X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Received: from wx-out-0506.google.com ([66.249.82.233] verified) by logan.com (CommuniGate Pro SMTP 5.1.12) with ESMTP id 2361193 for flyrotary@lancaironline.net; Sun, 30 Sep 2007 20:21:43 -0400 Received-SPF: pass receiver=logan.com; client-ip=66.249.82.233; envelope-from=msteitle@gmail.com Received: by wx-out-0506.google.com with SMTP id i27so2744217wxd for ; Sun, 30 Sep 2007 17:21:06 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; bh=bUEFXvJR0pVhS96wvFFls3ORWWefZgruYNZMHTbbavQ=; b=fBdM3HsPeamC/R4iLwGOk22tsu6g9b7UoOKV6Q81TMhiVwmVGgyFG7+pdlWSd5Gr9Lsq2PDqnySVVVx0RCgh941k/n48J5r/gLF4QJ3YJU06GeMOFghUwshvjOuukrPALUQPWyHkHNYt/tdult/sOfJAzHuki/bUHch8Wpzck+Q= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references; b=fmF0tGVfWxyKmtE6DwKNLHLF+KJWkkr+oGgoVVfxD9tIxawkUdybtceEfoDniSGRPe29ekztddpnnr+x0NK56+Xx9llCRBjzeYHBa3Re/A5bmXSfaw8y4hfCGfCz/KtNn1df2X0U/3+1alaO9BpWf7Ehnr8JaTrbWUPu53YGAyo= Received: by 10.90.31.2 with SMTP id e2mr3173906age.1191198066344; Sun, 30 Sep 2007 17:21:06 -0700 (PDT) Received: by 10.90.35.16 with HTTP; Sun, 30 Sep 2007 17:21:06 -0700 (PDT) Message-ID: <5cf132c0709301721q1e5ee6f2j26da3c3286db9783@mail.gmail.com> Date: Sun, 30 Sep 2007 19:21:06 -0500 From: "Mark Steitle" To: "Rotary motors in aircraft" Subject: Re: [FlyRotary] Re: Another cooling question In-Reply-To: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_8017_2145865.1191198066334" References: ------=_Part_8017_2145865.1191198066334 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Ed, Tracy, Lynn, George, etc , Thanks for the explanations, now I just need to find someone to explain it to me. ;-) I figured the numbers were wrong, just didn't know exactly why. I recalibrated my coolant sensors today using a candy "mercury" thermometer from Wally-World. The EM-2 sensor was ok at 212, but was reading high at ambient (84*). After recalibration it appears much closer throughout the entire range. The EFIS/One was also ok at 212* but was off further down the line. Using the candy thermometer, I went through the drill of heating the water to boiling, holding it there for everything to stabilize, then removing the heat and monitor the thermomter and EFIS/One as the temps fell. They should now be good to within +/- a couple of degrees. Next weekend I hope to do the same to the oil temp sensors. So, I had to go fly and try it out. I didn't get all the numbers, but coolant temps now show a spread of 15*. OAT was 24C and in an easy cruise (5200 rpm, 117 knts), the coolant temps were 160* out of engine and 145* return. So, if I had an operable cowl flap, I surely could have closed it up some to reduce my cooling drag. There were too many cumulus clouds to go above 2900', but the preliminary information looks promising. Also, I need to "open 'er up" and see what happens to the temps. Hopefully, the extra dynamic pressure will be more than enough to handle the extra BTU's. Oil temps settled in at 210* out of the engine, 185* return from oil cooler. I may experiment with reducing the coolant inlet duct and see if I can get the oil and water a bit closer. I also need to instrument each cooler with manometers to check for balance. But for now, after seven years of building, I'm happy that everything is staying in the green. Thanks to all for your input. This is definitely one for the archives. Mark On 9/30/07, Ed Anderson wrote: > > > > ----- Original Message ----- > *From:* Mark Steitle > *To:* Rotary motors in aircraft > *Sent:* Sunday, September 30, 2007 6:28 AM > *Subject:* [FlyRotary] Another cooling question > > > ED wrote: > > Mark, if you really had excess air flowing through your radiators the > coolant would drop more than 4 Deg F. In fact, the more air flow the more > coolant Delta T you would drop through the radiator. > > > That's exactly what I HAD thought, until I was told that the air could > pass through too fast and not pick up as much heat. This didn't make sense > to me. Maybe I wasn't listening closely and missed the point altogether > (wouldn't be the first time). > > > This is one of the oldest myths around - that air or coolant will flow too > fast to pick up the heat. It just IS NOT factual. The more mass flow you > have, the more heat you will carry away. It appears that some early > experimenters noted that if you slowed the flow of coolant through a > radiator that there was a greater temperature drop of the fluid than if it > flowed through faster. This apparently gave rise to the myth as you can > still find references to that experiment supporting the slower is better > myth. I once had an debate with a fererent believer in that myth, after > about 30 minutes of getting no where in convincing the individual of the > factual side, I resorted to this line. > > "So you claim that slow water cools better than fast water, the response > was "Yes", then I replied "If slower and slower water cools better and > better then stopped water must cool best - right?" A long silence, then > the individual hung up the phone. > > The fact is the coolant (in this example of slowing coolant through the > radiator) will indeed lose more heat to the air - if you slow its flow > through the radiator, because that slug of coolant spends more time > exchanging heat with the air. However, the slower flow also means you are > removing less heat from your engine - which is the real objective. > > We know that molecules of air transport the vast majority of the Heat > (there is a very small amount radiated away) in our installations through > contact with the metal of the radiator. The average speed of these > molecules (in air) is approximate the speed of sound (1100 feet/sec at sea > level). So any velocity of the macro airstream in our ducts and cores are > insignificant compared to the air molecules velocity. So speeding up this > air flow or slowing it down has no measurable effect on the frequency at > which the molecules contact the metal. Turbulent flow has more impact than > velocity change. Now changing the velocity of the flow does effect the > mass flow through the core and therefore our overall cooling effectiveness, > it just does not effect the "speed" with which the heat is transfer from > metal to air. > > > > What I DO know is that the air is flowing faster through the water > radiator than the oil radiator. (I'm not sure I have the ASI's hooked up > correctly, but they're both hooked up the same). I have a pitot behind each > radiator hooked up to two separate ASI's. In slow cruise, say 125-130 kts, > the water radiator ASI will read about 110knts and the oil ASI will read > about 90 kts. > > 110 kts would give you a dynamic pressure of approx 7.8 " H20. Now what > that is measuring depends on how your ASIs are hooked up. Since they are > differential pressure gauges they are measuring the difference between the > static pressure under your cowl and what ever reference their static side > accesses. If they reference the ampient outside air pressure (as you static > system does) then you are measuring Cowl pressure relative to ambient. If > there static lines are simply open to the cabin, then since cabin pressure > in normally a bit lower than ambient pressure, it would exaggerate the > reading a bit. > > But, in any case, it appears to me that you are measuring localized cowl > pressure. If you had a pitot tube before the core with its static > referencing ambient pressure, then it would be the dynamic pressure > (converted to a static pressure increase). If you had the pitot tube before > the core and the static accessing the cowl, then you would be measuring > pressure across your core. If you have the pitot tube under the cowl and the > static referencing ambient you would be measuring your cowl pressure. So it > depends on your static reference as well as where you have the pitot tube > position as to what you are measuring. > > > The way it was behaving before I opened up the exit, it appeared that > the air from the water radiator was trying to exit backwards through the oil > inlet. I say this because of how high the oil temps were reading. > I enlarged the cowl exit, and both the water and oil temps dropped > significantly. > > I would say your analysis is correct. Dennis also found that enlarging > his exit area improved the cooling. > > The ASI's are referencing the static port for these readings; should they > be referencing cowl or cabin pressure instead? Airspeeds readings seem > awfully high to me. > > Referencing the static port would then give you under the cowl pressure. > If you reference the cowl then you would be measuring the localized dynamic > pressure of the air (greater than existing under the cowl pressure) exiting > the core which I would expect to be small since your duct should have > converted most of the dynamic pressure to a static pressure increase before > the core. > Mark > (Going to the airport today to recalibrate temp sensors) > > > ------=_Part_8017_2145865.1191198066334 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline
Ed, Tracy, Lynn, George, etc
,
Thanks for the explanations, now I just need to find someone to explain it to me.  ;-)  I figured the numbers were wrong, just didn't know exactly why.
 
I recalibrated my coolant sensors today using a candy "mercury" thermometer from Wally-World.  The EM-2 sensor was ok at 212, but was reading high at ambient (84*).  After recalibration it appears much closer throughout the entire range. 
 
The EFIS/One was also ok at 212* but was off further down the line.  Using the candy thermometer, I went through the drill of heating the water to boiling, holding it there for everything to stabilize, then removing the heat and monitor the thermomter and EFIS/One as the temps fell.  They should now be good to within +/- a couple of degrees.  Next weekend I hope to do the same to the oil temp sensors.   
 
So, I had to go fly and try it out.  I didn't get all the numbers, but coolant temps now show a spread of 15*.  OAT was 24C and in an easy cruise (5200 rpm, 117 knts), the coolant temps were 160* out of engine and 145* return.  So, if I had an operable cowl flap, I surely could have closed it up some to reduce my cooling drag.  There were too many cumulus clouds to go above 2900', but the preliminary information looks promising.  Also, I need to "open 'er up" and see what happens to the temps.  Hopefully, the extra dynamic pressure will be more than enough to handle the extra BTU's. 
 
Oil temps settled in at 210* out of the engine, 185* return from oil cooler. 
 
I may experiment with reducing the coolant inlet duct and see if I can get the oil and water a bit closer.  I also need to instrument each cooler with manometers to check for balance.  But for now, after seven years of building,  I'm happy that everything is staying in the green.
 
Thanks to all for your input.  This is definitely one for the archives. 
 
Mark   

 
On 9/30/07, Ed Anderson <eanderson@carolina.rr.com> wrote:
 
----- Original Message -----
Sent: Sunday, September 30, 2007 6:28 AM
Subject: [FlyRotary] Another cooling question

 
ED wrote:
<snip>
 Mark, if you really had excess air flowing through your radiators the coolant would drop more than 4 Deg F.  In fact, the more air flow the more coolant Delta T you would drop through the radiator. 
<snip>
 
That's exactly what I HAD thought, until I was told that the air could pass through too fast and not pick up as much heat.  This didn't make sense to me.  Maybe I wasn't listening closely and missed the point altogether (wouldn't be the first time). 
 
 
This is one of the oldest myths around - that air or coolant will flow too fast to pick up the heat.  It just IS NOT factual.  The more mass flow you have, the more heat you will carry away.  It appears that some early experimenters noted that if you slowed the flow of coolant through a radiator that there was a greater temperature drop of the fluid than if it flowed through faster.  This apparently gave rise to the myth as you can still find references to that experiment supporting the slower is better myth.  I once had an debate with a fererent believer in that myth, after about 30 minutes of getting no where in convincing the individual of the factual side, I resorted to this line.
 
"So you claim that slow water cools better than fast water, the response was "Yes", then I replied "If slower and slower water cools better and better  then stopped water must cool best - right?"  A long silence, then the individual hung up the phone. 
 
The fact is the coolant (in this example of slowing coolant through the radiator) will indeed lose more heat to the air - if you slow its flow through the radiator, because that slug of coolant  spends more time exchanging heat with the air.  However, the slower flow also  means you are removing less heat from your engine - which is the real objective.  
 
 We know that molecules of air transport the vast majority of the Heat (there is a very small amount radiated away) in our installations through contact with the metal of the radiator.   The average speed of these molecules (in air) is approximate the speed of sound (1100 feet/sec at sea level).  So any velocity of the macro airstream in our ducts and cores are insignificant compared to the air molecules velocity.  So speeding up this air flow or slowing it down has no measurable effect on the frequency at which the molecules contact the metal.  Turbulent flow has more impact than velocity change.  Now changing the velocity of  the flow does effect the mass flow through the core and therefore our overall cooling effectiveness, it just does not effect the "speed" with which the heat is transfer from metal to air.
 
 
 
What I DO know is that the air is flowing faster through the water radiator than the oil radiator.  (I'm not sure I have the ASI's hooked up correctly, but they're both hooked up the same).  I have a pitot behind each radiator hooked up to two separate ASI's.  In slow cruise, say 125-130 kts, the water radiator ASI will read about 110knts and the oil ASI will read about 90 kts. 
 
110 kts would give you a dynamic pressure of approx 7.8 " H20.  Now what that is measuring depends on how your ASIs are hooked up.  Since they are differential pressure gauges they are measuring the difference between the static pressure under your cowl and what ever reference their static side accesses.  If they reference the ampient outside air pressure (as you static system does) then you are measuring Cowl pressure relative to ambient.  If there static lines are simply open to the cabin, then since cabin pressure in normally a bit lower than ambient pressure, it would exaggerate the reading a bit.
 
But, in any case, it appears to me that you are measuring localized cowl pressure.  If you had a pitot tube before the core with its static referencing ambient pressure, then it would be the dynamic pressure (converted to a static pressure increase).  If you had the pitot tube before the core and the static accessing the cowl, then you would be measuring pressure across your core. If you have the pitot tube under the cowl and the static referencing ambient you would be measuring your cowl pressure. So it depends on your static reference as well as where you have the pitot tube position as to what you are measuring.
 
 
 The way it was behaving before I opened up the exit, it appeared that the air from the water radiator was trying to exit backwards through the oil inlet.  I say this because of how high the oil temps were reading.  I enlarged the cowl exit, and both the water and oil temps dropped significantly. 
 
I would say your analysis is correct.  Dennis also found that enlarging his exit area improved the cooling. 
 
The ASI's are referencing the static port for these readings; should they be referencing cowl or cabin pressure instead?  Airspeeds readings seem awfully high to me. 
 
Referencing the static port would then give you under the cowl pressure.  If you reference the cowl then you would be measuring the localized  dynamic pressure of the air (greater than existing under the cowl pressure) exiting the core which I would expect to be small since your duct should have converted most of the dynamic pressure to a static pressure increase before the core .
Mark
(Going to the airport today to recalibrate temp sensors)
 

------=_Part_8017_2145865.1191198066334--