Variable Valve Events - Why?

[scarble]

Hello everybody, I'm new here. I've read a few threads on PH before and you seem like nice and well informed people.

I've got a bit of an odd question to ask and it's rather hypothetical, not relating to any specific vehicle or engine.

If you could, theoretically, have valves which opened and closed instantly and while open, offered zero resistance to flow, would there be any need for:
(i) Variable lift? Would you ever want to deliberately impede flow?
(ii) Variable timing. Would you ever want, for example, the intake valve to open before or after TDC.

Is the use of VVT purely due to valve opening/closing times?

This might well sound quite stupid and I apologise for that. I'm here to learn.

[Mr2Mike]

scarble said:

If you could, theoretically, have valves which opened and closed instantly and while open, offered zero resistance to flow, would there be any need for:
(i) Variable lift? Would you ever want to deliberately impede flow?

I can think of a couple of cases:
1) Impeding the gas flow can be used to increase gas speed which can be advantageous to keep fuel suspended in the air.
2) Variable lift can be used to throttle an engine

scarble said:

(ii) Variable timing. Would you ever want, for example, the intake valve to open before or after TDC.

I think the potential advantages would be significantly reduced by your magical valves, but there could still be gains from variable timing. It's possible to have over 100% volumetric efficiency in a normally aspirated engine, but to achieve this requires careful matching of valve timing with engine speed.

[scarble]

Ok, air speed = fuel suspended makes sense to me.
What if fuel is direct-cylinder-injected?

Mr2Mike said:

I think the potential advantages would be significantly reduced by your magical valves, but there could still be gains from variable timing. It's possible to have over 100% volumetric efficiency in a normally aspirated engine, but to achieve this requires careful matching of valve timing with engine speed.

could you elaborate?
Imagine that an intake valve opens instantly and completely at exactly TDC on the induction stroke and then closes instantly and completely at exactly BDC, reverse for exhaust valve/stroke.
Leaving aside emissions and EGR and such for now.

[anonymous-user]

Unless you can somehow invent "inertia free" air, then just having valves that open instantly is no real advantage! You need to accelerate and deccelerate the aircharge (and the exhaust flow), which, because they have mass, takes time. Varriable valve timing, is both about getting the timing correct with regard to valve pressure ratio, but also with regard to aircharge inertias.


You can reduce pumping losses with lower valve lifts at part throttle to choke the intake mass flow at the valve rather than in the plenum (throttled operation) this can give reduced "pumping" losses (BMW valvetronic et-al).

However, if you have the ability to instanty control valve lift, just controlling the valve timing would be sufficent (with the constraints of geometrical valve to piston clearances etc)

[scarble]

Max_Torque said:

Unless you can somehow invent "inertia free" air, then just having valves that open instantly is no real advantage! You need to accelerate and deccelerate the aircharge (and the exhaust flow), which, because they have mass, takes time.

I see, so an intake valve has to open before TDC to allow time for the air to accelerate?
Does intake valve timing become less of an issue with forced induction? Although I can imagine that as one cylinder closes and another opens the bulk of the pressure will be focused on the former piston or that area of the manifold?

You guys are awesome by the way.

Edited by scarble on Friday 11th March 19:00

[anonymous-user]

Contrary to popular belief, turbocharged engines do not "blow air" into the engine, they just increase the air density in the manifold (the actual pressure ratio across the engine is worse than for an NA engine) Because it is more dense, this air has more inertia than for an NA engine at WOT........

[scarble]

gah, making me look stupid now
Ok, I think I've got a better idea of this now anyway.

I feel like I sort-of knew all this already but it helps me to think to bounce things off other people.

Thank you.

I've got this old book, from 1933 called "Introduction to Internal Combustion Engineering" and among many things it has a drawing of a 1-valve per cylinder engine. Not 1-intake / 1-exhaust. 1 valve total. The two manifolds converge and there is a swing valve in the port that switches between the two.

If you had an intake manifold with a valve near where it diverges to two cylinders, could you use that to switch from one port to the other, would it reduce inertia?

I'm not sure on my terminology here, how do you differentiate the bit of pipe where the air comes in from the individual bits for each cylinder? Does it become a port once it is only feeding one cylinder?

edit: Maybe "reduce inertia" isn't quite right, will it clean up airflow? Does forced induction really not blow? In any form, not even a "blower"?

Edited by scarble on Friday 11th March 19:18

[davepoth]

Certain superchargers do blow air in, but they are in the minority.

[anonymous-user]

davepoth said:

Certain superchargers do blow air in, but they are in the minority.

depends upon the overall pressure ratio, if your boost pressure (actually intake runner "total" dynamic pressure) is larger than your exhaust manifold total pressure this can be the case. generally for road cars this is not the case. (as soon as you get a positive pressure ratio across your engine AND increase charge density, then you can make massive power (F1 turbo anyone?? ;-)

You have to ensure you use total pressure not static pressure (boost pressure is usually measured as a static pressure)

Edited by anonymous-user on Friday 11th March 21:41

[anonymous-user]

scarble said:

If you had an intake manifold with a valve near where it diverges to two cylinders, could you use that to switch from one port to the other, would it reduce inertia?

er, no !


you still have to bring the aircharge to a compete stop during the exhaust event, and then accelerate it back up to speed againt to fill the cylinder. (typical mean velocity of intake charge during the intake event is between 80 and 100 m/s)

Added to which, using a single valve system would undoubtly cause massive intake charge upheat (reducing charge density and hence intake massflow) and also result in large crevice volumes of exhaust gas, displacing some of the precious fresh air charge, leading to another reduction in vol eff!

[Pumaracing]

scarble said:

If you could, theoretically, have valves which opened and closed instantly and while open, offered zero resistance to flow, would there be any need for:
(i) Variable lift? Would you ever want to deliberately impede flow?
(ii) Variable timing. Would you ever want, for example, the intake valve to open before or after TDC.

Is the use of VVT purely due to valve opening/closing times?

Ignoring for the moment what might be meant be zero resistance to flow, it's easy enough to posit the hypothetical case of an actual poppet valve opening and closing instantaneously and seeing how an engine might run where such valves operated only at TDC and BDC.

C.F.Taylor in his books on the IC engine discussed what he called the Z factor which is a measure of performance derived from what I call the "Camflow Curve". Imagine an actual cam profile and an actual valve flow curve in CFM measured say every 50 thou lift increment until the valve reaches its peak flow.

It's not hard, certainly with computers, to interpolate the valve lift curve against the valve flow curve to arrive at a graph of flow versus crank degrees. So for example every degree or 10 degrees or whatever interval you opt for you look up the valve lift from the camshaft profile, then see what the valve is flowing at that lift and continue doing that until you have a graph showing valve flow CFM on the X axis and crank degrees from valve open to valve close on the Y axis.

The total area under that curve is a very good measure of potential engine power. Make that area bigger by fitting bigger valves, using better port shapes etc and the percentage gain in graph area translates quite nicely into extra power.

If you have such a computer program (I do) it's easy to compare the area under the Camflow curve of an actual real world cam profile against a theoretical valve that opens instantaneously at TDC to the same peak lift as the camshaft and closes again at BDC.

Taking a real flow curve and cam profile, from a standard road going Ford CVH XR2 as it happens, the results are quite surprising. The actual cam has a duration of about 260 degrees because it opens before TDC and closes after BDC and even though the theoretical cam opens instantly to the same peak lift for the whole of its 180 degree duration it only has about 10% more Camflow area under the curve.

Substitute a real cam of race duration, 300 degrees or so, and the theoretical instantaneous cam (TI cam) now has only about 85% of the Camflow area. So we can say that the TI cam can potentially generate a flow area, and therefore power potential, about the same as a fast road or mild rally real world cam.

However that's only scratching the surface of your exam question.

If you confine the valve events to TDC and BDC so there is no valve overlap then all sorts of nice things that happen in real engines fail to occur.

1) With no valve overlap there is no scavenging of the exhaust gas residuals in the combustion chamber so when the inlet valve opens the incoming charge faces a chamber full of old exhaust gas that must be at or indeed more likely considerably above atmospheric pressure. With an engine of average compression ratio that means a full cylinder at BDC with at least 10% old exhaust gas in it.

2) Pumping losses with an exhaust duration of only 180 degrees would be high.

3) With no valve overlap there is no opportunity for pulse charging using the exhaust manifold pressure waves to suck in more intake charge and generate greater than 100% volumetric efficiency. Real world race engines can operate at 130% VE or more. The TI cam engine has no hope of doing so.

4) With an inlet valve opening at TDC there is going to be no actual inflow into the cylinder until the piston starts to move down the bore and generate a depression. That wastes much of the total 180 degree cycle as the piston doesn't really move very far for quite a while either side of TDC.

5) In a real engine, especially a properly designed race one, it isn't really the piston that controls the cylinder depression. It's the exhaust system pressure waves that do that leading to high intake flow rates even when the piston isn't really moving very fast down the bore.

In summary the 180 degree TI cam engine would be something of a dog. It would be very tractable as there is no valve overlap so it would pull hard down to potentially zero rpm but it would have less power and torque than a conventional road cammed engine, poor fuel economy due to pumping losses and no chance of matching a race engine on power.

The question becomes a bit more interesting if you posit instantaneously opening valves that you can open before TDC and close after BDC which you will certainly need to do if you want to maximise engine power and take advantage of pulse tuning. However whatever you do there is no single valve timing strategy that can be optimum at every engine rpm. You'll always ideally want VVC. Longer duration for high rpm and vice versa.

[scarble]

Max_Torque said:

Added to which, using a single valve system would undoubtly cause massive intake charge upheat (reducing charge density and hence intake massflow) and also result in large crevice volumes of exhaust gas, displacing some of the precious fresh air charge, leading to another reduction in vol eff!

I actually meant a valve to split intake air into the individual ports but as well as the usual intake valves, I did not mean mixing exhaust/intake air. I was just wondering if there is any "waste" of energy by air flowing into the manifold or port or whatever for a closed cylinder. Although I guess that's far less of an issue if induction, even when "forced" is done by suction due to exhaust pressure wave/piston.
The whole idea sounds less good in the morning.

Great post Pumaracing, it all makes a lot of sense.

Thanks guys for being patient with me and sharing your vast knowledge. ICEs are tricky things to really wrap your head around. Can anyone recommend some good starting places? University, as you can probably tell, didn't really cover things like VVT and exhaust pressure waves

So if I understand correctly, this exhaust pressure wave thing would still apply in a forced induction engine (let's say turbocharged in this case) because forced induction doesn't actually "blow"?

[chuntington101]

Max_Torque said:

depends upon the overall pressure ratio, if your boost pressure (actually intake runner "total" dynamic pressure) is larger than your exhaust manifold total pressure this can be the case. generally for road cars this is not the case. (as soon as you get a positive pressure ratio across your engine AND increase charge density, then you can make massive power (F1 turbo anyone?? ;-)

You have to ensure you use total pressure not static pressure (boost pressure is usually measured as a static pressure)

Edited by Max_Torque on Friday 11th March 21:41

Max, how do you toal dynamic presure for both the intake and the exhaust? Also what way can you increase the intake manifold presure (dynamic) while reduing the exhaust presure (again dynamic)?

Thanks,

Chris.