ESE's works engine tuner

[adegnes]

I came to think of this post when I found out what a left side roller crankshaft bearing for a GasGas engine costs... those are wayyyy expensive in comparisson with the ball bearing used on the right side.
The intake situation on the std. engine is the same as the one Frits describes above, but as I´m modifying it for road racing, it will get a straight manifold as well as simetrical design left and right Vforce reedcages from an Honda ATV. (usually reedcages for twisted intakes have an asimetric design).

Asking for youguys opinion if this change in setup would open for the possibility of using a ball bearing instead of roller bearing.
The extra cost of the roller covers the expenses of an extra bearing change with some change leftover for a pint... and I like pints...
Just asking for your opinion.
In advance thanks.

I would keep the roller, If you ditch it you lose the floating crankshaft action and probably some power. That pint sounds tempting tho, and could be worth the power loss

[Frits Overmars]

according to a website, you need 33.31 cfm to make 48hp...now I know there are multiple variables that you need to take into account before assuming such a number, but doesn't 33 cfm seem rather low? I mean a 38mm carb can flow a shit tone more air then that (obviously the test pressure would need to figured). So what am I missing??? Why don't we run smaller carbs and smaller intake ports and ramp up velocity?

Because the intake doesn't flow all the time, and some of the time is does flow, it does it in the wrong direction. Sorry, no time for a real answer (and someone will probably soon chime in with a better explanation than I can provide anyway...)

Perfect answer Alex.

Shnaggs, why would you want to ramp up the velocity? If that were a good thing, you could reduce all cross flow areas to 1 mm² or even smaller. But somehow I doubt that it would help performance. I prefer to ramp up mass flow.
Anyway, increasing flow velocity by means of reducing the cross flow area is largely a fairy tale. Velocity comes from a pressure differential over a duct.
Lower the downstream pressure far enough and the flow velocity will reach Mach 1. Crank up the upstream pressure far enough and the flow velocity will also reach Mach 1 with the added benefit of a higher density in the flowing medium, i.e. more mass flow for the same Mach 1 flow velocity.

The fairy tale stems from the fact that with a large cross flow area there will be so much mass flow that the upstream pressure drops markedly even before the gas column in the duct is accelerated to the velocity corresponding to the initial pressure differential. So the velocity will never reach the value that it would reach with a smaller cross flow area. But who cares; we've got more mass flow.

Using figures from eight engines from Yamaha, Honda and KTM gave an average carb size constant of 33, ie. Throat diameter = √(hpx33). The individual numbers varied from 29.5 to 40, but even after eliminating these two the average remained at around 33, so I think this would be a safe starting point.

Johnno, this method will indeed give a safe starting point but you should keep in mind that averaging data from various engines will usually result in an outcome that deviates from the best engine's data.
Eliminating the extremes can also be useful, but I'd rather eliminate the engine with the lowest BMEP. And if you keep doing that, you'll end up with just one engine: the best one.

Asking for youguys opinion if this change in setup would open for the possibility of using a ball bearing instead of roller bearing.

I would keep the roller, If you ditch it you lose the floating crankshaft action and probably some power.

Keep them answers coming Alex .

[oldjohnno]

Johnno, this method will indeed give a safe starting point but you should keep in mind that averaging data from various engines will usually result in an outcome that deviates from the best engine's data.

I like that answer. A less diplomatic person would have said that using data averaged from various engines will usually result in a perfectly average engine.

[wobbly]

Anywhere you have asymmetric flow thru a reed entry you loose power - case or cylinder mounted.
Boyesen did some really nice blocks that had guide blades to turn the flow, but sadly this idea was offset by the crap two stage petals so hard to tell how good the flow idea was.

You would guess that having carbs angled down ( as most are ) would bias flow to the upper reeds as we are trying to do with soft petals on the top.
But no - put on a straight manifold and carbs that will work downdraft, and power goes up.
The reverse used to be true on ICC kart engines,they had the reed angled up under the cylinder, but with a bent manifold making the carb horizontal.
The first time I reversed the manifold and made the carb " updraft " ie normal to the intake flow direction it made a couple of Hp everywhere - a legal mod at the time,now everyone does it.

On an early version of the CPI Banshee cylinders the only way I could figure out in CAD how to fit in CR250 reed blocks onto the back of the cylinder without moving the studs
was to offset the blocks 8mm outward.
This gave severely asymmetric flow into the case and and when Calvin did an early retest with smaller reeds back on bore center, it gave +5 Hp.

[casal-fan]

Keep them answers coming Alex .

I would keep the roller, If you ditch it you lose the floating crankshaft action and probably some power. That pint sounds tempting tho, and could be worth the power loss

Roller it is then, no pints next week... damm
Thanks for the help, much apreciated!

[oldjohnno]

Anywhere you have asymmetric flow thru a reed entry you loose power - case or cylinder mounted...

Let's say you have a reed cage with an opening area that's around twice as wide as it is high, and enough frame clearance to align the carb properly. Would it be an advantage to use two smaller carbs side-by-side rather than trying to blend a single, bigger throat into the wide window?

[F5 Dave]

What about a 'third' carb at an oblique angle just to cover all your bases?

[peewee]

Roller it is then, no pints next week... damm
Thanks for the help, much apreciated!

theres a few places on ebay that have skf explorer rollers for a decent price. that's usually where I get mine http://www.ebay.com/usr/247industrie...53.m2749.l2754

[husaberg]

I would keep the roller, If you ditch it you lose the floating crankshaft action and probably some power. That pint sounds tempting tho, and could be worth the power loss

Congrads on the new minion.

[oldjohnno]

What about a 'third' carb at an oblique angle just to cover all your bases?

No, it has to be four. For symmetry.

Seriously though I think two would have advantages, especially with methanol. Each needle/seat would have to handle only half the volume of fuel and it'd be delivered through two main and needle jets. With alky it'd be easier to manage than a single carb I think.

[TZ350]

... especially with methanol ...

I just love 2t's and methanol. You can run them so rich raw methanol dribbles out the pipe's and the suckers still run.

[F5 Dave]

No, it has to be four. For symmetry.

Seriously though I think two would have advantages, especially with methanol. Each needle/seat would have to handle only half the volume of fuel and it'd be delivered through two main and needle jets. With alky it'd be easier to manage than a single carb I think.

My symmetry has three planes.

[adegnes]

Congrads on the new minion.

Thanks! He'll prove useful when time comes to overthrow the government!

[TZ350]

Ecotrons has sent me an updated 2T configuration file with both the VE and Alpha-N tables active. Ecotrons have also popped a new pressure regulator, some spare injectors and 2 bar MAP sensors in the mail, should be here in a week or so. Just enough time to get my EFI engine back together and in the bike.

Measuring the blowdown pressure pulse rather than the negative (since they should be proportional) gives you a head start on getting the fuel in. I have the pressure sample window set to 140-170° ATDC. The peak moves with RPM, because of the delay along the hose to the MAP sensor, so best to keep it as short and close to the port opening as possible. I have the injection start 10° after the sample window, so the latest pressure value is used to calculate the injector pulsewidth.

I am unable to choose where in the crank rotation I take the pressure sample. But Ecotrons allows me to choose several TPS/RPM break points for defining where the VE table swaps to the Alpha-N table. It does not have to be a straight line. It could be steps or even a rudimentary curve. so I will be able to target the problem areas in my series of dyno graphs.



The dyno graphs, all Alpha-N map, start at 20% TPS (green line) and go up in 10% steps 20 - 30 - 40 etc. Not much difference in power curves between 80 - 90 - 100% TPS.

All the EFI issues seem to be below 20 - 25% maybe that is where the pipe wave collapses.



I can use EngMod's predicted series of delivery ratios as a guide for the VE table.



This is my exhaust pressure takeoff for the MAP sensor. It is a one way valve with a small ceramic ball with only 0.010" movement, hopefully it wont break up being hammer'd around at 200Hz. The blue hose will have positive pressure in it and a small bleed hole melted into it with a hot pin for a pressure bleed down. Hopefully this will give the 2 bar MAP sensor something worthwhile to work with, that is hopefully a smooth positive pressure that reliably reflects changes in the mass air flow through the motor.

Nath88, thanks for the ideas, hopefully I can make them work for me.

[emess]

In the pic of the KZ10B intake the tuned length isnt quite as i drew the line from reed tip to bell end.
The length as should be used for sim accuracy is a point 2/3 of the distance down the block toward the tips ( the effective length of the reed port
opening varies as the petals are closed or fully open ) and the carb bell needs end correction for the actual reflection point off atmosphere.
In a PhD thesis paper i supervised we instrumented a YZ125 fully on a dyno at Uni for using fibre optics in the combustion chamber to control
fuel injection via the colour spectrum of the flame.
One thing discovered was the actual effective length of the intake, and it varied with reed tip opening, plus it showed that the end correction was
accurate when set at approx the intake curtain area length past the full bell diameter.
Without a bell, the end correction reflection point was the classically accepted 1/2 D of the intake pipe.

Wobbly - interesting, but it leaves me asking for more - how do you define or measure "curtain area length" ?

thanks