Return-Path: <13brv3@bellsouth.net> Received: from imf24aec.mail.bellsouth.net ([205.152.59.72] verified) by logan.com (CommuniGate Pro SMTP 4.2.5) with ESMTP id 484912 for flyrotary@lancaironline.net; Sun, 24 Oct 2004 13:09:26 -0400 Received-SPF: pass receiver=logan.com; client-ip=205.152.59.72; envelope-from=13brv3@bellsouth.net Received: from rd ([65.6.194.9]) by imf24aec.mail.bellsouth.net (InterMail vM.5.01.06.11 201-253-122-130-111-20040605) with ESMTP id <20041024170851.OBQ22603.imf24aec.mail.bellsouth.net@rd> for ; Sun, 24 Oct 2004 13:08:51 -0400 From: "Russell Duffy" <13brv3@bellsouth.net> To: "'Rotary motors in aircraft'" Subject: RE: [FlyRotary] Re: oil scoop? Date: Sun, 24 Oct 2004 12:09:11 -0500 Message-ID: <014401c4b9ec$2e328620$6101a8c0@rd> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0145_01C4B9C2.455C7E20" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.6626 Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 In-Reply-To: This is a multi-part message in MIME format. ------=_NextPart_000_0145_01C4B9C2.455C7E20 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: quoted-printable I'd suggest that you need some more information before you start = building a new duct/diffuser. Ed's suggestion about getting a handle on whether = there is enough airflow but poor distribution is a good one; and Ernest = suggestion of changing shape by some simpler means is also good. Hi Al, =20 Part of me wants to dive right into trying to fix this, but the more rational part says I should finish my last 4 hours while it's behaving = well. As a compromise, I might go ahead and smooth out the rear of the scoop, = to eliminate the flat wall back there. It might not be the full solution, = but I don't see any way it could hurt. At the same time, I can add some thermocouples.=20 Figure out some way to get an idea of flow distribution. If you are = going to measure exit air temp to determine delta T; you could get a rough = idea of flow distribution by measuring exit temp in 2 or 3 places - maybe divide = the area roughly in thirds, and measure temp in the center of each block. Exactly what I was thinking. I have 4 unused CHT inputs to the EM-2 = that are just coiled up under the cowl for tests just like this. I just have = to find a way to suspend some of them over the exit area of the core. Based on my scoop/diffuser shape, and the flow tests I did, my = conclusion about where the air is going is different than Ed's. My guess is that = there is flow separation and lots of turbulence at the front upper part of the duct, and most of the flow is actually going through the core near the = back end. It is also possible that the turbulence, (resulting poor pressure recovery) and the sharp turn into the core, may result in reduced inlet = flow (air spilling around the scoop. I have an even sharper turn that you do, so you may be right. Putting a steering vane in the middle of the core should help, regardless of which = end has the biggest problem. Maybe I'll do that too. =20 Your second version would probably result in less flow separation, and better pressure recovery, but there could still be issues of flow distribution and the need for some vanes. =20 I appreciate all the comments. =20 Rusty (need to go to the range now) ------=_NextPart_000_0145_01C4B9C2.455C7E20 Content-Type: text/html; charset="US-ASCII" Content-Transfer-Encoding: quoted-printable Message

I’d suggest that you need some more information = before you start=20 building a new duct/diffuser.  Ed’s suggestion about getting = a handle on=20 whether there is enough airflow but poor distribution is a good one; and = Ernest=20 suggestion of changing shape by some simpler means is also = good.

Hi=20 Al,
 
Part of = me wants to dive=20 right into trying to fix this, but the more rational part says I should = finish=20 my last 4 hours while it's behaving well.  As a compromise, I might = go=20 ahead and smooth out the rear of the scoop, to eliminate the flat wall = back=20 there.  It might not be the full solution, but I don't see any way = it could=20 hurt.  At the same time, I can add=20 some thermocouples. 

Figure out some way to get an idea of flow distribution.  If you = are=20 going to measure exit air temp to determine delta T; you could get a = rough idea=20 of flow distribution by measuring exit temp in 2 or 3 places - maybe = divide the=20 area roughly in thirds, and measure temp in the center of each = block.

Exactly what I was = thinking.  I=20 have 4 unused CHT inputs to the EM-2 that are just coiled up under the = cowl for=20 tests just like this.  I just have to find a way to suspend some of = them over the exit area of the core.

Based on my=20 scoop/diffuser shape, and the flow tests I did, my conclusion about = where the=20 air is going is different than Ed’s.  My guess is that there = is flow=20 separation and lots of turbulence at the front upper part of the duct, = and most=20 of the flow is actually going through the core near the back end.  = It is=20 also possible that the turbulence, (resulting poor pressure recovery) = and the=20 sharp turn into the core, may result in reduced inlet flow (air spilling = around=20 the scoop.

I have an even sharper = turn that you=20 do, so you may be right.  Putting a steering vane in the middle of = the core=20 should help, regardless of which end has the biggest problem.  = Maybe I'll=20 do that too. 

Your second version would probably = result in=20 less flow separation, and better pressure recovery, but there could = still be=20 issues of flow distribution and the need for some vanes. 

I appreciate all = the=20 comments. 

Rusty (need to go to the range=20 now)

------=_NextPart_000_0145_01C4B9C2.455C7E20--