What does "lift" do to air speed?

[Jim Ratto]

Curious what the consensus is on effects of increased lift on air speed through port/throat. For the longest time, the sales push on high-ratio rockers on stock/small cams was that the user would gain power all around, bottom to top, without increasing overlap, and without losing low and mid rpm torque/pull.
If you dial a cam profile @ lobe vs @ valve with 1.25, 1.4, 1.5 etc rockers, you'll see duration does change.
But speaking valve lift, meaning cam X and cam Y have same seat to seat duration, lobe center, but cam X lifts 0.500" and cam Y lifts 0.435", would the reduced curtain area of cam Y influence a higher air speed thru intake?
I've read if you give up lift, you can run more duration. And- if you increase lift, it's best to give up x-amount of timing.
Thanks

[Taylor]

How about pumping some numbers in here?  http://www.wallaceracing.com/machcalc.php

[Jim Ratto]

Interesting.
I have a formula I use from a cam class I took from Dema Elgin, it is actually published in AG Bell's book "Four Stroke Performance Tuning"
It uses air speed through port as a constant, dependent on chamber shape and cam profile then determines RPM @ whch peak power will be produced at, other variables being valve diameter and cubic cm per cyl.
In same publication, Bell discusses the "give and take" of increasing lift/reducing duration or vice versa, when it comes to building a powerband for semi-race or rally tune.

In recent years, there was another post on cam profiles and how well suited the Engle W series cams are to making good port velocity. I'm trying to figure out if that was an opinion or if there is truth behind that. The highest lift of any of the available W series "street cams" is right around .460-.470"
My thought process for the last 15 years has been lift is a variable determined (theoretically) by valve diameter and chamber design. And "mechanically" by P to V interference, spring type, life of heads.

Just thinking about things.

Thanks Taylor,

[modnrod]

Putting 1.4 rockers on a stocker does work all the way through the rev range without any holes, but only really because the stocker heads are so limited to begin with, it's a simple easy bolt-on Band-Aid.

I have sometimes found more duration and less lift on an intake limited motor, especially one with restrictive manifolding (like a centremount Solex........), combined with wide flat seats, makes a better average than high lift low duration. Single Port motors seem to like it even more.

I think maybe the air sort-of "skids" past the valve rather than flowing, maybe it fills the chamber better or burns better with the mixture motion this makes perhaps? 

[Peter]

Some guy told me long duration cams are a bit old technology...
He says high lift short duration gives best overall performance,
Modern engines dont have those heavy pushrods , heavy valves etc, so i guess thats why they can get away with it.
Like fk4x series, too bad they are bad for the valvetrain

[Martin S.]

Whenever someone mentions 'Lift' the video I made when Steve was assembling my motor comes to mind. I got lots of response to it. 'Slight lift' he says, ya right. He obv likes lots of it, and big valves too, the bigger, the better he says. Great reliable daily driven engine, nice smooth power and incredible torque was the result. http://youtu.be/xX3iaHUI-SA

[John Maher]

Curious what the consensus is on effects of increased lift on air speed through port/throat.

Port velocity is a function of flow and the cross sectional area of the port:
(CFM x 2.4) / Cross Sectional Area = Port Velocity
Note: CFM tested at 28" H2O, CSA in square inches, port velocity in feet per second.

The only influence the cam has in this context is to provide the various lift increments at which flow is measured. Looking at the equation above, an increase in cfm will result in higher port velocity. An increase in port size (CSA) with no gains in flow will reduce port velocity. Also velocity varies depending on where in the port it is measured e.g. the largest area (slowest port speed) tends to be immediately below the carb, while the throat or seat ID ought to be the smallest/fastest.

If you were designing an engine from scratch with specific requirements in mind, the cylinder heads (port velocity and cfm flow) would be figured out and optimised well in advance of cam choice.

In short, it's the cylinder head and intake system that determines port velocity, not the cam - assuming the cam has sufficient lift to exploit the head's max flow potential. If fitting higher ratio rockers gives you higher port velocity, it's because the extra lift delivered more flow.

[BeetleBug]

If you were designing an engine from scratch with specific requirements in mind, the cylinder heads (port velocity and cfm flow) would be figured out and optimised well in advance of cam choice.

Dam! Now you tell us 

[neil68]

When I was investigating which type of camshaft to install in my 2332 cc street/strip Beetle, one trend started to emerge. It became apparent that for a heavy, stock weight Beetle having a shorter duration and higher lift was recommended to improve 1/4-mile times.

[nicolas]

As much as i can understand what is posted, i am most definitely lost!

i always learned the valve should open immediately and completely (sort of what Koeninsegg does in their tests), in this regard the FK4x cams are 'better' as opposed to other FK- series cams, but is this true?

and what does the duration affect then? is their a way to match the 'need' for fuel and air of a particular engine (CFM) and airspeed to the cam profile? and if so, how is this done?

[Jim Ratto]

I played with your calculator site some today Taylor. My combination appears to be "choked" at least by the numbers. But at this stage of my life and where I am with the hobby and car... etc, I'm happier to know springs will live longer, along with guides, etc. The days of .580" + are behind me. It looks like cranking lift up in your calculator, but leaving all else alone, I would gain efficiency. At a cost, of course.

[Taylor]

Well I ran the numbers on mine before I ran it on they dyno .  It comes in at .47 mach.  It ended up running at above 100% at 6900rpm that's as far as I'm getting into it.  Don't want to chase the number anymore than that.  For low lift, it seams a smallish valve and relatively small CSA is what you're after Jim.

[Fiatdude]

Koeninsegg shows this because of their air operated valves can get away with it -- --

when operating a lifter and rocker valve assembly, you can have a ramp speeds that are too extreme/violent to be supported/contolled by the available springs we have -- I remember many years ago when the huge roller setups were out for SBCs turning 10,000 rpms and they came up with a another set of 'valve' springs that were installed in the lifter galley on top if the 'normal' popular triple springs that were used then......

So the answer to big lift is BIG valves at lower lift that flow the same

[John Maher]

As much as i can understand what is posted, i am most definitely lost!

i always learned the valve should open immediately and completely (sort of what Koeninsegg does in their tests), in this regard the FK4x cams are 'better' as opposed to other FK- series cams, but is this true?

and what does the duration affect then? is their a way to match the 'need' for fuel and air of a particular engine (CFM) and airspeed to the cam profile? and if so, how is this done?

On paper the FK4* concept is sound... less duration for better throttle response at low and mid-range rpm and gaining area under the curve by accelerating the valve off the seat more quickly. Only problem is FK4*s are hard on parts.

More duration means more overlap. At low engine speed you lose intake through the the exhaust, lowering Volumetric Efficiency and dynamic compression. ie. less performance. For a limited rpm range engine e.g. drag race, that extra duration can be a plus - the overlap period can pull more intake charge into the cylinder at higher engine speeds if exhaust and intake lengths are tuned to the right lengths.

So the answer to big lift is BIG valves at lower lift that flow the same

Depends what the engine is going to be used for. Most engines work best with the smallest, most efficient port and valve combination that's capable of delivering sufficient airflow to support the horsepower target you're aiming for. The bigger the valve, the more lift required for it to work effectively and often a bigger, lazier port to accommodate it.

More on efficiency and valve size vs lift below...
==========================================================

Port velocity isn't a constant. It varies depending on flow, and flow varies depending on lift.
Port velocity has different values depending on where it's measured - as explained in my post above (port velocity formula).
But the area that's probably of most interest is valve curtain area - because it's constantly varying, unlike the rest of the port.



The table and graph below might help. The data I've entered is based on a 42mm intake valve and port I prepared for a 2276cc street engine. Entries in the top left are self explanatory:
42mm valve
Seat ID: 90% of valve diameter (1.488")
Valve stem: 8mm (.312")

Lower left table (Flow Bench Data) shows the flow bench results... CFM in .100" lift increments.



The maximum flow area available is determined by the smallest cross sectional area (CSA) in the entire intake system - usually the seat ID. In this example 1.633 square inches (area at seat ID minus area occupied by the valve stem).

At low lift, the curtain area is much smaller, e.g. 0.362 sq inch @ .100" lift. Therefore cfm flow is low also but because of the venturi effect created by the multi-angle valve job and the back of the valve, efficiency (flow per sq inch) is very high. Adding valve lift allows more cfm flow but check how velocity drops off - curtain area is increasing proportionally more than CFM. This continues up to the point curtain area equals the seat ID area. This always occurs at a lift figure just short of one quarter of the valve's diameter, regardless of valve size. The actual figure is closer to 1/4 of the seat ID, in this example that's around 0.370" lift. As valve lift increases beyond this point, the curtain area obviously increases but it's now the seat ID that's the limiting factor rather than curtain area and any gains in flow from here on are a result of the valve getting itself out of the way.

Both the blue and red lines on the graph will look very similar in profile for just about any intake valve and port you care to measure... as lift increases, gains in flow start to taper off at a certain point (where this happens depends on the ports, seat and chamber design) and you'll always see lowest seat velocity when lift is around 0.25 x valve diameter. This is why bigger valves need more lift.... if you don't lift 'em high enough, velocity and flow will disappoint.

Looking at the Velocity ft/sec column gives an indication of how the head might perform. A good street engine tends to work really well with 290 to 300ft/sec. Very high rpm ram tuned n/a race engines need slightly less velocity, say 270-280 ft/sec.

An even better indicator is the CFM/sq inch column. This indicates the efficiency of the head. A straight tube with radiused entry and exits (theoretically ideal port) will flow 147cfm/sq inch @ 28" H2O. A cylinder head port with twists and turns, valve guide, guide boss, valve stem etc can be considered astounding if it makes 130+ cfm/sq inch range. Heads of that quality are extremely rare!

If you start your build with a pair of super efficient cylinder heads (excellent cfm/sq inch figures) with port velocity in the range suited to the type of engine being built, you have potential to make very impressive torque and horsepower per litre figures... as long as you choose the right cam...

... and induction

... and exhaust

...and etc etc  

[Zach Gomulka]

Isn't there a "rule of thumb", so to speak, that the valve lift should be somewhere in the neighborhood of 35% of the valve diameter? If so, does this apply any differently to the exhaust valve?

[max, Der Bahnstormerz]

Shouldn't we be looking at where lift occurs in relation to the crank angle and rod length. Surely this determines where we will see the biggest gains in HP

[nicolas]

thank you for clarifying all of this.

i still have a question on my mind which is related to this topic. if i look at the figures of an FK8 and an FK10 cam, the lift seems comparable, but the duration differs quite a bit. so    a FK 10 'misses' out on lift in my eyes. yet it seems to be a very commonly used cam. the FK8 on the other hand has a bit fallen out of favor, but non the less i have read about engines with it making 200+ HP. not shabby at all in my book. but it is not the HP that is my point here, but more the FK8 being a better 'package deal' for compacter duration and decent lift, without being a more strainy FK4x series cam. Is the overlap of the FK10 not holding back the performance?

[Jim Ratto]

I would think, Max, yeah max lift would want to be optimized vs where piston is in its travel from TDC to BDC. I remember from the class I took 15 years ago, we were handed a simple drawing and it depicted the rod @ a 90 degree angle to crank throw and written below that was "max piston velocity"
this was related to indexing cam profile against piston position.
I'm just asking questions because I thouight we could all use a new topic to throw around here and this does really interest me.
I think another area to consider is when does flow stop, and at what degree ABDC? Another book I have shows studies that were done, measuring cylinder pressure throughtout the intake stroke, actually before and after it, if we're talking textbook TDC>BDC condition. What raised my eyebrow was that cylinder pressure became higher than atmospheric as piston approached BBDC, and well ABDC, but with intake valve still open. A symptom of ram effect?
As far as max lift vs piston position, again I think (considering typical deck height etc) in the real world, we are only talking about "adjustability" of maybe 10-12 crank degrees? Most of the street engines I have set up, I will time cam to cross over intake lobe peak @ 104ATDC. One I set up @ 102ATDC (2276, 46 x 38, 10.6:1, 280' @ .050, made 213hp @ 7100/175 lb ft @ 5500).

In the realm of long living healthy street engines, made to get you to work on time, etc... how much of a loss in VE does a cam with under .500' cause? That's what's really got me thinking. In terms of valve train wear and tear vs incremental hp gains, in an engine that needs to come on hard @ under 3500, is there real merit in running 0.500"

[modnrod]

In terms of valve train wear and tear vs incremental hp gains, in an engine that needs to come on hard @ under 3500, is there real merit in running 0.500"

For a 1584 motor, a 35.5mm valve is 42% size to bore diameter. A bit small for outright hi rpm power, but still works OK for low speed, like your "under 3500" example above. Using Zach's 35% figure above as a reference, that means lift around .489".
If you had a 2276 motor though, 42% sized valve is actually 40mm, and the 35% lift mark is now .551".

So the lift you run is tied more in to the valve/bore diameter and port size/volume/specs than just the outright lift figure itself.
Part of what MHR was talking about above with L/D ratio and port velocity.

[Taylor]

Jim,  I believe your comment about cylinder pressure being higher than atmospheric while the intake valve is still open is true.  But the big question is when it happens.... The answer to that, as usual,  is that it depends.  Lol.  Depends on what?   Rpm, overlap, intake valve closing events, exhaust size, area under the curve......We can take connecting rod length out of the picture as it has no effect.   You could neither run a rod short enough or long enough to matter much.   I sort of tend to believe that an under lifted valve size will kill air speed and that if lower lift is desired than valve and port size should also be reduced.

[John Maher]

Max and Jim,

On most VW engines, peak piston speed occurs around 75° ATDC, give or take a degree or two.

Changing rod length has minimal effect... on the 2276cc engine I used as an example in the table above, switching from a 5.400" rod to 5.700" alters max piston speed crank angle by half a degree.

Fiddling with that kind of stuff comes way down the priority list.

[brian e]

John,
 Thanks for taking the time to share your knowledge with us. 

I don't have anything too add, but I was just wondering what program you used for the above graphs, and what simulator program you use. 

Thanks, I will now go back to sitting quietly and learning. 

Brian

[John Maher]

 wondering what program you used for the above graphs, and what simulator program you use.  

That graph's from Engine Pro. As a simulator it's not accurate - the camshaft side of things is way off. The flowbench data worksheet is neat but there's nothing in there that couldn't be replicated in Excel.

I've tried a few engine simulators. They can be useful but are also a great way of wasting a LOT of time. Best value for money program and the one I use most often is PipeMax by Larry Meaux. It's not a simulator exactly but it churns out a lot of useful info.

Before you spend any money on engine simulation software, realise the half decent ones require lots of data input i.e. detailed cam info, actual cfm figures, port dimensions etc. otherwise the results will be about as accurate as the average internet forum bench racing session: GIGO