ESE's works engine tuner

[Condyn]

Apologies in advance for questioning a very old, tried and trued idea.

With the concept of weight savings set aside, what is the purpose of the piston cutouts under the wrist pin? I understand they keep the transfer entrys un shrouded. I have read in this forum that the amount of mixture to enter the scavenging column does not exceed what is stored in the transfer passages. It is my interpretation that that also means there is no direct communication needed to the crankcase at BDC.

The reason I ask this question is because due to the piston cutaways when the piston is at TDC, there is a direct link from the intake port to the transfer windows in the cylinder. (on all engines I work with)
Would this not cause a turbulent reverse flow backwards down the transfer duct before the descending piston opens the transfer windows?

[andreas]

If the amount of gas in the tranfer ducts is all that enters the cylinder probably depends on how much volum they hold. Regardless, they need to be backed up by the crankcase, otherwise the vacuum in the ducts would be too great, and the delivery would vastly suffer.

[wobbly]

The statement about transfer duct flow into the cylinder was in response to the assumption that at BDC , when the pipe diffuser creates the max cylinder depression , there is communication of
this depression from the cylinder , thru the ducts, into the case - and this pulls mixture thru the reeds.
This cannot happen , as in every race engine I have seen in EngMod , the reeds are only just starting to open after the piston has risen to TPC.
Seeing this, led to the tuning trick of using the intake tuned length to create a + pressure ratio on one side of the petals , at the same time as the case goes negative.
An assumption is just an error waiting to be revealed.

Then looking at purely the mechanical aspects of the duct flow.
Taking the volume as that being from the port faces , down the ducts centerline , around the long turn floor radius in the case , and up to the piston cutaway/bore face ,
there is way more volume in these combined ducts than can ever be " transferred " into the cylinder - even with well over 100% VE.
Simple mechanical facts.
And yes , as the flow exits into the cylinder , there is not a vacuum " left behind " in the case , any pressure differential is instantly balanced - flow out = flow in , less any inefficiencies due to duct wall parasitics etc

Lastly , re TDC flows in or out of the ducts , yes in a cylinder reed when the inlet port is open , so are the transfer ports via the piston cutaway area.
But in this scenario , sure the transfers ports are exposed , but then so are the duct entries at the other end exposed to the same case volume - thus there is the same pressure delta at either end and no flow in or out can occur.

[andreas]

Question:
Why is a two or three stage converging baffle better, and how to fabricate it, like a parabel?

[wobbly]

The first parabolic converging cones I am aware of were developed for clutch pipes in KT100 Yamaha's.
These were spun cones, but had huge 38mm exit diameters , into an integral can muffler , so it was a bit of a weird design tangent that worked way better than a straight conic section.

The physics behind the idea is that the return wave amplitude is being boosted by the increasing cone angles , as the piston is rising in synch closing the port area down.
Rear cone angles depend upon the belly diameter/volume and the end use , dictating how much overev vs peak power / powerband width is needed for the application.
The bigger the belly , means more of the finite wave energy has ben consumed in the diffuser , thus steeper converging tapers are needed to generate the needed plugging efficiency.

Maybe a good place to start for a new application is to look at the R1 pipe rear cone I did 4 years ago , that has now been reused for the next 3 years in the R2 design , as I simply could not beat it in a sim or on the dyno.
This was iterated down over literally hundreds of sims and confirmed with 3 physical pipe variations on the dyno.

I would draw a straight cone and construct a table with length % and angle increase/decrease % per section.
The most critical section is the last, the steepest.
Its length and internal angle has an almost linear relationship to increasing peak power Vs overev bandwidth.
Too long and or too steep , and peak power increases dramatically - but then after peak it drops almost vertically , probably perfect for a CVT application.

[andreas]

Thanks Wobbly, that is seriously steep, and only 25/61 % header/ end of diffuser.
But how is to end of diffuser 61%, when belly and baffle are 9 and 25 % respectively?
With a 61mm ex-duct, I find to end of diffuser is 65.3%, and header 29.6%.

[wobbly]

Yes the header and diffuser % are abnormally short as this is a maxed out 125 with a straight line ignition that requires a very short pipe.
32% and 65 - 68% would be considered perfectly normal for virtually all configurations , except this particular case.
The belly and thus rear cone length % are dictated entirely by the end use requirement for peak power Vs overev , and vary enormously.
It peaks at 13800 with useable overev to 14800 , thus needing very carefully designed rear cones , and the very steep last diffuser , moved to the left , pumps up the front side from 11000 to 13000.
Another constraint is that going over 50Hp makes tuning very difficult for all but experts , and can end up needing excessive fuel to stop deto - not make Hp.

Useable range is quite narrow , starting at 10,000 WOT , thus a short header is needed to shift the whole powerband up and rightward.
Having a very good overev bandwidth means the possibility of deleting gearchanges on short straights or simply adding rear teeth to gain acceleration from the ratio , and then maintaining terminal speed with rpm capability.
Here is the result of the new engine , showing the suppressed peak used to generate a 1000 wide band now sitting right on 50Hp.

You are right about the % errors , I have no idea how I saved that incorrect layout from EngMod , here is the actual pipe numbers.

[andreas]

Very interesting.

[dutchpower]

Very interesting.

Changes in your pipe Andrea gif more
Can not compair whit std. pipe have no dimention of it

[andreas]

Dutchpower, this is a pipe you've tested?
I'm aiming for a tuned length of 820, starting at 35mm with a shallower header than the original drawing.
I have partial measures of a HGS pipe and probably a standard somewhere, are you asking to see those numbers?

[dutchpower]

Dutchpower, this is a pipe you've tested?
I'm aiming for a tuned length of 820, starting at 35mm with a shallower header than the original drawing.
I have partial measures of a HGS pipe and probably a standard somewhere, are you asking to see those numbers?

Do you have std. pipe drawing

[andreas]

Yes, I can measure it, Will take some time, because it is not at my place.

[lodgernz]

Wobbly, I'm about to build another pipe in my endless quest for more power from my Honda buckets, and I hope you can answer a question that has been puzzling me for a while.

Looking at your ultimate RS125 pipe that you published here a while ago, the length of each diffuser section, as a percentage of the total diffuser length is:
H1: 32%
H2: 52%
H3: 16%

I wondered how you came to these ratios. Was it purely from EngMod, or dyno trial and error, or just some Wobbly magic?

Imagine you were building a pipe for a bucket that had reasonable transfers, peak power at 13000 RPM and needed grunt out of corners but a decent pull to max RPM of, say 14,000, on a bike geared to reach that RPM at braking point on the fastest part of the (kart) track, which is usually less than 200 metres from the previous braking point.

How would you alter those ratios?
Would you still keep the steepest diffuser angle at H2?

[wobbly]

Lodger, I cant remember what the RS125 pipe was to be honest as I havnt worked on one for years now.
Repost it here so it jogs my memory.
But no from what I have learned about pipes needing good power range with no PV I have been making the last diffuser much longer and steeper as this pumps up the front side.
As you can see in the TM design.

The length of the header , combined with the angle of a short first diffuser sets the overev potential , shorter and steeper = more overev past peak.
This short first diffuser really only bypasses the possibility of creating a shock wave situation ( that destroys energy ) by having a way steep section directly off the header.

All of this guidance is iterated out in EngMod , then dyno tested , and of course I would say there is magic involved as well.
And then if you pinch down the cylinder exit to hit 0.8Mach , this mostly increases peak and overev potential , so one use of this is then to make the pipe longer to gain some more low end as well.

[lodgernz]

Here is the hell RS125 pipe design that has basically everything I know embedded in its concept, the 800mm length is for a 200* Ex only with a
proper ignition curve and powerjet switching..
The first 25 mm is an oval transition flange ( 41 by 32 ) in the duct, out to a 41 header.
The stinger nozzle is 23.2 with 25 stinger pipe.
Dual stage header, and steep mid diffuser create the deepest and widest depression around BDC at 13,000, with tons of overev power.
The 120mm Honda and Mota pipes wont even get close, especially with the final diffuser being the steepest,it creates the Ex port depression way too late in the cycle
to help at the natural peak of a 200* Ex around 13,000 and into the all important overev past 14,000.
Several details can be jigged to work better than this design, but thats for you guys with a good code to work with to figure out from looking closely at the pressure ratio traces.

This one Wobbly.