Variable valve timing

[Engineer]

Hy guys

I am a graduate student of mechanical engineering college on University of Ljubljana
And I was thinking abouth the theme 4 my diploma.
I have a question about wariable valve timing.I know that wariable valve timing system works by rotating (correcting)the whole cam shaft,al cylinders at once.
Now I would like to know what you think of the idea,of correcting all valves independantly.
I mean that if I want to correct just one or two cylinders,and the position of valves on other cylinders stays unchanged.
Please tell me if this would have any impact on perfomace,emisions or anything else.
What advantages and disadvantages would that kind of engine have.
It's just an idea,so please don't ask me why do I want to do so.
I thank you 4 your answers

Have a nice day!

[Ciro Pabón]

Well, I can tell you that my dealer showed to me a pamphlet on BMWs Valvetronic (if I remember well the name) which has dual valve control, which means it controls separately intake and exit valves but not individual cylinders (or so I think).

The pamphlet mentioned a 10% increased economy and emission reduction. The cars using this system do not have a conventional throttle, they are throttled through the intake valves timing variation. They talked about how this system avoids pumping losses, and attributed most of the economy, etc. to this fact, but I guess you already know that. However, this gives you a ballpark figure of how large a return you could expect from your investment.

I also know that some Lexus have electronic valves, so perhaps in this case each cylinder is controlled independently, but they'd need a lot of sensors, I guess.

[majicmeow]

Hey guys,

BMW's system uses Valvetronic to adjust the intake valve lift, which allows for the throttling of the engine to be done by the intake valves. The camshaft timing is an independent system that controls the exhaust and intake separately.

The electronic solenoid valve is not yet in any production cars, as far as I know...
The electronic system that Lexus uses (that Ciro mentioned) is there to provide a faster/backup valve adjustment in situations where oil pressure is limited (cold weather starting) and is still limited to adjusting the complete camshaft. It simply supplements the hydraulic system.

Seeing as the pistons on an engine are always in the same place in relation to each other, the need for independent adjustments would simply overcomplicate what is already possible. The camshaft adjustment on the current systems is fast enough and versatile enough to adjust for independent cylinders on a limited basis (in between firing cycles on the same bank of cylinders)

[manchild]

You'll find a lot of data for your theme here:

http://www.palivaventi.de/english/pali-Entry.html

http://www.motorlexikon.de/?I=6592

great e-book on the subject http://books.google.com/books?id=tz_wTd ... lt&f=false


First ever variable valve management - Louis Renault 1902

[Carlos]

An overview: Variable valve timing:
http://en.wikipedia.org/wiki/Variable_valve_timing

[riff_raff]

Engineer,

You have asked a question that would require many textbooks to fully answer.

Variable control of engine poppet valves can involve many variables. It can affect valve opening and closing points, duration of valve opening, lift of the valve, overlap of intake and exhaust valve openings, differential lift and timing between adjacent intake/exhaust valves in a common cylinder to effect gas flows, deactivation of all valves in certain cylinders to vary effective engine displacement, variation of intake/exhaust valve timings to alter dynamic compression ratio or produce Atkinson/Miller type combustion cycles, variation of intake/exhaust overlaps to change EGR rates, or deactivation of intake valves can even be used to improve engine braking (ie. diesel Jake brakes).

With regards to F1 engines, variable valve systems are not technically allowed. The camshafts themselves normally have some variation designed into the individual cam lobe phasing, from one end of the cam to the other. This is to compensate for shaft torsional deflections that become greater as the lobes get farther away from the shaft's driven end.

Personally, I think VVT should be allowed in F1.

[Tractorboy85]

Hi Engineer,
I did my thesis on testing an electrohydraulic valvetrain, so I guess I'm coming from the same place as you before doing my thesis...

Fundamentally due to the reciprocating nature of an internal combsution engine using camshafts to control the valve timing is a pretty convenient way of making sure that the engine always breaths at the same point in its operating cycle. To make each cylinder do something different I'm not sure has many advantages, ceratinly from a control and NVH point of view, due to the inter cylinder differences. Also can you imagine the amount of work that would be required to map an engine in any spehere that had different cylinder to cylinder valve timing?

Despite this there are a few interesting things you could play around with if you had complete control over each valve in a cylinder such as air charge motion within a cylinder to improve fuel consumption and emissions. Another obvious benefit is cylinder deactivation, which if you have a large enough engine, allows you to 'turn off' a certain number of cylinders, making the other cylinders work harder and in a more optimum part of their fuel consumption map. There are already quite a few mechanical solutions out there, including lost motion followers, however that doesn't mean you couldn't design your own?

However from a global engine point of view; variable valve timing certainly makes a lot of sense, the arguments for which have been laid out in many papers already. The big pitfall of the Valvetronic and other mechanical systems is the inflexability to de-couple valve opening duration with vavle lift. Although the valve timing in this sense can be phased relative to piston position using the fast acting VANOS phaser the system is by no means truly variable, although the chaps from BMW have gotten pretty close. Multiair or TwinAir which is currently being used in smaller FIAT engines is probably the next step on to a more fully variable system.

Electrohydraulic or purely electronic valvetrains as these offer the most amount of flexability for controlling IC engines in the future. Plus there is a whole load of interesting strategies you can pursue with these systems that mechanical variable systems can't do, such as opening the exhaust valve shortly during the intake valve opening period to meter internal EGR for example.

Hope this helps

[WhiteBlue]

Personally, I think VVT should be allowed in F1.

I think they will allow it for the 2013 season with the new rules. It would make no sense to push for fuel efficiency and deny the tools that would achieve it.

[tok-tokkie]

It is not a field I have followed. On CNC machine tools servo motors have taken over as the drives for the axes allowing precise control of each axis irrespective of the others by the computer. Separate servo motors to drive each cam shaft would allow very flexible valve timing. Initial position zeroing at start-up is a small complication but the consequences of power failure would be dramatic & extreme.

[Edis]

This thesis can perhaps be of some interrest: http://lup.lub.lu.se/luur/download?func ... Id=1749967

It's about pneumatic hybrids using the engine as an air compressor, made possible through free pneumatic valves. Chapter 4 describes variaous variable valve systems and control strategy.

As for adjusting valve phasing for individual cylinders I can't see any significant advantage with this. I would think that the next step is toward variable valve lift and then improved systems where you have greater flexibility controling both lift and timing more independently.

[riff_raff]

Edis,

With poppet engine valves, the relationship between duration and lift are not completely independent. In an ideal valve motion, the valve would instantaneously lift to its point of max flow and then close in the same manner. But due to physics, less duration usually means less lift, even with electrohydraulic or electromechanical actuation.

As tok-tokkie notes, pure EH or EM valve actuation systems are not generally fail safe. This is one major reason automakers have not yet fully embraced them.

riff_raff

[Mystery Steve]

Tractorboy,

Out of curiosity, do you have a link to your thesis anywhere?

[Tractorboy85]

Alas unfortunately my thesis was carried out in conjunction with an OEM within a institute at my Uni, so I had to sign a confidentiality agreement. As far as I know nothing that I did is in the public domain.

Riff raff you're correct on valve lift and duration being interlinked, but only to a point. Provided the elctical signal duration pulsewidth you use for controlling an electrohydraulic or electromagnetic actuated poppet valve is large enough to get you to max valve lift, then increasing the signal pulsewidth duration will give you londer valve opening duration. Of course shortening the signal duration from this point will mean the valve lift will be shorter and also the duration; and then they will be interlinked.

As for VVT in F1, I'm not sure it would make much sense. I've not actually seen the duty cycle for an engine being used around an F1 track, but I can imagine you either spend a lot of time at full load and the maximum speed, at idle, or going from a lower engine speed and load to the max speed full load condition.

One of the big drivers for VVT is that it doesn't compromise your idle quality (which was always the case for an engine with fixed valve timing) because you can move to min overlap conditions, which is normally at odds with your max power/torque condition where you typically have more overlap. Idle quality is only important because OEMs sell their products to millions of people that wouldn't put up with misfires or partial burn combustion events, which is probably something that isn't as important to F1 teams.

A secondary driver for VVT is part load emissions through internal EGR by using larger valve overlap, which is obviously a big issue for OEMs especially with ever stringent emissions regulations. Again I don't know the F1 rulebook but I assume there is nothing about reducing emissions.

The third issue I can forsee is related to the transient condition where you come out of a corner at part load and a low engine speed and go up to max engine speed at full load in a very short space of time. Again probably need to flesh this one out with some figures, but I can't see that there is a camphaser or camphaser technology at the moment that that is compatable with the camshaft accelerations in a transient condition. I think the best figures I have come across are around 400 CrankAngledeg/s with a electromagnetic phaser, whilst the fastest hydraulic phaser is around 330 CrankAngledeg/s. I guess also there is a parasitic loss effect with the additional hardware that is necessary to control the phasers, plus a weight penalty for the hardware that an F1 team might not want to incur. Whilst I agree that having VVT would allow you to optimise your performance at each speed at full load, is there a camphaser that’s quick enough to change your timings fast enough to make use of the technology?

I guess VVT from the point of view of throttle less load control would make much more sense, as you could significantly reduce your engine’s pumping work and increase the net energy output of the engine. However this again is only really applicable at part load, so probably not that relevant to an engine in F1.

Just a few thoughts, I might have missed something though as to why VVT would be applicable to F1? I obvisouly assumed you only would want to control the phasing of the intake and exhaust valves? What sort of config did you have in mind?

[Edis]

Edis,

With poppet engine valves, the relationship between duration and lift are not completely independent. In an ideal valve motion, the valve would instantaneously lift to its point of max flow and then close in the same manner. But due to physics, less duration usually means less lift, even with electrohydraulic or electromechanical actuation.

As tok-tokkie notes, pure EH or EM valve actuation systems are not generally fail safe. This is one major reason automakers have not yet fully embraced them.

riff_raff

With electromagnetic, electrohydraulic or electropneumatic free valves the valve lift profile have three phases. You have open, dwell and close. If the duration is long enough to lift and close you can increase the duration by simply increasing the dwell phase. This is much like how a fuel injector needle operate, the main difference being the much higher lift in the case of an engine valve.

Unlike a cam based system you can also open and close the valves whenever you want.

But EH components tend to be expensive and you need a system that works both in hot and cold environments when the viscosity of the hydraulic fluid will be very different. EM systems have issues with packaging and high seating velocities, so some sort of damping device is probably needed.

If each valve is equipped with a lift measurement device I don't think fail safety should be a big issue.

As for VVT in F1, I'm not sure it would make much sense. I've not actually seen the duty cycle for an engine being used around an F1 track, but I can imagine you either spend a lot of time at full load and the maximum speed, at idle, or going from a lower engine speed and load to the max speed full load condition.

With the current F1 engines I don't think the use of phasers would be possible, adjusting even a few degrees would probably mean that the pistons would collide with the valves. The short stroke, high compression ratio, high valve lifts and long valve lift duration means that the valves follows the pistons at a distance of just a few tenths of a mm when the piston is close to top dead center.

[PlatinumZealot]

Hy guys

I am a graduate student of mechanical engineering college on University of Ljubljana
And I was thinking abouth the theme 4 my diploma.
I have a question about wariable valve timing.I know that wariable valve timing system works by rotating (correcting)the whole cam shaft,al cylinders at once.
Now I would like to know what you think of the idea,of correcting all valves independantly.
I mean that if I want to correct just one or two cylinders,and the position of valves on other cylinders stays unchanged.
Please tell me if this would have any impact on perfomace,emisions or anything else.
What advantages and disadvantages would that kind of engine have.
It's just an idea,so please don't ask me why do I want to do so.
I thank you 4 your answers

Have a nice day!

I think it would be very useful in engines that disable some of the cylinders. I can see it reducing pumping losses.

If you are going to utiliize the rotation cam to control the valves, it is going to be tricky to have the cam shaft controll the timing all 8 sets of lobes separately (if it is a 4 cylinder engine).

As you can see below, the timing is controlled by an internal helical gear set.You need to have 4 separate sets of them since you want to control each cylinder individually. It is not going to fit in a conventional cylinder head.



Ultimately valve timing change the phase the valve lift. Agreed? So in essence it is controlling valve lift, just shifting that lift in relation to TDC or whatever reference.

It follows that it would make sense to use a method of fully controllable, and fully variable valve lift instead. So, you do not change the rotation of the cam. - let that stay fixed. And instead you put in place a independently controlled valve lifting mechanisms on each cylinder. You will get the effect of both cam-phasing and cam timing without needing 4 sets of cam timing gears (which take up too much space).

the unreleased "Advanced vtec" from Honda can fit your requirement. It is infinitely variable so you can get any valve lift you want within a desired range.
You just have to find a way to make it fit each cylinder individually and be independent.

http://www.vtec.net/news/news-item?news_item_id=659664