Perfect header length

Discussion in 'Performance Mods' started by jaguar, Nov 18, 2011.

  1. jaguar

    jaguar Well-Known Member

    Most people mistakenly think that to get the extra power from an expansion chamber is as easy as just bolting one on. Not so. let me explain. The distance from the exhaust port to the beginning of the baffle (the cone that reduces diameter with distance from the engine) is the most important. The sonic wave, first created when the exhaust blasts into the pipe, bounces off the baffle cone and produces extra engine power at a certain rpm range dependent on the distance from the engine. For the target peak rpm power it needs to arrive back at the exhaust port just as the piston, traveling upwards, closes the transfer ports. That return pressure wave increases the pre-combustion pressure for increased downward force on the piston during combustion (of fuel/air). If it arrives too soon or too late then the peak potential engine power is lost. Arriving too soon, due to a too short header pipe, it forces intake fuel/air back down into the crankcase thru the transfers and there remains less fuel to be combusted. Arriving too late, due to a header too long, there is less time for the wave to raise the cylinder pressure to the maximum possible and the best pipe effect is lost.
    How do you know what header length to use? Good question. It depends on ignition timing, cylinder compression, fuel mixture, exhaust port height, stinger diameter. Too many variables to accurately calculate. I can give a ballpark figure for a standard 48cc engine but its just a starting point. The right length could be longer or shorter than that. For quick easy testing I used rubber automotive hose for one or more test runs on a flat street to test for top speed. The hose was replaced once it was warped. After finally settling on a correct length I had steel tubing welded in place. Even being off by 1 inch can make a 1.5mph difference in top speed.
    If your header pipe is too long then it benefits an rpm range lower than the peak rpm the engine is ported for. So it would give good mid range power but limit top rpm power.
    If your header pipe is too short then you get neither good mid range power or good top rpm power. From all the photos I've seen I'd say most setups have too short a header. Most people get no more than a mean sound which makes them think there's more engine power.
    Where to start
    For a 48cc engine at sea level with a peak rpm of 6500 I would start at 37 inches (distance from beginning of baffle to piston face at exhaust port). If the exhaust port has been raised some then start at 39 inches. Here's a link for steel tubing almost the exact match for the header pipe diameter of the Grubee engine.

  2. Ghost0

    Ghost0 New Member

    Excellent post, I have often repeated the same information to our customers and is also mentioned in our installation manual. However, to state that the correct header length for a 49 cc engine is 37 inches is a bit misleading. The proper header length does depend on all the factors you mentioned but there is one critical one you left out. That is the actual design of the expansion chamber itself. The correct header length is dependent on the divergent and convergent cone angles as well as the diameter and whether it has any straight section in between the two cones. We do know from dyno testing that for our pipes expansion chamber design a good starting header length is around 12 inches. Again this is only a recommendation for our pipe and all the things you mentioned need to be considered as well as the gearing of the bike.
  3. jaguar

    jaguar Well-Known Member

    The distance I recommend as a starting point is not the header length, but the distance from the piston to the beginning of the baffle cone.
    The topic of this thread is misleading since the topic is not purely about header length. But when you go to set the correct piston to baffle distance it is the header length that you change.

    The starting length should be 29" not 37" according to Gordon Jennings who wrote that a good sonic wave speed to use is 1700 fps. I'm glad you questioned my post so that I had to research some more. I should know by experience but I live at 8500 ft altitude and have not experimented with pipes at sea level yet. Jennings paper is entitled "Do You Really Want to Know About Expansion Chambers?".

    Here's the math behind my result:
    With an exhaust duration of 140 degrees and a transfer duration of 110 degrees the duration between exhaust opening and transfer closing is 125 degrees. The period of one crank rotation at 6500 rpm is 9.23 milliseconds. 125/360 times 9.23 ms = 3.2ms
    At 1700 ft per second the sonic wave will travel 1700x.0032=5.44 ft which equals 65.3 inches. Since the wave travels to the baffle and then back to the cylinder it is necessary to divide that distance by two to get the correct distance between piston and baffle. 65.3"/2=32.6"
    But since that 1700fps is probably for higher compression better carburated engines then I think it's safe to multiply that number by .9 to arrive at 29.3"
    But then there is a huge difference between engines with the standard CDI and a true 2 stroke CDI because the standard one advances the spark way too early at high revs which ignites the fuel too early which insures that the exhaust is more "spent" by the time it exits into the exhaust pipe which means it is cooler in temperature which means the sonic wave is slower. So with a standard Grubee CDI the 1700fps number should be less. If it is actually 1500 then the length in question should be 25.8". This is the trouble with unknown variables which is why in the end you can only know by testing different pipe lengths. Start with a long length and shorten the header by 1/2" between each run. The run that gives you the best top speed is the one with the correct header length for your engine and pipe.

    A word of warning: Running the standard Grubee CDI with an expansion chamber set to the correct length may cause piston seizure because of the combined heating effects of overly advanced ignition with increased cylinder pressure at top rpms. If you have a standard CDI then its best to not have the pipe length corrected so that there won't be a sonic wave pressure boost at top rpms.
    Last edited: Nov 18, 2011
  4. Ghost0

    Ghost0 New Member

    Yes that is correct. The header is simply the exhaust manifold in most instances. We are both talking about the distance between the exhaust port and the beginning of the expansion chamber as this is the critical tuning dimension. All I was saying is that the design of the expansion chamber itself has a lot to do with the correct distance between the exhaust port and the start of the expansion chamber.
  5. jaguar

    jaguar Well-Known Member

    I agree but the most critical distance for top end power is where the beginning of the baffle is located. The most critical distance for mid range power is that of the beginning of the fist cone (ie: diffuser). For both distances to be correct a pipe needs to be designed just for the engine in question. Other than the Jaguar pipe I know of no such thing. People are selling pipes made for high rpm power pocket bikes that deliver a narrow powerband due to the steep "racing" angle of the baffle. The Grubee engines are low rpm and need a wide powerband for one speed street use. The exception would be if the bike has a shift kit and you don't mind shifting a lot to keep the engine "on the pipe".
  6. lazylightning@mail.r

    lazylightning@mail.r Active Member


    I'm wondering about the pipe material that I can use for the header, and of course a proper design for the chamber.
    I wanted to know if the header tube should be smooth inside or if it's okay to use the ribbed flexible staibnless tube, like this one:
    I am afraid that the ribs in the walls may disrupt the sound waves we are trying to tune. I have no access to welding equipment now and am considering trying to make an expansion chamber with screws, bolts and jb weld.
    I was even considering running the intake tube between carb and engine, through the inside of the exhaust pipe to heat the fuel mix before it enters the engine. Now I'm considering just attaching it to the outside of the expansion chamber in parralel so as to not disturd the inner dynamics of the expansion tube.
    So what about that header tube? Should it be smooth inside?
  7. jaguar

    jaguar Well-Known Member

    Yes it should be smooth inside.
    The fuel gets heated enough once the engine warms up. Your idea would overheat it making it expand and lowering engine power.
  8. lazylightning@mail.r

    lazylightning@mail.r Active Member

    Thank you for the insight into that. What is the ideal length of an extension tube between the carb and engine if there are no modifications to the ports or timing? Only the head has been sanded down to the fin structure surface for higher compression. I also polished the inside of the hemispherical combustion chamber of the head.
  9. jaguar

    jaguar Well-Known Member

    Around 3" more distance is ideal. That doesnt include the overlap of the tubing at both ends. For 1 inch overlap then the tubing needs to be 5 inches long. For even more boost you can weld a 6 inch long steel tubing (21mm dia.) into the exhaust pipe.
    Click onto my signature link to read more.
  10. cRazy_styleR

    cRazy_styleR New Member

    I have a question about shortening the header pipe of exhaust.
    I've red your text above.

    But what happens with rpm after pipe is shortened? For example, I have peugeot speedfight 2, tp trophy 70cc lc, yasuni r and some other tuning parts.. What will happen if I cut header pipe for 15cm ? Speedfight 2 has vertical cylinder position and has a really long header pipe.

    Friend also has (keewey focus) tecnigas trek and motoforce standard 70cc ac.. What would happend if he cuts pipe for 15cm?

    Will the rpm's go higher / lower and what about power band? Is it going to be wider or ?

    Last edited: Nov 10, 2012
  11. jaguar

    jaguar Well-Known Member

    Shortening the header raises the rpm peak, also shifts the powerband higher to match. does not widen it.
    Lengthening the header does the opposite.
    Now how much higher than the non-EC* you can get away with is unknown to me. On my engines I tune the pipe to peak out at 5-10% higher revs than what the engine does with just a straight header and muffler (standard). I suggest finding out that rpm peak first.

    EC: expansion chamber
  12. cRazy_styleR

    cRazy_styleR New Member

    Thanks a lot!

    I'm alo a bit confused (I don't know much about designing expansion chamber and timings) about pipe curves. For example, if I cut header pipe from yasuni r for about 6" (15cm), and notice that new powerband does not fit very well, or performance is a bit worse, may I weld pipe to old length (with cut piece, as it was before shortening) and get powerband to old "graph" ?

    I don't want to make mistake with no posibility to correct it :)
  13. jaguar

    jaguar Well-Known Member

    of course you can return it to how it was
  14. cRazy_styleR

    cRazy_styleR New Member

    I just measured my yasuni r pipe. Pipe looks like this:

    First mesaure is 27cm (10.5 inch), before weld, and after weld is 32cm (12.6 inch). And after second weld, it is 33cm(13 inch) diamter.
    Why are there different diameters of the pipe, when (difuzor) cone is way further? How would engine behave if that pipe was the same diameter (width) a whole way to difuzor cone ?
    By difuzor cone I mean on cone which is first on expansion chaber.

    Also, will performance change if only (without any other modifications) pipe after expansion chamber and before silencer is longer ? And what if that pipe is bigger diameter than stock?

  15. jaguar

    jaguar Well-Known Member

    having the header expand is something useful for high rpm engines. But the increased diameter where the diffuser begins leaves the designer less ability to have a healthy diffuser angle unless the belly diameter is increased.
    the questions you ask lead me to believe you haven't read my info on expansion chamber theory. Please read it completely (following all the links) and then get back to me. I charge a small fee to design pipes if you need that done for you.
  16. mikejeffries

    mikejeffries New Member

    The period of one crank rotation at 6500 rpm is 9.23 milliseconds.

    Is this correct?
  17. jaguar

    jaguar Well-Known Member


    yes that's correct
  18. cRazy_styleR

    cRazy_styleR New Member

    You're right, I missed your link to book. I saw there was formula for d3 wich is the stinger length.

    Somehow, it's confusing to me. Here's my point of view:
    Gases back of the exhaust cane, others go through stinger and silencer.. Gasses which went through the stinger are not returning into chamber, they are leaving through silencer, or are they? Don't understand that part - How does the longer stinger affect to higher back pressure?

    I want to modify my exhaust, but don't have some strong expirience. That's why I have so much question ..

    It would cost a lot of money while that pipe came to me. :(
  19. jaguar

    jaguar Well-Known Member

    I don't make pipes for people, only consult to help them in their own construction.
    smaller diameter stinger and/or longer stinger increases flow resistance which increases back pressure.
  20. cRazy_styleR

    cRazy_styleR New Member

    Ok. :)

    So, basically the longer stinger causes bigger back pressure which will then be faster, am I right?

    I red that midsection is there to define and set correctly back pressure time to cylinder. So, I'm thinking, if back pressure comes to cylinder earlier than it should be, increasing the midsection would set this timings right, right?