ESE's works engine tuner

[wobbly]

Expanding bellys have various effects as noted above.
But in pipe design it allows you to increase the effect of the reverse cone, without making it shorter.It also increases the volume and can reduce the propensity of the wave to create a shock front ( killing energy) as it reverses back toward the diffuser.
This is seen as a quite noticeable blip in the sim, example below seen at around 20* before TPC, of the pressure ratio at the EX port

[bucketracer]

Dad was relating a story, told to him by Speedpro about how he tried a RamTube on his Suzuki 50. Apparently it worked really well at idle, which was the RamTubes resonant frequency and again at the higher harmonics as the engine was reved up, but it killed the engine in between resonant peaks.

The peaks were remarkable but the down sides out weighed the benefit of the peaks and there didn't seem any real way of using the idea. A variable length inlet might work but it would be hard to have it long enough for the bottom and then make it short enough for the top end.

Here is another idea about how to change the resonant length of the RamTube, from http://www.phys.unsw.edu.au/jw/woodwind.html#sound



Vary the length of the reflected wave by opening holes in the side of the RamTube like they do with Wood Wind Instruments.



By having holes at odd distances along the tube, the harmonic frequencies in the RamTub could be kept in step with the engines RPM.

The holes could be opened and closed by RC Servos.

True it would make TeeZees bike six feet wide but as long as he sticks to the outside of the track all should be good, or maybe he could put an outrigger wheel on the RamTube and enter it as a side car. Seriously the idea has got to be worth looking at.

Norton Rotary GP bike with variable length inlet http://www.youtube.com/watch?v=48foOHK2tyA

And a bit of a speed comparison http://www.youtube.com/watch?v=kBg86...eature=related

[speedpro]

Or you could look at an R1 intake system

[Moooools]

Dad was relating a story, told to him by Speedpro about how he tried a RamTube on his Suzuki 50. Apparently it worked really well at idle, its resonant frequency and again at the harmonics as the engine was reved up, but it killed the engine in between resonant peaks.

The peaks were remarkable but the down sides out weighed the benefit of the peaks and there didn't seem any real way of using the idea. A variable length inlet might work but it would be hard to have it long enough for the bottom and then make it short enough for the top end.

Here is another idea from http://www.phys.unsw.edu.au/jw/woodwind.html#sound



Vary the length of the reflected wave by opening holes in the side of the RamTube like they do with Wood Wind Instruments.



By having holes at odd distances along the tube, the harmonic frequencies in the RamTub could be kept in step with the engines RPM.

The holes could be opened and closed by RC Servos.

It sounds good when you express it like that, but in reality opening holes does not work for an engine.

Opening holes works for STANDING WAVES. Standing waves are standing because they have no velocity.

The air coming into a carburetor does need velocity (or else it wouldn't work). So the standing wave model does not apply.

__________________________________
However it can be more closely modeled with other principles of SHM (Simple Harmonic Motion).

A spring oscillates a mass (m) at a frequency proportional to 1/m. In other words: the larger the mass the lower the frequency.

In this case the mass is the air inside the inlet tube to the airbox, and the spring is the compressing and expanding air inside the airbox.

Therefore if the inlet tube is made smaller (width or length) it will oscillate at a higher frequency for the top end, and if the inlet tube is made smaller it will oscillate slower for the bottom end.

You do not, however, need a tube that is shortest at idle and longest at the top end. You would still have the resonant peaks that occur when the engine is at 1/2,1,2,3,4 or 5 times the resonant frequency.

So all you would need is to have a pipe set up for say 4000RPM at full length and this would cover 1000rpm, 2000rpm, 4000rpm, 8000rpm and 12000rpm at full length.
In between these points it could be shortened with a servo to the correct length and be put back out to full length as it reaches the next point.

This would give you maximum benefit at 4000RPM as it would be one oscillation per revolution, whereas 8000rpm would be one oscillation every second revolution and at 8000 rpm it would only be in step every third revolution.
__________________________

Anyway, that is my spiel over. I just love physics like that.
If I have made any mistakes in there let me know, and I will beat myself up.

Enjoy

[bucketracer]

This would give you maximum benefit at 4000RPM as it would be one oscillation per revolution, whereas 8000rpm would be one oscillation every second revolution and at 8000 rpm it would only be in step every third revolution

Ok Mooooools, sounds good, now if the pipe is say 1m and correct at 4,000rpm what would the lengths need to be for 5, 6 and 7,0000?

Would that be, if 4K=1m then 5k=0.8m 6k=0,67m 7k=0.57m and then back to 8k=1m again, 8k is a half fundamental and 9k=0.8m 10k=0,67m 11k=0.57m all for halves and then back to 12k=1m and 13k=0.8 for third waves.

Not to sure about my numbers but if they are right I can see how your idea would work.

The fundamentals will have the most effect and the halves less so but the coming on the pipe more than makes up for it. If I have understood all this correctly it looks like it could be made to work and is worth a try.



Looking at TeeZees graph, I think that fundamentals need to be around the 4-5-6-7 mark.

Next time TeeZee has his bike on the dyno, Thomas and I will play with some tubes to see if we can find some sort of resonating length to start with.

[Moooools]

It is too hard to do the maths on it. But I just found this good link with all of the theory. It pretty much backs up my on-the-fly model of the physics ivolved.

http://motorcycleinfo.calsci.com/Airboxes.html

And also just realised that you would not even have to have the steps in the inlet.

If can you make the airbox small enough (not too small though) it will increase the spring constant of the air inside. This means a shorter length pipe can be used and which would result in a shorter range of motion need for the variable length pipe.

If you cannot get the airbox small enough then the inlet pipe can be made wider so that the is more change in volume for a given change in length. I can't be arsed doing the calculations but they are all at the link above.

[Bert]

Looking at TeeZees graph, I think that fundamental need to be around the 6-7 mark.

Next time TeeZee has his bike on the dyno, Thomas and I will play with some tubes to see if we can find some sort of resonating length to start with.

I do wander if improvement could be resolved with good low-mid range jetting (more likely needle shape and slide cutout) and ignition; given the performance above this range.

Maybe an exhaust powervalve could be in order??? I'm sure you guys are talented enough to machine / install a blade type valve into the GP to adjust the port heights for more bottom end.

Moooools; I think there was a Java app floating around a few years back that parelled the info on that webpage. I have a funny feeling the boys covered that back around page 80?? around the time the euro boys poped in.

Also I stumbled across this website (Pistons).
http://www.strikeproducts.com.au/pistons.asp
Kartng pistons and one offs... Quoting another forum on the background:
The guy in Perth that does the Strike pistons is Ken Seber. He used to work for Orbital engines and set up the gravity casting facility there. When he was made redundant he took on the piston casting setup and started making kart pistons. Realy did make a nice part....

[bucketracer]

Thanks Bert and Mooools this is all good stuff, next time the bike is up on the dyno we will try a few bits of pipe and see what happens.

[bucketracer]

But I just found this good link with all of the theory. It pretty much backs up my on-the-fly model of the physics involved. http://motorcycleinfo.calsci.com/Airboxes.html

Had a look at the link and maths.

They suggest about 300mm for 10,000rpm in a 4-Stroke (depending on diameter) so that's 5,000rpm for a 2-Stroke, now we have something to start with.

[wobbly]

There is no "formula" that can directly generate a tuned length for the intake of a 2T as there are too many extra variables involved.
Helmholtz doesnt apply and neither does organ pipe theory as the tuned tube is connected to a varying volume, with a varying intake end area
In a rotary valve there is another scenario going on and that is the standoff issue.
The actual "real" reflection point of an intake tube is about 1/2D past its end point, this is reduced alot by having a tapered and or a bellmouth entry, but the waves bouncing up and down the tract - still hits "soft" atmosphere out past the carbs end.
That is why there is a "fog" of fuel sitting outside the carb when the effect is working.
Thus in practical terms we find that with a RV, its all but impossible to get the intake short enough to have the second harmonic ( the most useful of the tuned lengths) in tune with the engine "forcing frequency" when in the powerband, along with reducing the deleterious effects on carburation caused by too much stand off.
Aprilias have a 42mm carb hanging off the RV cover with as short a body as they can, with no "manifold" at all really.
In a reed valve setup,tuned for peak power around 12000 rpm, we find that the best tuned length is around 135mm from the reed tip, to the end of the bellmouth works best.
What this does is hard to explain but easy to show you.
The wave bouncing up and down the intake length, when its "in tune", will arrive at the reeds as a + going pressure ratio, at the same time as the case drops below atmospheric, due to the piston rising - plus the pipe diffuser sucking like hell on the Ex Port around BDC.
When this happens the reeds have + on one side and - on the other.They open real fast, with little "flow energy" being waisted, and Hp is made big time for free.
Here is a graph of a very winning 125 kart engine, the only change being to use a 20mm shorter intake rubber that hit the 135mm rule, spot on.
The sim curve shows the intake pressure ratio crossing zero at the same time as the case, but with opposite signs,right at the point in this case when the PV closes and the engine is coming on the pipe as well.
Using the second harmonic spreads this effect over the widest band with the highest useable effect.

[kel]

Interesting stuff but left me slightly confused, didnt the piston port motors rely on calculated tuned intake lengths? I understand helmholtz dosent apply to reed motors as the reeds effectively block the reverse flow, but surely the rotary disk, even though they are usually forced to have very short intake lengths can utilise helmholtz and tuned length. From memory Kawasaki ran a system called ram tubes on some of their rotary valve racers. Is it not possible we were seeing the benefit of helmholtz when TZ increased the GP’s crankcase volume? i.e. a bigger crankcase volume lowers the Helmholtz-frequency (as does a longer intake length). The introduction of the bellmouth may have helped as well I guess.
I am curious as to the Aprilia RSA’s intake runner; there must be one as they run a pressurised airbox?
ps. Love the simulator shot

[wobbly]

Well you see its something lots of closet professors delight in doing - trying to assign hard numbers to a highly variable system, but as in many cases like this they get shagged up the bum by reality.
Increasing case volume decreases the Helmholtz resonance frequency, but big volumes work better at high rpm, due to the pipe having greater volume to suck on.
Works in reverse, oh dear.
The rotary valve engines are more affected by the intake resonance creating big standoff effects, thus the intake length is best when as short as possible, as this reduces the resonance "tuning" to a point above the engines forcing frequency.
Forget intake length tuning - it doesnt work. edit ( in a racing engine with closing at 80* or more ), and even worse is that the standoff issue cant be included in a sim,to show the bad effect on fueling, overcoming any advantage gained via "ram tuning" length.

Intake airbox theory is quite different.It is basically a simple Helholtz, but again, in practical terms it is all but physically impossible to create a big enough volume that is "correct".
The intake diameter and length are critical, and do conform to theoretical analysis.
Reed intake length tuning does work well in a sim, where the length and case pressure effects can be accurately predicted,but as to creating a simple theoretical relationship to give a correct tuned length for any engine config, i just cant see it ( love to be proved wrong, get Dr Aitken on the job).

[wobbly]

Just to add a little more to the intake tuning scenario,the TD and TZ piston port engines had an "insulator" spacer on the cylinder about 25mm wide.
We all thought this did what it was called ie kept heat off the carb,but no.
Shorten that spacer by 5mm and it was absolutely impossible to jet the carb.
Even blipping the throttle you saw huge standoff at the bellmouth.
Make it longer to pick up bottom end power, and for every 1 Hp gained at 8000 you lost 2 at 10000.
So again, reality fucks over theoretical analysis.
The shorter length was technically correct for the rpm,but was impossible to utilize.

Edit, the other thing I forgot to add before re Mr Helmholtz.
Its all very well making an assumption that the intake and the case vol is in fact a Helmholtz - the problem is there are 3 others doing his finest.
The case connected to the transfer ducts, the transfer ducts connected to the cylinder volume, the cylinder volume connected to the exhaust duct, and the exhaust duct connected to the pipe volume, and the pipe volume connected to the stinger.
How many did I say, anyway we have a multiplicity of ducts and volumes, trying to define one as independant of the rest is suspect to say the least.

[TZ350]

Interesting, in the hand activated trial of the V Tec there is the beginings of some extra power at the top. Not sure what contributed this, the vetec added an extra 25mm to the inlet tract for 195mm total or possibly the little kick I gave the end of the divider better directs the flow behind the conrod as the incoming air/fuel enters the crank case.



This is the RC Servo that I am going to use to close the V Tec below 7,000 rpm. There is a small battery pack and a servo tester set to the closed position. The plan is that below 7k the Ignitech activates the relay and allows the pulses from the servo tester to tell the servo to close. Above 7k the relay is open and there are no pulses going to the servo motor so it automatically returns to the neutral position and opens the V Tec.



Rotary Valve side shows the divider coming right up to the valve. In this view the valve rotates ant-clockwise. The inlet is 45 deg wide and the V Tec closes it 10 deg earlier, I would have liked 30 but this is how the first cut turned out.

Everything is securely glued in place so now we can try a different rotary valve and get into some serious tuning for more power spread and explore water injection.

Next time I make one of these I will try and reposition the divider so it closes earlier, then I can have a realy radical closing point for extra top end power while keeping the wide power spread.

[bucketracer]

Rotary Valve side shows the divider coming right up to the valve. The inlet is 45 deg wide and the V Tec closes it 10 deg earlier. Then I can have a realy radical closing point for extra top end power while keeping the wide power spread.

In the old Rotax tandem twin, going from 85* ( stock) to 88* is like night and day in overev performance, but carburation becomes finicky and alot of bottom end power is lost.

Hmmm now that TeeZee has found that the V Tec cleans up the bottom end carburetion I think I can see where he could be headed with this variable inlet closing idea...........