Speedmaster head hypothetical
Here's a test I ran this morning with the swirl meter on the bench. I've been thinking about the relationship between swirl and SSR flow separation. I think it's sort of a chicken and egg relationship. Actually the separation probably comes first, but the swirl may actually feed the separation and it snowballs.
Anyway, look at the graph for a few interesting points. On this port/head, the flow separation really starts about 0.400" lift. This may also be the point where flow on the cylinder wall side of the valve is maxed out because of shrouding (it may actually happen a little earlier than 0.400). Look what happens to the swirl at that point. It takes off. So, at 0.400" lift the flow on the SS and the shrouded sides of the valve start decreasing while the flow on the long side and cyl center side of the valve increases quite rapidly. Long side and cyl center side of the valve is where the daylight is. That is the straight shot/line of sight area where the air 'wants' to go. Increasing flow in this area increases swirl. Look what happens to CFM as swirl increases rapidly (0.400-0.700 lift). CFM starts to decrease rapidly.
Look at another section of the graph from 0.150-0.400 lift. In this area swirl is almost constant. This is the section of the flow curve (the CFM line) that is the most steep. The more steep the flow curve (the slope of the line), the more flow per inch of valve opening. That is a positive thing. Flow has not yet separated on the short side and shrouding has not significantly cut off flow on the cyl wall side. So in this section it appears flow is moving around the entire perimeter of the valve quite well.
So we see an inverse relationship between flow and swirl. If swirl increases rapidly, flow does not. Where flow increases rapidly, swirl does not. And this makes sense from an energy standpoint. Increases in flow and increases in swirl both take increases in energy or area. So every move of the valve (by 0.050" in this case) gives so much additional potential energy to work with. That 0.050" movement of the valve can yield a large increase in flow or a large increase in swirl. It will not yield a large increase in both simultaneously. Both can increase simultaneously, but not at a huge rate (there is a limit to the positive slope of the lines).
Darin Morgan repeatedly states that too much swirl is not good for performance. It may be good for emissions and efficiency, but not performance. On this head, the cause of too much swirl is flow separation and shrouding. Darin preaches keeping velocities as equal as possible around the entire perimeter of the valve. It's hard to do.
Here's a test I ran this morning with the swirl meter on the bench. I've been thinking about the relationship between swirl and SSR flow separation. I think it's sort of a chicken and egg relationship. Actually the separation probably comes first, but the swirl may actually feed the separation and it snowballs.
Anyway, look at the graph for a few interesting points. On this port/head, the flow separation really starts about 0.400" lift. This may also be the point where flow on the cylinder wall side of the valve is maxed out because of shrouding (it may actually happen a little earlier than 0.400). Look what happens to the swirl at that point. It takes off. So, at 0.400" lift the flow on the SS and the shrouded sides of the valve start decreasing while the flow on the long side and cyl center side of the valve increases quite rapidly. Long side and cyl center side of the valve is where the daylight is. That is the straight shot/line of sight area where the air 'wants' to go. Increasing flow in this area increases swirl. Look what happens to CFM as swirl increases rapidly (0.400-0.700 lift). CFM starts to decrease rapidly.
Look at another section of the graph from 0.150-0.400 lift. In this area swirl is almost constant. This is the section of the flow curve (the CFM line) that is the most steep. The more steep the flow curve (the slope of the line), the more flow per inch of valve opening. That is a positive thing. Flow has not yet separated on the short side and shrouding has not significantly cut off flow on the cyl wall side. So in this section it appears flow is moving around the entire perimeter of the valve quite well.
So we see an inverse relationship between flow and swirl. If swirl increases rapidly, flow does not. Where flow increases rapidly, swirl does not. And this makes sense from an energy standpoint. Increases in flow and increases in swirl both take increases in energy or area. So every move of the valve (by 0.050" in this case) gives so much additional potential energy to work with. That 0.050" movement of the valve can yield a large increase in flow or a large increase in swirl. It will not yield a large increase in both simultaneously. Both can increase simultaneously, but not at a huge rate (there is a limit to the positive slope of the lines).
Darin Morgan repeatedly states that too much swirl is not good for performance. It may be good for emissions and efficiency, but not performance. On this head, the cause of too much swirl is flow separation and shrouding. Darin preaches keeping velocities as equal as possible around the entire perimeter of the valve. It's hard to do.