"Free valve technology" in dirt bikes?

[rhinoracin]

If this is as good as these people say it is, and manufacturers start building engines like this, street bikes could be using this technology one day. But would this also benefit single cylinder dirt bikes?

And there's a part where it says 2-4 stroke switching...hmm

 

 

 

Edited May 8, 2014 by rhinoracin

[bikedude987]

very cool tech.  And very applicable to most car engines with the addition of a high volume air compressor.  

However, on a dirtbike, you still need high volume/high pressure compressed air and a likely high pressure oil supply.  So you've just offset your weight savings from removing the cams and cam chain by adding an air compressor.

 

Probably the reason most car mfg's haven't added this to production cars.  I'd bet the weight/cost is a wash, despite obvious performance and efficiency benefits.  

[grayracer513]

Not really very "new" overall.  Several F1 engines have been using this kind of concept for quite a while now.  It evolved from the idea of using compressed air as a valve spring (a spring is subject to loss of tension if vibrated at the wrong frequency, and the mass of the moving half of the spring has inertia that contributes to valve float.  a volume of compressed air has neither of these problems).  Volvo, Honda, and Lotus that I've heard of have tinkered with cam-less designs.  Honda's was said to idle smoothly at 600 RPM, and was docile as an Accord 4 door around the pits, in spite of the fact that it produced competitive horsepower at 20K.  The valve timing was varied in real time based on the the demand.

 

Conceptually, it's the ultimate.  The practicality is all that remains to be worked out.  Kind of a lot of "stuff" to haul around with you, though, and the compressor has to be powered up before the engine will run, so belt driving it is not a complete solution.   On the other hand, Ducati did campaign an engine that used pneumatic valve "springs" some years ago, so it's obviously not totally impractical. 

 

The ability to switch to two-stroke cycle operation would depend on there being a supercharger, since the intake ports are in the head.  The power, most of the exhaust, and nearly all of the intake cycles have to be able to happen on the same down stroke, and with the poppet valve layout, only a supercharger can accomplish that.  But it's cool.

[k.g]

So here's a silly question. What happens when you lose a connection at one cylinder? Does it cut fuel to that cylinder? I'd assume each line would need to be separate, and not a common rail system.

Why not and electromagnetically controlled valve system? No need for prerequisite air before start. It could be timed from the crank position.

[Johnny_On_The_Spot]

The big problem with these designs that energy is not returned to the system.

 

On a normal cam & spring design, torque is required to ramp the valve up the cam and compress the spring.  After peak lift is achieved the spring "drives" the camshaft and returns its energy to the timing drive and then to the crank. 

 

On a pneumatic/ hydraulic/ electric system there are large parasitic losses - whether they are offset by the performance gains of the additional valve duration depends on the design.

[1987CR250R]

Not really very "new" overall.  Several F1 engines have been using this kind of concept for quite a while now.  It evolved from the idea of using compressed air as a valve spring (a spring is subject to loss of tension if vibrated at the wrong frequency, and the mass of the moving half of the spring has inertia that contributes to valve float.  a volume of compressed air has neither of these problems).  Volvo, Honda, and Lotus that I've heard of have tinkered with cam-less designs.  Honda's was said to idle smoothly at 600 RPM, and was docile as an Accord 4 door around the pits, in spite of the fact that it produced competitive horsepower at 20K.  The valve timing was varied in real time based on the the demand.

 

Conceptually, it's the ultimate.  The practicality is all that remains to be worked out.  Kind of a lot of "stuff" to haul around with you, though, and the compressor has to be powered up before the engine will run, so belt driving it is not a complete solution.   On the other hand, Ducati did campaign an engine that used pneumatic valve "springs" some years ago, so it's obviously not totally impractical. 

 

The ability to switch to two-stroke cycle operation would depend on there being a supercharger, since the intake ports are in the head.  The power, most of the exhaust, and nearly all of the intake cycles have to be able to happen on the same down stroke, and with the poppet valve layout, only a supercharger can accomplish that.  But it's cool.

Volvo Honda and whatever have tinkered with but Ford, Caterpillar, and International have been doing it since the early 90's.  The 7.3L Powerstroke engine was the fist to use hydraulics to operate fuel injectors.  The technology was developed by International.  Caterpillar uses it on their smaller engines (9L and under).  It has been used all of Fords diesel engines since 1993.  International has developed a version of the same technology that allows for camless control of valves.

Edited May 9, 2014 by 1987CR250R

[grayracer513]

Operating fuel injectors at 2500 RPM is a little different deal than controlling valve timing at more than 15K, though, wouldn't you say?

[1987CR250R]

Operating fuel injectors at 2500 RPM is a little different deal than controlling valve timing at more than 15K, though, wouldn't you say?

 

Considering most electronic engines have 3-5 injection events per power stroke at 2500 rpm and valves only have to open once every other revolution.  I would say not so different.  Take a look at International.  They have built a fully camless 4-stroke engine.  I personally hate the hydraulic unit injectors (HEUI) but they do work and the same system does work for valve control too.

Edited May 9, 2014 by 1987CR250R

[grayracer513]

Considering most electronic engines have 3-5 injection events per power stroke at 2500 rpm and valves only have to open once every other revolution.  I would say not so different. 

 

And an injector moves how much weight over what distance, as compared to 50 grams or better of exhaust valve over 10mm or more at a similar frequency?

[1987CR250R]

Well they use 3000 psi through an 8-1 intensifier to make 24,000 psi.  The force is available.  The volume flow is certainly an issue.  The system really exists as a band-aid to bring older engines into emissions compliance.  The camshafts were not strong enough to support such high firing pressures.  Never the less it is a very capable and adaptable system.  It would be interesting to see what could be done if they used a separate less viscous oil as the driver instead of crankcase oil.

 

Another reason for the use of the system is the ability to pop the injector open more quickly than a cam can which eliminates some of the dribble on the initial opening.  This would certainly be beneficial for valve control as well because the faster you can open the valve the more air you can breathe into the engine within the allowed duration (making more torque).  That was one of the principal reasons roller lifters gained favor because they allowed faster opening ramps for the valves.

Edited May 9, 2014 by 1987CR250R

[grayracer513]

That was one of the principal reasons roller lifters gained favor because they allowed faster opening ramps for the valves.

When Chevrolet started using roller tappets in their V8 engines in the mid '90's, it made a huge positive difference in engine performance even with all the emissions requirements, for that very reason. 

[EnglertRacing]

When Chevrolet started using roller tappets in their V8 engines in the mid '90's, it made a huge positive difference in engine performance even with all the emissions requirements, for that very reason.

Actually.

Slower acceleration but greater velocity than a large flat tappet unless an inverse radius or hollow flank profile is used.

[EnglertRacing]

The big problem with these designs that energy is not returned to the system.

On a normal cam & spring design, torque is required to ramp the valve up the cam and compress the spring. After peak lift is achieved the spring "drives" the camshaft and returns its energy to the timing drive and then to the crank.

On a pneumatic/ hydraulic/ electric system there are large parasitic losses - whether they are offset by the performance gains of the additional valve duration depends on the design.

At max rpm I don't think its putting much if any energy back into the camshaft.

The spring is struggling to close the valve and the follower just barley touching thr camshafts.

http://s163.photobucket.com/user/GrocMax/library/valvetrain%20videos?sort=3&page=1

[grayracer513]

Actually.

Slower acceleration but greater velocity than a large flat tappet unless an inverse radius or hollow flank profile is used.

 

Degree one of them out. You'll find the same opening/closing range used as with flat tappets, but more dwell time at full opening.  Draw your own conclusions as to the tappet acceleration. 

 

Flat tappet systems are limited by the fact that the lobe must not contact the lifter face other than squarely on the face.  Running the lobe contact over the edge is pretty much instant doom.  Roller tappets can support angular contact out to 30 degrees or more.  I know the point you're trying to make about the radius of the roller being a factor, but that's compensated for.

[EnglertRacing]

 

I actually have about 400 graphs and master lobes for flat tappet cams circa late 1960s in in my shed right now they came with my camshaft grinder ?

In my conversations with the previous owner Jim From Davis marine in Anaheim 100 years ago, who's been my friend for a long time the roller cams require an inverse radius to beat (out accelerate) a flat tappet off the seat to x amount of lift.

 

Now this is all dependent on roller/tappet to core diameter. but for good old pushrod ford or chevy with typical core sizes and and mushroom lifters, and standard size rollers.... what these graphs depict are about right.

 

Mr. Jim the previous owner of my cam grinder keeps asking me go to lunch with him and Mr. Ed Iskenderian.  I should take him up on this because Ed is really, really old but unfortunately i don't get to do anything besides go to work :[

 

Im not a proponent of flat tappet cams or anything.

If anything I like my cams roller with a inverse radius.

[grayracer513]

Now compare the flat tappet cam for a '94 5.7L Chevy V8 "R" from a truck to a '98 5.7L "R" cam, then the cams from a '92 Z-28 and a '98 Z-28.  Apples to apples.

 

Aftermarket performance cams for the automotive world are a different situation.

[Johnny_On_The_Spot]

At max rpm I don't think its putting much if any energy back into the camshaft.

The spring is struggling to close the valve and the follower just barley touching thr camshafts.

http://s163.photobucket.com/user/GrocMax/library/valvetrain%20videos?sort=3&page=1

 

Obviously if you run them at a high enough speed the valve will float - but this is outside normal operating ranges anyway.  Your average car is really only used between 1500-4000 rpm realistically, and if the system is designed properly for a bike it shouldn't float either.

 

but it is a fact that the energy is returned to the cam drive, minus inertial and frictional losses.  Look at "CTA" phaser actuation, the cam torque is actually what is used to index the cam, not oil pressure.  

Edited May 16, 2014 by KX250Fmotoracer

[grayracer513]

There is "some" energy returned to the cam by the valve springs at low RPM, but it is limited to the energy required to compress the spring, which is in reality a very small part of the drive losses in a camshaft.  At anything beyond about 3000 RPM (1500 at at the cam), the rotational resistance caused by the angular contact with the lifters and valve train inertia (which reduces the energy returned by the spring even further) far exceed that returned to the cam, even without valve float.

[EnglertRacing]

another way of looking at it is.

the valve is not closed for free the energy is used to close the valve is put into the spring when the camshaft opens the valve.

so sure some small amount of energy is put back into the cam from the spring. but non anywhere near what is put in. the valve doesn't close for free...

valve springs get hot, wheres the heat come from?

work put into them

 

and another way of looking at it is. if the valve doesn't approach float about 1 rpm after the rev limiter then your closing ramp is too slow, and you don't have the area under the curve that you need.

 

there is huge advantages to having nearly infinite control over valve timing and lift.

first off, you don't need a throttle plate any more, throw that in the garbage.

now under all operating conditions the ports pressure is closer to atmospheric and the density is higher

at part throttle on a conventional system, there will be manifold vacuum, using a variable lift/timing system

your basically at atmospheric pressure everywhere in the port and just opening the intake valve 0.01"

 

your "cam profiles" are limitless

you can have a map of profiles for each throttle positions and rpm

 

 

 you can more closely approach a digital or square wave than a lift curve,, where the valve is instanly open, which may be great for the intake but

engines are finicky. its actually been seen that they are liking slower ramp rates on the exhaust which still boggles my mind why they dont like more area under the curve.."hey little thumperoo why do you like a slow exhaust cam" i dont know.

 

but all of this would be so easy to sort out with a cam less system.

the benefits of basically having nearly 100% torque at all rpms due to the timing beign right for all engine speeds far outweighs the fact that a spring puts a fraction of its energy back into the cam at running speeds

[Johnny_On_The_Spot]

There is "some" energy returned to the cam by the valve springs at low RPM, but it is limited to the energy required to compress the spring, which is in reality a very small part of the drive losses in a camshaft.  At anything beyond about 3000 RPM (1500 at at the cam), the rotational resistance caused by the angular contact with the lifters and valve train inertia (which reduces the energy returned by the spring even further) far exceed that returned to the cam, even without valve float.

 

Do you have anything to back this up?  

 

The torque into the cam exceeds the torque returned to the cam, but by a surprisingly small amount.  I used to work at a tier 1 cam phaser supplier, so I have a a decent handle on these systems - there is certainly a negative and positive "torque signature" from a cam on a running engine.  They are actually surprisingly efficient.  (Forgive the terrible video - but this gets the idea across:

)

 

There are boatloads of data available on parasitic losses from cam drives - here is an example of one on pg 114.  You can easily see the positive torque as the spring extends: http://books.google.com/books?id=oSgu2fFZVeoC&printsec=frontcover&dq=Tribological+Analysis+and+Design+of+a+Modern+Automobile+Cam+and+Follower&hl=en&sa=X&ei=Ndh2U6aGAsSTqgaFq4CYDQ&ved=0CC0Q6AEwAA#v=onepage&q=Tribological%20Analysis%20and%20Design%20of%20a%20Modern%20Automobile%20Cam%20and%20Follower&f=false

 

Yes, there are losses - springs arent 100% efficient and convert some energy into heat.  High spring preload causes excessive friction when the cam is on the base circle, etc.

 

One of the easiest ways to make the system much more efficient is to use a "variable spring" that only requires high spring force at high rpm - this reduces most of the parasitics at low load (where the largest percentage of valvetrain losses occur)  

Edited May 17, 2014 by KX250Fmotoracer