Kevin Cameron writeup

[rsss396]

Pipe Temp and Airbox Pumping
 By Kevin Cameron
 Winter 2000 - 2001
 A cold pipe can keep your clutches from hooking up
 Snowmobile airboxes and their intakes are being built as resonant systems

Let’s cover two subjects in this issue - exhaust pipe temperature effects, and resonant airboxes.
 
A cold pipe can keep your clutch from hooking up, because its lower resonant frequency peaks the engine at lower rpm than the clutch is set for. That's worth avoiding. Pipe temperature effects can also produce inconsistent and therefore useless dyno results - also worth avoiding.
 
The airbox used to be just an intake silencer and a place to put the air filter. Now it's much more than that, so read on before you gut or toss your box. Just as is being done on new cars and motorcycles, snowmobile airboxes and their intakes are being built as resonant systems. When the airbox is resonating strongly, driven by the engine's suction pulses, its rapid internal pressure fluctuation covers a range of plus and minus 10-15%. This is just like the resonance of a bottle when you hum into it. If your engine's intake events run in step with the positive side of this resonance, it's just like getting a 10-15% supercharge boost for free. That's worth having. And what if you modify your engine, raising its peak-power rpm beyond the range of the airbox resonant frequency? There is a simple relationship you can use to alter airbox frequency by changing the length and/or diameter of the airbox intake pipe(s). That's worth having.
 


PIPE TEMPERATURE EFFECTS
 
Why are pipe temperature effects important now? Back in the 1970s no one had ever heard of such a thing. What's different now?
 
In the search for horsepower, tuners have built pipes with ever-larger suction horns. Back in the 1970s, it was common for the ratio of header-pipe minimum cross-sectional area to center section area to be in the range of six or seven to one. If the header pipe was 40 mm, the center section was around 100 mm. As tuners have learned that stronger pipe suction and bigger crankcase volumes make more power, they have increased pipe suction by increasing this area ratio to values over ten to one and still climbing. This means that a 40 mm head pipe now feeds into a chubby center section of more than 125 mm in diameter.
 


The 2001 Arctic Cat ZR 500 is the first consumer sled to feature exhaust pipe temperature based ignition timing compensation. A temperature sensor fitted to the center section of the exhaust pipe feeds data to the ignition system, which selects the best timing curve to stabilize the engine performance for maximum performance and consistency.



What's the point? As center sections have grown fatter, their surface area has increased just as much, so any cold air reaching pipe surfaces cools them much more strongly than it did 20-30 years ago. Also and obviously, the use of separate pipes for each cylinder increases total pipe surface area even more. Therefore if you have a big bunch of black steel bananas under your sled's hood, you have the makings of a pipe temperature problem.
 
It works like this. Your engine's rpm of peak torque comes at the rpm where the return pressure wave from your pipe's final cone arrives just as the exhaust port is about to close. Fresh charge that has been pulled by pipe suction through the cylinder and out into the first part of the header pipe is now forced back into the cylinder by this reflected wave. This supercharge is what produces maximum torque in the engine. Its arrival time is determined by two things - the distance from exhaust port to the tuning point in your reflector cone, and the speed of sound in your pipe.
 
The length of your pipe doesn't change, but the speed of sound inside it very definitely does. Because sound is propagated as a wave through actual collisions of gas molecules with each other, its speed depends directly on how fast those molecules are, on the average, moving. The temperature of a gas is a measure of that average molecular velocity. Pressure is much less important, because it does not affect molecular velocity.

We aren't talking about a small effect here. In the very hot gas that jets from the cylinder just as the port opens, the speed of sound may be 2500 feet per second (fps) instead of the 1087 fps that is the velocity for so-called standard conditions. In the coolest part of your pipe, where the gases have expanded considerably, sound velocity may still be over 1400 fps.
 
Exhaust flow is extremely turbulent, and turbulence is a highly effective way of increasing heat flow between a fluid and the walls of its container. Therefore anything cooling the outside of the pipe also cools the gases flowing inside it. Those of you with dyno experience know that cold-pipe and hot-pipe dyno runs give different power, and at significantly different rpm. Now think about whether or not the "weather" around your pipes is the same on track or trail as it is inside the hot dyno room. Same? Or maybe pretty different?
 
Please bear in mind that when I refer to pipe temperature, I do not mean exhaust gas temperature, or EGT. EGT is measured in the head pipe, up as close to the exhaust port as possible, by a probe that sticks into the pipe to measure gas temperature only. EGT gives us valuable information about combustion. Pipe temperature, on the other hand, is just what it says it is - the temperature of the metal pipe itself. We are interested in pipe temperature because it tells us whether or not the pipe is hot enough to put the rpm of peak torque where it belongs. We measure pipe temperature on the outside of the fat center section of the pipe because this is the region - with all its large surface area - that is slowest to heat up and fastest to cool off.
 
Remember the ‘99 Arctic Cat ZR 440 race sled? It had a hot/cold rocker switch on the bar. At the start of a race, when the pipes are cold, the "cold" position switched the timing to a retarded map, thereby dumping more heat into the pipes to correct for their low temperature. The higher EGT of retarded timing also brought pipe temperature rapidly up to operating range. Once this was achieved, the switch would be flicked to the "hot" position, which would reduce EGT and give maximum hot-pipe performance.
 


It all started with the 1999 Arctic Cat ZR 440 race sled. A handlebar mounted thumb-rocker switch allowed the driver to select a “hot” or “cold” position to manually change the ignition timing to help build pipe heat for stronger acceleration.

For the 2000 ZR 440, Arctic made this system automatic as their "EPTS", or Exhaust Pipe Temperature System. Five ignition maps, covering the requirements of the pipe in stages as it warms up, are selected by the ignition computer, based upon a temperature-sensing thermocouple on the pipe center section. When the pipe is hot, the ignition is retarded approximately 5 degrees. A retarded ignition releases more heat into the pipe. This does two things. First, it makes the pipe work right now by compensating for the cooling effect of the cold pipe metal with hotter exhaust gas. Second, that hotter exhaust gas quickly heats up the pipe. As the pipe heats up and the speed of sound in the gases in it rises, the rpm of peak torque rises. To keep it from rising beyond the clutch set-up, the ignition computer switches in successively more advanced timing maps to reduce EGT, thereby keeping the rpm of peak torque close to constant.
 
Now the 2001 ZR 500 consumer sled will have a similar EPTS system. Does the retarded timing lose horsepower? No, it gains in power-to-the-ground because more is gained by keeping the engine working with the clutch than is lost by not pushing the timing to the bitter end. The curve of ignition timing versus torque is more like a round-topped hill than a spike - a little retard has a pretty small effect.
 
I spoke with Arctic Cat engineer Greg Spalding. He said, "Three or four years ago I looked at getting the pipes hotter to accelerate faster on our 440. Initially I tried a retard button, with some special Kokusan (ignition manufacturer) equipment capable of retarding the ignition as much as 40 after top. We found that twelve degrees after top held it at 5000 - that really put some heat into the pipes - like 1300 degrees. Then we translated that into the hot/cold rocker switch, and now the five-map system with the pipe temp sensor."

[rsss396]

REAL WORLD VERSUS THE DYNO
 
Nine years ago 500 cc Grand Prix motorcycles appeared at the US GP with carbon-fiber cases over their exhaust pipes. When I asked what the purpose of the cases was, I was told that they brought dyno and track performance closer together, and prevented the pipes from cooling off during off-throttle sections of track enough to hurt acceleration.
 


Aluminum clamshells and insulating heat wrap are widely used to provide a greater level of consistency to pipe temperatures. As an added benefit, the “shell-noise” of the exhaust system is also reduced.
 
We have all seen dyno setups in which the pipes stick out of the engine every which way - one right in front of a cooling fan, another somewhere else. Back in the days of small-diameter pipes this used to work semi-OK, but now that fat pipes lose heat so fast and take so much time to warm up, dyno testing has to make allowance for pipe temperature and/or cooling effects. If it does not make such allowance, any gains made on the dyno may not translate to the snow.
 
Another effect many of you will be familiar with is inconsistent performance from aftermarket twin or triple pipes. The people developing such competition-oriented pipes assume their users will be hard on the throttle all the time, keeping the pipes hot and therefore working as they should. Users who aren't hard on the throttle may be disappointed, because big individual pipes cool right off under those conditions, under-peaking the engine.
 
Why worry if the engine under-peaks by a few hundred revs? Here's why. Every system on the engine is designed for operation in a specific rpm range - the intakes, the airbox resonance, the porting, and the clutch. If the engine doesn't reach the rpm these systems are designed for, don't expect them to work at their best. You want a symphony from your engine, not Dixieland.
 
Savvy tuners are now realizing that they need to know how hot pipes must be in order to work right. Working perfectly in a hot dyno room during summer development sessions is no guarantee that those pipes will reach that temperature with big clouds of Wyoming powder snow hissing onto them, mixed with frigid 20-below air. Solving this problem means A - measuring the temperatures of pipes when they are working properly, and B - taking steps to make sure they can reach that same temperature on your sled, in the field. That means either (a) creating an automatic system as Arctic has done, or B - measuring pipe temperatures in the field and finding effective means of raising them to the design point - limiting air access to the cowl, wrapping the pipes or parts thereof, or installing pipe covers.



Exhaust system center sections have grown increasingly fatter over the past few years. This increases the pipe suction and resulting power output, but the increased surface area makes the exhaust more susceptible to the affects of cooling.

BACK TO THE AIRBOX
 
Any hi-fi enthusiast knows that woofer enclosures work best when the resonant frequency of the enclosure is nicely centered on the speaker's response range. The enclosure usually consists of a sealed volume with the speaker installed in one of its walls, and an opening, called a reflex port, cut into the enclosure. A resonant system consists of a mass, which vibrates back and forth against the restraint of something flexible, like a spring, with an excitatory force to drive it. In the case of the speaker enclosure, the mass is the air in and within one diameter's distance of the reflex port. The spring is the compressible air inside the enclosure. The system is set into vibration by the amplifier, driving the speaker cone back and forth as a piston.
 
In the case of an engine's intake airbox, the mass is the air in the airbox inlet pipe(s). The "spring" is the compressibility of the air in the box. The excitatory force - a very powerful one - is the endless sequence of strong engine intake suction pulses from the carburetors. The airbox must not have any significant leaks, as the throttled, back-and-forth airflow through them acts like a hand on a vibrating bell (anyone who's ever tried to play low notes on a valved wind instrument knows what a killer leakage is). The airbox inlet pipe is usually made with a smooth bellmouth on either end to reduce flow losses. Carburetors or throttle bodies must likewise seal positively to the box. When a system like this gets to humming, the pressure inside it vibrates rapidly plus and minus 10-15% of atmospheric pressure. In fact, the humming is so powerful that in many cases a sub-resonator is placed near the atmosphere end of the inlet, to prevent radiation of this powerful honking sound to the outside. EPA objectors are always waiting there with calibrated sound meters and spectrum analyzers at the ready.
 
How can you adjust the resonant frequency of your airbox if you raise your engine's peak-torque rpm with pipes or porting? One way is to invest $30,000 or so in professional wave dynamics software like Ricardo "Wave", running on a $10,000 Sun workstation. Probably on the right back street in Hong Kong you can pick up a pirate copy for $25, but which street is it?
 


The airbox inlet tubes, or “horns”, are specifically designed to provide a resonance that can increase the total airflow by up to 10-15%. Removing these can cause the engine to loose power and increase the intake noise.

We're so used to the idea that problems have to be solved with silicon logic that we forget about steel and aluminum solutions. “Wave” is great if you have a tricky fuel mixture glitch with #7 cylinder in your Ford NASCAR engine. But with a simple formula that tells us which variables push the airbox frequency which way, and by approximately how much, we can devise dyno experiments that will get us the answers we need - without those expensive Cathay-Pacific coach tickets.
 
Here is the formula.
 (Airbox Frequency), squared, is proportional to inlet pipe area/(airbox volume X inlet length)
 
This is useful because it shows us that if we want to raise airbox resonant frequency, we must increase inlet pipe area or decrease airbox volume or inlet pipe length.
 
AN EXAMPLE
 
If our present engine is a twin, giving peak torque at 8200 rpm, that is 8200/60 = 137 revolutions per second, or 137 X 2 = 273 suction pulses per second. Unless there is some special problem, the airbox will be designed to resonate near that frequency.
 
If we now want to raise peak torque revs by 10%, to 9020 rpm, we must also raise airbox frequency by a similar amount. If we raise airbox frequency by 10%, its square will increase by 1.1 X 1.1 = 1.21 times, or 21%. That means that whatever is on the right-hand side of the equation must also increase by a factor of 1.21. Take your pick.

You can:
 a- increase inlet pipe area 21% (that is, increase its diameter by 10%) or,
 b- decrease airbox volume by 21% or,
 c- decrease inlet pipe length by 21%
 
Because these systems generally work better the bigger you make the airbox, we won't try  B  Since we are raising revs and power, increasing inlet area looks pretty good, so we could choose option A - increasing inlet pipe area. However, option C - would appear to be the easiest. Before we go to the dyno, we'll make up a few airbox inlet pipes to give us some test choices. Then we can run through our tests quickly and zero in on the sweet spot. Each end of the box inlet pipe should have a smooth bellmouth.
 
Likewise, go carefully before removing internal airbox "furniture". Assume nothing, but test with each change to understand its effect. Airbox designs are sophisticated now, so their internal features often have functions.
 
Any resonant system always has anti-resonances. In the case of an airbox, that is an rpm at which the engine breathes from the box when pressure is at the low part of its cycle. What if there's an anti-resonance right where you want your clutch to engage? Of course you could imagine a system with a variable-length inlet pipe to deal with this, but the easy way is just to kill the anti-resonance by opening a big hole in the airbox. Systems of this type are in use on certain sports motorcycles. When the engine runs near the rpm of the anti-resonance, the engine control computer tells a little motor to open the airbox port. When it revs up, the motor closes the port.