Two-Stroke Engine Tuning: Maximizing Performance

[Ayrity]

hey all, I just completed putting a small 2 stroke on my old bicycle, yes majorly geared down, I am actually rather pleased with it. I used an internally geared 3 speed hub on the rear wheel so i could shift back down to 1st gear after being in 3rd and coasting to a stop (you don't have to have the wheel moving to shift gears, unlike derailers) and basicly welded a 20in wheel rim to the small sprocket off of the hub attached to my 24 in wheel, giving me the reduction i needed.

anyway, having completed the drivetrain, i would like to now move into getting the most out of my 23cc 2 stroke engine. after all my reading, I realized that the biggest problem they have is the fact that they waste some fuel during each cycle out of the exhaust port. Racing dirtbike engines use what they call expansion chambers in the exhaust pipe to reflect sound waves back at the engine, and timed right, they push the unused fuel/air mixture back into the engine. this also helps the compression ratio. take a look at the bottom of this linked page if you want a visual http://www.vf750fd.com/blurbs/stroke.html

I had 2 ideas about how to tune these engines in a better way. the first is an adjustment to the expansion chamber design. on this page http://www-staff.lboro.ac.uk/~elvpc/bikes.html?http://www-staff.lboro.ac.uk/~elvpc/progs/expch.html you can see that all the pipe bits diameters and lengths are all dependant on the RPM that the engine is wanted to run best at (giving a better bottom, middle or top end power band). my idea is to create the different sections of piping of the expansion chamber to be variable in both their lengths and diameters, and have it vary depending on the rpm of the engine, thus giving you the best timed reflection back at the engine. this could be accomplished much in the same way afterburner nozzles change their diameter, and as far as length goes, have the pipes overlap one inside the other and depending on what the equations give you for a particular section of the chamber, have a spring of a certain spring constant attached between the 2 sections (smaller constants for a length that should change more relative to the other lengths, etc).

my second idea may be more of a question, so let me ask it. Why isn't there just a valve on the exhaust port that shuts just before the fresh unburned fuel mixture starts to flow through it? it would prevent the bad gas "mileage" and help control pollution and increase compression ratios as well, and it wouldn't even have to deal with the pressure of the combustion like in a normal 4 stroke engine, so it could be pretty small and light. perhaps a disk valve? anyway please let me get your feedback on these things, thanks.

 

[Cyrus]

Why isn't there just a valve on the exhaust port that shuts just before the fresh unburned fuel mixture starts to flow through it?

To save weight is my guess.

 

[Ayrity]

the way i see it, it wouldn't be much weight at all, maybe 1/10 of a pound

 

[turbo]

the way i see it, it wouldn't be much weight at all, maybe 1/10 of a pound

There is a lot of stuff that has to be added to the engine to make valves work, like cams, cam followers, valve return springs, etc, etc and these all drain mechanical power. A properly-designed 2-stroke can out-perform most 4-strokes. The Kawasaki 500 of the 1970's demonstrated this quite nicely, as did the Yamaha RD350 (I had one!), leaving 4-stroke bikes with much larger displacements in the dust.

As for the expansion chambers - when exhaust leaves the cylinder and heads out the pipe, there's lots of complex stuff going on. One of the interesting things is that when the exhaust pops out the end of the pipe, it creates an acoustic "reversion pulse" that can propagate back up the exhaust pipe and restrict the free flow of the next exhaust cycle. There is a whole industry built around designing exhaust systems for motorcycles to deal with this and other isues.

 

[Ayrity]

There is a lot of stuff that has to be added to the engine to make valves work, like cams, cam followers, valve return springs, etc, etc and these all drain mechanical power. A properly-designed 2-stroke can out-perform most 4-strokes. The Kawasaki 500 of the 1970's demonstrated this quite nicely, as did the Yamaha RD350 (I had one!), leaving 4-stroke bikes with much larger displacements in the dust.

thats why i thought a disk valve would be good, just a spinning disk with a hole cut into it, have a 90 degree turn of the axle off the disk, and then put a belt from the axle to the drive shaft, then time it correctly tune it using trial and error even.

 

[brewnog]

thats why i thought a disk valve would be good, just a spinning disk with a hole cut into it, have a 90 degree turn of the axle off the disk, and then put a belt from the axle to the drive shaft, then time it correctly tune it using trial and error even.

Any idea of the pressures involved, and the amount of pressure a spinning disc is likely to be able to contain? And how exactly would you time it, bearing in mind that one of the key features of a 2 stroke is its simplicity and lack of auxiliary mechanical systems.

In addition to pressure pulse tuning (and Lotus (I think) are currently experimenting with variable geometry exhaust manifolds) you might want to research scavenging, piston bowl shape and prechamber design. Where most people think of little whizz-bang scooter engines when "2-stroke" is mentioned, all the largest Diesel engines for power generation, pumping, and marine applications are 2 strokes, and achieve very high power densities and efficiencies through careful consideration of these effects.

 

[Ayrity]

thats true, weren't the old diesel-electric subs 2 stroke? anyway, my idea has evolved a bit since i went through a couple boring classes today haha, i will make a drawing and scan it in if anyone is interested.

the good thing about the disk valve is that it would be an easy instal, run off of the power of the engine, and be able to be easily timed and adjusted, with just a pulley and belt from the crank or drive shaft. also the good part is that it would only have to contain the pressurized gas/air mixture, but not the combusted gasses at the time of combustion like the valves in a 4 stroke do.

anyway, if anyone wants to see i will scan in my drawings after I am done with them. i think I am going to call it a 2.5 stroke engine, with the 0.5 stroke representing an over-scavenging with just air, but no fuel, therefor not wasting any of the fuel or polluting as much.

 

[turbo]

thats why i thought a disk valve would be good, just a spinning disk with a hole cut into it, have a 90 degree turn of the axle off the disk, and then put a belt from the axle to the drive shaft, then time it correctly tune it using trial and error even.

If you introduce valves of any type, you introduce complexity, weight, and mechanical friction, with is a drag on performance. Typically the best 2-stroke engine performance is acheived when a properly-tuned exhaust system works to scavenge the exhaust and resist the exhaust flow of the about-to-be-burned air/fuel mixture, in effect acting like a little compressor/supercharger. This is done by controlling the geometry of the exhaust to take advantage of the acoustic reversion pulses. If you close the exhaust valve during this part of the cycle, you lose the extra fuel/air compression supplied by the positive-pressure reversion pulse. I used to help my neighbor's son tune his dirt bike for performance (kind of a fill-in father figure) and when the expansion chamber got into a certain resonance with the engine, the torque and power-band were incredible. Did I mention that the bike was really throaty and loud when that happened?

These pulses are a factor in 4-stroke engines, too, but most often (due to the valving) the drive for performance will lead you to go to a stepped (larger and larger diameter) exhaust, with some kind of internal flanging to supress positive reversion pulses, and perhaps wrap the exhaust pipes in heat-resistant tape to keep the pipes very hot, which aids in scavanging. If you know how to cut and weld steel and you are willing to fabricate exhaust systems, you may be able to achieve a lot with a small investment. Inventing a new valving system for 100+ year mature technology might be a bit of an uphill climb. There's nothing wrong with trying if you think you've got a great idea, though. Good luck!

 

[brewnog]

How would you time it? How would you achieve dwell? How would you ensure that you would not inhibit trapped mass? Any idea of pressures?

 

[NateTG]

2-stroke diesel is a very different animal in terms of efficiency because diesels don't put fuel into the cylinder until the moment of ignition, so it's only air that's used to scavenge the exhaust.

For small engines (like the one you've got) the goal is typically to get high power density and simplicity which 2-stroke nicely provides.

When IC's get very large (think ship) the power density of two-stroke is also very attractive. AFIAK this 2-stroke is the most efficient production IC in the world: http://www.bath.ac.uk/~ccsshb/12cyl/

Since fuel injection and sophisticated valve control can mitigate many of the problems associated with 2-stroke engines, it does seem a bit odd that we don't see high-power 2-strokes in performance cars.

 

[brewnog]

That's fantastic! Makes my 64 litre V16s look tiny!

A lot of engines of that scale aren't designed to be stopped and started, - they're even serviced while they're running (including replacement of bearing shells!). Suppose it makes sense when you have to think about lighting fires inside the bores just to get enough heat in the cylinders to start them.



As for 2-strokes not being used on cars, I think that the fuel economy, emissions and noise issues are still pretty overwhelming. Direct injection can help (and removes the need to mix fuel with oil) but you need so much silencing (particularly when using pressure-pulse tuning) that exhaust lines end up being really quite restrictive, and the injection equipment puts such a high cost onto the engine that they become unviable economically. Reliability and durability are also issues, anyone got any service intervals for the highly strung 2 stroke bike engines? Or should we think back to the old Saabs and Trabants?! Having said all that, Honda and Ford currently have 2 stroke research programmes, so perhaps we might see them reappearing for medium-sized applications.

 

[Ayrity]

there is my hand drawn design. as you can see it has a fuel injector, and that means that the scavenging can be done by just air, wasting no fuel, however, the disk valve aids in keeping both higher compression ratio and a longer power stroke. it works like this. starting with how the picture shows it, the spark ignites, sending the piston down, right after the top of the piston reaches the bottom of the exhaust port, and just before reaching the top of the air intake port (seen on the left of the engine) the disk valve allows the exhaust out (this is the longer power stroke as compared to the average 2 stroke), letting it equal normal atmospheric pressure, then as the piston reaches the bottom of the intake air port, the disk valve shuts and allows pressure to be built up again in the cylinder. you should note that this means there is a bit of overlap when both valves are open, this time will be used to scavenge the exhaust completely out of the engine and the disk valve will be timed to shut again exactly when the unburned air begins to come out of the exhaust pipe.

then the piston will begin traveling back up, sucking in fresh air in past the reed valve, and injecting the fuel (the gas can be injected way before the ignition happens unlike diesel because of the spark plug, this allows the gas time to atomize with the air). the disk valve is closed this entime stroke, allowing for a slightly higher compression ratio than the average 2 stroke which would have had to wait for the piston to seal it. the air fuel mixture is pressurized, spark is ignited and it begins again.

i realize that a valve on the engine may add a bit of complexity, but it could be very very lightweight and does not have to suffer high heat because it only seals the cylinder during the weakest compression time, and during combustion, the piston also helps to seal the exhaust outlet, protecting the valve from the combustion. therefor the valve would add almost no extra drag or friction.

what do you guys think? any comments?

 

[turbo]

what do you guys think? any comments?

The mechanisms for timing the valve would add weight, the fuel injection system would add weight and rob power, and most importantly, your valve would probably have to be lightweight and would be subject to erosion from hot exhaust gases. Maybe if you could devise a lightweight adjustable-timing drive train to run a lightweight ceramic disk valve with fair sealing capacity, you'd have a chance. Intuitiviely, light metal disk valves would fail pretty quickly from overheating unless there is a way to conduct the heat away from them faster than they accumulate it. With a hot 2-stroke at full chat, this may be hard to achieve.

 

[Ayrity]

i understand that the fuel injection and valve would add some weight, but i don't need a carberator anymore so that removes some weight, and also, this would be way lighter than any 4 stroke with the same power output, and far cleaner than any average 2 stroke, that said, i think the addition and slight power robbing of turning a valve and fuel injection is an easy price to pay for that kind of engine. it could go faster than a normal 4 stroke, and further (better mpg) than a normal 2 stroke.

i do see what you mean about the valve material though, perhaps ceramic is the way to go

 

[brewnog]

Again, have you thought about dealing with the in-cylinder pressure? The moment the piston moves past the outlet port, you'll have a LOT of pressure trying to blow that flimsy, poorly sealed valve right out of your exhaust system. You don't just have an instant pressure pulse at the moment of ignition, the pressure continues to rise at an incredible amount. You only need to look at the design of conventional valvegear on a 4-stroke (or the pressure pulse effect exploited on a small 2 stroke) to appreciate that. Also, why does it not have to suffer high heat? It's dealing with exhaust gases!

Am I missing something with this? You've obviously thought about it, but that was my concern from the out.

 

[Ayrity]

brewnog, i was thinking that as the top of the piston moves down past the bottom of the exhaust port, the hole in the disk valve would begin to open, and it would be all the way open by the time the air inlet port was all the way uncovered. this is the overlap of both ports being open, helping to scavenge. i don't know if there is something that could hold in the pressure at that point, but it would only have to hold it in at almost the smallest point of pressure, because the piston has moved almost all the way down, almost at the end of its power stroke.
i guess it does have to deal with high heat of exhaust gasses, what i meant to say was that it doesn't have to deal with the initial explosion of combustion.
and as far as heat goes, if its found that the material can't take that much heat, it could always be flipped so the axle is on the bottom (below the exhaust pipe) and have it dip one side into oil or something.

 

[Gigabyte111]

if its found that the material can't take that much heat, it could always be flipped so the axle is on the bottom (below the exhaust pipe) and have it dip one side into oil or something.

ever seen a bucket of oil burn? i would sugest a fin system simmilar to that used to cool engines without raidiators. your 2-stroke probably has this on the cylender head. or if you do want to use the above idea, use a nonflamable substance such as water. one more thing. the people who built this engine most likely knew what they were doing. don't go trying to reinvent the wheel. try it like it is. anything you add will probably just have a neagative effect. if you really want good preformance, use the gasoline with the highest octane rating you can find.

 

[NateTG]

Well, let's start with:
You're not lubing the crank-case anymore.
(You're not lubing the crank-case anymore.)
It's important enough to write twice. Your 2-stroke motor has lubricant mixed in with the fuel so that, as it goes through the crank case, it lubricates. Unless you set up an alternative lubrication system, you'll be siezing up.

As a benefit, you no longer need to use 2-stroke fuel, so you can actually get something that won't wreck your injector right away. Of course, getting the right injector, and getting it to work right is going to be an adventure in its own right.

With the geometry you've drawn, closing the exhaust valve on the power strok will make the exhaust gasses pressurize the crank case so you'll have trouble getting air in, and closing it on the compression stroke will make the piston push air/fuel back into the crank case. So I don't see a whole lot of benefit there unless you're also willing to stick a valve on the intake port.

 

[Ayrity]

first of all, i would lube the crank case, of course. probably use a thicker oil in the bottom of the crank, that way it wouldn't tend to blow up into the cylinder with the air.

youre right, i know nothing of fuel injectors, but they work on a lot of engines, and i bet i could get it to work on this one with some experimenting.

maybe my draving doesn't make it clear, but i have said it a few times now, the valve would be opened just before the piston uncoveres the intake air port, so just before the intake port is uncovered, the pressure in the cylinder is at 1 atmosphere of pressure, so NO exhaust would be forced into the crank case where the air comes from.

 

[Cliff_J]

The hole in your disk would need to look more like a slot. But why bother? If all you're trying to do is create a one-way flow, you could use another reed valve. If you have a situation where the incoming mixture is only air, you can afford to shovel it out the exhaust port in droves without hurting fuel consumption, but you will have less in the combustion chamber. You may want to copy most 2-cycle diesel applications that use a turbocharger (or supercharger or both) to pressurize the incoming air charge. Yes, more complex but definitely higher potential for power/weight.

You will note there are very very few examples of direct injection gasoline applications. Meaning parts would be expensive and difficult to come by.

You could easily replace the bearings on the crank and rod with sealed units and eliminate the need for constant lubrication.

 

[Ayrity]

thanks for the tips on lubrication. i was thinking of trying to modify an existing 2 cycle i have by just drilling a hole in the top next to the spark plug and threading it and putting in a small 4 cycle fuel injector, like from a lawn mower. the reason for the disk valve and not a reed valve is that i want the valve to be closed as soon as the piston moves to block the intake port, this way it can increase the compression ratio without the use of a turbo or supercharger. like in a regular 2 cycle, as the gas and air mixture is injected, it is pushed upwards by the piston, but since the exhaust port is higher than the intake port, it doesn't start compressing the mixture until it passes by the exhaust port, sealing it. with a disk valve, you can seal it early on the compression stroke, and open it late on the power stroke, letting it last longer and extracting more power out of each stroke. Is this thinking wrong? i understand it would be hard to keep everything cool, but in principle is my thinking correct on these ideas?

 

[NateTG]

Don't get me wrong, I'd definitely like to see a more developed 2-stroke gasoline motor. It's just that I'm pretty sure you'd be better off starting with a 4-stroke or, even better, starting with something like:
http://rbowes1.11net.com/dbowes/
(especially if you add a super/turbo charger)

 

[Ayrity]

interesting idea. any clue why this hasnt been wide spread on automobiles yet? i know i should start with something like that, but i have 2 or 3 2 strokes just sitting around and i want to try to use them, and make them better, so i was hoping for some help/constructive crit on my idea specifically thanks

 

[NateTG]

interesting idea. any clue why this hasnt been wide spread on automobiles yet? i know i should start with something like that, but i have 2 or 3 2 strokes just sitting around and i want to try to use them, and make them better, so i was hoping for some help/constructive crit on my idea specifically thanks

Electronic valves on production cars are a year or two away.

2-strokes are notorious for being hard to break. I don't think you'll get a big performance increase, but it won't hurt to try.

 

[kleinjahr]

Interesting idea. I do wonder how you will seal the valve seat so there is no blowpast. Perhaps rather than a disk some type of ball valve? Rather than a straight through hole, use one that is cut to give proportional opening. I have seen hand valves, in plastic, of this type but not in metal.

 

[Ayrity]

actually since nate showed me that link with the electronic valves, i was thinking of using one of those... maybe use it in conjunction with the magneto on the flywheel to have it activate?

on another note, does anyone know how i could rig up a fuel injection system? i know how they work, and how the ecu controls them, do you think it would be possible to use a system lifted from a car? use the 02 sensor and stuff like that?

 

[Cliff_J]

Note that the electronic valves are moving like 2mm.

An easier place to get an EFI system more closely matched would be a motorcycle, snowmobile, or ATV. They are closer in displacement and number of cylinders so the injector size would be in the ballpark so the pulse width would still be large enough for good atomization. Also they would be likely speed density so with a MAP and IAT you'd be set once you adjusted the tables to compensate for your displacement. How long they'd last inside the combustion chamber when they are not designed for it might be measured in 2-cycles too though.

You're more likley to be able to use a mechanical injector for direct injection with better success, they have been used in diesel applications for years with success and are designed for this type of operation. But how to retrofit one of those is another massive problem to itself, but an old gas fuel injection system (GM had mechanical gasoline injection on the infamous 57 Chevy 283HP 283CID engine and on Corvettes and so on) might lend itself better. The all metal construction seems a lot more robust than plastic...

 

[Ayrity]

hmm i like the idea of the mech injection, any details or sites that could explain better how that works? and also could i find a mechanical system in the general size i would need for my small engine? i have tried looking for 2 stroke diesels on like ebay, and there are none... and one more question, is gasoline injection different than diesel injection? how so?

 

[GENIERE]

hmm i like the idea of the mech injection, any details or sites that could explain better how that works? and also could i find a mechanical system in the general size i would need for my small engine? i have tried looking for 2 stroke diesels on like ebay, and there are none... and one more question, is gasoline injection different than diesel injection? how so?

I believe Orbital (Australian Company) uses some of the 2-stroke injection ideas you discuss in their engines.

Two stoke diesels have been around for a while, at least since WW2. To fit the diesel engine into an aircraft wing, the Germans used pairs of pistons. They compressed the fuel/air mixture between the pistons (no head) with a crankshaft at both ends of the cylinder yielding a very flat engine. Later a triangular setup using multiple banks of 6 pistons (3 cylinders) was used in high-speed boats and locomotives.

 

[Cliff_J]

http://www.hilborninjection.com/product.asp?Id=42&CatId=35

Welcome to Fuel Injection Engineering Co., Inc. We are the original designer and sole manufacturer of the HILBORN fuel injector. We provide a complete product line of HILBORN fuel injectors, fuel pumps, and related parts to fit a variety of engine types and racing applications. We also offer technical information, service and flow testing of all HILBORN injection systems.

HILBORN fuel injectors have been successfully used in all types of racing including oval track, drag, dry lakes, super modified, off-road, motorcycle, tractor pulling, hydroplanes, and Indy Lite series.

Google works well
http://www.auto-solve.com/mech_inj.htm

Wiki is always good, even math to figure out pulse width
http://en.wikipedia.org/wiki/Fuel_injection
http://en.wikipedia.org/wiki/Gasoline_Direct_Injection

Oddly, the wiki article isn't using SI, and you'll find a couple things about injectors different (some in CC instead of lbs but the conversion is nearly 10 so you can just move the decimal) and also the pressure affects the fuel delivery. The mechanical gets a little tricky to tune since its not a computer value, and without working on diesels its a pretty rare sight to even see a mechanical injected vehicle (even at a car show) so that valuable tuning information might be tricky to come by for mechanical injection.

 

[swedish_lunacy]

Ayrity,
Good on you for thinking outside the square. It's great to see people like you rolling up their sleeves and getting into it. Two strokes can produce tremendous power under the right conditions.

1) If you are thinking about fuel injection then use electric rather than mechanical. You WILL need the added control and granularity that it affords you over mechanical systems. EFI is very reliable. You will need the following: MAF sensor, fuel rail, fuel pressure regulator, injector, computer ( try and get one that will handle fuel AND spark ). Can be quite cheap these days. Remember the power is in the tune, tweak, tweak and more tweaking

2) Use a MAF rather than a MAP air flow sensor. Might be hard to get a small enough MAF but it is useful in that it requires much less aggro to setup as it is not so concerned about plenum geometry etc.

3) Don't bother with an O2 sensor, it is only required in "closed-loop" mode. Your EFI will run on a "fixed" map, ie "this much air needs this much fuel". The EFI will tell you how much air is coming into the motor ( mass in kg's or lb's ) and your map just adds fuel to suit. This way if your mods allow more air to come through, more fuel will be added automatically.

4) Have a look at Yamaha's EXUP system which is basically a RPM-variable exhaust geometry setup.

5) Have a look at Yamaha's "power-valves". http://www.dansmc.com/powervalve.htm
http://en.wikipedia.org/wiki/2-stroke_power_valve_system
These were an eccentric lobe that changed the exhaust port geometry based on RPM. They changed the timing by adjusting the height of the top or bottom ( can't remember which ) of the exhaust port cutaway on the side of the cylinder wall. The lobe would slowly occlude or reveal more of the port as the rev's changed. NOTORIOUSLY unreliable due to the fact that rotating parts need to be gasketed and are subject to high particulate content and extreme heat. FYI, lobe was rotated by cables run from a servo elsewhere on the bike

Look at nitro for more power, it is non-parasitic and it will afford the single biggest gain. Your motor needs much more fuel when it is running a shot of NOS so get a system that injects NOS and more fuel simultaneously

 

[Cliff_J]

Lunacy - a few followup points:

1) He's talking about direct injection, it will take a very sturdy injector to survive directly inside the combustion chamber.

2) Most small systems are speed-density, very few use a MAF. There are even quite a few GM V6s that only used speed density!

Oh, and nitro is usually short for "Nitromethane" as in Top Fuel Dragster with oxygen already in the fuel (also used in scale model engines of fractional displacement) and NOT for Nitrous Oxide. Slang might be "the bottle" or "on juice" or NO2 and so on but nitrous is widely understood and a small grammatical omission can lead to big issues. Stoichometric for nitromethane involves a lot less oxygen!

But it isn't the intent of the poster to simply increase his short-term power output for a drag race or something like it, he's trying to improve the efficiency of the 2-stroke motors he has. Little different approach even if a higher specific output is the goal.

 

[Ayrity]

wow great info thanks

 

[Ayrity]

thanks for all the ideas guys, cliff is right about my goal, so i think NO2 is out, but i loved the rest of your ideas lunacy. I appreciate all of your inputs on the fuel injection, but this is the area where i get a bit lost, i don't know much about this kind of stuff, like the computer control unit etc. like cliff, what does "speed density" mean? also i was thinking, is it at all possible to just swap the head off of a fuel injected engine, put it on a 2 stroke block (so it has the air injection set up in the crank), or otherwise convert backwards from a 4 stroke to a 2 stroke? it would already have a separate oil set up, and it would have the valve on the exhaust there and ready to adjust, obviously it would take more than this, but basicly, just readjust the valves, figure a way to have the power stroke pump fresh air to scavenge the exhaust, and then just double the injection and spark? is this worth thinking about?

 

[NateTG]

thanks for all the ideas guys, cliff is right about my goal, so i think NO2 is out, but i loved the rest of your ideas lunacy. I appreciate all of your inputs on the fuel injection, but this is the area where i get a bit lost, i don't know much about this kind of stuff, like the computer control unit etc. like cliff, what does "speed density" mean? also i was thinking, is it at all possible to just swap the head off of a fuel injected engine, put it on a 2 stroke block (so it has the air injection set up in the crank), or otherwise convert backwards from a 4 stroke to a 2 stroke? it would already have a separate oil set up, and it would have the valve on the exhaust there and ready to adjust, obviously it would take more than this, but basicly, just readjust the valves, figure a way to have the power stroke pump fresh air to scavenge the exhaust, and then just double the injection and spark? is this worth thinking about?

The speed density stuff talk is because EFI systems can be more efficient if they know the mass of air in the cylinder so there are sensors for measuring the air flow into the cylinders. There are a variety of methods for measuring - speed density refers to one of them.

Modern 4-stroke engines generally have all of the valves at the top of the cylinder, and don't use the crank case for aspiration or as a fuel pump. Some do have superchargers which are effectively air pumps used to improve the compression - but you can use them to scavenge in addition to aspirating.

Realistically, you've got to replace or retool the air, fuel, and timing systems but it's likely to be easier than retrofiting valves onto a 2-stroke.

Apparently orbital does offer modern 2-stroke and direct injection tech...
http://www.orbeng.com.au/orbital/nonflash_index.htm

 

[Ayrity]

ya i was thinking that might be easier, but i can't find anywhere online, google or ebay any sort of smallish 4 stroke direct fuel injected engine...

 

[Cliff_J]

Ayrity - the job of the fuel management is to try to get the mixture as close to ideal as possible and NateTG did a great job explaining this. Here is one article (of many a google search found) with more detail:

http://www.carcraft.com/techarticles/24549/

Now though, you've completely lost me on your ideas...and purpose, a modern 2-cycle engine can be tuned for incredible performance and it seems you're trying to reinvent a wheel that is round and oval.

You're not going to be able to use a cylinder head off a 4-cycle unless you can match the chamber volumes otherwise you'll affect the compression ratio. Also, the valves would need to be actuated by a camshaft that needs to be driven off the crankshaft, things not found on a 2-cycle. The time in the machine shop just went up dramatically and I don't see the value in it.

We want to injest a mixture, combust it, let it expand to perform work, and then expel it as best possible. The 2-cycle simply let's the events overlap and typically suffers ill effects only at low speed operation where the timing of events isn't very good.

If we have a 10:1 CR and a 100cc engine, we know that if we could fill it with 100cc of mixture then when compressed it would be 10cc and we could figure out the mixture temp (ideal gas law) to make sure we don't have detonation. But the problem is this assumes 100% volumetric efficiency, which is unlikely for a 4-cycle. So if there's only 90cc of mixture in the cylinder, then when compressed to 10cc its only an ECR of 9:1 in that case. However, if properly tuned, a 2-cycle with its intake tuned properly and some mixture in the exhaust temporarily getting pushed back into the combustion chamber, it may be stuffing in 105cc for a volumetric efficiency of 105%!

If you understand harmonics, its a resonance in terms of the behavior and timing. Which means its a frequency dependant event, and will have a different response depending on frequency. And while you can examine it in step-by-step detail, the events all need to be managed and factored at the frequency of operation.

And you won't find much at all on direct injection, its not much more than a very small niche technology on the cutting edge. Its more esoteric than braking assist techologies or some of the other really new and expensive stuff finding its way on production vehicles.

 

[swedish_lunacy]

Lunacy - a few followup points:

1) He's talking about direct injection, it will take a very sturdy injector to survive directly inside the combustion chamber.

2) Most small systems are speed-density, very few use a MAF. There are even quite a few GM V6s that only used speed density!

Oh, and nitro is usually short for "Nitromethane" as in Top Fuel Dragster with oxygen already in the fuel (also used in scale model engines of fractional displacement) and NOT for Nitrous Oxide. Slang might be "the bottle" or "on juice" or NO2 and so on but nitrous is widely understood and a small grammatical omission can lead to big issues. Stoichometric for nitromethane involves a lot less oxygen!

But it isn't the intent of the poster to simply increase his short-term power output for a drag race or something like it, he's trying to improve the efficiency of the 2-stroke motors he has. Little different approach even if a higher specific output is the goal.

Cliff, you are absolutely correct. Teaches me a lesson for posting after a night of bleary-eyed, hard work . Funny thing is I'm normally the first to point out the difference between Nitro ( Nitromethane ) and Nitrous Oxide ( which is actually N2O ). You're not wrong regards stoich for Nitromethane, 1.7:1 ratio means lots of fuel flowing into those top-fuellers ! Actually real world ratio is even worse due to them running quite richer than stoich.
Agree that N2O is probably not what Ayrity is after. In fact first and foremost I would invest my money/time in an optimised expansion chamber and exhaust combined with an electronically controlled ignition system. Then I would branch into a variable geometry exhaust system ( maybe variable geometry port as well - AKA Yamaha power valves, note valve is a bit of misnomer ).

Direct injection - DI:
ooops didn't notice Ayrity was talking about DI. Not for the faint hearted. It would definitely be too tricky to tackle in a do-it-yourself fashion. Injector durability and fuel pressure required are two limiting factors. The latter requirement probably representing quite a large parasitic loss

EFI would still be good for very fine control but I haven't seen many EFI small capacity 2-strokes ( only monstrous 3.5L outboard motors )

Regards MAF vs MAP:
A MAP's output will need to be tuned/calibrated/calculated with respect to both displacement and volumetric efficiency ( the later being a moving target throughout the rev range in a 2-stroke ). MAF just tells you how much air is going in full stop, regardless of VE, displacement, etc. But yeah, chances of finding a MAF small enough is pretty slim

 

[Ayrity]

well that explains why i can't find anything on the direct injection. one the other comments, youre right, I feel like i am going a bit in circles, but i am only trying to figure out the best way to attack my idea, the one of having direct injection and using the valve on the exhaust. I was getting the feeling that drilling into the head of the 2 stroke engine and setting up a fuel injection system was going to be very hard to do, so i looked for another alternative, like changing a 4 stroke into a 2 stroke, that way i already have the valve on the exhaust and a fuel injection system.

ive been playing around with the idea here, and i think it doesn't have to be direct injection as long as the injector is right next to the intake port, and timed correctly so that it shoots out the gas at the last moment, and make sure the exhaust valve is closed first. so something like, having the fuel injector at the bottom of the downstroke of the piston, that way it could inject seperately from the air intake, so no fuel is wasted on scavenging, and it is low enough so that it is not right next to the spark plug when the combustion happens, in fact when the exhaust touches it, it will be at 1amu, becuase the exhaust valve will always be opened before the injector is exposed in its new "port"

what do you think would be the best way to get this idea to light? (and just a second question, i can't seem to figure out what difference there is between a diesel fuel injector and a gasoline injector, it does seem that a diesel injector can withstand the combustion...)

 

[brewnog]

A Diesel fuel injector will squirt at much, much higher pressures than a petrol injector, - it can not only mechanically withstand and seal under in-cylinder pressure conditions, but it's able to meter and inject fuel at incredibly high pressures (we wouldn't bat an eyelid to run at 200MPa line pressure). Petrol injectors probably max-out at 100MPa, and that's for direct injection. Conventional injection is far less than this.

 

[Ayrity]

hmm ok thanks. would it be a bad idea to think about using a diesel injector for this engine idea? do you think it would be difficult to tap a hole for an injector and install it into the head of the engine? what about the changing a 4 stroke engine into this kind of 2 stroke engine? the only reason i bring that up again is because of the dedicated lubrication system. or is there an easier way to make a 2 stroke have a separate oil bath for the crank?

basicly, if you were going to attempt to make this engine, which way would you choose to try and why? thanks

 

[swedish_lunacy]

If you want to build an injected two stroke, have a look at the marine industry two strokes. Those outboards run some pretty cool systems. I think they may have a separate oil injector as they run condition-dependant variable fuel/oil ratios. Lubing the crank via any other method will probably requie quite a lot of engineering effort.

I think traditional DI is going to be pretty tricky but I think that an injector just above your intake port might be a good alternative. Diesel injectors tough as nails so will probably be allright. Not sure what the performance of a diesel inj is if run at low pressure, the fuel may not atomise properly.

Also need to look at the parasitic losses of having to pressurise the fuel. It may not be efficient enough on such a small scale.

PS: Have a look at the following, http://www.dolmar.com/755.php twenty four chainsaw motors in a motorcycle chassis

 

[brewnog]

Parasitic losses: On the 4 stroke direct injection Diesels I work with, each unit injector takes ~2 tonnes of force to actuate, when running balls out. Bearing in mind that it's doing this 15 times per second, and there's one per cylinder, you can see that they're pretty tough things.

 

[Ayrity]

they do have small, like 4 and 5 hp four stroke diesel engines though. maybe i could lift off the injection system off of one of those.

by the way, how come they don't make smaller 2 stroke diesels? is it the cost of the turbo/supercharger needed would be as much as the engine? why can't diesels use the downstroke of the piston to push their air into the cylinder (my guess is its not enough compression, but i don't know).

a benifit of maybe modifying a diesel engine is i could use biodiesel to run my bike :) i have experimented with biodiesel and had some good results, so maybe that's a better plan...modify a regular smallish 4 stroke diesel into a 2 stroke? what do you guys think of that?

again, I am worried about the lubrication system... this is why I am leaning to 4 stroke to modify

 

[Averagesupernova]

Ok I've been reading over this for the last couple of days and my memory isn't perfect so I may repeat something that has already been said.
-
Ayrity, let me get this straight. You basically want to stay with a 2 stroke design that fully scavenges the exhaust out of the combustion chamber yet doesn't waste fuel out the exhaust port right? A 2 stroke engine more or less RELIES on the incoming air-fuel mixture to help push the exhaust out. So, if you want to scavenge the combustion chamber without wasting fuel you need a blower. Here is what I would do:
-
Conventional porting arrangement with the exhaust slightly higher than the intake port. BUT, the intake port is only going to be porting fresh air into the engine and under pressure by the blower. The intake ports ONLY function is to scavenge. The exhaust port would probably be a very small distance above the intake port. Only enough to avoid the exhaust blowing back into the intake port. This distance could probably be less than the average conventional two-stroke design.
-
Next step, getting fuel into the combustion chamber:
-
Ok, here's an idea that isn't really anything new, but maybe just a little bit. Put a valve in the piston. Not a reed or poppet valve, but a slide valve with its own port that feeds out the bottom of the piston. Stay with a carburetor and reed valve to feed the air-fuel charge into the crankcase.
-
Here is how the valve in the piston would work: It could be improved on or simplified, but I'll explain it in such a way to best illustrate it. Instead of having one connecting rod for the piston, have 2. Yep, you read it correctly. They are spaced slightly to allow for linkage to fit between them. The linkage goes from the crankshaft journal up to the piston. The journal on the crank is the same for the outside 2 rods, but different in between. It is offset ground. So as the crank turns, the piston goes up and down and the linkage goes up and down relative to the piston. The linkage is attached to it's own miniature piston within the regular piston. This second piston is the one that covers and uncovers the port within the main piston which forms a valve. This way when the crank is at a certain position the linkage is pulled away from the piston which opens the in-piston valve. The offset journal on the crank is positioned in such a way as to open the in-piston valve just as the exhaust port is covered up when the piston is on the way up. The one problem I can see with this idea is that the combustion chamber is sealed up when the in-piston valve opens. Also, since the trapped fresh air in the combustion chamber WILL be part of the combustion, it seems that the charge coming up through the in-piston valve will need to be a fairly rich mixture. It seems like a properly designed exhaust system should draw enough air out of the combustion chamber to leave it in a slight vacuum in order to help get the fuel up through the in-piston valve. One other thing that could be done is to use a 2-stage blower. I'm not sure of which specific one, but older planes, bombers I think, used this. Basically the crankshaft turned a turbine on the exhaust pipes to help draw out the exhaust as well as turning a turbine on the intake for obvious reasons.
-
I know it sounds like a strange animal and I wouldn't expect you to make it work, but this IS the place to discuss it. So everyone feel free to shoot this down or modify it or whatever. BUT, before you get too carried away with the bashing, valves in the piston HAS been done before. The radial engine where the block spins and the crank was stationary used a valve in the piston. The fuel was piped in through the crankshaft. I'm not sure about fresh air. If I recall correctly, it had ONE speed of wide open.

 

[Cliff_J]

Averagesupernova - little complex, but like the idea since it moves away from DI for now, that technology is just a little bleeding edge right now. Oh, and those old stationary crank airplane motors had another advantage - there's no major reciprocating parts, the pistons travel in a circle. And on the TV program I saw it on, the German version simply cut a number of ignition pulses to idle and then let all the pulses through for normal operation - idle or wide open, like an on/off switch! I guess the plane could turn one direction on a dime, but like a freight train in the other.

Ayrity - turbos and superchargers cost a lot of money because of the precision work needed. A 2-cycle involves casting 3 main engine parts, a crank/rod/piston with a couple bearings and its operational with 10 machined parts that require maybe two dozen machine operations. Simple and super cheap with surprisingly good power output for the displacement and weight and cost.

But Ayrity, you are certainly fascinated with 2-strokes even if to a extreme.

The 2-stroke effectively gives up some of its power stroke to begin the exhaust stroke as they overlap, and the intake stroke also begins while the exhaust stroke is still occurring and so on. You're talking about trying to tweak those a little more by adding a ton of technology.

Now the biggest question is how to make that technology pay for itself. A military vessel needs high-power and lightweight so they use turbines. A transport truck needs to be economical on long trips but have power to navigate steep grades and varying loads, so a turbo is able to make it more economical. For a dirtbike or motorcycle, pretty tough economics, especially with the new emission standards even in off-road parks. Just the silencer/muffler on a 2-stroke can steal 20% or more of the power to comply with noise requirements.

 

[Ayrity]

supernova, great thinking, i really do apprecicate some new ideas like that. i do see what you mean with the valve in the piston. i was thinking about doing the same kind of thing actually, only i was going to do it like this... allow just air to come into the crank and have it push out all the exhaust, but at the last moment of the downstroke of the piston, have a valve switch that puts the air coming from the crank through a different tube that has the carb on it, and also leads to the intake port, again, like in Novas idea it would probably have to be a pretty rich mixture. if any of that was unclear to anyone i would be happy to draw it out and scan it, I am probabbly going to do it for my own benifit anyway haha.

Cliff, its not that I am obsessed with 2 strokes, its just, when i put my mind on something, i work with it until it gets where i want it, or until i run out of money, or until i don't have the tools to do it yet. in which case i put them away for later working :) thanks for humoring me though haha, i love this place

 

[Averagesupernova]

Ayrity, I've given a bit more thought to my idea. I figured why have a slide valve in the piston? Why not just make it so it slides in the block along side of the main bore? It is still run off of the crank. The slide valve would be closer to a spool valve though since it would control a scavenge port as well as a port that let's fuel in. This would give us a chance to go to indirect fuel injection right ahead of the spool valve which would probably guarantee that enough fuel-air charge gets into the combustion chamber. Basic cycle timing would be the same as my previous example. Also, you could have a conventional pressure oil system in the bottom end so there is NO FUEL MIX. So in this idea contrasted to my previous you have a trade off of going to a more conventional piston/rod assembly and eliminating the reed valves and a carburetor while going to a fuel injection system and a slightly more complicated valve. My previous example had a valve, so adding a bit of complexity to it isn't that big of a deal. Of course the simplest is no valve, but to achieve what you want I don't think it is possible. Lubrication for the valve is now a consideration though since there won't be oil going around it at all. Just fresh air and fuel. You may want to consider an oil metering pump for it the same way the Wankel rotary meters oil for lubrication. I think you would also be guaranteed to be going away from the need for the 2 stage blower.

 

[Ayrity]

ya, that's basicly what I was thinking of doing, but just your idea uses an injector, and mine a carb i think.

*************************EDIT: SUPERNOVA- would you mind drawing your idea its a bit unclear to me... sorry thanks

 

[Averagesupernova]

That's SUPERnova to you buster! I'll give it a whirl. I've never attached anything before. So we'll see. About the spool valve lube, I suppose with a conventional oil system oil could be pumped to the valves and since they don't pass exhaust I think a seal could be made to stand up just fine.