ESE's works engine tuner

[husaberg]

I don't think VSK stems from somebody's initials; VSK has always the pre-designation of Rotax two-stroke expansion pipe drawings. This was already the case when I first got involved with Rotax engines, in 1978. I don't remember the VSK-number of the 1978 works pipe (I've got it at home in Holland somewhere) but I do remember the maximum diameter: 100 mm. Being a young smart-ass, I did some calculations of my own and according to those, that diameter should be 128 mm. Ridiculous of course, but I built it anyway. It was a missile. It took Honda all the way until 1992 before they brought out their first fat pipe (on the RS125 A-kit). And Aprilia did it in 1995 when Jan Thiel went there and carried my Rumi pipe.

Frits without being to prying is the total pipe volume then more important than the actual baffle cone and diffuser angles appear to be.....

[Frits Overmars]

I would say the total pipe volume is not that important - not in itself anyway. What's important is the way exhaust gas energy is converted into suction by the diffuser, while still saving some of that energy for the positive return pulse.
It's about finding a balance. It's no use sucking a litre of fresh mixture through the cylinder into the header if only a quarter of that mixture gets pushed back.
And it's no use either if the positive return pulse not only shoves all the washed-throught mixture but also a large amount of hot spent gases back into the cylinder.

[husaberg]

I would say the total pipe volume is not that important - not in itself anyway. What's important is the way exhaust gas energy is converted into suction by the difussor, while still saving some of that energy for the positive return pulse.

Thanks I think I understand, but what I was trying clumsily to push at was that the angles are a representation of arriving at volume within the tuned length.
but yes on further thoughts the angles are I guess just as important.

I seen something just on a net troll .
I haven't had the chance to read it yet mind you.
looks to be based on Blair data........at the first glimpse
http://www.dragonfly75.com/motorbike/ECtheory.html


This bit I did notice though was about putting a parallel section in the baffle cone to broaden the power.
http://www.dragonfly75.com/motorbike/BRW.html

[Muciek]

Thanks

I seen something just on a net troll before,
I haven't had the chance to read it yet mind you.
looks to be based on Blair data........at the first glimpse
http://www.dragonfly75.com/motorbike/ECtheory.html
This bit I did notice though was about putting a parallel section in the baffle cone to broaden the power.
http://www.dragonfly75.com/motorbike/BRW.html

I have posted link to this in 2013
http://www.kiwibiker.co.nz/forums/sh...post1130623219

He did updated this exceel sheet few times, but dunno if this works , I tried to put there a pipe which I built for engine and I know how it works and outcome in graphs was way different....

[husaberg]

I have posted link to this in 2013
http://www.kiwibiker.co.nz/forums/sh...post1130623219

He did updated this exceel sheet few times, but dunno if this works , I tried to put there a pipe which I built for engine and I know how it works and outcome in graphs was way different....

Sorry I missed it........

[wobbly]

Always amazes me when people make huge jumps of blind faith when trying to explain some fantastic " new " idea they have concocted.
The 50mm piece of tube added to the tailcone will of course smear out the timed return wave, but what was completely ignored was the tuned length is now 50mm longer.
I am sure I could make an even better power curve with a "proper " pipe design that was 50mm longer - if the engines powerband shape actually needed that in the first place.

The volume of the pipe is not something that can be used as a number to shoot for, what is important as Frits said, is the end use use of the finite energy imparted down the header from the
Ex port.
The fatter the belly, and thus the steeper the diffuser, the greater energy that is used to create a depression at the port around BDC.
Thus less energy can be used to create the stuffing pulse.
This can be countered somewhat by a steeper rear cone, and this is what does happen in reality, where I have seen huge 145mm belly 250cc pipes with 30* rear cones.
This approach means the rear cone length is reduced, but the steep angle gives at least some amplitude to the event.

I was sceptical when I was told about " Mr " VSK who worked at Rotax, as I spent several years when in England working at JL pipes trying to get to talk to this mythical pipe
designer - no one knew who actually did do the designs, so I never got to meet him.
But whoever it was, the two stage header was a feature of those later pipes, and must have been dyno tested in house, for them to have been published.
Every pipe I have run thru EngMod makes better power with that design so to me its a no brainer nowdays.

[husaberg]

Always amazes me when people make huge jumps of blind faith when trying to explain some fantastic " new " idea they have concocted.
The 50mm piece of tube added to the tailcone will of course smear out the timed return wave, but what was completely ignored was the tuned length is now 50mm longer.
I am sure I could make an even better power curve with a "proper " pipe design that was 50mm longer - if the engines powerband shape actually needed that in the first place.

The volume of the pipe is not something that can be used as a number to shoot for, what is important as Frits said, is the end use use of the finite energy imparted down the header from the
Ex port.
The fatter the belly, and thus the steeper the diffuser, the greater energy that is used to create a depression at the port around BDC.
Thus less energy can be used to create the stuffing pulse.
This can be countered somewhat by a steeper rear cone, and this is what does happen in reality, where I have seen huge 145mm belly 250cc pipes with 30* rear cones.
This approach means the rear cone length is reduced, but the steep angle gives at least some amplitude to the event.

I was sceptical when I was told about " Mr " VSK who worked at Rotax, as I spent several years when in England working at JL pipes trying to get to talk to this mythical pipe
designer - no one knew who actually did do the designs, so I never got to meet him.
But whoever it was, the two stage header was a feature of those later pipes, and must have been dyno tested in house, for them to have been published.
Every pipe I have run thru EngMod makes better power with that design so to me its a no brainer nowdays.

Is it possible to indulge my idle curiosity and run it with a sim of a pipe/engine combo with the tuned length altered to see what it actually does? assuming it fits within the parameters of the sim.

I had a look at his pipe calculater and it appears to be similar in factors to a program I seen years ago based on Blair, think it was on a Kart site

[wobbly]

Yep already done it.
Here is a WR200 sand drag engine.
Baseline curve,50mm tube added to the rear cone, then 25 added to the diffuser and 25 added to the tailcone.
The long pipe is nowhere near optimised ( it detonated in the midrange ) but as you can see a "proper " pipe works hugely better than simply adding
a lump into the tailcone - dumb.

[husaberg]

Yep already done it.
Here is a WR200 sand drag engine.
Baseline curve,50mm tube added to the rear cone, then 25 added to the diffuser and 25 added to the tailcone.
The long pipe is nowhere near optimised ( it detonated in the midrange ) but as you can see a "proper " pipe works hugely better than simply adding
a lump into the tailcone - dumb.

Thanks Wob, I will test your patience further, what happens if the tuned length (including adding the tailpipe lump was kept to the same value) as the optimised version.
It looks odd that the curve shape remains similar than the just tuned length alteration would suggest.
I know it will likely be poos I just want to see what it on its own does.
might I suggest blue black and yellow cause I am colour blind as well a bit thick.

[wobbly]

Na, the original info tried to indicate that the tube added to the rear cone was awesome, then ignored the tuned length change.
But if I add 50mm to the rear cone, then the header and diffuser will be way too short to be correctly proportioned, if I reduce them by an equal amount - so then we are not
dealing with the "same" pipe at all.
I could see on the screen what was happening near peak power - the front 2/3 of the pipe acted like normal, except the return pulse was much lower, follower by
a long slow rolloff of the pulse after the Ex port was closed.
So the thing sort of acted like a normal design, just with a shit rear cone that was working way too late in the cycle to be effective.
I tried parallel spacers in the diffuser and the rear cone many years ago when I spent 3 months full time on the dyno to design a totally new KT100 direct drive pipe.
They didnt work in that application, but what I ended up with was seriously weird having the steepest diffuser right after the header.
The so called Wobbly 3 Cone pipe I built over 2000 of them all by hand, and the things are still searched out by people wanting to win KT races.
So here is the baseline Vs the long rear with 50mm removed from the diffuser - even dumber.

[Larry Wiechman]

I tried parallel spacers in the diffuser and the rear cone many years ago when I spent 3 months full time on the dyno to design a totally new KT100 direct drive pipe.
They didnt work in that application, but what I ended up with was seriously weird having the steepest diffuser right after the header.
The so called Wobbly 3 Cone pipe I built over 2000 of them all by hand, and the things are still searched out by people wanting to win KT races.

I've heard of this, but never seen one. Can you teach a little more about this concept? Direct drive would seem to be a hellish challenge to a pipe design.

[wobbly]

The KT100 in direct drive application has to pull from 8,000 to 16,000 - a tough call, and against a huge number of pipe designs that were
mostly derived from 100cc european kart engines.
The KT100 Ex port exit is way too big, so I made the spigot a reducing taper, this then exited into a normal widening tapered header.
One popular variant for high rev motors started with a bigger diameter at the flange face - creating a reversion step as well as a smaller nozzle.
The length spacers are between the pipe entry and the header exit.
I tried spacers up at the flange but they could never match the power of the reducing, expanding concept.
The pipe has 3 diffuser cones, the steepest at the front - then 2 rear cones inside the main body with varying bleed hole centres along its length.
The bleed hole centres and the body volume can be used to tune out dips in the torque curve.
I found early on that the KT is thermally limited to around 17Hp in direct drive configuration,with any more power on track it would overheat after 2 laps and instantly deto at peak torque,
but responded every time to removing peak and pumping up both ends of the powerband.
No other design has come even close to the powerband width of this setup,with two header nozzles available for the engines that were setup to rev to 16,000 + Vs those setup with more mid
power and only reving out to 14,000+.
They won hundreds of NZ titles, and no I wont be building any more.

[Larry Wiechman]

The thermal limitations are a very real problem. I could never understand the logic of the rear facing exhaust port. The carb side is cooled by air, the exhaust side is cooled by accident.
I've considered making a stainless header, (lower conductivity) that extended into the exhaust duct, with the objective of reducing the duct diameter and insulating the casting from direct exposure to hot gases. The downside would be what the hot tube would do to the returning, fresh charge.
Do you find it easier to design a pipe for a kart equipped with an oil clutch? I think a high RPM, slipper clutch would be an advantage on a bucket.

[wobbly]

I made a spigot/flange with a CNC tongue that pushed into the port bottom to fill the silly S bend in the floor.
But the stewards deemed it illegal.
A cluch pipe is a doddle to design in comparison, needs a big flat torque curve between 10,000 and 13,000 from where the clutch locks in ,to the relatively low peak rpm point.
Also tried an aluminium spigot with an oval to round transition, but it didnt make any more power and wore out quicksmart.

[richban]

Nope, this in effect means the trigger is too close to TDC, but if you need more advance to get the timing lines spot on, then add some timing in the channel
trim boxes down the bottom of the ignition curve page.
You can use these to check each cylinders timing when under load on the dyno, means you can correct for crank wind up and also a crank that hasnt been assembled in phase correctly.
Doing this means you know for sure both cylinders are actually firing when the curve says it should.

Forgot the timing light to do this dame it. Now with squish back in the happy place and a more accurate 15 degrees of timing at peak she has gained more power. But the big breakthrough was opening the power valves sooner. That gave a massive 8 more toques. That should give a marked improvement in corner exit. ( If i am brave enough to pull the trigger early ) Thanks again to all that have offered time and advice as this bike has come together. Time to polish it ready for a summer of racing.