Only for those who (Dave? Bill?, Rusty?.....) wish to punish themselves with more minutia on air flow in ducts/diffusers.
A month or two ago, I posted some slides extracted from a university study/course on the effects of airflow separation inside a duct. This involved the boundary layer which appears to act somewhat different inside the constraining walls of a duct as compared to its free flow across an airfoil. The cause of flow separation in the duct being the pressure build up by the expansion in duct area which led to two counteracting forces.
The pressure build up actually helps the boundary layer stay attached to curving duct wall - for a time. But, this same pressure that helps "push" the boundary layer against the duct wall also slows down the boundary layer which ultimately leads to flow separation and reversal.
This "understanding" led me to my "pinched duct" design to accelerate the boundary layer and cause it to penetrate further into the higher pressure area before separation. I also inferred that this effect was what made the Streamline duct of K&W so effective. No claim was made that the pinched ducts were anywhere nearly as effective as the Streamline duct, but were an attempt to meet a space constraint.
Most attempts to use the streamline duct in a space too small involves truncating the duct from the inlet end. However, while this does tend to preserve some of its effectiveness, if the distance is very short (like my 3 -6 inches) the large expose core area likely increases cooling drag considerably. So I decided to keep the inlet small (unlike what truncating the streamline duct would have resulted in) but to pinch it down to keep the boundary layer velocity high resulting (hopefully) in further penetration down the duct before flow separation occurred.
Some questioned my interpretation (always a smart thing to do, as I only had one short course in aerodynamics as an Electrical Engineering student - so my attention was probably not as keenly focused as it should have been {:>)). In any case, I went looking for the source document so that any interested could read it and draw their own conclusion.
The original source for this material was
http://www.me.dal.ca/site2/courses/mech3300/5_Separation.ppt. However, they have (as Universities frequently do) apparently rotated material presented and removed this briefing from their website. This leaves the .html portion I saved when first reading the presentation which is attached.
I was only partially successful in providing the source - in that I found the original script that went with the slide presentation - but unfortunately the slides are not present with it. I do have a number of the slides I had previously extracted (used in my presentation) but since they could be "tainted" by my "explanation" of the slides, I will not present them.
I reviewed the script again and still believe my interpretation is correct, but others should have the opportunity to decide for themselves. We do that sort of thing on this list.
But, if you do decide differently, please don't tell my pinched ducts {:>)
Ed