Pipe dreams.

Derek, Dan,
There needs to be synchronicity between all of the sequential events throughout the entire engine cycle. Assuming that all of the air going into the engine equals that which exits the engine then one element that does not blend harmoniously will hamper the effective use of that mass of air.
Inserting two inches into the parallel section of the pipe will indeed alter the t/l, but it will also affect the proportions of the remainder of the pipe, if indeed those two inches are needed? The header and diffuser will be positioned further from the reflection point of the reverse cone. The t/l can only be fully effective when working with a correctly proportioned, dimensioned and positioned diffuser. The diff controls the drawing of mixture from the case into the cylinder, too long a header delays that draw later in the transfer cycle. The effect is to shorten the useful transfer phase with the diff trying to pull from a closed port. The reverse is also true.
The peak of the suction (negative) reflections from the diff needs to be around bdc where the piston is stationary. There is no point in having the most aggressive suck on the case with the piston partially obscuring the transfer ports.
The header needs to accommodate the exhaust pulse with a shallow taper that does not reflect reverse-sign waves. The diff effect begins when the piston starts to expose the top edge of the transfer port. Suction reflections start to emanate from the short but larger first angle of the diff, the most aggressive angle has the longest length to work around bdc and the third angle shallows down to match the parallel section that has the largest diameter. There is no point in having large angles when the port is only partially open, you simple loose efficiency.
Whatever t/l you arrive at the header needs to terminate at 33% of that t/l measured from the piston face, and the diff, 66% from the piston face. So you can see that just sticking two inches into the parallel section might cure one problem but create another. Those two percentages can of course be slightly adjusted but don`t stray too far. For a 1000mm t/l, a 5% error is a long way.
What is required is a long, phased suction on the crankcase to pull the maximum quantity of fresh mixture into the cylinder. How you choreograph the expulsion of spent gas and retain the fresh whilst not mix the two is of course, something else again!

Trevor

On the other hand!

What all Bantam tuners should seek to achieve is a broad band to the power range delivered by their engine. This is especially relevant as we have only three gears with which to get that power from the crank to rear wheel, and accelerate away, torque then becomes everything!
The characteristics of available power delivery are heavily influenced by the exhaust pipe and the pulse resonance.
At the engine`s optimum rpm fresh mixture is sucked from the case via the transfer ports, and some goes directly into the exhaust duct. The reflected pressure pulse from the rear cone pushes this fresh mixture from the duct back into the cylinder and the exhaust port then closes. So far so good, pretty standard stuff. However, at around 2/3 of this optimum torque rpm the returning pulse arrives when the transfers are still open, and, the fresh mixture is returned to the cylinder and on into the crankcase. The consequences of this are pretty dire for engine power. Not much of a problem if your engine has six gears and a power valve, drop a couple of gears get the revs up and the all is well again, we haven’t got six gears!
If you want more revs you need more time/area, and the higher the exhaust port goes the stronger the exhaust pulse is, and the more the disruptive influence is on scavenging at the 2/3 torque rpm. As the gas then has less time to expand and cool, the duct will be hotter and the gas lurking there will remain hotter and the density of the next scavenge cycle mixture will be less and you make less power!
Once more, it is a case of what was wrong initially can`t be made up for later on.

So serious was the situation with the Aprilia engine and a power band that only began at 10,000rpm with the power valve fully closed. The exhaust port was not then fully open until 12,000rpm, closing the throttle would reverse-flow so much that combustion deposits exited the carb and detonation rapidly set in. Max revs, throttle wide open, no problems!

Trevor

the question then must be ............. what set of timings / events make a broad powerband with high power torque ?

also , what psi must be left above the piston (less than crankcase psi I assume) to allow the pumping through the transfes?, would a widening of the top 1/4 of exhaust duct let the pressure drop commence quicker and thus give the down stroking piston a little more to shove?

is crankcase pressure measurable ?

Dan,
Case pressure is both measurable and calculable, an example of calculation is the RS 125. At transfer opening pressure is 1.4atm this then increases to 1.43atm as the reverse flow gets going, but quickly falls to 1.06atm at bdc, and goes negative at 15*abdc: this is a simulation result. By installing a high speed pressure transducer to the crankcase and connecting to a computer, pressure fluctuations can be measured. This is the sort of stuff that QUB are so good at.
The pressure in the cylinder just prior to transfer opening is almost entirely dependent on the exhaust function, this includes the port and duct that lead to the pipe. Blow-down STA is paramount, so yes, max out the port area prior to transfer cycle, leaving the lower part narrow will help to discourage short-circuiting. A lower exhaust port height allows for a greater pressure drop before transfer opening, after saying that the bigger the ex. port the more the scope for scavenge losses, it`s all a compromise.
It is true that once in the effective power band the engine becomes a resonator and not a pump, but case pumping can help a lot to get things moving so to this end it is worth eliminating all of those dead areas in the case particularly the further away from the transfer ducts that they are. Typically the recessed flanks of a standard crank disc are an example, there is little chance of the mixture trapped there to be moved to the cylinder with the case wall sealing off the escape route. Concentrate that mixture close to the transfer ducts and performance goes up. Holes in crank discs function as dampers and expand and contract with pressure fluctuations and only contribute to the overall measured static volume.
I personally would not go to 130* transfer timings, particularly if adequate time/ area can be achieved by port number and width. You need to keep timings moderate, and to achieve the required blow-down for the rpm you select. I have always done my sums for the w/c engine at 10,500 rpm, one transfer timing just nudges 130* the remainder rather less, the exhaust duration is 192* It is a nice torquey motor capable of pulling high gearing; 14x54 at Lydden. Another thing I do is to use slightly longer pipe dimensions than the "formulas" dictate, to keep the power band wide.
The big bore, long rod Bantams have features that hinder power making and absorb a lot of energy that the 54mm bore 125s aren't handicapped by, and as I keep banging on about, the lack of a sensible ignition system makes things much worse!

Trevor