Return-Path: Received: from [24.25.9.101] (HELO ms-smtp-02-eri0.southeast.rr.com) by logan.com (CommuniGate Pro SMTP 4.1.8) with ESMTP id 2904437 for flyrotary@lancaironline.net; Wed, 24 Dec 2003 18:48:08 -0500 Received: from o7y6b5 (clt78-020.carolina.rr.com [24.93.78.20]) by ms-smtp-02-eri0.southeast.rr.com (8.12.10/8.12.7) with SMTP id hBONm4Ac000237 for ; Wed, 24 Dec 2003 18:48:06 -0500 (EST) Message-ID: <000a01c3ca77$fa011600$1702a8c0@WorkGroup> From: "Ed Anderson" To: "Rotary motors in aircraft" References: Subject: Re: [FlyRotary] Re: Streamline Vs Parabolic Ducts/Diffusers Date: Wed, 24 Dec 2003 18:45:13 -0500 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0007_01C3CA4E.10E04960" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1106 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 X-Virus-Scanned: Symantec AntiVirus Scan Engine This is a multi-part message in MIME format. ------=_NextPart_000_0007_01C3CA4E.10E04960 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable =20 ----- Original Message -----=20 From: Tracy Crook=20 To: Rotary motors in aircraft=20 Sent: Wednesday, December 24, 2003 4:57 PM Subject: [FlyRotary] Re: Streamline Vs Parabolic Ducts/Diffusers Reply #2.. =20 Ed, Are you sure the labels are not backwards on the two duct = pressure recovery charts? Tracy Labels are correct, however, I used two different Y scales for = the dynamic pressure on the two charts. The dynamic pressure values are = correct but the unequal scales appear to give the parabolic curve the = greater dynamic pressure recovery, when in reality it is the streamline = duct. The Y axis on the right is for dynamic pressure. I just posted the two charts with their dynamic pressure scales = equal and I think it is clearer now. Ed Anderson Ahh! that explains it. The only thing missing (for me) is that = I don't have a mental picture of the streamline duct shape. In fact, I = may have it wrong for the parabolic as well. I picture the duct area = for the parabolic going up like the chart for the streamline shows it. I also thought a 7 degree duct was a simple straight sided duct = with sides that diverge at a 7 degree angle from the centerline. Maybe = I'm more confused than I though : -) BTW, this is great stuff Ed, keep it coming! Tracy Again, my failure to communicate is probably at the root of the = problem. The other (blue curves) shown on the graphs are not the shape = (coordinates) of the ducts walls, rather they represent the area of the = duct at that x coordinate point. So the shapes might appear to present = the wrong duct coordinates - but they are not duct coordinates (:>), = they are the duct area at that coordinate (in square feet).=20 The reason I did area is that I was using the continunity of = mass equation to derive the air velocity at that x coordinate using = p*V*A (DensityxVelocityxArea) which holds constant throughout the = airflow in the cooling system. So I needed to calculate the duct area = at each X coordinate in order to compute the velocity at that point = which of course I used to calculate the dynamic pressure at that point. I probably need to throw in the curve for the duct wall = coordinates as well as that would make it clearer. By the way, I have two data point on the pressure drop across = the GM cores. You flight with Paul and one done by a gent on a test set = up. Yours measure around 7.25 inches h20 and the test stand gave 7.5 = inches h20. =20 I had calculated a pressure drop coefficient for the GM core = based on the dimensions (including space between fins, etc). If I = assume your ducts were closer in performance to the parabolic ducts then = the difference between the measured and my calculated ranged from 4 - = 7%. Since few of us probably have the low loss of the streamline ducts. = I think that is a fairly good correlation between calcuated pressure = drop and measured. I am going to assume that is close enough for me to continue to = see if I can develop an "optimum" design with the GM core and then = assuming we don't have room for optimum what effect which changes might = have. Anyhow, Merry Christmas to you and Laura. My daughter and = son-in-law just came in with our grandsons, so all for tonight. Ed Anderson ------=_NextPart_000_0007_01C3CA4E.10E04960 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
 
 
----- Original Message -----
From:=20 Tracy = Crook
Sent: Wednesday, December 24, = 2003 4:57=20 PM
Subject: [FlyRotary] Re: = Streamline Vs=20 Parabolic Ducts/Diffusers

 
Reply #2..      
  Ed,  Are you sure the labels are not backwards on = the two=20 duct pressure recovery charts?
 
Tracy
 
 Labels are correct, = however, I used=20 two different Y scales for the dynamic pressure on the two = charts. =20 The dynamic pressure values are correct but the unequal scales = appear to=20 give the parabolic curve the greater dynamic pressure recovery, = when in=20 reality it is the streamline duct.
The Y axis on the right is for = dynamic=20 pressure.
 
 
I just posted the two charts = with their=20 dynamic pressure scales equal and I think it is clearer = now.
 
Ed Anderson
 
Ahh!  that explains = it.  The=20 only thing missing (for me) is that I don't have a mental = picture of the=20 streamline duct shape.  In fact, I may have it wrong for = the=20 parabolic as well.   I picture the duct area for the = parabolic=20 going up like the chart for the streamline shows it.
 
I also thought a 7 degree duct = was a=20 simple straight sided duct with sides that diverge at a 7 degree = angle=20 from the centerline.  Maybe I'm more confused than I though = :=20 -)
 
BTW, this is great stuff Ed, = keep it=20 coming!
Tracy
 
 
 
Again, my failure to communicate = is=20 probably at the root of the problem.
 
The other (blue curves) shown on = the=20 graphs are not the shape (coordinates) of the ducts walls, = rather they=20 represent the area of the duct at that x  coordinate = point. =20 So the shapes might appear to present the wrong duct coordinates = - but=20 they are not duct coordinates (:>), they are the duct area at = that=20 coordinate (in square feet). 
 
The reason I did area is that I = was using=20 the continunity of mass equation to derive the air velocity at = that x=20 coordinate using p*V*A (DensityxVelocityxArea) which holds = constant=20 throughout the airflow in the cooling system.  So I needed = to=20 calculate the duct area at each X coordinate in order to compute = the=20 velocity at that point which of course I used to calculate the = dynamic=20 pressure at that point.
 
 I probably need to throw = in the=20 curve for  the duct wall coordinates as well as that would = make it=20 clearer.
 
By the way, I have two data = point on the=20 pressure drop across the GM cores.  You flight with Paul = and one=20 done by a gent on a test set up.  Yours measure around 7.25 = inches=20 h20 and the test stand gave 7.5 inches h20. 
 
I had calculated a pressure drop = coefficient for the GM core based on the dimensions (including = space=20 between fins, etc).  If I assume your ducts were closer in=20 performance to the parabolic ducts then the difference between = the=20 measured and my calculated ranged from 4 - 7%.  Since few = of us=20 probably have the low loss of the streamline ducts. I think that = is a=20 fairly good correlation between calcuated pressure drop and=20 measured.
 
I am going to assume that is = close enough=20 for me to continue to see if I can develop an "optimum" design = with the=20 GM core and then assuming we don't have room for optimum what = effect=20 which changes might have.
 
Anyhow, Merry Christmas to you = and=20 Laura.  My daughter and son-in-law just came in with our = grandsons,=20 so all for tonight.
 
Ed Anderson
 
 
 
------=_NextPart_000_0007_01C3CA4E.10E04960--