Cam timing vs runner length
06-04-2019, 06:37 PM
#1
Cam timing vs runner length There's been a lot more discussion about valve events on the forum recently. There have also become a multitude more intake manifold options available to us in recent years, and even some clever people modifying intakes to suit their needs.
I started thinking about the ridiculousness of cam timing in relation to how fast an engine is actually turning.
At 6500 RPM the crankshaft is turning 25 microseconds per degree.
25 microseconds.
We are splitting hairs talking about how a 47 degree IVC will perform so much better than a 44 degree IVC because XYZ - which in some cases is very true, but the reality of the situation is a bit mind boggling.
.000025 seconds per degree. Three degree change in cam timing is .000075 seconds.
The thought that an extremely significant difference in airflow occurs because of a .000075 second difference in valve timing breaks my mind, but the results are impossible to argue with.
I have more to talk about which starts to look at intake manifold runner length, but I'm out of time, and kind of want to stretch this out a little bit to see if anyone has any comments about what I've written.
I started thinking about the ridiculousness of cam timing in relation to how fast an engine is actually turning.
At 6500 RPM the crankshaft is turning 25 microseconds per degree.
25 microseconds.
We are splitting hairs talking about how a 47 degree IVC will perform so much better than a 44 degree IVC because XYZ - which in some cases is very true, but the reality of the situation is a bit mind boggling.
.000025 seconds per degree. Three degree change in cam timing is .000075 seconds.
The thought that an extremely significant difference in airflow occurs because of a .000075 second difference in valve timing breaks my mind, but the results are impossible to argue with.
I have more to talk about which starts to look at intake manifold runner length, but I'm out of time, and kind of want to stretch this out a little bit to see if anyone has any comments about what I've written.
06-05-2019, 03:13 AM
#2
Lets say you have an LS3 with a stock intake.
If you have a camshaft with the same valve events and overlap with the only difference being a 44 IVC and then you try a 47 IVC.
The 47 IVC would allow you to rev higher and ultimately make more power.
The IVC of 44 wil give you better low end but will not hang on as well up top and will make less power,
If you have a camshaft with the same valve events and overlap with the only difference being a 44 IVC and then you try a 47 IVC.
The 47 IVC would allow you to rev higher and ultimately make more power.
The IVC of 44 wil give you better low end but will not hang on as well up top and will make less power,
06-05-2019, 05:24 AM
#3
Moderator
iTrader: (4)
Join Date: Jul 2014
Location: My own internal universe
Posts: 10,446
Received 1,836 Likes
on
1,145 Posts
That is why velocity > flow.
Flowing 861.7 cfm is great, but you do not have minutes. You have milliseconds.
Also pressure waves in the manifold are at the speed of sound. At the speed of sound, that one degree is 8mm more of the air column getting in. There degrees is about an inch -- assuming the timing of the pressure wave is optimal, which is where runner length vs rpm vs cam timing come in. You reach peak torque at peak cylinder airmass, which is when the runner length, IVC, and pressure waves sync up perfectly. Runner length determines what rpm that occurs.
Flowing 861.7 cfm is great, but you do not have minutes. You have milliseconds.
Also pressure waves in the manifold are at the speed of sound. At the speed of sound, that one degree is 8mm more of the air column getting in. There degrees is about an inch -- assuming the timing of the pressure wave is optimal, which is where runner length vs rpm vs cam timing come in. You reach peak torque at peak cylinder airmass, which is when the runner length, IVC, and pressure waves sync up perfectly. Runner length determines what rpm that occurs.
06-05-2019, 11:02 AM
#4
That is why velocity > flow.
Flowing 861.7 cfm is great, but you do not have minutes. You have milliseconds.
Also pressure waves in the manifold are at the speed of sound. At the speed of sound, that one degree is 8mm more of the air column getting in. There degrees is about an inch -- assuming the timing of the pressure wave is optimal, which is where runner length vs rpm vs cam timing come in. You reach peak torque at peak cylinder airmass, which is when the runner length, IVC, and pressure waves sync up perfectly. Runner length determines what rpm that occurs.
Flowing 861.7 cfm is great, but you do not have minutes. You have milliseconds.
Also pressure waves in the manifold are at the speed of sound. At the speed of sound, that one degree is 8mm more of the air column getting in. There degrees is about an inch -- assuming the timing of the pressure wave is optimal, which is where runner length vs rpm vs cam timing come in. You reach peak torque at peak cylinder airmass, which is when the runner length, IVC, and pressure waves sync up perfectly. Runner length determines what rpm that occurs.
I was doing my own math after I came up with the 25 microsecond per degree of crank rotation number and decided to look at the speed of the pressure wave in the intake tract. I was very surprised to find that as you said, at the speed of sound 1 inch takes ~3 times longer than one degree of rotation.. about 76 microseconds to travel 1 inch.
It just points further to the importance of the complete system, and if constrained - such as using a stock intake - some of the steadfast rules that start popping up with cam timing may very well be a bandaid or compromise rather than an optimal selection.
06-05-2019, 11:35 AM
#5
Moderator
iTrader: (4)
Join Date: Jul 2014
Location: My own internal universe
Posts: 10,446
Received 1,836 Likes
on
1,145 Posts
I calculated it. Speed of sound is 330 m/s or 330,000 mm/s. Times .000025 is 8mm. X3 is 24m approx 1-inch.
But that math only works if the pressure wave is timed just right -- peaking at the valve right before IVC. The port is not moving air at the speed of sound. Can figure it out if you know CSA.
If I use port info I have at hand, 410 cfm = 708,500 CI per min = 11,800 CI per sec
Port area 3.2 sq in makes 3690 linear inches or second air column. That is way over simplified because port flow is dynamic. You could use that number to get 0.28 inches for the added three degrees. X 3.2 sq in to get 0.89 cubic inches of air. All using average speeds and cross sections.
Likely not representative but a conversation piece
But that math only works if the pressure wave is timed just right -- peaking at the valve right before IVC. The port is not moving air at the speed of sound. Can figure it out if you know CSA.
If I use port info I have at hand, 410 cfm = 708,500 CI per min = 11,800 CI per sec
Port area 3.2 sq in makes 3690 linear inches or second air column. That is way over simplified because port flow is dynamic. You could use that number to get 0.28 inches for the added three degrees. X 3.2 sq in to get 0.89 cubic inches of air. All using average speeds and cross sections.
Likely not representative but a conversation piece
06-05-2019, 12:24 PM
#6
TECH Addict
So is there a known formula/correlation between CID, CSA, and runner length that determines the optimal IVO event?
It seems that the pressure waves should be fairly predictable, as waves often form a pattern of consistent repetition. And if their velocity is known to be relatively consistent, then any given intake should be fairly predictable... like all unported Fast 102's with long runners provide whatever volume of air in the stack, with whatever frequency of pressure waves. And these factors would be easy to find for all of the most common intakes. But where would this information be used as part of a formula?
Still have to consider the displacement of the engine, and the CSA of the heads being used, right?
Does the pull of the piston influence the air speed at the valve? Such as two identical displacements, but with different bore/stroke dimensions?
It seems that the pressure waves should be fairly predictable, as waves often form a pattern of consistent repetition. And if their velocity is known to be relatively consistent, then any given intake should be fairly predictable... like all unported Fast 102's with long runners provide whatever volume of air in the stack, with whatever frequency of pressure waves. And these factors would be easy to find for all of the most common intakes. But where would this information be used as part of a formula?
Still have to consider the displacement of the engine, and the CSA of the heads being used, right?
Does the pull of the piston influence the air speed at the valve? Such as two identical displacements, but with different bore/stroke dimensions?