X-Virus-Scanned: clean according to Sophos on Logan.com X-SpamCatcher-Score: 2 [X] Return-Path: Received: from ms-smtp-03.southeast.rr.com ([24.25.9.102] verified) by logan.com (CommuniGate Pro SMTP 5.1.8) with ESMTP id 2029900 for flyrotary@lancaironline.net; Mon, 07 May 2007 08:08:10 -0400 Received-SPF: pass receiver=logan.com; client-ip=24.25.9.102; envelope-from=eanderson@carolina.rr.com Received: from edward2 (cpe-024-074-103-061.carolina.res.rr.com [24.74.103.61]) by ms-smtp-03.southeast.rr.com (8.13.6/8.13.6) with SMTP id l47C6wcI021916 for ; Mon, 7 May 2007 08:06:59 -0400 (EDT) Message-ID: <000701c790a0$69f24280$2402a8c0@edward2> From: "Ed Anderson" To: "Rotary motors in aircraft" References: Subject: Re: [FlyRotary] Re: Pinched Duct Date: Mon, 7 May 2007 08:08:24 -0400 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0004_01C7907E.E28ABA50" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.3028 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3028 X-Virus-Scanned: Symantec AntiVirus Scan Engine This is a multi-part message in MIME format. ------=_NextPart_000_0004_01C7907E.E28ABA50 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I understand, Al. I always appreciate your input and viewpoint. =20 One reason I throw out the information and my "logic" is so that others = can examine and critique - it would be nice if I knew it all, but, sigh! = I am under no delusion that I do. You certainly bring up food for = thought and I would suggest any who may considered adopting the = "Pinched" duct give serious consideration of your viewpoint. External diffusion is undoubted a factor - as it is in most forward = facing ducts . I do agree that any time you accelerate air your incur = greater loses, my take was that accelerating the boundary layer delayed = separation and the adverse effects so caused. That increased skin = friction was a far lesser evil than flow separation as indicated by the = NACA study. As to whether the boundary layer penetrates further due to increased = velocity is probably dependent on the influence of the rate of pressure = rise vs the rate of boundary layer deceleration. Is that relationship = linear or non-linear? Given the number of variables, I suspect = non-linear but don't know. The rapid expansion of the streamline duct configuration would tend to = indicate (to me at least) delaying separation, by keeping the airstream = energy high, until just before the core pays dividends. Now whether my = ratio of inlet, pinched and core area successfully accomplish that is = simply unknown at this point. =20 Using this approach, I have been able to reduce my inlet area and make = use of a very constrained space for a duct. Again, I make no pretense = that it could possibly be as effective as a full up streamline duct for = instance, but on the other hand I believe it creates less cooling drag = than a streamline duct truncated sufficiently to meet my space = constraints. It might well be that its success is largely due to = external diffusion as you suggest , but I would have expected that = effect on all my ducts. The pinched duct does cools my installation better than my previous = approaches to a duct and has enabled me to reduce my radiator inlet area = by 33% less area than previous duct attempts. But, your may be = correct, I may have just gotten lucky on this duct shape and it is = simply "less-bad" than my previous attempts. =20 I do agree that all of this is conjecture from both view points and = only instrumentation, data collection and analysis might clarify what is = actually happening. =20 As always, appreciate your viewpoint. Thanks again, Al Ed =20 ----- Original Message -----=20 From: Al Gietzen=20 To: Rotary motors in aircraft=20 Sent: Monday, May 07, 2007 1:30 AM Subject: [FlyRotary] Re: Pinched Duct Ed; Well, ED, the more I look at this idea, the more trouble it gives me; = so I will express a dissenting opinion. I never argue against what = works; but my conclusion here is that - assuming it does work better in = your case than a nice entrance and a continuously expanding duct of some = reasonable shape - is due to the fact that you have external diffusion = (buildup of pressure in front on the cowl entrance) and turbulent flow = at the entrance behind the prop. So without doing some flow testing, = I'd suggest caution in using this approach for inlet scoops in front of = radiators in other locations. Accelerating the air costs you dynamic pressure and increases friction = losses, and (depending on entrance and frictional losses) 'pinching' the = area by more than about 30% would cause serious choking of the flow, = resulting in most of the air spilling around the entrance. The more = rapidly expanding area after the pinch and more rapid pressure increase = is at least as likely to result in flow separation as a less rapidly = expanding area, and a slower pressure increase. And since you are now = expanding from a smaller high velocity area, I'm gonna guess that the = fractional area of turbulent flow would be larger. The concepts you put together make sense in themselves, but it seems = to me a little like adding apples and oranges to get more apples. So, = more proof of concept is required; well, at least to convince me. Have you made any measurements of the flow distribution at the core = exit? Just my opinion, and, of course nothing personal. Best, Al (off to Baja for a few days of relaxation) ------=_NextPart_000_0004_01C7907E.E28ABA50 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
I understand, Al.  I always appreciate your = input and=20 viewpoint. 
 
One reason I throw out the information and my = "logic" is=20 so that others can examine and critique - it would be nice if I knew it = all,=20 but, sigh! I am under no delusion that I do.  You certainly = bring up=20 food for thought and I would suggest any who may considered adopting the = "Pinched" duct give serious consideration of your=20 viewpoint.
 
External diffusion is undoubted a factor - as it = is in most forward facing ducts .  I do agree that any time = you=20 accelerate air your incur greater loses, my take was that =20 accelerating the boundary layer delayed separation and the adverse = effects so=20 caused.  That  increased skin friction was a far lesser = evil than=20 flow separation as indicated by the NACA study.
 
  As to whether the boundary layer = penetrates=20 further due to increased velocity is = probably=20 dependent on the influence of the  rate of pressure rise vs the = rate of=20 boundary layer deceleration.  Is that relationship linear or=20 non-linear?   Given the number of variables, I suspect = non-linear but=20 don't know.
 
The rapid expansion of the streamline duct = configuration=20 would tend to indicate (to me at least) delaying separation, by = keeping the=20 airstream energy high, until just before the core pays = dividends.  Now=20 whether my ratio of inlet, pinched and core area successfully accomplish = that is=20 simply unknown at this point. 
 
Using this approach, I have been able to reduce = my inlet=20 area and make use of a very constrained space for a duct.  Again, I = make no=20 pretense that it could possibly be as effective as a full up = streamline=20 duct for instance, but on the other hand I believe it creates less = cooling drag=20 than a streamline duct truncated sufficiently to meet my space=20 constraints.  It might well be that its success is largely due to = external=20 diffusion as you suggest , but I would have expected that effect on all = my=20 ducts.
 
The pinched duct does cools my installation = better than my=20 previous approaches to a duct and has enabled me to reduce my radiator = inlet=20 area by 33% less area than previous duct attempts.   But,=20 your may be correct, I may have just gotten lucky on this duct = shape=20 and it is simply "less-bad" than my  previous = attempts. =20
 
 I do agree that all of this is conjecture = from both=20 view points and only instrumentation, data collection and analysis might = clarify=20 what is actually happening. 
 
As always, appreciate your viewpoint.  = Thanks again,=20 Al
 
Ed
    
----- Original Message -----
From:=20 Al = Gietzen=20
Sent: Monday, May 07, 2007 1:30 = AM
Subject: [FlyRotary] Re: = Pinched=20 Duct

Ed;

 

Well, ED, = the more=20 I look at this idea, the more trouble it gives me; so I will express a = dissenting opinion.  I never argue against what works; but my = conclusion=20 here is that - assuming it does work better in your case than a nice = entrance=20 and a continuously expanding duct of some reasonable shape - is due to = the=20 fact that you have external diffusion (buildup of pressure in front on = the=20 cowl entrance) and turbulent flow at the entrance behind the = prop.  So=20 without doing some flow testing, I=92d suggest caution in using this = approach=20 for inlet scoops in front of radiators in other = locations.

 

Accelerating the=20 air costs you dynamic pressure and increases friction losses, and = (depending=20 on entrance and frictional losses) =91pinching=92 the area by more = than about 30%=20 would cause serious choking of the flow, resulting in most of the air = spilling=20 around the entrance.  The more rapidly expanding area after the = pinch and=20 more rapid pressure increase is at least as likely to result in flow=20 separation as a less rapidly expanding area, and a slower pressure=20 increase.  And since you are now expanding from a smaller high = velocity=20 area, I=92m gonna guess that the fractional area of turbulent flow = would be=20 larger.

 

The = concepts you=20 put together make sense in themselves, but it seems to me a little = like adding=20 apples and oranges to get more apples.  So, more proof of concept = is=20 required; well, at least to convince me.

 

Have you = made any=20 measurements of the flow distribution at the core = exit?

 

Just my = opinion,=20 and, of course nothing personal.

 

Best,

 

Al (off = to Baja for=20 a few days of=20 relaxation)

------=_NextPart_000_0004_01C7907E.E28ABA50--