ESE's works engine tuner

[Flettner]

Thankyou Frits.

[wobbly]

I have found many times that a 2T racebike on a DynoJet has near on 12.5% losses to the crank power as seen in Engmod if you input realistic data , most importantly Combustion Efficiency , and Exhaust Duct Temp.
So that takes into account tire friction , chain loss ( probably the same as a sprocket power test ) gearbox gear ratio pair and primary pair losses.
This also takes account of the DynoJet fudge factor , built into the software since day one, as they simply didnt believe how bad a Harley really was ,and still is in reality.

[husaberg]

This also takes account of the DynoJet fudge factor , built into the software since day one, as they simply didnt believe how bad a Harley really was ,and still is in reality.

I seem to remember something like a either a little more less like 30 HP (or was it 30kw?) for a 883 for the mid 90's? potato potato.....
i have never ridden a std one but i once rode a highly tuned 1200 one that had had an additional 12g spent on it that was nearly as fast (in the straight line anyway) as a std 1980's RZ350.


In context in the fifties a decent 650 brit twin could knock out 40HP plus, a decent 500 Goldie single nearly also could do the same.
They also steered a shit load better and weighed 100 pound less.

[SwePatrick]

After using the FOS 2T power correction algorithm in my latest dyno tests , that gives much more realistic power numbers ( for example high mBar ambient air - I had an unbelievable 1040 the other day )
I have discovered that JIS D1001 correction gives almost identical results.
This might be useful on dyno software where the FOS code cant be inputted.
Ive got no idea what the correction data is for this , but it does seem to be much closer than any of the others commonly available.

I tried to search but couldn´t find any info.
What are the advantages with FOS correction?

Rgds

[andreas]

Patrick, it was discussed a few pages back.

[wobbly]

The usual SAE and J corrections were specifically designed for 4T use.
The 2T responds very differently to air pressure and ambient temp - more especially toward the extremes.
They are nowhere near accurate at 10*C or 35*C , nor at 900 mBar or 1040 mBar pressure and seem to give abnormally low Hp at the reference conditions as well.

This makes a joke of me trying to gain 1/2 Hp when the correction for the day is already 2Hp wrong.
Means I have to do a " stock " or baseline test every day to get anything like an accurate comparison - and this is often just not possible as the baseline run I did yesterday with a particular
cylinder configuration, I cant replicate as Ive just spent a day grinding the ports.

[diesel pig]

I have recently became very interested in Frits's 24/7 Inlet system and I am now kicking myself for not saving all Frits and others posts on the subject. The problem is my Google Fu is so weak I am doing no good finding the posts now. If only a great Google'er like Master Husaberg could do it, who could probably do it in no time. Oh well ever mind.

[F5 Dave]

I seem to remember something like a either a little more less like 30 HP (or was it 30kw?) for a 883 for the mid 90's? potato potato.....
i have never ridden a std one but i once rode a highly tuned 1200 one that had had an additional 12g spent on it that was nearly as fast (in the straight line anyway) as a std 1980's RZ350.


In context in the fifties a decent 650 brit twin could knock out 40HP plus, a decent 500 Goldie single nearly also could do the same.
They also steered a shit load better and weighed 100 pound less.

The 1340 evos pushed out a fairly impressive 53hp I seem to recall, which at the time was the same as my RZ350 on Sayles dynojet with a head skim and roadbike port clean. There may have been a weight disadvantage but we never measured which bike was heavier .

That market dried up for him as he unhelpfully increased power a little by removing screaming budgie go faster parts and jetting. Then Wellington motorcycles bought a new type of dyno that produced the correct type of Hardley horsepower numbers well in excess of his DJ.

[SwePatrick]

The usual SAE and J corrections were specifically designed for 4T use.
The 2T responds very differently to air pressure and ambient temp - more especially toward the extremes.
They are nowhere near accurate at 10*C or 35*C , nor at 900 mBar or 1040 mBar pressure and seem to give abnormally low Hp at the reference conditions as well.

This makes a joke of me trying to gain 1/2 Hp when the correction for the day is already 2Hp wrong.
Means I have to do a " stock " or baseline test every day to get anything like an accurate comparison - and this is often just not possible as the baseline run I did yesterday with a particular
cylinder configuration, I cant replicate as Ive just spent a day grinding the ports.

Thanks for the answer(i DID read the earlier discussion)
But i was asking kind of sloppy, i wanted to know more why the FOS correction corrects more than any other correction.

Rgds.

[Frits Overmars]

i wanted to know more why the FOS correction corrects more than any other correction.

Attachment 351453

Did you read the two SAE papers I mentioned in the PDF, Patrick?

[husaberg]

I have recently became very interested in Frits's 24/7 Inlet system and I am now kicking myself for not saving all Frits and others posts on the subject. The problem is my Google Fu is so weak I am doing no good finding the posts now. If only a great Google'er like Master Husaberg could do it, who could probably do it in no time. Oh well ever mind.

....................

That's Herr Helmholtz messing up things for you, Wob. https://en.wikipedia.org/wiki/Helmholtz_resonance
The cycle from a positive crankcase pressure differential accelerating the transfer flow to a negative case differential slowing that flow velocity back to zero again takes a fixed amount of time. At low revs this time is less than the total transfer opening time, so the cycle continues, performing more than half a sine wave during the open transfer period.
It's the same phenomenon you can observe with piston-controlled inlet systems, where at low revs blowback can occur although the piston is still on its way up towards TDC, so the crankcase pump is still trying to inhale.

In theory the reed valve inlet system ought to prevent the crankcase pressure from dropping below atmospheric. In practice the reed valve needs this pressure drop before it will even think about going to work.
Neils gibs-controlled rotary inlet system offers the best of both worlds. Although.... you may remember Roland Holzners 24/7 experiment at Modena a couple of years ago. Roland realized it without any electronics because the CIK would have vetoed that. His system worked, but with the karting world being the karting world, the CIK vetoed it anyway, like it had done with Rolands synthetic inlet inserts some time before. If the CIK doesn't like you, they'll let you know...

Not even a free breakfast, I'm afraid. The problem is not the symmetry of the port timing; it's the duration of the timing. Asymmetry might have worked in the long-gone era of tiny crankcase volumes but these days Helmholtz will hardly be impressed by what the piston happens to be doing.

The video tells it all: blowback from the 24/7 carburetor whenever its throttle is opened, wich is only to be expected considering the Helmholtz resonance frequency there: conventional crankcase volume, big carb diameter, short inlet tract, low revs and infinite inlet timing.
Revving the engine higher before opening the 24/7 carb, like Alex is proposing, will help, but the best (and simplest) solution is to increase the inlet tract length.

I love watching "Two Stroke Stuffing" on YouTube and seeing how Alex Degnes is enjoying his hobby and at the same time is doing a wonderful job at promoting the two-stroke engine. Of course I also enjoy the way he makes good use of information, like the transfer layout in his present cylinder which mirrors my scavenging concept, his plan to incorporate my 24/7 inlet timing, and now his new cylinder with symmetrical scavenging.The latter will give him a lot more blowdown angle.area than his unbridged exhaust port experiment, and a lot more transfer angle.area to go with the enhanced blowdown.
I can't wait to see his next videos. I don't mind admitting that from time to time it's not only a trip down memory lane for me, but also a shameless ego trip .
Attachment 348909Attachment 348908Attachment 348906 Attachment 348907 Attachment 348910

I fully agree, Adreas; I'm just writing it down step by step.
Duct volume V = cross flow area A * duct length L
The mass m in our duct is proportional to the duct volume V
m ≈ V
m = A * L
The pressure differential p over this duct exerts a force F= p * A upon mass m
acceleration a is force / mass ;
a = F / m
a = (p * A) / (A * L)
a = p / L
In English: the acceleration of a mass in a duct is proportional to (pressure / length) . If you double the length, the rate of acceleration will be halved, and vice versa.


While we are at it, let us take a look at the amount of mass that is moved through the duct in a given amount of time.
First, let's make some assumptions in order to keep things simple (my hobbyhorse, as y'all may have noticed).
1: The pressure differential over our duct is constant (in reality it would drop because of mass moving from the high-pressure side to the low-pressure side of the duct).
2: The duct ends in a port which is fully opened during t seconds, after which it is suddenly closed.
The length of the gas column that passes through the port, is ½ * (acceleration a) * (time t)²
Now if we compare a duct with length L to a duct with a length 2L the acceleration in duct L will be twice as high as the acceleration in duct 2L, as we have seen above.
This means that with the same pressure differential and the same amount of open port time, the long duct will fill the cylinder only half as well as the short duct. It's a great reason for using a short inlet duct which allows us to use a short opening time.

The 24/7-inlet idea arose in 1974 when the rotary inlet disk of my racer stuck open and it just kept running, as long as I kept the engine in the power band. It meant that I rould ride back to the pits instead of pushing through the grass alongside the track side (which I had already done often enough...).
This experience proved what I already knew in theory: that the pumping effect of the crankcase only serves to start the engine; as soon as it runs in the power band, the exhaust pipe takes care of all the gas movement.
So how do you build a rotary disk that stops in the open position when the engine revs reach the power band?
I had no idea...
But then why would you need a rotary disk anyway? Because it flows better than a reed valve? That doesn't matter when you only need it to start the engine. And a reed valve is much simpler to open...

Below you see an old prototype with a typical design error: it is too complicated. You do not need two gear-coupled reeds; one reed that swings out of the way would be enough. And it can be operated by the same servomotor that operates the exhaust power valve Very Happy .

Maybe not, indeed. My bike was rotary but that didn't stop it from running the wrong way around. Carburation was way off, of course, with a disk that opened 65° before TDC and closed 40° before BDC, but it ran anyway.
Heck, it even ran when the disk hexagon was smashed and the disk stuck open permanently. It was the birth of the 24/7 idea.

There are practical limits at both sides of the inlet timing. You can open the inlet as soon as the pressure in the crankcase has dropped to the pressure level upstream of the inlet disc.
Open it any earlier and you will lose crankcase pressure to the outside world; open it any later and the crankcase pressure will drop more than it needs to, slowing down the transfer flow.
The Helmholtz resonance of the blowdown+scavenging cycle determines how many milliseconds after exhaust opening the pressures upstream and downstream of the inlet disc are equal.
This fixed amount of time means that at low revs you could open the inlet even before BDC. But this is impractical for two reasons (unless your name is Neil Hintz).
A: The inlet needs to open later as the revs increase.
B: Opening the inlet exactly when the pressures upstream and downstream of the inlet disc are equal means that there wil be a zero-pressure opening signal sent to the carburetor, and carburetors don't respond well to something they don't notice.

Summary: opening the inlet too early or too late will both cost power, but opening it too early will cause more problems than opening it too late.

When should we close the inlet? When the inlet flow is slowed down to zero by the ever-rising crankcase pressure. Close it any earlier and you will miss out on the last bit of crankcase filling; close it too late and you will lose mixture flowing back from the crankcase towards the outside world, passing the carburetor on its way out and taking some more fuel with it. This same amount of now overly-rich mixture will be sucked in again at the next inlet cycle, passing through the carburetor for the third time, taking some fuel with it for the third time and trying to drown the engine.

Summary: don't close the inlet too late. For a well-developed racing engine 85° after TDC will be about the limit. If you are really power-hungry, you can try a couple of degrees more, but it may harm your lap times.

If you do find more power past 85°, you may instead consider returning to the 85° closure timing and fitting a bigger carburetor.
By the way, 85° after TDC is also considered the maximum foolproof timing for a piston-ported inlet system.

In case you are interested in the Aprilia RSA125 inlet system: cylinder capacity=124,8 cc; crankcase volume @ TDC=675 cc; inlet timing= 142,5°/88°; carburetor diameter= 42 mm.

So this is it? Not quite yet, because closing the inlet too early can also give strange effects, as I discovered when I put an experimental distance piece between the carburetor and the inlet disc.
This increased the inertia of the inlet column, so this column came up to speed slower, and was subsequently cut off by the closing inlet disc while it was still flowing towards the crankcase at a respectable speed. The sudden closure caused a high collision pressure upstream of the inlet disc and the mixture column bounced back like you wouldn't believe. It filled the entire dyno room with a mixture mist within seconds, and the tiniest spark might have blown the roof off. Ever since this experience I distinguish between blow-back, which is a nuisance, and bounce-back, which is outright dangerous.

The simplest solution for totally asymmetric and variable inlet timing is the common reed valve; it doesn't need any governing. It won't flow as well as a rotary valve,
but below the power band that doesn't matter. And once we're in the power band we can hold it open 24/7 as you say. A simple servo motor can do that; we can even use the exhaust power valve servo if there is one.
No need for any electromagnets. I doubt if those could be fast enough for cycle-to-cycle operation anyway. It's not like lifting an injector needle 0,1 mm off its seat;
a reed valve should move about a hundred times further, but at the same frequency as that injector needle....

I doubt that Muhr. I don't know when Tassinari started producing its Vforce reeds, but I do know that my first 24/7 prototype dates from the previous millennium .
Attachment 336062

Joking apart, I think it's possible that a conventional reed starts behaving like a 24/7 inlet when the engine frequency rises above the reed's natural frequency,
but it will still offer a lot of flow resistance. Better swing the reeds out of the way altogether.

Running it: yes. Starting it: no.
That is why my 24/7 inlet system has a hinged reed that can be shut for starting the engine in the conventional reed-mode, and that can be swung out of the way, once the engine is running in the power band and the exhaust pipe suction is taking care of things.

The advantages of a 24/7 inlet are an unrestricted mixture flow with better symmetry than a sidemounted rotary valve and minus the friction of that rotary valve.
A 24/7 inlet is no use at all below the power band, as you have already found out. But in the power band it can give good cylinder filling, and power in the overrev tends to be better than in the case of a rotary valve, unless you do a Flettner with variable rotary valve timing on the fly, but controlling that will definitely be more complicated than controlling a 24/7 valve. By the way, starting a 24/7 engine without any starting valve at all may be a pain; you've found that out as well.
That is why I designed it with a simple reed that swings out of the way, once the engines runs in the power band.
Experimenting with different inlet tract lengths is the way to go. In theory you could also experiment with different carb diameters and different crankcase volumes,
but changing a length piece is a lot simpler.

Carb-wise there are no special requirements. But I've got some other pointers for you.
Try to accommodate the 24/7-housing in the reed cavity of the crankcase, so you can switch between both systems. Then, build and run the engine with a normal reed, so you can be sure that there are no hidden problems.
The necessary inlet tract length depends on crankcase volume, carb diameter and rpm of maximum power. This rpm is established by the cylinder's angle-areas,
so it is more or less fixed, and also there's only so much you can do to enlarge the crankcase volume, but you are free in your choice of carb diameter.
As a guideline, try SQRT(cylinder capacity x rpm of max.power / F). This factor F should be about 1200 (for conventional inlet systems F should be about 900).
This set-up will leave the inlet tract length as the value to play with. And it may need to be quite long, maybe 350 mm from carb bellmouth to 24/7-flap,
and even longer if you use a big carb, so it will be wise to allow for quickly-exchangeable distance pieces.

Alex, that bolted-together accessory that you used to explain the effects of twisting, is a great idea to make clear what is going on.

PS: If you over-pinched your crankshaft, you can of course open it up again with a chisel and a hammmer, but I prefer to use a 'reversed vice'. That's just an M8 or M10 bolt, shortened so much that it will fit between the crank webs, and a nut. Unscrewing the nut will give you much more control than hitting the chisel with a hammer.

PS2: I suppose it's past sunday over there, but anyway: Happy Birthday ESE!

I am not at liberty to say any more about it than I already did here, Haufen. You may want to search for 'Modena' and 'Roland Holzner'.OK, this is a question I can answer because it was obvious for all the world to see that the Modena 24/7 uses the normal homologated 30mm Dellorto carb.
My fears that injection would be necessary were not confirmed. The Modena runs fine with a carb, notwithstanding the fact that they also have a working injection system on the shelf.

The 24/7 system depends on the pipe even more than a conventional inlet system does, but the pipe itself does not need to be altered.
Crankcase volume, inlet length and carb diameter all play a role; none of those values are critical, but their combination should yield a suitable Helmholtz frequency.
The most practical way to go about this would be to start with a fairly big crankcase volume and a fairly small carburettor, vary the inlet length and see what happens.

The international kart governing body FIA-CIK considers electrickery a thing of the devil; they do not even allow variable ignition or electric power valves and power jets. And I'm convinced that they will find a way to block the 24/7 Modena even though the valves are purely mechanically operated.
Modena and the CIK don't exactly see eye to eye; after Modena introduced its synthetic flow-guiding inlet insert, the other kart engine manufacturers conspired and the CIK announced that the insert would not be accepted again at the next homologation round, although it's just a piece of plastic that every manufacturer could make at next to no cost. In fact it was initially developed by Modena so the complex milling of the transition from the reed cavity to the transfers could be skipped. My suggestion to use a plastic with low thermal conductivity also contributed a bit to engine power.

Didn't I already tell? I thought I did. The direct inlet version evolved into the 24/7 version which produces 10% more power than the regular Modena engine (with the same obligatory 30 mm carb, pipe, compression volume etc.)
BTW Ken, nice analysis.

Great minds think alike, Ken. The picture below shows the onset of an experimental Modena engine, created by my friend, Modena technical director Roland Holzner.
Attachment 312417

If you take this idea further and concentrate on eliminating its drawbacks, as Roland did, you'll get this:

Attachment 312418 Attachment 312419

That's right: 24/7
http://www.kiwibiker.co.nz/forums/sh...post1130867376

Thank god, finally a remark that I can comment on without violating the Ryger non-disclosure agreement.
Yes, you could employ an additional reed. But why would you? The 24/7 valve itself is a reed; how much simpler do you want things to be?

Roland has one huge handicap: the rulebook. The carburettor diameter is restricted to 30 mm. That of course influences all other dimensions in his inlet system.
A second handicap is that none of these kart engine manufacturers has unlimited funds. Roland had to avoid upsetting the beancounters so he worked with what he had.
Besides, even if money would not be an obstacle, Kart engine manufacturers cannot change an engine at will because of homologation issues.
Rolands primary goal is to get his fuel injection approved, which won't be easy because he already rubbed some stone-age officials up the wrong way with the synthetic inlet inserts in the present Modena engines. These were homologated all right, but subsequently the officials let Modena know that these inserts won't be accepted in future engines. Why? All I can think of is envy.

In the long run: yes, I think so. But there are many people here working on rotary inlets and for them it is simpler to carry on than to switch to 24/7.
On the other hand, if you are already working on a reed valve engine, converting it to 24/7 is much simpler than converting a rotary valve engine.
And finally I am in a position that I don't need to push everyone into 24/7; it's already on the move: http://www.kiwibiker.co.nz/forums/sh...post1130808805

That was my initial fear too, Smitty. That was one of the reasons I was so keen on dropping the carburettor and switching to fuel injection. But the Modena tests have shown that a single carburettor can handle the switch from a closed to an open 24/7 inlet without any problems. And Modena have also got a conventional reedvalve kart engine running quite nicely on injection so they know what they should be looking for.

No maybe's about it: more complex and more difficult to tune. Are you sure that deep in your heart you're not a double-overhead camshaft man, Smitty?

Clearance should be about 0,4 mm. And the disk should be able to float until it rests against either side of the disk chamber. That makes setting it up a whole lot easier.
Remember the 12-speed twin-disk Kreidlers? They had both disks pressed onto the crankshaft. Getting them well-sealing yet friction-free was a nightmare.

The main advantage of EFI on a kart is that the motorcycle carbs everyone is using were never designed to handle acceleration in four directions. It's the one factor that gives more trouble on kart engines than all other factors put together.
Riley Will of BRC is already achieving good results with EFI and so is Roland Holzner of Modena engines in Italy. But injection has yet to be homologated on 125 cc karts, which could be a special problem for Modena. Their heat-insulating inlet insert was already more advanced than the ultraconservative FIA-CIK officials would have liked to see and you can imagine how they feel about Modena's experimental 24/7 engine which does not seem to care about their obligatory 30 mm inlet restriction.

As long as that tract is not as short as it could be, I will call it long.
and don't tell you won't be able to get to that top disk cover bolt .

Where else? Imagine that you are in a test cabin and the total air entrance to the cabin is a 24 mm orifice. You can be sure that the air flow through that orifice will be 24/7, not just 220° out of every 360 crankshaft-degrees. That's 60% better...

It can be, but this will inevitably mean that there will be circumstances where the resonance works against the engine rather than with it.
With EFI you don't need to worry about fuel dropping out of the mixture due to too low a flow velocity, so make that plenum a big as you can; think Samsonite.

At the time Jan was thinking about the RSA inlet disk position and he wanted to find out whether the direction of crankshaft rotation had much influence on the inlet flow. So he put a standard kart engine on the test bench, measured the power, started it in reverse and measured the power again. 'Reverse' , meaning the crank rotating backward, gave one extra horse.
I discussed this with Roland Holzner of Modena Engines the other day and he told me that they'd already performed the same test at Rotax. Roland said that the negative effect wasn't so much due the counter-flow induced by the crank webs, but mainly due to the fact that the fat end of the con rod was directly in front of the reed case at the moment of initial inlet flow.

Here's some more Helmholtz-encouragement:

Regarding induction systems: build a short induction tract and visit a test bench.
You can forget about calculations. The formulas you find in various books are all based on the Helmholtz resonator. It made me write a simple story, called:
Helmholtz blues
"A Helmholtz resonator consists of a volume connected to a duct". That is what Wikipedia tries to tell you.
But that is a Helmholtz resonator in its simplest form; one that you won't find anywhere in an engine.
What you do find in an engine is an intake tract with a variable cross-sectional area. This tract is from time to time connected to a variable volume (the crankcase) through a very variable window (the intake port / reed valve / rotary inlet).
The crankcase is connected to a number of transfer ducts with variable cross-sectional areas, which are from time to time connected to a very variable volume (the cylinder) through a number of very variable windows (the transfer ports).
The cylinder is from time to time, through a very variable window (the exhaust port), connected to an exhaust pipe with a very variable cross-sectional area who at the same time doubles as a volume. This pipe volume is constantly connected to a big volume (the outside world) through a tailpipe with a constant cross-sectional area and constant entry and exit windows (thank God, finally someting that's not variable).
We call this a compound Helmholtz resonator .
The various papers also tell us how to calculate the resonator's frequency:
" frequency = speed of sound / (2*pi)* Sqr ( cross-sectional area of the neck / ( volume of the resonator * effective neck length ) ) ".
O yes, the speed of sound... It is dependent on temperature, which is not really constant in the intake tract and the crankcase, rather variable in the transfer ducts and very variable in the cylinder and the exhaust pipe.
Now the above frequency formula is not exact; it is an approximation that is usable as long as the volume of 'the' tract is very small compared to the resonator's volume.
So when engines are concerned, that formula goes very far out the window.
Who said gas dynamics is simple dull?
PS:
It's a similar story with acoustics. That is a sub-branch of gas dynamics, simplified with a lot of assumptions that are acceptable as long as the sound pressure does not exceed a certain limit. The wave pressures in a two-stoke exhaust exceed that limit by a factor of thousand. Bye bye, acoustics...

For me: FOS symmetrical schavenging, 24/7 inlet system, continuously variable transmission; not even remotely related to a Rotax based single.

Helmholtz frequency, my dear Watson. The smaller crank will give the pipe more crankcase volume to breathe from. This larger volume will also lower the resonance frequency of the inlet system, hence the drop at high rpm. It can be compensated with a shorter inlet tract, a bigger carb diameter and a later inlet closure. To put it real simple: if you have more volume, you need more time to fill it. But more time, i.e. later closure, has adverse effects at low revs. So my preference is a shorter tract (unless you go the 24/7-way and employ a reed for the low revs, and swing it out of the way at high revs).

I would stay away from air-cooled racing engines at all costs.Then again, I would find a way around the carb diameter limitation; take a look at www.pit-lane.biz where I explained the backgrounds of my 24/7 inlet system.

[Vortex]

Hi guys,
On my dyno software i have a few options concerning the correction, starting with "off" means no correction from ISO and DIN to 95/1/CE.
By default the 95/1/CE is in use together with a connected meteostation installed in the dyno room and unfortunately without any possibility entering values manually.
Now i'm trying to convince the manufacturer in Italy that at least i would like to take the temp. in the airbox and not in the room as the reference.
What do you think about, 1. the used correction and 2. the place where the reference temperature is taken?

Thank you in advance
Vortex

[diesel pig]

It's like the Great Master is reading my mind. If only I had a mind worth husaberg reading.

[ken seeber]

Hewzer, thanks for putting up the collation of 24/7 by Frits.
Absolutely legendary stuff.

[Frits Overmars]

Hewzer, thanks for putting up the collation of 24/7 by Frits.
Absolutely legendary stuff.

Boy, did I write all that stuff? I suppose I've earned a piece of apple pie with my coffee