 Techline-April09-H-D 110" CVO - Part IV: The cylinder liners and other issues (cont.)
This article concludesin the next thread.
Two Types of EFI
There are two main types of electronic fuel injection systems used to monitor the various sensor readings, such as ambient temperature, engine temperature, throttle position, manifold air pressure, crank position, and others. Once a stock part is replaced, like an upgraded exhaust, high-flow air cleaner, or longer duration/higher lift cams, these changes will disturb the factory-installed air/fuel calibration settings that closely link with ignition tables in the electronic control module (ECM). In order for your Harley to run properly with parts changes, adjustments must be made to give the ECM the necessary instructions regarding how much air and fuel are required. This is not easy to do.
Carburetors require mechanical jetting changes, which are easy. EFI systems need electronic adjustments made either to the ECM or the signals they send out to the fuel injectors. The ECM fuel and coupled ignition tables are fixed. Up until recently (1995 to 2006), Harley-Davidson, as well as most of the aftermarket, chose open-loop EFI for the ECM to receive, interpret, and control fuel settings from sensor readings. The vast majority of aftermarket ECM modification systems used to alter fuel settings are limited to specific changes without further adjustment. They cannot adjust on their own. Therefore, anytime a change is made, such as an exhaust, neither the factory nor the aftermarket calibration tables are valid any longer. Sometimes this will be injurious to the engine if a lean condition develops. Other times the fuel will become too rich and wasteful.
It is possible to rewrite the stock ECM calibration tables, but hardly practical. Most aftermarket units allow a choice of settings or modified tables for specific changes, but they need changing every time another modification is made. The automotive industry has been using closed-loop EFI forever, favoring it over open-loop EFI. Harley-Davidson finally made the switch over to closed-loop operational fuel control in 2006 with the Dyna series. It carried this over into the full 2007 lineup including Sportsters. The reason for the change is the addition of one of the now-important sensors, the O2 or oxygen sensor.
This sensor monitors O2 content in the exhaust stream as it heads out of the exhaust system. It provides real-time monitoring and diagnosis of the air/fuel ratio, which is sent to the ECU (ECM) for analysis. With this system, the fuel delivery can correct automatically to the factory calibration tables, which are based on EPA dictates. Whether one agrees with the preset calibration tables or not, the point is that they become accurate in the predetermined, and thus correct, air/fuel ratio. Many aftermarket exhaust companies have begun the EPA switchover by supplying their exhausts with threaded openings known as bungs for optional O2 sensors.
Self-tuning ignitions, like the Terminal Velocity, can now recalibrate over 500 times per second to adjust to not only any changes the rider may make, like an air cleaner, but to constantly changing load conditions that occur while riding. These would include adding a passenger or trailer, going up hills, going down hills, accelerating, decelerating, or simply cruising. The ignition no longer cares either, as it simply reads the exhaust stream, analyzes the parts-per-million O2 content, and delivers constant information in milliseconds to the ECM so that a decision can be made, again in milliseconds, to determine fuel requirements instantly. It will also compensate for altitude changes, lowering fuel as elevation increases because of less air and increasing fuel as elevation approaches sea level. This is new to the motorcycle industry, but again, not the automotive, especially in racing circles. Some pretty big names use this technology such as BMW, Mercedes, Honda, and the domestic Big Three. Furthermore, Formula One racing also uses this trick.
Oxygen Sensors
We need more gas than what we get in a 14.7:1 mixture for cooling the combustion chamber, as well as power. Harley-Davidson uses narrow band O2 sensors that work well in both stock and most performance applications. The exception is with supercharging and turbocharging applications, two examples of forced induction where wideband O2 sensors are necessary to deal with higher boosts and rapidly changing rpm levels. Forced induction compresses air above atmospheric pressure without the low pressure area necessary for natural aspiration, thus increasing the intake air charge. Compression forces more air into the engine, providing a boost. With the additional air in the boost, more fuel can be injected into the charge, and the power of the engine is increased substantially and almost immediately. Lambda air/fuel ratio can change too fast for narrow band sensors to respond adequately, perhaps leading to engine damage. This is why faster-reacting, wideband O2 sensors are a good idea in these instances. Lower boosts of 6 to 10 pounds can still use the narrow band O2 sensors without dangerous spikes in mixture leanness.
A closed-loop EFI requires oxygen sensor(s) to feed exhaust composition lambda to the ECM. It functions at 14.7:1 in cruising situations, as well as at hot idle. A closed-loop system needs a full circle to reach its mandated air/fuel ratio. The full circle ends with the ECM determining the pulse width of the fuel injectors, which is how long they stay open. This decides the amount of fuel injected. The engine goes through its strokes until completing the exhaust cycle. The O2 sensors measure lambda and send a signal back to the ECM. The ECM will increase, or more likely in the real world, decrease fuel injector pulse width to reach a lambda of 1, which is equivalent to a stoichiometric air/fuel ratio.
An open-loop system functions in accelerating and cold idling situations, particularly during engine warming after startup. It’s unlikely that open-loop EFI will reach an air/fuel ratio of 14.7:1. However, the likelihood of a richer condition of around 12.0:1 with a lambda of less than 1 (12.0:1 —: 14.7:1 = 0.816 Lambda), will exist with open loop. An air/fuel ratio of 12.0:1 is the 2006 and later H-D engine-default, rich-idle mixture. When the ECM decides the engine has warmed sufficiently, it will put the idle into a closed loop and seek a 14.7:1 ratio. Blipping the throttle or otherwise taking the engine past 1200 rpm from an idle of 1000 (plus or minus 50 rpm) will engage the closed-loop mode and heat the engine further.
Open-loop EFI is not EPA-functional for the cruising air/fuel mixture ratios on the newer bikes. Accelerating and rich idling cannot use the lean air/fuel ratio used in stoichiometric closed-loop EFI because the engine will stall at idle and stumble while accelerating, generating even hotter injurious temperatures. These operating conditions require a richer air/fuel ratio.
Problem Is Oil Seepage
As I have said, the CVO 110 rear gasket leak develops typically beginning at 1,800 miles. The problem is one of an oil leak and not a compression leak, which is an odd situation for a head gasket. Usually when a head gasket is compromised in any way, a compression leak will follow. Most times, a compression leak will destroy the head gasket integrity from the inside out. I haven’t heard of any CVO 110 rear head gasket leaks developing into a full-blown head gasket leak, although I think it would happen if left unattended. All the cases I have heard of began early in the warranty period. Therefore, the struggling dealers, who at this point don’t have adequate information or fixes, are dealing with CVOs in the early stages of leakage. Parts replacement will not solve this problem, but this is the only option the dealers have now, outside of revamped gaskets that cannot smother the symptoms, let alone attack the cause.
A compression leak is a serious mechanical problem that needs addressing immediately. Compression leaks result from a blown head gasket that allows the combustion chamber to release red-hot gases into the atmosphere. These gases act like a blowtorch, and, if left unattended, will destroy the gasket surface of both the head and cylinder, creating a progressively larger pathway by melting surrounding metal.
The head gasket oil leak originates at the return oil passage, and goes outward through, above, or below a compromised head gasket. An affected rear head gasket will typically begin to weep an oil mist. The weeping progresses into an actual oil leak. As more miles go by, there’s oil dripping down on top of the primary case and the entire back of the rear cylinder and head, with a resultant baked-on oil residue that will drive any new H-D owner nuts. Oil will travel up or forward because of wind eddies at speed. In addition, the wind and gravity will also push and pull the oil down and back, as one would expect. The gasket from the return passage inward remains sealed, and contains the cylinder’s burning fuel and compression, as designed.
The factory uses an O-ring (#11273 on 1999-2003 bikes, depending on model) for Twin Cams to seal the oil inside the return oil passage. For whatever reason, the factory discontinued the O-ring. This O-ring fits around the hollow dowel that inserts into the head oil return passage and down into the cylinder return hole. It prevents an oil leak every time. The hollow dowel acts both as a locating pin for the mating head and cylinder, and as an oil leak preventive measure. At Heavy Duty Cycles, we still use these O-rings on all our performance work where possible, as well as normal rebuilds on TC 88s and TC 96s. It is not possible to use an O-ring on the TC 95s or TC 103s because both utilize a stock 3-3/4" TC 88 or TC 96 sized-cylinder bored out to 3-7/8" to gain 95" and 103" respectively. There is not enough room for the O-ring on the compression side of the dowel. The same situation exists with the CVO 110 4" bore.
Another clue exists here. Why do the 95" and 103" cylinders not leak oil? The Motor Company has upgraded the 4" bore head gasket from the original #16801-07 to one delineated by #16801-07A in an attempt to rectify this leakage situation. At the time of this writing, the Motor Company was redesigning the gasket once again, and there are rumors of a #16801-07C gasket in the works. If true, this means four different gaskets to attack a symptom. The new gaskets have not met with success in alleviating the problem. Of course, how could they? Fixing a symptom does not fix the cause.
In my book, Donny’s Unauthorized Technical Guide to Harley-Davidson 1936-2008, Volume I: The Twin Cam, I ruled out the head gasket surface since I was not aware of the head causing leakage problems in other applications. Well, I take that back. I have now seen and measured 0.007" warping on the rear CVO 110 head gasket surface, but none on the front head’s gasket surface. This glaring clue leads, once again, to the initial cause of excess heat. Interestingly, in our experience, 116" and 124" S&S Cycle engine gaskets do not fail. S&S uses an oil passage return Viton O-ring (#50-8034). Viton is a great material to use for engine-sealing O-rings, as oil or heat does not compromise it. Again, there is much speculation and Internet chatter that the rear cylinder steel insert is moving and separating from the cylinder to cause this leak. The reader already knows the truth of this speculation. If it were occurring, there would be more severe problems. Cylinder cutaway pictures demonstrate what is really going on. However, it’s easy to believe there is some degree of separation from the aluminum cylinder. I can hook my fingernail on the cylinder sleeve where it meets the cylinder and the liner. Under a magnifying glass, it looks as if there has been a disturbance between the liner and the cylinder body. In addition, impregnating oil seeps out between the two when heat is applied.
This article concludes in the next thread. Check back issue for pix and extra information. |
Linear Mode
