Mailing List flyrotary@lancaironline.net Message #40216
From: Ed Anderson <eanderson@carolina.rr.com>
Subject: Diffuser Efficiency under two different conditions
Date: Wed, 7 Nov 2007 10:24:06 -0500
To: Rotary motors in aircraft <flyrotary@lancaironline.net>
Only for those reallllllyyyy fascinated by airflow in ducts/diffusers {:>). 
 
  For a  time I had been puzzled by reference in literature to an "ideal" or at least maximum expansion ratio for the inlet/outlet area of a diffuser.  Much of the literature point to a maxim divergence angle from the inlet to exit of a diffuser of  7-11degrees.   On the other hand there is literature that points to such diffusers as the Streamline diffuser which has a divergence angle approaching 65 degrees and with 84% efficiency as being great.  So how can it be both ways??. 
 
It became apparent after a lot of reading that the difference appeared to be whether there was a "resistance" to the airflow right at the exit of the diffuser such as a radiator/heat exchanger. 
 
For conditions (such as a wind tunnel) where there is no resistance (radiator) at the exit of the diffuser, separation will occur if the divergence angle is greater than aprox 11 deg (depending on type of diffuser and some other factors.  But when a resistance (radiator) is present things are different.  I think I puzzled out why that happens, but that is just my speculation so I won't bore you folks with that.  Instead I recently found some documentation that gives some hard facts about the two diffuser conditions.
 
  I recently found an Naca study that specifically made this point even with charts which showed the divergence angles of diffuser with and without resistance and their efficiency.  In fact, it even states that with a resistance (radiator) SHORTER inlets to a diffuser is apparently better than longer.
 
Also there is reference to the fact that if the boundary layer is thinner (rather than thicker) that separation caused by the adverse pressure gradient of a diffuser is delayed.  A thinner boundary layer is equated to a boundary layer with a higher velocity and one that does not travel far (and therefore slow down and grow into a thick boundary layer). 
 
These two findings of short inlet and higher velocity boundary layer having some benefits (if in front of a radiator) offers some encouragement that my "Pinched" ducts may in fact have some merit. 
 
But, regardless of that point, I though some of you might find this interesting.  I extracted two sections which hopefully is good enough to read.  If you want a copy of the report (high speed only), send me an e mail
 
Second jpg with diffuser with resistance coming in next e mail.
Subscribe (FEED) Subscribe (DIGEST) Subscribe (INDEX) Unsubscribe Mail to Listmaster