Thinking About Valve Events + Piston Velocity Together: How?

[CGT]

there is no meaningful change in piston dwell at TDC but there is at BDC, where it doesn't matter.

Why doesn't it matter?

[Ron E]

This would be so much easier if I could visualize it and see an animation of the piston and cam movement...


Adam

You may like EA pro. It has a "see engine" feature that has the piston and valve action simulation.

[Stan Weiss]

The green line is the intake cylinder head flow mapped on to the intake cam profile.

The red line is the exhaust cylinder head flow mapped on to the exhaust cam profile.

The blue line is the piston flow demand cfm

The cyan and magenta lines are the scaled piston velocity.

Stan

ab-cam-flow-piston-flow-piston-velocity.gif

[ptuomov]

Here's a follow-up question.

Suppose that the short-block is built, cast in stone, not easily changed. Suppose further that this short block has maximum rpm of x (say, 6800 rpm), and all things considered, the peak power is therefore desired at y (say, 6300 rpm). Take these as given. Also take the volumetric efficiency at peak power rpm as given (say, 105% or something).

Let's say we've also picked the cam. We know where we want the IVO because of idle, emissions, and low rpm drivability, etc. considertions. We know roughly where we want the IVC given the rpms. We also know the peak air flow demand crank angle, around say 78 degrees ATDC. Since the tappet diameter, valves, etc. are given, we also know that maximum acceleration and maximum velocity of the intake lobe. So we know how much we can lift the valve at 78 degrees ATDC. Say it's 8mm (four valve head). Take all that as given as well.

Armed with these assumptions, how should I think about trading of head flow in CFM@28" vs. coefficient of discharge CD for the head at the maximum air flow demand crank angle lift? We know the peak air flow demand, and we know the valve lift at that crank angle. But there are presumably a lot of things one can do to the valve and the port that will either increase or decrease CFM@28" and CD. So how do I optimize CFM vs. CD?

[Erland Cox]

With only 105% VE I believe that your biggest gain are outside the cylinder head.
I have never gained VE, torque or hp by using a much better head than one that is good enough.
It is all the parts around the head that make it shine.
More lift with the same duration, moving the cam to find the sweet spot, exhaust and intake manifold tuning and ignition.
And using as much compresion as possible.

Erland

[Turbo231]

I got this off the internet. Have not checked it though. I am currently integrating it into an Excel sheet that will calculate piston flow rate. If I get ambitious, I may try to incorporate valve events into the spreadsheet.

slider crank.xls

[Stan Weiss]

Here's a follow-up question.

Suppose that the short-block is built, cast in stone, not easily changed. Suppose further that this short block has maximum rpm of x (say, 6800 rpm), and all things considered, the peak power is therefore desired at y (say, 6300 rpm). Take these as given. Also take the volumetric efficiency at peak power rpm as given (say, 105% or something).

Let's say we've also picked the cam. We know where we want the IVO because of idle, emissions, and low rpm drivability, etc. considertions. We know roughly where we want the IVC given the rpms. We also know the peak air flow demand crank angle, around say 78 degrees ATDC. Since the tappet diameter, valves, etc. are given, we also know that maximum acceleration and maximum velocity of the intake lobe. So we know how much we can lift the valve at 78 degrees ATDC. Say it's 8mm (four valve head). Take all that as given as well.

Armed with these assumptions, how should I think about trading of head flow in CFM@28" vs. coefficient of discharge CD for the head at the maximum air flow demand crank angle lift? We know the peak air flow demand, and we know the valve lift at that crank angle. But there are presumably a lot of things one can do to the valve and the port that will either increase or decrease CFM@28" and CD. So how do I optimize CFM vs. CD?

That should be about a 2.25:1 rod stroke ratio, which would think should cause a little difference in your cam.

Stan

[ptuomov]

We also know the peak air flow demand crank angle, around say 78 degrees ATDC.

That should be about a 2.25:1 rod stroke ratio, which would think should cause a little difference in your cam.
Stan

I just ballparked the example. In practice, I'd actually compute the maximum piston velocity crank angle and then add to that the time in crank angle degrees that it takes for the pulse to travel at sound of speed from the piston to the valve. Is that your methodology for inferring my rod to stroke from 78 degrees ATDC being the maximum flow demand at the valve? If so, what did you assume my stroke was to get the distance from piston to valve?

[Stan Weiss]

Did not guess anything about your engine. Just looked at this chart I made some years ago.

Stan

ab-rsrcdmpv7x7.gif

[ptuomov]

Did not guess anything about your engine. Just looked at this chart I made some years ago.

Good chart, thanks for posting. Just to ballpark it, I added a couple of degrees to that for the time that it takes to for the sound to travel from piston to valve.

I didn't do any exact calculations. One crank degree at 6300 rpm would take 0.00002645502 second. At 100C speed of sound, that amounts to about 10mm per crank degree. The piston is about 35mm in the hole and there's maybe another 5mm to go to the valve (?), so that would amount to 4 crank degrees. Since I'm not sure how the wave develops, it could be something else, too.

[David Redszus]

The green line is the intake cylinder head flow mapped on to the intake cam profile.

The red line is the exhaust cylinder head flow mapped on to the exhaust cam profile.

The blue line is the piston flow demand cfm

The cyan and magenta lines are the scaled piston velocity.

Stan

ab-cam-flow-piston-flow-piston-velocity.gif

Using Stan's calcs and graphics, the important curves are:
The green line is the intake cylinder head flow mapped on to the intake cam profile.
The blue line is the piston flow demand cfm

Note the location of the piston flow demand peak compared to the head flow curve.
The maximum head flow does not occur at maximum valve lift. So flow bench data using max lift
can be misleading and wasteful of intellectual resources.

Per Blair, valve Cd will change with pressure ratio; so how does one derive flow from only 28"?

[ptuomov]

Per Blair, valve Cd will change with pressure ratio; so how does one derive flow from only 28"?

It probably does, but not that much, if we're talking about say lift = 20% of valve head diameter. Right?

[David Redszus]

Per Blair, valve Cd will change with pressure ratio; so how does one derive flow from only 28"?

It probably does, but not that much, if we're talking about say lift = 20% of valve head diameter. Right?

To be accurate, all flow measurements must utilize air mass flow, not CFM or velocity.

The CD will change with both pressure ratio and flow area ratio. The normal ranges expected in an engine would result in a Cd variance in excess of 17%, which makes meaningful measurement meaningless. If the correct values of Cd are not used.

While lift is often expressed as some percentage of valve head diameter, the preferred method is to express lift as a percentage of maximum lift. For a 2" valve with a max lift of 25% or 0.5", full lift would be 1.0 while half lift would be represented as .5 lift. So says prof Blair.

For a very comprehensive treatise on floe coefficients (and whats wrong with them) see Blair, Simulation of the four stroke engine, Chapter 3, Discharge coefficients in a four stroke engine. Not an easy read.

Given the engine geometry and cam lift curve, I'm sure Stan could calculate and plot the flow curves very accurately.

[pastry_chef]

Anyone have the equation to obtain the point of maximum piston speed from rod and stroke?

[ptuomov]

Per Blair, valve Cd will change with pressure ratio; so how does one derive flow from only 28"?

It probably does, but not that much, if we're talking about say lift = 20% of valve head diameter. Right?

To be accurate, all flow measurements must utilize air mass flow, not CFM or velocity.

The CD will change with both pressure ratio and flow area ratio. The normal ranges expected in an engine would result in a Cd variance in excess of 17%, which makes meaningful measurement meaningless. If the correct values of Cd are not used.

While lift is often expressed as some percentage of valve head diameter, the preferred method is to express lift as a percentage of maximum lift. For a 2" valve with a max lift of 25% or 0.5", full lift would be 1.0 while half lift would be represented as .5 lift. So says prof Blair.

For a very comprehensive treatise on floe coefficients (and whats wrong with them) see Blair, Simulation of the four stroke engine, Chapter 3, Discharge coefficients in a four stroke engine. Not an easy read.

Given the engine geometry and cam lift curve, I'm sure Stan could calculate and plot the flow curves very accurately.

So you're also telling me that I'm doing it wrong?

In the example case, we're talking about 10.2mm peak lift cams that have about 8.95mm lift at peak piston speed crank angle. So we're talking about 88% of peak lift. In this case, the flow bench measurements at that 8.95mm valve lift line up with about constant coefficient of discharge as pressure is varied. (I'm not talking about varying pressure on a valve that is just cracked open, or varying valve lift.) Let's be clear here, I am talking about trading of flow vs. Cd at a given lift that is 24% of the valve area and 88% of the peak lift. When I say "flow" or "CFM@28"" I mean SCFM@28", so that should be pretty closely related to mass flow.

Any insights to the original question now that I'm hopefully asking it right so I don't get bounced to the back of the club line for wearing sneakers?