X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Received: from cdptpa-omtalb.mail.rr.com ([75.180.132.120] verified) by logan.com (CommuniGate Pro SMTP 5.2c1) with ESMTP id 2464331 for flyrotary@lancaironline.net; Fri, 09 Nov 2007 21:19:34 -0500 Received-SPF: pass receiver=logan.com; client-ip=75.180.132.120; envelope-from=eanderson@carolina.rr.com Received: from edward2 ([24.74.103.61]) by cdptpa-omta04.mail.rr.com with SMTP id <20071110021856.USL5837.cdptpa-omta04.mail.rr.com@edward2> for ; Sat, 10 Nov 2007 02:18:56 +0000 Message-ID: <001a01c82340$0d955a30$2402a8c0@edward2> From: "Ed Anderson" To: "Rotary motors in aircraft" References: Subject: Re: [FlyRotary] Re: Total,duct, Ambient or Velocity???? Date: Fri, 9 Nov 2007 21:18:58 -0500 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0017_01C82316.246942F0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.3138 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3138 This is a multi-part message in MIME format. ------=_NextPart_000_0017_01C82316.246942F0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable George, here you are getting into something we have not discussed in = depth. Two equations/laws of fluid dynamics are involved. Bernoulli's equation = and an equation called the law of continuity. This equation relates to = the fact that you don't create or lose mass in the duct, so the mass = flow is a constant everywhere in the duct. The mass flow is frequently = shown as the product of air density*cross section area*air velocity =3D = mass flow or simply p*A*V The equation goes something like this, the p1A1V1 (mass flow at point = 1) =3D p2A2V2 (mass flow at point 2). Since the air is normally = considered to act like it is incompressible at the lower speeds we are = talking about, that means the density p1=3D p2, so we can drop them = from the equation for this explanation. That leaves us with A1V1 =3D A2V2 or the product of the area and = velocity at point 1 is equal to the area and velocity at point 2 in the = duct. Now if A1 =3D A2 then V1 has to equal V2 for the two sides of = the equation to be equal. But, what if A2 =3D 2* A1 or the cross = section area of point 2 is made twice the cross section area of point 1. = Then if A2 =3D 2*A1, we can substitute 2*A1 for A2 in the equation and = we have the following. Taking A1V1 =3D A2V2 and substituting we have A1*V1 =3D (2*A1)*V2. So = what does that tell us about the air velocity at point 2 now that we = have doubled the cross section area there? =20 Well solving the equation for the new V2, We can call the new velocity = at point 2 V2n (for V2 new) with V2o being the old velocity at point 2. = So we have V2n =3D A1V2o/(2*A1) Now we can cancelled the A1 in the = numerator and denominator on right side of the equation leaving V2n =3D V2o/2 This shows us that the new velocity at point 2, V2n = is 1/2 the old velocity (V2o) at point 2 or V2n =3D 0.5V2o So what this says is the velocity starts changing (slowing in this case = and the pressure increasing ) as soon as the cross section area A2 = starts to increase from A1. The process continues until the area stops = expanding (or the kinetic energy of the moving air has all been = converted to a static pressure increase) and that is where the process = is finished as the duct/diffuser has expanded to its maximum area. = Actually, this process happens with both nozzles and diffusers just the = opposite way. Its derived from the Bernoulli equation and the = continuity law. So if you had a duct whose cross section area continued to expand for a = distance of 2" or 20" or 200" then theoretically the pressure would = continue to build and the velocity to decrease until all of the kinetic = energy of the moving air has been converted to pressure increase. This = is all theoretical, there are losses and turbulence and etc, that makes = a difference, but you get the ideal. It depends on your specific = diffuser dimensions. Think of it this way, George, some wind tunnels have diffuser which = expand over 10's of feet while some microscopic cooling systems have = diffusers measured in 10th's of an inch. Now aren't you sorry you asked {:>)? Ed ----- Original Message -----=20 From: George Lendich=20 To: Rotary motors in aircraft=20 Sent: Friday, November 09, 2007 4:52 PM Subject: [FlyRotary] Re: Total,duct, Ambient or Velocity???? Ed and Al,=20 This is all good info me, it either confirms, clarifies or informs. The straw concept is a timely reminder of pressure differentials, a = good example IMHO. One thing I would really love to know is - at what point in the inlet = duct does the dynamic flow change to static pressure. I would assume = this would vary with different shaped ducts and different dynamic flow ( = airflow speed).=20 Your opinions on this or guesstimates ie 1", 2" or 3" from the face of = the rad, would be of great interest to me. George (down under) Hi Al, Not picky - some good points as always . Yes, I agree, = generalization does have its pit falls, but on the other hand I think = they can help promote a conceptual understanding which can be refined = (through study and experiments) to meet a particular situation. As we = know, cooling airflow is attempting to balance conflicting aerodynamic = and thermodynamic principles. =20 I also agree that much of this stuff addresses the "Perfect = theoretical duct" out of necessity as there is only one perfect duct but = many, many implementations that fall short of perfect. So its more of = a conceptual goal to be aimed for - it may never be achieved, but = provides at least guidelines. But,this is just my opinion of course. Actually, I disagree, you can not "suck" air though anything. You = may create a partial pressure difference with the fan, but it is the = higher pressure air on the other end of the duct that pushes or "blows" = air through the duct into the area of lower pressure {:>) . =20 But, semantics aside, yes, I agree, lower exit pressure is what you = are after and that does not always equate to larger exit duct area. In = fact, if the air heated by the core flows through a nozzle it might even = produce thrust and lower exit pressure using a smaller exit. But, in = general, I still believe that in most of our cases, we are short of the = level of duct design that would reliably permit that. What we need is = someone to invest in one of those $$$$ Computer Fluid Flow software = programs and see what they would reveal. Ed ----- Original Message -----=20 From: Al Gietzen=20 To: Rotary motors in aircraft=20 Sent: Friday, November 09, 2007 1:09 AM Subject: [FlyRotary] Re: Total,duct, Ambient or Velocity???? It would seem "reasonable" that a low pressure area at the exit = will help flow through a duct - no argument on that point. What the = report appeared to say is that the after a certain point opening the = exit area wider does not appear to have any additional benefit. (Exit = "area" and exit "pressure" are not interchangeable terms) That if the = duct is capable of "using up" all of the kinetic energy in your air flow = by obstructions, pressure drops and friction losses then enlarging the = exit does not necessarily add to the flow. Remember you can not suck air through a duct, you can only blow it = through. (Of course you can suck air through a duct - I do it after (and = sometimes before) every flight with the fan I have on the back side of = the radiator) So in effect if the straw is pinched you can "suck" on it = all you want but it won't increase flow {:>). =20 If I understood the report, it appears that enlarging the exit = area beyond the frontal area of your core provides little if any = additional benefit. That does not mean cowl flaps never work or provide = benefit. In fact it appears that the better your duct, the more = benefit the cowl flaps appear to have, the worst your duct, the lesser = benefit - just the opposite of what you might think. Ed; Don't mean to be picky, but some of these generalities are making = me nervousJ. These things are applicable only when the duct/diffuser is = operating at max efficiency - which is rarely the case. Lot's of good info. Thanks. You're right; it's some kind of = magic, and you don't know for sure until you built it and try it. Al -------------------------------------------------------------------------= --- 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_0017_01C82316.246942F0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
George, here you are getting into something we = have not=20 discussed in depth.
 
Two equations/laws of fluid dynamics are = involved. =20 Bernoulli's equation and an equation called the law of =  continuity. =20 This equation relates to the fact that you don't create or lose = mass in the=20 duct, so the mass flow is a constant everywhere in the duct.  The = mass flow=20 is frequently shown as the product of air density*cross section area*air = velocity =3D mass flow or simply p*A*V
 
The equation goes something like this, the=20  p1A1V1 (mass flow at point 1) =3D = p2A2V2=20 (mass flow at point 2).  Since the air is normally considered to = act like=20 it is incompressible at the lower speeds we are talking about, =  that means=20 the density  p1=3D p2, so we can drop them from = the equation=20 for this explanation.
 
That leaves us with A1V1 =3D = A2V2 or the=20 product of the area and velocity at point 1 is equal to the area and = velocity at=20 point 2 in the duct.  Now if A1 =3D A2 then = V1=20 has to equal V2 for the two sides  of the = equation to be equal.   But, what if A2  =3D 2* = A1=20 or the cross section area of point 2 is made twice the cross = section area=20 of point 1.  Then if A2 =3D 2*A1, we can = substitute=20 2*A1 for A2 in the equation and we = have the=20 following.
 
Taking A1V1 =3D A2V2 and = substituting we=20 have A1*V1 =3D (2*A1)*V2.   So what does that = tell us=20 about the air velocity at point 2 now that we have doubled the cross = section=20 area there? 
 
Well solving the equation for the new V2, We can = call the=20 new velocity at point 2 V2n (for V2 = new) with=20 V2o being the old velocity at point 2.  =
 
So  we have V2n = =3D=20 A1V2o/(2*A1)  Now we can cancelled the = A1=20 in the numerator and denominator on  right side of the equation=20 leaving
 
V2n  =3D =20 V2o/2    This shows us that the = new=20 velocity at point 2, V2n is 1/2 the old velocity=20 (V2o) at point 2 or   V2n =3D=20 0.5V2o
 
So what this says is the velocity = starts changing=20 (slowing in this case and the pressure increasing ) as=20 soon as the cross section area A2 starts to increase from = A1.  The=20 process continues until the area stops expanding (or the kinetic energy = of the=20 moving air has all been converted to a static pressure increase) =  and that=20 is where the process is finished as the duct/diffuser has expanded to = its=20 maximum area.  Actually, this process happens with both nozzles and = diffusers just the opposite way.  Its derived from the Bernoulli = equation=20 and the continuity law.
 
So if you had a duct whose cross section area = continued to=20 expand for a distance of  2" or 20" or  200" then = theoretically the=20 pressure would continue to build and the velocity to decrease until all = of the=20 kinetic energy of the moving air has been converted to pressure = increase. =20 This is all theoretical, there are losses and turbulence and etc, that = makes a=20 difference, but you get the ideal.  It depends on your specific = diffuser=20 dimensions.
 
Think of it this way, George, some wind tunnels = have=20 diffuser which expand over 10's of feet while some microscopic cooling = systems=20 have diffusers measured in 10th's of an inch.
Now aren't you sorry you asked=20 {:>)?
 
Ed
 
----- Original Message -----
From:=20 George=20 Lendich
Sent: Friday, November 09, 2007 = 4:52=20 PM
Subject: [FlyRotary] Re: = Total,duct,=20 Ambient or Velocity????

 Ed and Al,
This is all good info me, it either = confirms,=20 clarifies or informs.
The straw concept is a timely = reminder of=20 pressure differentials, a good example IMHO.
 
One thing I would really love to know = is - at=20 what point in the inlet duct does the dynamic flow change to static = pressure.=20 I would assume this would vary with different shaped ducts and = different=20 dynamic flow ( airflow speed). 
Your opinions on this = or guesstimates ie 1",=20 2" or 3" from the face of the rad, would be of great = interest to=20 me.
 
George (down under)
Hi Al,
 
Not picky - some good points as always = .  Yes, I=20 agree, generalization does have its  pit falls, but = on the=20 other hand I think they can  help promote a conceptual = understanding=20 which can be refined (through study and experiments) to meet a = particular=20 situation.  As we know, cooling airflow is attempting to = balance=20 conflicting aerodynamic and thermodynamic  principles. =20
 
I also agree that   much of this = stuff=20 addresses the "Perfect theoretical duct" out of necessity as there = is=20 only one perfect duct but many, many implementations=20  that fall short of perfect.  So its more of a = conceptual=20 goal to be aimed for  - it may never be achieved,=20 but provides at least guidelines.   But,this is=20  just my opinion of course.
 
Actually, I disagree, you can not "suck" air = though=20 anything.  You may create a partial pressure difference with = the fan,=20 but it is the higher pressure air on the other end of the duct that = pushes=20 or "blows" air through the duct into the area of lower = pressure =20 {:>)  .  
 
 But, semantics aside, yes, I agree, = lower exit=20 pressure is what you are after and that does not always equate to = larger=20 exit duct area.  In fact, if the air heated by the core flows = through a=20 nozzle it might even produce thrust and lower exit = pressure using a=20 smaller exit.  But, in general, I still believe that in most of = our=20 cases, we are short of the level of duct design that would reliably = permit=20 that.  What we need is someone to invest in one of those $$$$ = Computer=20 Fluid Flow software programs and see what they would = reveal.
 
Ed
----- Original Message ----- =
From:=20 Al=20 Gietzen
To: Rotary motors in = aircraft=20
Sent: Friday, November 09, = 2007 1:09=20 AM
Subject: [FlyRotary] Re: = Total,duct,=20 Ambient or Velocity????

 

It =  would=20 seem "reasonable" that a low pressure area at the exit  will = help=20 flow through a duct - no argument on that point.  What the = report=20 appeared to say is that the after a certain point opening the exit = area=20 wider does not appear to have any = additional benefit. (Exit =93area=94 and exit =93pressure=94 are = not interchangeable=20 terms)=20 That if the duct is capable of "using up" all of the = kinetic=20 energy in your air flow by obstructions, pressure drops  and = friction=20 losses then enlarging the exit does not necessarily  add to = the=20 flow.

 

Remember you can=20 not suck air through a duct, you can only blow it = through. (Of course you can suck air = through a=20 duct =96 I do it after (and sometimes before) every flight with = the fan I=20 have on the back side of the radiator) So in effect = if the=20 straw is pinched you can "suck" on it all you want but it won't = increase=20 flow {:>).  

 

If I = understood=20 the report,  it appears that enlarging the exit area beyond = the=20 frontal area of your core provides little if any additional = benefit. =20 That does not mean cowl flaps never work or provide benefit.  = In fact=20 it appears that the better your duct,  the more benefit the = cowl=20 flaps appear to have, the worst your duct, the lesser benefit - = just the=20 opposite of what you might think.

Ed;

 

Don=92t mean to=20 be picky, but some of these generalities are making me=20 nervousJ.  These=20 things are applicable only when the duct/diffuser is operating at = max=20 efficiency =96 which is rarely the case.

Lot=92s = of good=20 info. =20 Thanks.  You=92re right; it=92s some kind of magic, and you = don=92t know=20 for sure until you built it and try it.

Al

No virus found in this incoming message.
Checked by AVG = Free=20 Edition.
Version: 7.5.503 / Virus Database: 269.15.24/1115 - = Release=20 Date: 7/11/2007 9:21 AM
------=_NextPart_000_0017_01C82316.246942F0--