter the engine’s power output enough to require different gearing.

First time at a new track. To choose final gearing, pick a track that is similar for which you do have gearing info. Your goal is not perfect gearing, just to be able to stay out for the entire session.

Ed’s gearing data is based on Bridgestone tire circumference
Dunlops are the same size (Michelins are not).

Internal Gearing:
One of the coolest things about a TZ is the fact that you can alter the internal gearing without dismantling the engine (that, and that alternate ratios actually exist!). Where external gearing mostly determines the top speed of the bike, internal gearing allows you to be in the right gear for each corner.

The parts catalog lists 2 alternate ratios for each transmission gear (a “short” and a “tall”). Kit gearboxes provide 5 alternate ratios for each gear. Rich Oliver has 8 ratios for each gear.

Kit parts are anything you can buy that is beyond what comes standard (part numbers from your manual or manuals from prior years) Example: B,C,D are alternates (A,E,F,G are kit).

Whatever comes on the bike, and any spares that come with it. Each year you might get slightly different spares from the previous years. On the newest bike, Yamaha is giving out different combination of alternate gears. The reason is that they can make different gears available. If they only made the same there gears available, then all the others would be really expensive. Allows “economies of scale.”

Choosing Internal Ratios:
When Ed first rides a track, he makes notes after every session regarding the gearing for each corner. He uses a downward pointing arrow to indicate a downshift, and a G with a number to indicate what gear he’s in. He works his way through each corner for the entire track.

Then, he compares his gear usage for all corners on the track. Inevitably there will be compromises. For example, one corner may require a high 2nd, while another may require a standard. Knowing what gear you are in for each corner is sometimes difficult, but it is the only way to know how to alter the internal ratios.

In Ed’s own words: “I pay attention to how many downshifts I have to make for each corner, so I’m always in the same gear for that corner. When accelerating, I always upshift at the same RPM.

For each corner, I mentally note if the gearing is too tall or too short for that corner. For example, if I think 2nd is too short, on the next lap I’ll try 3rd. If it is too tall, then I know I need to change to either a tall 2nd or a short 3rd.

When I’m back in the pits looking at my notes, I have to make compromises. If one corner needs either a tall 2nd or a short 3rd, and another needs a standard 3rd, I know I have to install a tall 2nd. In the beginning it’s hard.

It is better to be on the track with tall gearing and speed up to that gearing. The trick is if you start out with gearing that is so tall you can’t get up to speed, then you didn’t help yourself.

If it is a wash or I’m ambivalent, I’ll always choose the taller gearing. That way, if I get faster, it works better. If I get faster with short gearing, then I’m hurting my bike.”


Rino Fender:
On most bikes, the radiator is aerodynamically only slightly better than the proverbial barn door. More air is going *to* the radiator than can possibly go *through* it because the engine is in the way. The extra air piles up in front of the radiator and produces something akin to the bow wave in front of a barge. Aerodynamicists call this a stagnation point.

The Rino fender sits out in front of the radiator, parting the air so some of it never gets to the radiator. This creates a smaller stagnation point. Enough air reaches the radiator so there is no effect on cooling.

The Rino fender proved itself to Ed during a practice at Willow Springs. In back-to-back-to-back (Rino, stock, Rino) testing, Ed found that with the Rino fender he was able to go through turn 8 with the throttle pinned. With the stock fender, he had to back off slightly. Ed believes he was running wide with the stock fender because the pressure at the stagnation point was lifting up the fairing and taking weight off the front wheel. This testing made a believer out of Ed, and the Rino fender has not been off his bike since.

Ed says the change is immediately apparent, remarking that the bike is quieter (not the engine noise, but the air noise). He goes on to say you can feel the effects of the Rino fender most strongly in really fast corners like turn 8 at Willow Springs and turns 1 and 2 at Brainerd. (Turn 1 at Brainerd follows a mile-long straightaway.)

The following story supports Ed’s faith in the Rino fender and the manufacturer’s claim for an increase in top speed ranging from 4 to 6 MPH. At an AMA practice at Brainerd in ????, Ed felt his engine was only “average fast” — nothing like Rich Oliver’s bike. Rich was going 4 – 5 seconds per lap faster than Ed during this particular practice session.

At one point in the session, Rich went through the turn preceding Brainerd’s mile-long straight and came upon Ed at exactly the right time to pick up his draft. With the throttle pinned, Rich was going *much* faster than Ed. Yet after pulling out of Ed’s draft, he was unable to pass.

Rich couldn’t get past Ed on the straight. He had to get back in line and draft. Rich did not pass Ed until turn 2, which was about 1.5 miles away from where he first caught Ed’s draft.


Crash Parts:
TZs are light and consequently crash very well. So well, in fact, that Ed is not sure the teflon sliders that some racers install are worth their small weight penalty. Ed says that TZ crash parts are standard fare pretty much the same as any other roadracer. Things like levers, footpegs, shifter and brake pedals, handlebars, master cylinder, fairing stays, throttle tube, cables — and any small peripheral stuff you could rip off in a crash.

The way to minimize your initial spares investment is to go racing with TZ buddies. Everyone probably has an assortment of spares that came with the bike. Not everyone is going to need every part every weekend. The idea is to trade and share. Slowly build up your spares based on what you and your buddies break.

For National events, Ed even carried complete, stickered, spare bodywork. This is unnecessary for the club racer. Any fairing damage that can’t be temporarily fixed with duct tape is probably not the only thing keeping you from racing. Often there are much more serious problems elsewhere.

Oddities: (parts that won’t interchange)
The frames changed in ’96. So ’96 and later exhaust pipe hangers are in a different place than the earlier bikes.

The later crankshafts are backward compatible, but the ’91 with the lighter rods, will not fit in the later cases because the OD on the big end of the rod is just enough larger to interfere with the cases.

Throttle Cables:
Ed recommends upgrading older bikes to ’96+ throttle cables. These go out in front of the fork tubes instead of behind them. This lessens the amount the cables get tugged and stretched, consequently, the carbs should stay in sync longer. In order to install the longer cables, you will have to rotate the throttle tube so the cables go out the front instead of down.

Pressurized Airbox:
You can put the ’99 bodywork on the older bikes, but this requires using the ’99 upper fairing stays also.

You can use the ’99 lower airbox with the ’96 upper air box to build a ram-air bike. The ’99 upper airbox will not fit in the older subframe. Get the ’96 bodywork, do a little fiberglass work.

Make an extension on bottom of airbox, there are some kit parts that attach to the fairing lower to make an airbox there.

The ’97 ignition is an upgrade for earlier bikes. They are likely to be cheap because the ’97 started to handle funky.

Boring older 38mm carbs to 39mm makes a huge difference. The results are more midrange, better throttle response, and more top end — there is no downside, you don’t give up anything. But this mod is not worth doing on 2000 and newer.

In ’99, the kit carbs available which are downdraft. That is a noticeable improvement.

Chassis Setup

Car guys laugh at bike setup saying we can’t even change the final drive ratio without affecting the wheelbase. We just have to do the best we can with what we’ve got.

Ed’s advice is to: “Pay a suspension specialist. It is worth the money. Buy the right spring to get the washer stack to operate in the right part of the damping oil.”

Lacking that, here are a few ballpark values:

front sag: (with 1/2 load of fuel) 35mm
rear sag: (with 1/2 load of fuel) 25mm

For a 160# body weight use:
0.76 kg/mm fork spring
425# shock spring

Start with the standard “air spring” (standard oil level). If the fork bottoms, raise the oil level. If the fork chatters, lower the oil level. For the TZ, 1 cc of oil is 1mm of oil level in fork.

Springs are too soft, people compensate with too much preload. Example: enter turn 3 at willow, flicks to turn 4. Bike jumps off the ground too much preload.


Steering Damper:
Don’t buy a stock one because you can’t get it rebuilt. Ed likes Storz (tell story). Backwards mounting makes the damper harder to damage in a crash.

Ed likes to clean the wheels before taking them to the tire guy. Having been on the other side of the transaction, Ed says he did better work on clean wheels as opposed to filthy ones.

When asked what he thought about running tires backwards, Ed replied that in Hawaii he would buy a used tire; run it for a while; flip it and run it down equal to the other side; flip them again and run it all the way to the bottom; and then flip it again and run it to the bottom on that side too. If you’re at a track with huge loads, or trying to do Nationals — it’s probably not a great idea. But if your lap time is 30% off the fast guy, you are *totally* safe.

There is no difference at all in how they work — they are still round going both ways! The only issue is that the tire comes to an end and the one piece of rubber overlaps the other piece of rubber, they construct it so that load is pushing the rubber down, rather than peeling it up. With a 0-degree belt, that is not even the case. And now they are making tires with multiple stripes across the side, so it is totally not an issue.

Ed says a lot of guys at Willow Springs put their tire on backwards initially, and then run it “the right way around” for the race.

At most club tracks where the speeds are lower, you can switch back and forth without any noticeable effect.

Riding Techniques

Finger on the Clutch:
Remember Ed’s story about crashing due to an air-leak induced engine seizure? He says that the old racer’s advice about having one finger on the clutch would not have been much help: “Even if I’d had my whole hand on the clutch, there would not have been time. It was instantaneous.”

This is not to say there aren’t situations in which having a finger on the clutch is not beneficial. There are a lot of occasions where pulling in the clutch will save your leathers. Ed puts it like this: “You’re on the gas, the bike is accelerating well, and then it just slows down (making a rrrrr sound). When you feel it slowing down, get that clutch in right away. As you squeeze the clutch, blip the throttle — if it revs fine, there was probably nothing wrong. If something is wrong, it will quit running even though you blip the throttle. Getting the clutch in will prevent the rear wheel from locking.”

Ed also maintains it’s possible to be fooled. He says, “If you have a tail wind pushing you hard, and you begin braking, some TZs seem like they just quit running for a second.” The tuner will hear “It stalled going into turn three, so I stopped.” Upon restarting the bike, nothing wrong can be found. It runs perfectly. It’s a mystery.


Torquing Fasteners:
You must torque, at least, the following fasteners: head, base, crankcases, clutch hub, and rotor nut. As Ed puts it, “The smarter you are, the more times you’ll use the torque wrench.”

Safety Wire, Special Items:
There are a few special areas need attention on a TZ. Firstly be sure to safety wire the throttle cables at the carb and the throttle tube. If you don’t the cable can pop out of the adjuster making it impossible to completely shut the throttle (a dangerous situation). Use the fine (0.020″ diameter) safety fire for this task.

Also, puts a wrap of safety wire around all three power valve wheels (one on the servo motor). Although a wheel’s set screw may still come loose, at least it won’t fall out completely. Those set screws should be checked periodically.

Detonation Sensor:
Ed has never used one; never had a need for one; never damaged a TZ from detonation.

Airbox Cover:
Yamaha says it is for a “wet race”, but Ed says to leave it on all the time. It helps keep rocks out, and permits a slightly higher airbox pressure. Remember, you are tuning your 250 as two separate 125cc engines — do the best you can for both!

Ignition Pickup Coils:
A cracked ignition pickup can cause a high-speed misfire because the pickup coil can shake thereby changing the timing. The pickup coils cannot be purchased separately (a complete magneto CDI costs over $1000). The solution is to epoxy the pickup coil to repair the crack.

Exhaust Springs:
New exhaust springs have a bend at the pipe end. This makes them extremely difficult to install. Just put the spring on backwards. New springs are black. Old ones are chrome.

In-Bike Crank Installation:
Ed learned this trick from Mark Vanderwerf. When installing a crankshaft while the engine is still in the bike, put the crank into the front case half, and then bolt the case halves together. Otherwise, it is too easy to end up with a connecting rod sticking out the opening for the reed rather than the opening for the rod.

Taking the bike off the stands (and tire warmers) to start it for warmup is time-consuming, a pain, and allows the tires to cool significantly. Instead, start the bike on the stands with an electric drill. Chuck-up a cut-off 3/8″ extension fitted with a 19mm socket for the crank nut. (Tul-aris used an automotive starter.)

Seat Comfort:
Install two pieces of foam. Otherwise old guys can’t bend their knees enough to get onto the bike. You want to buy the old seat foam part number. The newer seat foam (which has cross hatches in it) costs twice as much. To install the foam peel back and cut off half the backing paper. Align the paper side first by making contact. Then push the adhesive side down. Finally pull the paper off the other side and attach it.

Bicycle Shorts:
Malcolm Smith Racing (MSR) sells “impact shorts” for $32.95. They have plastic on the hips. MSR also sells a limited-availability V- top padded short for $64. (They should probably be called the “kit” shorts.)

Cleaning Brake Pistons:
Although it is tempting to use a spray contact cleaner to clean brake pistons, Ed learned that some of the factory teams use something far less harsh (something like Simple Green).

Reasoning that chemicals were harsh and perhaps hard on the seals, Ed now cleans the pistons with a toothbrush and a squirt bottle of soapy water. He gain access to the pistons by first removing the pads and then pumping the pistons out a little bit. After the wash, he rinses with another squirt bottle containing tap water.

Quick Shifters:
Also known as electronic ignition cutouts — this is the thing to get once you have everything else. Ed says it is not uncommon to hear, “Everything was going great until my quick shifter took a shit.” But never once has he seen a guy stand up on the podium and say, “I won this race because of my quick shifter.”

Andy Edwards says they make the tranny parts wear out faster.

And for a 250, the only one that works is the KLS which costs $1000.

Aftermarket Exhaust:
If your stock pipes are undamaged — run them. However, if you destroy your pipes, don’t buy new stock ones — replace them with Jolly Motos.

Jolly Motos rev higher ’96 pipes seems to be better than the previous ones. Ed’s 2000 would not rev past 11,500. Jolly Motos will rev to 14,000.

There is another pipe maker that copies the kit pipe design for the other side.

Quick Disconnect Fuel Line:
Ed says they are fine, but he does not have one. Instead, he discards the clips from the stock neoprene hose. If your bike is older, you may want to get a new pliable hose. Buy a nice stock molded one from Yamaha.

If you do choose to use a quick disconnect. Carry spare replacement o-rings, as race fuel tends to cause them to swell.

Glossary (definition of terms)…

Pre-Ignition: When a hot glowing piece of metal in the combustion chamber ignites the mixture before the spark plug does at the proper time.
Pre-Detonation: No such thing. Comes from confusing pre-ignition and detonation.
Free (unladen) Sag
Race (laden) Sag
Ride height
Plug chop
Heat cycle
Case Packing
Cylinder Porting

Ed Sorbo opened the first sportbike-only motorcycle shop in Hawaii in 1984. In 1998 he moved to the mainland to compete in AMA 250cc GPs until their demise. These days, Ed…. WHAT NOW?

Jim Hubert operates Deus Ex Machina Engineering, Inc., an electronics consultancy. He was responsible for the power valve controller and certain data acquisition hardware on Rob Tuluie’s home-built unlimited GP bike, the Tul-aris. Jim’s rolling laboratory is a TZ250 fitted with a data acquisition system of his own design.

Author’s Note:
All errors and omissions are mine and mine alone. I invite readers to share their experiences

Contact me . I’ll write a follow-up article based on whatever I learn.

Out Takes

“Often it is more important what you don’t do, it is more important than what you do do.”

Critical subplot: thinking versus riding.

Do you have a graph of the ignition advance curve?

I know you prefer EGO, but have you ever measured EGT? I am set up to measure both on both cylinders.

What do you think of running tires “backwards” to get additional mileage out of them? I do that all the time. Ed.

Where do you deviate from the manual’s recommendations?

The deck height obviously affects the squish clearance.

Had you gone 138.5, you really could not say if the change helped or hurt — likewise if you had gone 137.5. On the other hand, had you only managed 137.0 mph, you could say that the change probably was detrimental.

In A. Graham Bell’s __Two-Stroke Performance Tuning__, the author asserts: “…no two engines respond to RAD changes exactly alike. Usually small displacement, high rpm road race engines and liquid cooled engines in a high state of tune are most affected by a change in air density.” Does that sound like a TZ, or what?

RAD effects jetting in two ways:
1) There is a different quantity of air to “burn”.

2) The mixture strength changes

lower RAD: we HAVE to lose power. On a normally-aspirated engine there is nothing we can do regain this loss.

is now too rich. The rich mixture causes a power loss as well. By rejetting, we can make the engine run the best it can given the relative air density.

Weather stations don’t give a direct readout in RAD. So you have to carry a calculator, and maybe some lookup tables, or a laptop.

or even a 3% change in jetting because of the next reason

These reflections can increase the engine’s volumetric efficiency.

Have to interpret the throttle position related to RPM. Revved out in top gear.

Average your gauge to everybody else’s. After all it is called, “relative” air density, not “absolute” air density.

When I get back to the pits, I make a worksheet which spells out, for each corner, with how many downshifts I’ve made.

I make those notes and compare the gear usage for all corners on the track.

create reflections of acoustic waves so as to produce peak power at a particular engine rpm. Designers use this

phenomenon to create pressure pulses within the exhaust system at the appropriate time to yield increase peak power.

much of the information is general in nature and therefore applicable to other years (and manufacturers) as well.

front of the bike that is bigger than the bike. The rest of the air has to go around the stagnation point which is bigger than the bike. It is making your bike less aerodynamic. It makes the stagnation point smaller.

You would not have a cassette gearbox if you didn’t need one.

On balance, there is little net gain.

Can burn excess in tow vehicle. I WOULD NOT. Tetraethyl lead
Heywood p. 653

Some environmentalists foam at the mouth even hearing the name Thomas Midgley, who discovered not only TEL but also ozone-depleting CFCs.

Every time I ride the bike I write a page about the track….

For example, you can have the engine tighten up. Ed says he’s hurt a crankshaft that way.

It is the same reason that allows a resistor plug to help control EMI, and it has to do with increasing the “rise time” of the spark itself.

’96 bodywork will have more clearance for spark plug.


Due to the difficulty of typesetting formulas accurately, the following are presented in the C computer language format.
Motor Octane Number:
MON = 100.0 + 28.28*T / (1.0 + 0.736*T + sqrt(1.0 + 1.472*T –

Where T is milliliters of TEL per US gallon (1ml of TEL contains 1.06 grams of lead) p. 473 Heywood
Standard Deviation:
Relative Air Density:

RAD = 1734.7 * P / (T + 460)
Where P is barometric pressure in inches of mercury, T is temperature in degrees F.

willow 29.92 – 2.6 = 27.32 @ 100F RAD = 84.5

pikes peak 29.92 – 5.3 = 24.62 @ 45F = 84.5

(100+460) / (45+460) = 1.109

27.3 / 24.62 = 1.109
Humidity Contribution:

Saturation Vapor Pressure in millibars, given Temperature in degrees C.
// Adapted from Honeywell website

SVP = 6.1378 * exp( (17.502 * T) / (240.9 + T) );

saturation_percent_water = saturation_vapor_pressure / 10.0;

humidity_correction = (RAD * saturation_percent_water(T) * RH) /100.0;

corrected_RAD = RAD – humidity_correction;
Ignition timing from Advance
Advance from Ignition Timing
Speed in a gear
Andy Edwards Questions:
Removed Tape improved speed: aerodynamics or lower coolant temp?