X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Sender: To: lml@lancaironline.net Date: Thu, 18 Feb 2010 21:58:33 -0500 Message-ID: X-Original-Return-Path: Received: from web36906.mail.mud.yahoo.com ([209.191.85.74] verified) by logan.com (CommuniGate Pro SMTP 5.3.2) with SMTP id 4128914 for lml@lancaironline.net; Wed, 17 Feb 2010 21:13:37 -0500 Received-SPF: none receiver=logan.com; client-ip=209.191.85.74; envelope-from=chris_zavatson@yahoo.com Received: (qmail 92880 invoked by uid 60001); 18 Feb 2010 02:13:01 -0000 DomainKey-Signature:a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type; b=dS1iYZ1qUzT/OScDHo22nf++q/y9Scw38Uwtru4uyPItXJRCqraz0dDdaEhFymb+FkHH6cA22BM95Xg0Aqk+2XWjEB4709UfvnoDgBXCQ8snAMtuzWe2L+gXzF5v2Fg/dXEAnChhy9WPnBhDQjMUGkngckJInGWIscwDCkI1OO8=; X-Original-Message-ID: <249157.92327.qm@web36906.mail.mud.yahoo.com> X-YMail-OSG: 6w.gSVMVM1nZOmFINZBoL3FgMjyKAwGKtMnmxZV73kz2GG_B_pjYaCKFrGA4ewNav.1DsMQmZmGpYQEw9t8RVrjLbRpEzimKrmqWH47nqATSw9fRa76JuOiL2QZqfHGd8rpnE30QoYvXCtWX3Q19oFM7yq8YHrzdxZkyib_y2g9IOmGIf2evJYvU52bdH4ackPN25qBiQefg.i.BaIUKh5me2Kni2heL9meWlBVcZuNTyJmmFN7o2BeG89gJn6hVY.DBfrO7CYSNeuvOM4DsZXo7UC4e2FWs60eun2iNQkRaNp439AbR_5usMDR5YMk_fIg7aCgBq1wXV61fhfycqNmpjHCidLAYQTEno3b762lQm4PUkZdBB2eoJ0em0OOE4w-- Received: from [149.32.224.33] by web36906.mail.mud.yahoo.com via HTTP; Wed, 17 Feb 2010 18:13:00 PST X-Mailer: YahooMailRC/300.3 YahooMailWebService/0.8.100.260964 X-Original-Date: Wed, 17 Feb 2010 18:13:00 -0800 (PST) From: Chris Zavatson Subject: Re: [LML] Re: Cowl pressure X-Original-To: Lancair Mailing List MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="0-322544877-1266459180=:92327" --0-322544877-1266459180=:92327 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Gary,=0AThe inlet pitot tube was set up to measure dynamic pressure.=A0 It = recorded the difference between total and static - just like the aircraft p= itot tube.=A0 At first I was surprised by the readings, but the=A0numbers= =A0were confirmed to within a few percent by computing mass flow via all th= e pressure measurements.=0A=0AThe 360 inlets are so small that it drives up= the inlet velocity and prevents the more common external pressure recovery= .=A0 Most inlets, including=A0those on=A0subsequent Lancair designs, are mu= ch larger relative to the airflow required by the engine.=A0=A0This does po= se a problem for the 360 in terms of pressure recovery since it=A0relies so= heavily on internal diffusion.=A0 The upside is that there is virtually no= disturbance to the external flow.=0A=0AFor the diffusers, constant angles = were ruled out simply due to space and geometric constraints.=A0 Even-thoug= h we have extended prop hubs, more length is really needed to conform to an= y conventional rules of thumb on diffuser angles.=A0 I too couldn't find mu= ch published on bell mouth diffusers.=A0 The bell mouth, with appropriate s= urface roughness, is an attempt to keep flow attached as long as possible.= =A0 There was no expectation of keeping it completely attached except along= the upper surface.=A0 If the flow is slowed sufficiently before separation= , losses can be brought to acceptable levels.=A0 Note that the non-standard= inlets ahead of the=A0diffusers already start the diffusion process with m= uch more controlled geometry.=A0 If your sharp exit has sufficiently slowed= the flow, I imagine it would have the same effect.=A0 Starting with slight= ly larger inlets would help immensely in this regard.=0A=0AThe exit of the = 360 is in need of serious help.=A0 The firewall to exit transition is a har= d 90 degree corner.=A0 Subsequent Lancair designs incorporated a nice exit = tunnel to help get the air out in a more controlled fashion.=A0=A0In the 36= 0, the=A0lower engine mount attachment sits right where an exit ramp would = be located.=A0 As such, the exit area is over-sized to account for its inef= ficiency.=A0=A0On my plane,=A0the area could be cut down as=A0there is more= than enough air flow, but there is one minor problem.=A0 I am carburetted = and therefore have a cowl scoop that starts pointing back up towards the fu= selage once clearing the carb area.=A0 I can't increase this angle=A0any fu= rther since flow=A0already separates from the scoop=A0at very low=A0angles = of attack=A0(at speeds=A0above ~=A0190 KIAS).=A0 Additional difficulties in= cleaning up the exit are created by the retracting nose gear penetrating t= his area.=A0 I do have some ideas for the exit, but they have=A0been placed on the back burner=A0pending completion of other improvement projec= ts.=A0 If and when I get a new exit up and running, I will instrument the e= ngine compartment again for a good comparison.=A0=0A=0AChris Zavatson=0AN91= CZ=0A360std=0Awww.N91CZ.com=0A=0A=0A=0A=0A________________________________= =0AFrom: Gary Casey =0ATo: lml@lancaironline.net=0ASe= nt: Wed, February 17, 2010 11:38:55 AM=0ASubject: [LML] Re: Cowl pressure= =0A=0A=0AChris,=0AThanks for posting the link to the most excellent report.= =A0But I have a couple of questions: =A0You said you used a "pitot" probe = in the inlet and reported the velocity at about 100% of aircraft velocity. = =A0But aren't you really measuring total pressure, which is the sum of the = velocity and static pressures? =A0So some of the pressure recovery could ha= ve been ahead of the inlet, which I think is a good place to have it. =A0Th= at reduces the amount of recovery that has to be done in the diffuser. =A0I= 've seen two(at least) recommendations as to the shape of the diffuser sect= ion. =A0Conventional wisdom seems to specify a constant angle to prevent se= paration at the downstream portion. =A0Another recommendation was to use a = bell-mouth exit as you have done. =A0I couldn't follow the rationale for th= e bell-shaped exit so I used a nearly-constant angle with a sharp exit. =A0= Comment? =A0Finally, I see that the pressure below the engine is very low a= nd I would expect to see some losses as the low-velocity air exits the cowl. =A0Wouldn't it be = better to construct a small-area converging nozzle to provide for the exit?= =A0The lower cowl pressure would go up, but that would provide the energy = to accelerate the air to something closer to the free-stream velocity. =A0T= rue?=0AThanks again for the excellent information.=0AGary=0A=0A=0A=0A______= __________________________=0ARe: [LML] Cowl pressure=0AFebruary 16, 2010 7:= 15:25 AM MST=0AFrom:=0A"Chris Zavatson" =0ATo:=0A= "" =0APaul,=0AThat one doesn't have an easy=A0singul= ar answer.=0AFigure 15 in the link below shows required mass flow and press= ure drop and how this relates to CHT.=A0 Unfortunately it=A0doesn't cover t= he most desirable CHT range.=0AYou can see what I measured on my plane in F= igure 13.=0Ahttp://www.n91cz.com/Pressure/PlenumPressure.pdf=0A=0AChris Zav= atson=0AN91CZ=0A360std=0Awww.N91CZ.com=0A=0A=0A --0-322544877-1266459180=:92327 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable
=0A
Gary,
=0A
The inlet pitot tube was set up to measure d= ynamic pressure.  It recorded the difference between total and static = - just like the aircraft pitot tube.  At first I was surprised by the = readings, but the numbers were confirmed to within a few percent = by computing mass flow via all the pressure measurements.
=0A
&nbs= p;
=0A
The 360 inlets are so small that it drives up the inlet vel= ocity and prevents the more common external pressure recovery.  Most i= nlets, including those on subsequent Lancair designs, are much la= rger relative to the airflow required by the engine.  This does p= ose a problem for the 360 in terms of pressure recovery since it relie= s so heavily on internal diffusion.  The upside is that there is virtu= ally no disturbance to the external flow.
=0A
 
=0A
= For the diffusers, constant angles were ruled out simply due to space and g= eometric constraints.  Even-though we have extended prop hubs, more le= ngth is really needed to conform to any conventional rules of thumb on diff= user angles.  I too couldn't find much published on bell mouth diffuse= rs.  The bell mouth, with appropriate surface roughness, is an attempt= to keep flow attached as long as possible.  There was no expectation = of keeping it completely attached except along the upper surface.  If = the flow is slowed sufficiently before separation, losses can be brought to= acceptable levels.  Note that the non-standard inlets ahead of the&nb= sp;diffusers already start the diffusion process with much more controlled = geometry.  If your sharp exit has sufficiently slowed the flow, I imag= ine it would have the same effect.  Starting with slightly larger inle= ts would help immensely in this regard.
=0A
 
=0A
Th= e exit of the 360 is in need of serious help.  The firewall to exit tr= ansition is a hard 90 degree corner.  Subsequent Lancair designs incor= porated a nice exit tunnel to help get the air out in a more controlled fas= hion.  In the 360, the lower engine mount attachment sits ri= ght where an exit ramp would be located.  As such, the exit area is ov= er-sized to account for its inefficiency.  On my plane, the = area could be cut down as there is more than enough air flow, but ther= e is one minor problem.  I am carburetted and therefore have a cowl sc= oop that starts pointing back up towards the fuselage once clearing the car= b area.  I can't increase this angle any further since flow = already separates from the scoop at very low angles of attack&nbs= p;(at speeds above ~ 190 KIAS).  Additional difficulties in = cleaning up the exit are created by the retracting nose gear penetrating this area.  I do have some ideas for the exit, but they h= ave been placed on the back burner pending completion of other im= provement projects.  If and when I get a new exit up and running, I wi= ll instrument the engine compartment again for a good comparison. =0A
 
=0A
Chris Zavatson
=0A
N91CZ
=0A360std
=0A=0A

=0A
=0A
=0A= From: Gary Casey <casey.= gary@yahoo.com>
To: l= ml@lancaironline.net
Sent: Wed, February 17, 2010 11:38:55 AM
Subject: [LML] Re: Cowl pressure

=0A
=0A
Chris,
=0A
Thanks for posting the link to the most exc= ellent report.  But I have a couple of questions:  You said you u= sed a "pitot" probe in the inlet and reported the velocity at about 100% of= aircraft velocity.  But aren't you really measuring total pressure, w= hich is the sum of the velocity and static pressures?  So some of the = pressure recovery could have been ahead of the inlet, which I think is a go= od place to have it.  That reduces the amount of recovery that has to = be done in the diffuser.  I've seen two(at least) recommendations as t= o the shape of the diffuser section.  Conventional wisdom seems to spe= cify a constant angle to prevent separation at the downstream portion. &nbs= p;Another recommendation was to use a bell-mouth exit as you have done. &nb= sp;I couldn't follow the rationale for the bell-shaped exit so I used a nea= rly-constant angle with a sharp exit.  Comment?  Finally, I see t= hat the pressure below the engine is very low and I would expect to see some losses as the = low-velocity air exits the cowl.  Wouldn't it be better to construct a= small-area converging nozzle to provide for the exit?  The lower cowl= pressure would go up, but that would provide the energy to accelerate the = air to something closer to the free-stream velocity.  True?
=0AThanks again for the excellent information.
=0A
Gary
=0A
=0A
=0A
=0A=0A
=0A
=0A
=0A

Re= : [LML] Cowl pressure

=0A
February 16, 2010 7:15:25 AM MST
=0A
=0A
=0A
=0A
From:=0A
=0A
"Chris Zavat= son" <chris_zavatson@yahoo.com>
= =0A
= =0A
= =0A
=0A
To:
=0A
"" <lml@lancaironline.net>
=0A
=0A
=0A
=0A
=0A
Paul,
=0A
That one doesn't have an easy singular answer.
= =0A
Figure 15 in the link below shows required ma= ss flow and pressure drop and how this relates to CHT.  Unfortunately = it doesn't cover the most desirable CHT range.
=0A
You can see what I measured on my plane in Figure 13.
=0A<= DIV style=3D"MARGIN: 0px">http://www.n91cz.com/Pressure/PlenumPressure.pdf<= /DIV>=0A
 
=0A
N91CZ
=0A
360std
=0A=0A
 
=0A


=0A=0A --0-322544877-1266459180=:92327--