Put down the NY Times and get ready for some light reading about air filtration for Ducati Superbikes.
The topic of modifying the air intake system has been discussed as extensively as what fuel octane to use and what's the best oil. Here's the basic issues, so you can decide for yourself whether to throw out the stock air filters.
Your engine is prevented from making more horsepower mainly because of the difficulty of getting enough air. You can easily add more fuel if you can get more air through the engine. Superchargers and turbochargers are the common way of doing this in automobiles. Four valves per cylinder is often used in modern motorcycles.
The governing air flow restriction is the throttle body diameter, the size of the intake and exhaust valves, and camshaft lift and duration so if any upstream flow area is smaller, it'll just see a higher air velocity as air passes through it. If you introduce a flow obstruction such as an air filter into the upstream air flow you'll get a reduction in air pressure at the throttle body and flow into the engine will be reduced. A dirty air filter will be the biggest flow obstruction.
There has been much discussion about opening up the front air intake openings in the fairing and to remove the decorative grill in order to reduce flow restriction. I've done it, and it's probably a good idea to open up this space, but the fact is that air velocity will simply increase to supply enough air if you leave it alone.
Similarly, there's been a lot of questions regarding removal of the rubber blocks in the air runners of later year bikes. Again, these blocks are nice smooth venturis that simply accelerate the flow through them on the way to the airbox but are designed to restrict air flow very little. They really don't hurt anything if you leave them in. Their purpose is to reflect sound waves so as to reduce induction noise to achieve overall legal limits.
Before we get to the question of air filters, we first need to consider the air intake system design. One reason for using an airbox and intake runners is to direct cool air to the engine. This design approach also results in a small pressurization of intake air that increases with bike speed. It also reduces the volume of the intake flow noise.
But there's also a performance benefit because the airbox is a Helmholtz resonator. That is, a resonance effect occurs when you connect an enclosed volume of a suitable size and shape to an engine's intake stacks causing the air inside to resonate at a frequency that results in pressure peaks coincident with the cylinders' intake stroke frequency (at a particular RPM.) This can theoretically increase power by 10-15% within a particular RPM range by boosting airflow into the engine. Air boxes need to be well sealed and stiff in construction to maintain these resonance characteristics. A simple way to illustrate this is to blow across the mouth of an empty bottle. The sound you hear is the natural frequency of a Helmholtz resonator.
If you change the shape and free volume of the airbox, you change its resonance frequencies, and the engine RPM at which it responds to enhance filling the engine with air. For example, a larger Corse airbox is tuned for optimum filling a at higher RPM where race bikes normally operate, but street bikes usually do not.
An airbox also functions as a plenum, a space where the air velocity is reduced so as to eliminate turbulent air flow prior to being smoothed and accelerated down the velocity stacks. In fact, if you just place your finger anywhere near the edge of the top of the velocity stack you'll see dyno power drop off due to the disturbed air flow pattern.
The air in the plenum is also considered "free air." That is, its already passed through the air runners and filters so it can be supplied to the engine without any flow restriction. If you use the auto industry's standard calculation of air required for "nil" vacuum restriction within the air intake system, you should have at least 130% of engine capacity in available air volume between the throttle butterflies and the air filter element.
All of this tells us why Ducati places the Superbike air filters in the air runners. This location avoids lowering the frequency of the airbox (by not filling up a large portion of the airbox volume with a bulky foam filter) and prevents disturbing the airflow near the velocity stacks as well as improving throttle response by maintaining a large free air volume between the filter location and the velocity stacks.
This is also why Ducati didn't use over-the-velocity stack bellmouth style filter. This location doesn't meet that 1,300 cc. plenum volume needed to avoid degrading throttle response (which a dyno doesn't measure BTW.) Dyno tests say you get somewhat less peak horsepower with these filter types but on a stock bike they don't seem to make a lot of difference.
Another point to consider is that unless you have customized a FIM chip on a dyno in an attempt to match the flow and resonance characteristics of an over-the-bellmouth filter you'll have to use a chip that was developed using the stock filters.
The computerized engine management system uses a fixed fuel injection metering scheme controlled by the EPROM chip that was developed in combination with the stock intake/exhaust configuration. Unless you install a programmable FIM chip and sort out any changes on a dyno with a knowledgeable operator/programmer you won't get optimum (low-end?, midrange?, high-end?, power?, throttle response?) performance.
Ducati engines are the result of countless hours of development work on a dyno. They are designed as a unit with air flow and exhaust flow optimized together for each engine configuration. When you change things like intake and exhaust configurations it's a hit-and-miss proposition. Many times you make improvements in one RPM range and performance at other speeds decreases. Without dyno before-and-after checks, multiple changes can produce confusing results.
Also, seat-of-the-pants tuning can be misleading. If a change, say, reduces mid-range performance at the expense of high end power the engine feels more peaky and this feels like an improvement. But is it? Strange dips in dyno horsepower and torque curves can and do occur at speeds where you spend most of your time riding.
Regarding the large foam in-airbox filters, the people at Sigma Performance (www.sigmaperformance.com) distributors of arguably the best aftermarket chips for superbikes recommended using the stock filters. They state that all FIM chips are made and initially tuned with the stock filters, and they highly recommend staying with the stock units.
This is because they say they have found that a new CLEAN stock air filter gives 2-5% better performance than any of the aftermarket filters and they have only seen reductions in performance when using over-the-bellmouth filters. As a type, they change the air box resonance and require about a 3% leaner mixture to get back some of the power. They say that these filters have flow characteristics that aren't well matched to the fixed metering of the stock fuel injection system.
So it seems that the main advantage of the of a over-the-bellmouth filter location in the airbox is that it allows only clean air into the motor. The fuel tank-to-airbox seal is notoriously poor on superbikes, and will often allow dirt into the airbox bypassing the stock filters.
So, if you see bits of sand and grit and dead bug parts in the bottom of the airbox one solution is to replace the stock tank-to-airbox seal with thicker stick-on foam tape, or use the updated seal from the 1999 and-up bikes.
Access to the stock filters requires some effort so if you don't clean them regularly an aftermarket filter may be a better choice for you.
Further, over-the-bellmouth designs, however, do reduce the chance of engine damage in case of a crash. IF you crash, AND the gas tank becomes dislodged from the air box breaking the seal, AND debris gets into the air box these designs can prevent engine ingestion and further damage caused by the crash. Your insurance company will thank you.
So even though the stock filter location is better, that's not to say that the stock filter is best - they do have significant drawbacks themselves. They use a single density foam that has a uniform pore size and they are run dry (i.e. without applying a dirt retention oil.) So, they flow air well, but they're not a very efficient filter and a lot of smaller dirt particles get through to the engine. This accounts for most of the dirt coating the inside of your airbox.
So, the stock filters won't stop any dirt particle smaller than the foam cell size unless they are run with a filter oil. The stock system is actually a two stage filter if you consider the debris/insect-filtering effect of the air duct inlet screens.
As a type, foam air filters are a reasonable balance between good airflow, dust holding capacity, and filtration efficiency for small particles. The way that better foam air filters work is simple. Open cell polyurethane foam is wetted with a specially developed sticky oil. The sticky oil is suspended in the path of the dirty air on the strands of the web-like cell structure of the foam. This makes it difficult for small dirt particles to pass through the depth of the filter without sticking to the strands. Larger particles are trapped if they are bigger than the distance between the cell strands themselves.
As the outer wetted strands become loaded with dirt particles and no longer sticky, the wetted strands downstream continue trapping dirt, until the entire foam thickness is utilized. Also, as dirt particles build up on the strands the space between strands decreases, further increasing the filter efficiency by trapping the smaller dirt particles that initially could pass between the strands in a clean filter. This approach prevents the surface loading and air restriction that single-stage paper filter elements experience and consequently extends the service life of the air filter element. Finally, when the filter is sufficiently dirty to stop trapping small particles and clogging reduces airflow, it can conveniently be washed, re-oiled, and re-used. When oiling a foam air filter, use as little as possible, but do get adequate coverage.
Some manufacturers use foam as part of a multi-stage filter design. Stage 1 might be a screen or a coarse foam layer that stops larger debris from clogging up the stage 2 filter layer. Stage 2 would be a medium or fine pore foam to collect smaller particle that the stage 1 layer missed. Stage 3 could be a finer foam or cotton gauze layer that catches particles down to micron sizes. One or more of the Stages can be oiled to increase filter efficiency, and the whole thing, if made sturdy enough, can be cleaned, re-oiled and reused.
The only other (than stock) filter that doesn't reduce air box volume is made by Sportsbike (Australia). It mounts at the air tube entrance to the air box. It uses two pre-oiled, clean and reuse, single density foam filters. It's main drawback is that has a smaller cross-sectional area (about 7 sq. in. vs. stock 32 sq. in.) and dirt holding capacity than any other design (except for the tiny MadDuc "things" that mount on the velocity stacks.)
I think that it's important to be skeptical of the marketing hype by aftermarket filter manufacturers and their failure to provide ANY comparative dyno results that demonstrate that their product provides some performance advantage, or at least no loss in performance.
Modern high performance motorcycle engines are the result of countless hours of development work on a dyno by the manufacturer. They are designed as a unit with air flow and exhaust flow optimized together for each engine configuration. Purposely (say to save costs) using an unnecessarily restrictive air filter in their design will decrease performance relative to their competitors and their marketing advantage.
So, where are the dyno charts from the manufacturers of the aftermarket air filters? If they really improve performance over stock filters across the RPM range then it's really a marketing advantage to release their design development dyno charts. Without evidence to the contrary, think It's safe to conclude that it is not to their advantage to release any dyno charts or comparisons with other vendors. The aftermarket air filter market for motorcycles seems to be built on hype by the manufacturers and by the profits to their sellers. They compete on hearsay and testimonies from "happy" customers or recommendations from their own vendors- not on proof of superior performance.
Some Final Words on Air Filters and Performance
An air filter is not motorcycle performance equipment, it is protective equipment. It's function is to prevent dirt from entering the engine and damaging it. Dirt, by nature, is very abrasive and gets caught between parts that require a precision fit to function correctly.
An air filter that is selected for use on a race bike most often is not a good choice for use on a street bike. A race bike's function is to provide maximum performance and to finish (win) the race. Often the life of a factory racing team engine is practicing for, and finishing, one race. Then it's rebuilt to restore clearances by replacing any worn parts.
So, an air filter that is used on a race bike is selected using different priorities than one that is selected for use on a street bike. It primarily has to minimize any adverse effect on engine performance while still preventing the engine from ingesting dirt from a controlled racetrack environment. It has to capture and hold enough dirt without reducing intake air flow (clogging) to finish the race. It has to be accessible enough to be changed or cleaned quickly under racetrack conditions.
A street bike air filter, on the other hand, needs to function for thousands of miles in a variety of dusty conditions before cleaning or replacement. Consequently, it needs to hold a lot more dirt, and doesn't need to be nearly as accessible. Performance degradation is important but is still secondary to filter life.
The smaller the dirt particles captured by the filter, the better. Some designs and materials are better at this (filter efficiency) than others.
Whatever approach you or the filter manufacturer take, the underlying issue is that you need to CLEAN the filter before accumulated dirt reduces airflow sufficiently to reduce engine performance and economy.
This link http://ducatigarage.netfirms.com/filtertest.html goes to some testing done by the folks at the Ducati Garage/Neuspeed. This testing was done with a CAR air filter, not with Ducati filters!!