X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Received: from mail14.syd.optusnet.com.au ([211.29.132.195] verified) by logan.com (CommuniGate Pro SMTP 5.2c1) with ESMTPS id 2462261 for flyrotary@lancaironline.net; Thu, 08 Nov 2007 17:12:32 -0500 Received-SPF: none receiver=logan.com; client-ip=211.29.132.195; envelope-from=lendich@optusnet.com.au Received: from george (d211-31-71-103.dsl.nsw.optusnet.com.au [211.31.71.103]) by mail14.syd.optusnet.com.au (8.13.1/8.13.1) with SMTP id lA8MBiMt028494 for ; Fri, 9 Nov 2007 09:11:46 +1100 Message-ID: <002801c82254$5d4f1440$67471fd3@george> From: "George Lendich" To: "Rotary motors in aircraft" References: Subject: Re: [FlyRotary] Re: Total,duct, Ambient or Velocity???? Date: Fri, 9 Nov 2007 08:11:50 +1000 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0025_01C822A8.2DFC03C0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 X-Antivirus: avast! (VPS 0657-0, 12/12/2006), Outbound message X-Antivirus-Status: Clean This is a multi-part message in MIME format. ------=_NextPart_000_0025_01C822A8.2DFC03C0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Ed,=20 If I'm understanding you correctly, it appears that you need dynamic = pressure ( flow) that turns into a high static pressure (at the Rad = face). To maintain this high static pressure, the dynamic flow must be free of = turbulence, which is associated with flow separation from the duct = walls. Hence the need for proper divergent angles. There must be good pressure drop across the Rad, not too high or you = lose heat transfer, not too low as to create excess drag. There must be = some turbulence within the duct fins to enhance heat transfer, but not = too much as to create restrictions. I still feel a low pressure area behind the rad would be beneficial.=20 George ( down under) Even in the Naca studies they often 'fess up that theoretical = considerations must give way to practical installation considerations = {:>). From what I have recently read, theoretically if you could do = your exit the best way, you might even get a small thrust benefit - at = least enough to overcome the cooling drag. However, I think the best = most can do is simply provide an unimpeded exit flow and minimize = losses. =20 There is some interesting information on usefulness of cowl flaps and = why they some times do not seem to make any difference. I don't claim = to fully understand it all, but it appears that once your losses in the = duct exceed a certain limit - opening up or even creating a low pressure = region at the exit does not promote more air flow through the duct. = There is only so much energy in the air velocity to turn into dynamic = pressure and if your losses in the duct total up to your dynamic energy = limit then nothing you do at the exit will improve the flow. At least = that is the way it appears to this old brain. But, it sure keeps an old brain from freezing up completely trying to = understand some of this. I personally believe that all of the = literature is pretty clear that the best thing you can do with your duct = work is to prevent flow separation in the diffuser.=20 Cooling goes down and drag goes up - not what we are looking for. = Its now finally clear why some of the reports quote 7-11 deg as max = diffuser divergence angles (2theta) and others show good diffuser = performance up around 60 deg divergence. The reason for the two = (seemingly conflicting) different findings is two different diffuser = configurations. One with no resistance behind it and one with = resistance (radiator). Another important basic is to set down and figure out the air mass = flow you must have to handle your critical cooling regime (full power = climb out?). That then drives your inlet size, the size cooler you need = - and as they say - is the basis from which all else flows(pun = intended). But as you say how many of us do that. I find that it is often similar differences that can/do end up = confusing those of us who are ignorant but trying to understand and = apparently find conflicting findings in these reports. You reallllllyyy = have to read them carefully from end to end. Ed ----- Original Message -----=20 From: wrjjrs@aol.com=20 To: Rotary motors in aircraft=20 Sent: Thursday, November 08, 2007 10:28 AM Subject: [FlyRotary] Re: Total,duct, Ambient or Velocity???? Ed, It seems like a cogent discription Ed. I have been studying the = problem for some time. I like your no core example, much cheaper but it = will only fly once. (And for a short time!) The question I have been = pondering is, does it really help us to consider a exit ducting to = direct our exit flows. The data you presented seems to indicate that it = does. The dynamics of the pressure drop across the core contain = compromises related to the efficiency of the heat exchanger, flow of the = water in it and air through it. Many of the designs I see lately pay = very little attention to the exit and re-merging the flow. In = core-in-the-standard-inlet systems such as yours the exit ducting may = not be practical. This is a problem I have see with the Eggenfellner = Sabaru installations as well. At least the rotary can have some exit = area without the cylinders right there in the way! The exit question = tends to favor the chin scoop. The problem is that this has always = proven to be a high drag choice. Currently I'm favoring a vertical side = radiator (or radiators) ducted to the outside (cowl) blowing into the = engine area with a diversion duct to turn the air towards the normal = rear bottom exit. Possibly with a cowl flap for climb. These have never = been easy choices. Often we intend an elegant solution, only to be = rebuffed by the need for hoses, wires, and exhaust pipes and other = unimportant stuff like that. ;-)=20 Thanks for all your research, Bill Jepson -----Original Message----- From: Ed Anderson To: Rotary motors in aircraft Sent: Thu, 8 Nov 2007 5:05 am Subject: [FlyRotary] Re: Total,duct, Ambient or Velocity???? Hi Bill, It is my opinion, based on my limited knowledge of the topic, that = dynamic pressure in the duct is the most significant factor. If you = don't have it - you have no flow. If you do have it you will have flow = but you could have significant Major losses - that's why you may need = other types of pressure measurements to figure out the problem. In = fluid flow talk, they appear to refer to loss of energy through wall = friction as a major loss as it is not recoverable (but this is minor at = our speeds) , while trades between dynamic and static in the duct result = in "minor" losses which may or may not really be minor. Here is my understanding, you would like to convert dynamic energy = to static pressure increase in front of the core as that slows down the = velocity reducing drag and tends to give you more even velocity = distribution across the core (assuming little or no separation of flow = from the duct walls). You would like the greatest pressure drop across = the core which results in the highest velocity through the core tubes = generating turbulence for better heat transfer. However, there is a balancing point, more pressure drop generally = means better heat transfer from metal to air, however, it also generally = means less mass flow because of the resistance. Too much pressure drop = =3D too little mass flow and overheating, too little pressure drop =3D = great mass flow but higher duct drag and less heat transfer per unit = time which can also lead to overheating. =20 I like to use this example to emphasize the point. You would get = maximum pressure drop by placing a solid board across the duct - = however, the air flow would be nil and cooling likewise. On the other = hand, if you remove all obstructions in the duct (including the core) , = the pressure drop would be nil, the airflow would be maximum but = cooling would still be nil. The only significant difference is the no = core approach is cheaper and causes less drag {:>) In any case, all the literature I have read seems to indicate that = the difference in pressure between the inlet and out let of the duct is = a (if not THE) key factor. That dynamic pressure is the only thing = (assuming no fans/blowers) that will move significant air through the = duct. Since this dynamic pressure is referenced to the dynamic pressure = available in the freestream flow as that is what it starts out as, I = personally think referencing dynamic pressure measurements to ambient = air is what we are mainly interested. This is rather than referencing = it to the duct static pressure as shown in the diagram. But, you have = to remember this is all from the guy who has not done any duct = instrumentation. But, my reason for focusing on dynamic pressure is that you can = infer a lot from your duct dynamic pressure readings about what is going = on in the duct. If your dynamic pressure is down, then your static = pressure is up and vice versa. If you have dynamic pressure then you = have flow while static pressure does not necessarily tell you that. =20 However, it all really depends on what you are trying to figure out = on what measurements you take. It would appear if you know how to interpret what you are measuring = then all provide some useful information. That's about the extent of my limited knowledge. Ed ----- Original Message -----=20 From: WRJJRS@aol.com=20 To: Rotary motors in aircraft=20 Sent: Thursday, November 08, 2007 12:28 AM Subject: [FlyRotary] Re: Total,duct, Ambient or Velocity???? Ed, The slide is a good way to explain the various references. I = am still confused as to what will give you the "best" data. The static = in duct pressure compared to the total, or to the velocity? It probably = doesn't matter if you use the same method all the time. Bill Jepson -------------------------------------------------------------------------= - See what's new at AOL.com and Make AOL Your Homepage. -------------------------------------------------------------------------= --- Email and AIM finally together. You've gotta check out free AOL = Mail! -------------------------------------------------------------------------= ----- No virus found in this incoming message. Checked by AVG Free Edition.=20 Version: 7.5.503 / Virus Database: 269.15.24/1115 - Release Date: = 7/11/2007 9:21 AM ------=_NextPart_000_0025_01C822A8.2DFC03C0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
 Ed,
If I'm understanding you correctly, it = appears that=20 you need dynamic pressure ( flow) that turns into a high static = pressure=20 (at the Rad face).
To maintain this high static pressure, = the dynamic=20 flow must be free of turbulence, which is associated with flow = separation from=20 the duct walls. Hence the need for proper divergent angles.
 
There must be good pressure drop across = the=20 Rad,  not too high or you lose heat transfer, not too low as to = create=20 excess drag. There must be some turbulence within the duct fins to = enhance heat=20 transfer, but not too much as to create restrictions.
 
I still feel a low pressure area behind = the rad=20 would be beneficial. 
George ( down under)
Even in the Naca studies they often 'fess up = that=20 theoretical considerations must give way to practical installation=20 considerations {:>).  From what I have recently read, = theoretically if=20 you could do your exit the best way, you might even get a small thrust = benefit=20 - at least enough to overcome the cooling drag.  However, I think = the=20 best most can do is simply provide an unimpeded exit flow and minimize = losses. 
 
There is some interesting information on = usefulness of=20 cowl  flaps and why they some times do not seem to make any=20 difference.  I don't claim to fully understand it all, but it = appears=20 that once your losses in the duct exceed a certain limit - opening up = or even=20 creating a low pressure region at the exit does not promote more air = flow=20 through the duct.  There is only so much energy in the air = velocity to=20 turn into dynamic pressure and if your losses in the duct total up to = your=20 dynamic energy limit then nothing you do at the exit will improve the=20 flow.  At least that is the way it appears to this old=20 brain.
 
But, it sure keeps an old brain from freezing = up=20 completely trying to understand some of this.  I personally = believe that=20 all of the literature is pretty clear that the best thing you can do = with your=20 duct work is to prevent flow separation in the = diffuser. 
 
 Cooling goes down and drag goes up - not = what we=20 are looking for.  Its now finally clear why some of the = reports =20 quote 7-11 deg as max diffuser divergence angles (2theta) and others = show good=20 diffuser performance up around 60 deg divergence.  The reason for = the two=20 (seemingly conflicting) different findings is two different diffuser=20 configurations.  One with no resistance behind it and one with = resistance=20 (radiator).
 
 Another important basic is to set down = and figure=20 out the air mass flow you must have to handle your critical cooling = regime=20 (full power climb out?).  That then drives your inlet size, the = size=20 cooler you need - and as they say - is the basis from which all = else=20 flows(pun intended).  But as you say how many of us do = that.
 
I find that it is often similar differences = that can/do=20 end up confusing those of us who are ignorant but trying to understand = and=20 apparently find conflicting findings in these reports.  You = reallllllyyy=20 have to read them carefully from end to end.
 
Ed
----- Original Message -----
From:=20 wrjjrs@aol.com=20
To: Rotary motors in = aircraft=20
Sent: Thursday, November 08, = 2007 10:28=20 AM
Subject: [FlyRotary] Re: = Total,duct,=20 Ambient or Velocity????

Ed,
It seems like a cogent discription Ed. I have been studying the = problem=20 for some time. I like your no core example, much cheaper but it will = only=20 fly once. (And for a short time!) The question I have been pondering = is,=20 does it really help us to consider a exit ducting to direct our exit = flows.=20 The data you presented seems to indicate that it does. The dynamics = of the=20 pressure drop across the core contain compromises related to the = efficiency=20 of the heat exchanger, flow of the water in it and air through it. = Many of=20 the designs I see lately pay very little attention to the exit and=20 re-merging the flow. In core-in-the-standard-inlet systems such as = yours the=20 exit ducting may not be practical. This is a problem I have see with = the=20 Eggenfellner Sabaru installations as well. At least the rotary can = have some=20 exit area without the cylinders right there in the way! The exit = question=20 tends to favor the chin scoop. The problem is that this has always = proven to=20 be a high drag choice. Currently I'm favoring a vertical side = radiator (or=20 radiators) ducted to the outside (cowl) blowing into the engine area = with a=20 diversion duct to turn the air towards the normal rear bottom exit. = Possibly=20 with a cowl flap for climb. These have never been easy choices. = Often we=20 intend an elegant solution, only to be rebuffed by the need for = hoses,=20 wires, and exhaust pipes and other unimportant stuff like that.=20 ;-) 
Thanks for all your research,
Bill=20 Jepson

-----Original Message-----
From: Ed Anderson=20 <eanderson@carolina.rr.com>
To: Rotary motors in aircraft=20 <flyrotary@lancaironline.net>
Sent: Thu, 8 Nov 2007 5:05=20 am
Subject: [FlyRotary] Re: Total,duct, Ambient or=20 Velocity????

Hi Bill,
 
It is my opinion, based on my limited = knowledge of=20 the topic, that dynamic pressure in the duct is the most significant = factor.  If you don't have it - you have no flow.  If you = do have=20 it you will have flow but you could have significant Major = losses -=20 that's why you may need other types of pressure measurements to = figure out=20 the problem.  In fluid flow talk, they appear to refer to loss = of=20 energy through  wall friction as a major loss as it is not = recoverable=20 (but this is minor at our speeds) , while trades between dynamic and = static=20 in the duct result in "minor" losses which may or may not really be=20 minor.
 
Here is my understanding, you would like to = convert=20 dynamic energy to static pressure increase in front of the core as = that=20 slows down the velocity reducing drag and tends to give you more = even=20 velocity distribution across the core (assuming little or no = separation of=20 flow from the duct walls).  You would like the greatest = pressure drop=20 across the core which results in the highest velocity through the=20 core tubes generating turbulence for better heat = transfer.
 
  However, there is a balancing point, more = pressure=20 drop generally means better heat transfer from metal to air, = however, it=20 also generally means less mass flow because of the resistance.  = Too=20 much pressure drop =3D too little mass flow and overheating, too = little=20 pressure drop =3D great mass flow but higher duct drag and less heat = transfer=20 per unit time which can also lead to overheating.  =
 
I like to use this example  to emphasize = the=20 point.  You would get maximum pressure drop by placing a solid = board=20 across the duct - however, the air flow would be nil and cooling=20 likewise.  On the other hand, if you remove all obstructions in = the=20 duct (including the core) , the pressure drop would be nil, =  the=20 airflow would be maximum but cooling would still be nil.  The = only=20 significant  difference is the  no core approach is = cheaper=20 and causes less drag {:>)
 
In any case, all the literature I have read = seems to=20 indicate that the difference in pressure between the inlet and out = let of=20 the duct is a (if not THE) key factor.  That dynamic pressure = is the=20 only thing (assuming no fans/blowers) that will move significant air = through=20 the duct.  Since this dynamic pressure is referenced to the = dynamic=20 pressure available in the freestream flow as that is what it starts = out as,=20 I personally think referencing dynamic pressure measurements to = ambient air=20 is what we are mainly interested.  This is  rather than=20 referencing it to the duct static pressure as shown in the=20 diagram.  But, you have to remember this = is all=20 from  the guy who has not done any duct = instrumentation.
 
But, my reason for focusing on dynamic = pressure  is=20 that  you can infer a lot from your duct dynamic pressure = readings=20 about what is going on in the duct.  If your dynamic pressure = is down,=20 then your static pressure is up and vice versa. If you have dynamic = pressure=20 then you have flow while static pressure does not necessarily tell = you=20 that. 
 
However, it all really depends on what you are = trying to=20 figure out on what measurements you take.
It would appear if you know how to interpret = what you are=20 measuring then all provide some useful information.
 
That's about the extent of my limited=20 knowledge.
 
Ed
 
 
----- Original Message ----- =
From:=20 WRJJRS@aol.com=20
To: Rotary motors in = aircraft=20
Sent: Thursday, November = 08, 2007=20 12:28 AM
Subject: [FlyRotary] Re: = Total,duct,=20 Ambient or Velocity????

Ed, The slide is a good way to explain the various = references. I am=20 still confused as to what will give you the "best" data. The = static in=20 duct pressure compared to the total, or to the velocity?  It = probably=20 doesn't matter if you use the same method all the time.
Bill Jepson
 



See what's new at AOL.com and Make AOL Your = Homepage.
Email and AIM finally together. You've gotta check out free AOL Mail!

No virus found in this incoming message.
Checked by AVG Free = Edition.
Version: 7.5.503 / Virus Database: 269.15.24/1115 - = Release Date:=20 7/11/2007 9:21 AM
------=_NextPart_000_0025_01C822A8.2DFC03C0--