X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Received: from n6.bullet.re3.yahoo.com ([68.142.237.91] verified) by logan.com (CommuniGate Pro SMTP 5.2.15) with SMTP id 3765534 for flyrotary@lancaironline.net; Mon, 20 Jul 2009 08:10:50 -0400 Received-SPF: none receiver=logan.com; client-ip=68.142.237.91; envelope-from=casey.gary@yahoo.com Received: from [68.142.237.90] by n6.bullet.re3.yahoo.com with NNFMP; 20 Jul 2009 12:10:15 -0000 Received: from [216.252.111.168] by t6.bullet.re3.yahoo.com with NNFMP; 20 Jul 2009 12:10:15 -0000 Received: from [127.0.0.1] by omp103.mail.re3.yahoo.com with NNFMP; 20 Jul 2009 12:10:15 -0000 X-Yahoo-Newman-Property: ymail-3 X-Yahoo-Newman-Id: 484618.43630.bm@omp103.mail.re3.yahoo.com Received: (qmail 560 invoked by uid 60001); 20 Jul 2009 12:10:14 -0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1248091813; bh=21PW3CGwVHI2GyRh4oc2aFaUaUgMVCkrOdcJ3oHADhs=; h=Message-ID:X-YMail-OSG:Received:X-Mailer:References:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type; b=q/WyrTu1ZErrsYiK+cm9c9999o5HIwiYWpIO0PhpNvO4GnVpecuz3GyVhBeTYR2i7qc6/TBAs+0+8OPql3rV8bZzfrzc9aJ5TZRh6n0RgAN3r9dpQ4XcOTMbN50A+AMHftnfllX0Ht76VGU024gh70Uk/4hVrsVl0H91gH6j7bo= DomainKey-Signature:a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:X-YMail-OSG:Received:X-Mailer:References:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type; b=1pAIn2sOU3WdykyLalogu8jEinSbR9URm0INzCYH1xTdnbdhaUzlRWd3H1r5TgVlC12ZApjKN4tto/W8xKljSBflpXUf0LiPRrra98NsHJkbCV0NnSim33vbdtb23ikiHmci+FPIirp0MnyBPUFwuDpFmyITuUoxMGj9NSQdP80=; Message-ID: <515702.82897.qm@web57505.mail.re1.yahoo.com> X-YMail-OSG: yp6shWgVM1mSANi0nsrd.W.ECx7B3CyKz3C6R6UdnvxlayHk4SeILWyKj9jRJwXgJpH1tCDoi3ATa_Su.u0NZ43YyrMxe3Ex4fbkWjCEvhMowgPRhyOFr9QYXqDt2b3XFVInis9GwQO0WU5Qm_H7qmnGr6WnU2.Goza8JxZG._7n7_y9wCg.6ugPQ_L096lq3KTkWAVRRW.jmHbCdF11sFVnlVUVb3UXCZ99r7qJibePO41lHSLLwRLscZXM8h6hfqskB_K6.A7MVmkIiH8D5Dya Received: from [97.122.186.184] by web57505.mail.re1.yahoo.com via HTTP; Mon, 20 Jul 2009 05:10:13 PDT X-Mailer: YahooMailRC/1358.22 YahooMailWebService/0.7.289.15 References: Date: Mon, 20 Jul 2009 05:10:13 -0700 (PDT) From: Gary Casey Subject: Re: K&W Stream Duct Explanation To: Rotary motors in aircraft In-Reply-To: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="0-234358593-1248091813=:82897" --0-234358593-1248091813=:82897 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable I'm afraid I don't understand the concept (that's all right - there are lot= s of concepts I don't understand) and it seems to be the reverse of the con= ventional approach embodied in numerous applications, from jet engine inlet= s to instrumentation venturis. The conventional thinking is that as the ve= locity drops the flow is more likely to become detached since there is less= kinetic energy pushing the molecule forward and more pressure force pushin= g it back upstream. The approximation for the "perfect" shape then turns o= ut to be a straight tapered wall at about a 10-degree angle, terminating in= a sharp edge. The idea of an inlet is then to recover a significant porti= on of the pressure ahead of the inlet and then recover as much as allowed b= y the straight-wall diffuser and finally ignore the rest. Typically 50% is= recovered ahead of the inlet (that's about a 25% velocity reduction) and a= nother 50% can be recovered in the diffuser. That's a 75% total recovery, not too bad. You can get better than that if the length availab= le in unrestricted. Take a close look at a subsonic jet engine inlet and n= ote that the first part of the inlet is pointed inward and that's because a= t cruise flight the inlet is oversize and the airflow is deflected outward = ahead of the duct. The leading edge is rounded because at low airspeed the= inlet is the then too small and the air is flowing inward to get to it, re= quiring the rounded shape to avoid separation. Inside the duct it is more = or less a constant diverging angle, rounded off just ahead of the fan blade= s. The shape is modified because of the area of the fan hub "spinner". Ta= ke a look at a carburetor venturi and you'll usually find the the diffuser = increases in area and then is truncated, not bell-shaped. The K&W picture = of the "bad" diffuser shows the angle increasing to the point the air flow = separates - just reduce this angle and it won't separate and that would, I would think, produce the best recover in the length available. T= hen one might have to add baffles to distribute the low-velocity air unifor= mly across the core and that can be done with little energy loss.=0AJust so= me observations=0AGary Casey =0A=0A=0A=0A________________________________= =0A=0A=0A =0AHi Gang=0A =0AI came across a abstraction of K &W=E2=80=99s= =0Atheory of the Streamline duct that I put together a number of year ago. = =0AWhile it is my interpretation and therefore subject to error, I have not= found=0Aanything in the intervening years that indicates a misunderstandin= g. =0AWhile it is not exactly light reading, I think it is easier to follow= than K &W.=0A =0AIn any case, I thought I would make it available to promo= te=0Aunderstanding and discussion {:>)) of one approach to cooling our inst= allations.=0A =0AThis paper targets four topic areas:=0A =0A=09* Diffusers = In General =0A=09* StreamLine Duct =0A=09* Importance of Inlet to Core area= ( Ai/AB ratio) =0A=09* Truncation of StreamLine Duct =0A =0A =0ABest Regar= ds=0A =0AEd=0A=0A=0A=0A --0-234358593-1248091813=:82897 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable
I'm afraid I don't understand the concept (that's = all right - there are lots of concepts I don't understand) and it seems to = be the reverse of the conventional approach embodied in numerous applicatio= ns, from jet engine inlets to instrumentation venturis.  The conventio= nal thinking is that as the velocity drops the flow is more likely to becom= e detached since there is less kinetic energy pushing the molecule forward = and more pressure force pushing it back upstream.  The approximation f= or the "perfect" shape then turns out to be a straight tapered wall at abou= t a 10-degree angle, terminating in a sharp edge.  The idea of an inle= t is then to recover a significant portion of the pressure ahead of the inl= et and then recover as much as allowed by the straight-wall diffuser and finally ignore the rest.  Typically 50% is recovered ahead of the inl= et (that's about a 25% velocity reduction) and another 50% can be recovered= in the diffuser.  That's a 75% total recovery, not too bad.  You= can get better than that if the length available in unrestricted.  Ta= ke a close look at a subsonic jet engine inlet and note that the first part= of the inlet is pointed inward and that's because at cruise flight the inl= et is oversize and the airflow is deflected outward ahead of the duct. &nbs= p;The leading edge is rounded because at low airspeed the inlet is the then= too small and the air is flowing inward to get to it, requiring the rounde= d shape to avoid separation.  Inside the duct it is more or less a con= stant diverging angle, rounded off just ahead of the fan blades.  The = shape is modified because of the area of the fan hub "spinner".  Take = a look at a carburetor venturi and you'll usually find the the diffuser increases in area and then is truncated, not bell-shaped.  T= he K&W picture of the "bad" diffuser shows the angle increasing to the = point the air flow separates - just reduce this angle and it won't separate= and that would, I would think, produce the best recover in the length avai= lable.  Then one might have to add baffles to distribute the low-veloc= ity air uniformly across the core and that can be done with little energy l= oss.
Just some observations
Gary Casey 

=0A=0A=0A =0A=0A=0A =0A=0A =0A=0A=0A=0A=0A=0A=0A=0A
=0A=0A

Hi Gang

=0A=0A

 

=0A=0A

I cam= e across a abstraction of  K &W=E2=80=99s=0Atheory of the Streamli= ne duct that I put together a number of year ago. =0AWhile it is my in= terpretation and therefore subject to error, I have not found=0Aanything in= the intervening years that indicates a misunderstanding. =0AWhile it = is not exactly light reading, I think it is easier to follow than=0A K &= ;W.

=0A=0A

 <= /font>

=0A=0A

In any case, I thought I w= ould make it available to promote=0Aunderstanding and discussion {:>)) o= f one approach to cooling our installations.

=0A=0A

 

=0A=0A

This paper targets four topic areas:

=0A=0A

 

=0A=0A
    =0A
  • Diffusers= In General=0A
    • =0A
  • StreamLine Duct
    • =0A
  • Importance of Inlet=0A to Core area ( Ai/AB=0A ratio)
    • =0A = Truncation of=0A StreamLine Duct =0A
=0A=0A

 

=0A=0A

=   =0A=0A

Best Regards

=0A=0A  

=0A=0A

Ed

=0A=0A


=0A
=0A=0A=0A=0A --0-234358593-1248091813=:82897--