ESE's works engine tuner

[TZ350]

The piston to head tapping that caused the vibration frequency that fooled your deto sensor into thinking there was deto has disappeared.

Fooled the deto sensor and me, Yes ... totally.

Towards the end, I was starting to wonder why the motor hadn't expired after hours of rattling away on the dyno.

[seattle smitty]

[QUOTE=Frits Overmars;1130810900].
Using a sleeve would mean varying the exhaust timing as well. (quote)

Why? This doesn't have to be a sleeve like an older 2-stroke had, an iron sleeve whose finished I.D. forms the cylinder bore. It can made as a ring, sitting in a wide slot just barely below the exhaust ports. Depending on the intake port shapes and spacing, you could have this ring-sleeve rotate rather than move up and down, to reduce the port area, even the port opening height (sorry I can't make a sketch).

Such a ring would not interfere at all with most of the heat transfer path, from TDC down to just below the bottom of the exhaust ports. Below that, yes it's a barrier. And the ring itself would get hot and transfer heat into the intake charge. I can imagine some possible partial remedies, but maybe they would not overcome the problem. I do think that current water jacketing design in conventional 2-stroke cylinders has room for improvement, if more cooling is wanted. You remember what the big "goldfish bowl" water jacket of the Harley/Aermachi RR250 looked like . . . suppose it had been cast with cooling fins INSIDE the water jacket, greatly increasing heat dissipation over the exhaust port . . . if that's what you need (I'm an outboard guy, and our cooling requirements are more easily dealt with). One thing about that engine as it was, was that the cylinder head had what in effect was a big single cooling fin around its out edge, hanging far into the enormous water jacket. IF you want a lot of cooling in your cylinder head, that's a way to get it.

(As usual, conceptualizing is easy and fun; executing is hard, LOL.)

[Yow Ling]

You got me curiouser and curiouser. Are you thinking sliding or hinged transfer roofs? Me, I've even considered reducing the piston stroke while the engine was running, raising the BDC position while maintaining the TDC position, so the transfers would only partly open (I'd probably had too much to drink).
I hope you will come up with something much, much simpler.

How would you change the stroke when its running ? The second part I can follow maybe using a junk headand moving the cylinder up or down

[husaberg]

Attachment 307279

How would you change the stroke when its running ? The second part I can follow maybe using a junk headand moving the cylinder up or down

Attachment 290165

Crecy Engine .

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


For variable transfer height this in configured in reverse.



or even a Yamaha spool valve.

later
http://forums.autosport.com/topic/74...w-engine/page-

[Frits Overmars]

How would you change the stroke when its running ?

Like in the pictures below (there are multiple solutions). But I dropped the idea almost as soon as I got it: too complicated and certainly not suitable for high revs.

[seattle smitty]

No, NOT like that. Your deal does NOT need a sleeve that moves at engine speed. It only moves as fast as your engine climbs its power curve, or comes back down. Outboard racers don't pull their trombone pipes in and out twelve thousand times a minute! Maybe on a bike or shifter kart you want your port-timing to change in every gear, as the engine climbs (or descends) the power curve, so you can't change the port-timing manually, as outboarders do their pipes, but still this is NOTHING like trying make another mechanical movement at engine speed, at every stroke. You don't run the guillotine-valve in your exhaust port in and out at twelve thousand times a minute. A port-timing valve (for that matter, like a variable-timing rotary valve) only has to respond as fast as that guillotine-valve in the exhaust. Okay, my notion involves doing this with a partial "sleeve," but it is NOT like a cylinder sleeve or the sleeves in an ancient sleeve-valve engine.

[136kg136ps]

With the roof angle changing,the area changing it seems the volume will necessarily change as well?

Given the small clearance to the exhaust ducts,I'd keep the actuation and directional control method simple to stay compact.

A hinged piece(shaped to fit up into the transfer port roof as close to the bore as possible),hinged at the outside radius where there is more meat in the cylinder.Moving with a simple rod actuating it. A rod which runs perpendicular to the bore inside the cylinder and being actuated from outside the cylinder via rocker arms(or some other mechanism) in turn connected to each other to ensure they are coordinated in motion.

My experience is gas turbines which have variable vanes (many) situated around the turbine whose ends protrude into the turbine gas flow but are controlled from the outside.They are situated radially and act rotationally but the principle is similar.

This gives us sealing the hinged piece within the duct and actuator within the cylinder as one problem.I can't see any others yet.Other than a few pounds of extra mass.

[speedpro]

I envisage something like a "comma"(,) shaped piece lying horizontal forming the roof of the transfer port. It would pivot on a shaft through the part that would normally be the top part of a written comma. In the position with the comma fully raised it would form a smoothly contoured transfer duct outer wall and port top/roof. Pivoted to a lower position obviously the roof slope would decrease as would the proximity to the cylinder wall. Most likely neither change would be an issue when it happens which would be at "part open" or closed throttle. I don't see it being too difficult to do which makes me wonder why nobody has done it yet.

[ken seeber]

Couple of things that can be observed from the FOS layout. I did a rough sketch of the ports based on a 54 stroke, 120 rod crs, 190 ex duration and 130 transfer duration .
1. As the floor of all the exh ports are, say 2.5, above the top of the transfers (compared to a conventional RSA style layout where central exh floor is approx. @ BDC), it can be seen that the piston needs to only be around 41 long. Taking advantage of this, the transfer passage entry “roof” can be raised to correspond with the piston @ BDC. This then allows the “floor” of the entry to be raised. What this means is there is a much greater mixture flow higher up the piston, exposing the piston to more cooling plus more lubrication for the pin bearing. The bulk mixture flow at the max flow rate instance, say @ BDC, would be improved because one is not drawing from the more shielded inside of a longer piston. The piston would also be much lighter due to not having the long conventional skirts.
2. If the exh outlets were fore and aft (ie square to the crank axis), this would mean that the piston pin bore might not uncover an exh port, eliminating any exh to transfer port short circuiting.
3. Given that we have tons of transfer port area, I thought it would be reasonable to consider having the floor of the transfer ports a few mm above BDC. For a 2 mm step, this would give a 50 deg duration of no interference to the transfer flow by the piston edge. With the piston relatively lower to the transfer port floor, it also means that, with a well-designed coolant passage inside the “cup handle”, the ring would be closer to the coolant passage. Additionally, because of the symmetry of all the transfer passages, this coolant passage would be simple to implement.
I’d envisage a vertical reed block between the crank webs. While I do recognize a reed is not as good as a disc valve, perhaps with the Frits 24/7 feature, this disadvantage is no longer relevant.
Personally I’m keen on the DAS (Direct Atmosphere System), but can see that it would be hard to implement, say in CIK and seemingly all karting controlling bodies where the opportunity for technical advancement is about zero. Gotta say the freedom that was there in 125 GP racing, was a wonderful thing.
I’d imagine Frits is aware of all these things, but it is interesting all the same.
What also is abundantly clear is that there is stuff all room between the floor of the exhaust and the roof of the transfers, particularly when they overlap as can be seen in a top down elevation. So, when considering throttling (or controlling the mixture flow rate) via lowering the roof of the transfers there is precious little real estate to incorporate valves etc. There are other methods of course. Still, I have the feeling that right now there are lots of dudes on the case at present, so let’s wait and see.
Anyways, better get into the New Year’s eve party thing.

[breezy]

Couple of things that can be observed from the FOS layout. I did a rough sketch of the ports based on a 54 stroke, 120 rod crs, 190 ex duration and 130 transfer duration .
1. As the floor of all the exh ports are, say 2.5, above the top of the transfers (compared to a conventional RSA style layout where central exh floor is approx. @ BDC), it can be seen that the piston needs to only be around 41 long. Taking advantage of this, the transfer passage entry “roof” can be raised to correspond with the piston @ BDC. This then allows the “floor” of the entry to be raised. What this means is there is a much greater mixture flow higher up the piston, exposing the piston to more cooling plus more lubrication for the pin bearing. The bulk mixture flow at the max flow rate instance, say @ BDC, would be improved because one is not drawing from the more shielded inside of a longer piston. The piston would also be much lighter due to not having the long conventional skirts.
2. If the exh outlets were fore and aft (ie square to the crank axis), this would mean that the piston pin bore might not uncover an exh port, eliminating any exh to transfer port short circuiting.
3. Given that we have tons of transfer port area, I thought it would be reasonable to consider having the floor of the transfer ports a few mm above BDC. For a 2 mm step, this would give a 50 deg duration of no interference to the transfer flow by the piston edge. With the piston relatively lower to the transfer port floor, it also means that, with a well-designed coolant passage inside the “cup handle”, the ring would be closer to the coolant passage. Additionally, because of the symmetry of all the transfer passages, this coolant passage would be simple to implement.
I’d envisage a vertical reed block between the crank webs. While I do recognize a reed is not as good as a disc valve, perhaps with the Frits 24/7 feature, this disadvantage is no longer relevant.
Personally I’m keen on the DAS (Direct Atmosphere System), but can see that it would be hard to implement, say in CIK and seemingly all karting controlling bodies where the opportunity for technical advancement is about zero. Gotta say the freedom that was there in 125 GP racing, was a wonderful thing.
I’d imagine Frits is aware of all these things, but it is interesting all the same.
What also is abundantly clear is that there is stuff all room between the floor of the exhaust and the roof of the transfers, particularly when they overlap as can be seen in a top down elevation. So, when considering throttling (or controlling the mixture flow rate) via lowering the roof of the transfers there is precious little real estate to incorporate valves etc. There are other methods of course. Still, I have the feeling that right now there are lots of dudes on the case at present, so let’s wait and see.
Anyways, better get into the New Year’s eve party thing.

Do we know what bore or stroke Frits had in mimd? thoughs little areo engines which run at really high sort of ticking over speeds( sure ive read 14000 rpm whilst not under load on the ground) could be looked at and scaled up perhaps. No wonder cvt would be very helpfull at thoughs kind of rpm values (P. L. A..N. when you can only raise the rpm to produce more power) A sleeve, recessed around the cylinder sleeve, which moves up and down when required around the cylinder sleeve to fully expose the transfers and vary the exhaust ports hieght was what came to mind. The sleeve could be made of a steel alloy material and raised and lowerd by magnetism.

[136kg136ps]

Speedpro,exactly.The comma shape describes what I was thinking perfectly.It would seem the side that forms the port wall would have to conform in shape to the port.I do worry about turbulence at the edge as well as caused by a rapid change in volume and shape.

Since it's not in a high temp environment ,mounting could be pretty simple.One could even retrofit a conventional existing cylinder to test out various mounting methods.Just make the comma,grind out a recess in the transfer port roof,figure a way to have it rotate around a pin and then on to actuation.

[seattle smitty]

Six transfer ports. So six push-pull rods opening and closing your "comma-valves"? And how do you actuate them, what sort of linkage to six pushrods all around the cylinder? It could be worked out, but seems more complex than my notion of a sliding or twisting sleeve.

With my ring-sleeve, if you make it move up and down, you only need two pushrods, one on either side, ganged together outside the cylinder, OR if you make rotate sideways to change port area/timing, you have a rack-and-pinion on the bottom edge, so just one rod, with a bellcrank outside the cylinder, OR, move it with a little cam. The movements are short. At least they are slow (relative to engine speed).

And although your comma-valves are not "in" a high temp environment, they are very close to one, could pick up a lot of heat, could transfer that heat to the intake charge. This is also a problem for my sleeve scheme.

Happy New Year, y'all!!!

[136kg136ps]

smitty,I agree your method would likely have fewer parts and failure points.

I believe in the moveable roof scenario there would be better directional control and there would likely be less than 10 parts per transfer port all very simple shapes other than the piece that fits into the transfer port.A powervalve style motor could be run off the ignition to vary the position with a variety of inputs TPS,MAP,EGT etc...After seeing how convoluted some PV actuation systems are I am surprised no one has tried a variable transfer port with success.

Imagine the rocker arms arranged with the actuated arm shorter than the arm connected to the motor so that each arm could be connected to another.They would be rocking around the circumference not up and down axially.

Sealing the actuating rod would be simplest if there was no penetration of a water cavity but even if so,it's low pressure water at a moderate temp.

There would be no or little thermal issues.

Compared to a http://www.geaviation.com/marine/eng...litary/lm2500/
our 2 strokes are the epitome of simple,even the oval bore H@n&a foul strokes with 8 valves would be.

You can see on the link above where the variable stator vanes are and get an idea of how simple actuating multiple parts are.After safety wiring all those parts together as well as every cannon plug,fuel,air and hydraulic line I got to know them far to intimately.

[seattle smitty]

[QUOTE=136kg136ps;1130811327] . . . and there would likely be less than 10 parts per transfer port
There would be no or little thermal issues. (end quote)

That's 60 parts. Looking at a photo of one of Frits' prototypes, it's not easy to envision how you would lay out all of your linkage. (but not easy to package my sleeve, either).

Why do you see no thermal issues? Each of your six valves will rotate up against the back of an exhaust port floor, leaving no room for any water jacketing. (Again, this is an issue with my idea, though possibly addressable by various means, maybe including ceramic coating the exhaust port floors, although that has a drawback or two.)

[136kg136ps]

Without the moveable part that metal would be in direct contact with that same exhaust floor.With the moveable part,the part is in direct contact with a pin,there would be a small air gap around the part allowing it to move freely so it should be actually be cooler than originally.Albeit the big end would be closer to the exhaust,the heat would be conducted less efficiently to the comma via air and a pin it rotates on than if it were 1 contiguous piece of metal.I assume it would of course be cooled by the transfer stream as well.

In the diagrams Frits has provided there is no water passage I can see near the bore between the exhaust floor and transfer roof.The comma shape as speedpro suggested in fact tapers as it approaches the bore due to the lack of metal there for much of anything.I may be wrong but I did look first to see if it impinged on any water passages.

As far as parts counts,KTM,H0n$@ and Suzuki had similar amounts for their PV systems.One could go the route of spring loading the comma piece lowering the count but giving up precise control.

Thanks for taking the time to help hash this out,I only wish I had the ability to spend the $ and time knocking together a model to test.