X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Received: from ispmxmta05-srv.alltel.net ([166.102.165.166] verified) by logan.com (CommuniGate Pro SMTP 5.0.7) with ESMTP id 964317 for flyrotary@lancaironline.net; Wed, 01 Feb 2006 20:35:43 -0500 Received-SPF: pass receiver=logan.com; client-ip=166.102.165.166; envelope-from=montyr2157@alltel.net Received: from Thorstwin ([67.141.70.54]) by ispmxmta05-srv.alltel.net with SMTP id <20060202013456.QVUQ8731.ispmxmta05-srv.alltel.net@Thorstwin> for ; Wed, 1 Feb 2006 19:34:56 -0600 Message-ID: <001801c62798$f1ab3780$01fea8c0@Thorstwin> From: "M Roberts" To: Subject: sutability of NPG for rotary engine use Date: Wed, 1 Feb 2006 19:35:26 -0600 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0015_01C62766.A6E857F0" 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 This is a multi-part message in MIME format. ------=_NextPart_000_0015_01C62766.A6E857F0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I wouldn't worry about the theory behind mdot*Cp*deltaT too much. This = has been tested thoroughly. As long as your Cp number is correct the = calculation will be nuts on. The place you get into trouble is measuring = the constants for Cp or for a heat transfer coefficient. Your analysis = looks good to me Ed. The point Ernest made is a valid one. Cp is per = unit mass. A more dense fluid will transfer more heat per volume flow = than a less dense fluid. Bill S. also makes some good points. The main thing I have to add is: it would be nice to know what the = convection coefficient is for NPG. That is what gives the heat transfer = between the hot metal and the coolant. A more viscous fluid would tend = to have a thicker boundary layer and less turbulence. That could cause = problems. The turbulence and mixing of the boundary layer help to = transfer heat.=20 I also would be cautious about the vapor pressure. Boiling is not a bad = thing. It is a good thing. The heat transfer coefficient for a phase = change (liquid to gas) is infinite. This helps to cool a hot spot. We = are talking about sub cooled boiling here where the bulk liquid is = cooler than the boiling point. Locally the liquid boils and transfers = all the heat the metal can move. The limiting factor is actually the = metal conduction for this case. The bubbles of vapor are cooled by the = surrounding coolant and collapse. Put a pot on the stove and watch as = you transfer from sub-cooled to nucleate and finally bulk boiling. You = can see the process happen.=20 Both bulk and nucleate boiling are to be avoided. Sub cooled boiling, a = thin boundary layer and turbulence are all good things. NPG strikes out = on all three. In addition it requires more power to pump and the = pressure drop through the evaporator core type coolers at low temps is = suspect.=20 In short: The 13b was developed to use water/glycol as a coolant. To properly validate NPG you need a dyno and a lot of thermocouples, = plus a way to measure the mass flow of the coolant, pressure drops, pump = power, and the heat transfer coefficient.=20 Anybody got that laying around in their hangar? Do you want to be a guinea pig? I would not use NPG. Monty ------=_NextPart_000_0015_01C62766.A6E857F0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

I wouldn't worry about the theory = behind=20 mdot*Cp*deltaT too much. This has been tested thoroughly. As long as = your Cp=20 number is correct the calculation will be nuts on. The place you get = into=20 trouble is measuring the constants for Cp or for a heat transfer = coefficient.=20 Your analysis looks good to me Ed. The point Ernest made is a valid one. = Cp is=20 per unit mass. A more dense fluid will transfer more heat per volume = flow than a=20 less dense fluid. Bill S. also makes some good points.

The main thing I have to add is: it = would be nice=20 to know what the convection coefficient is for NPG. That is what gives = the heat=20 transfer between the hot metal and the coolant. A more viscous fluid = would tend=20 to have a thicker boundary layer and less turbulence. That could cause = problems.=20 The turbulence and mixing of the boundary layer help to transfer heat.=20

I also would be cautious about the = vapor pressure.=20 Boiling is not a bad thing. It is a good thing. The heat transfer = coefficient=20 for a phase change (liquid to gas) is infinite. This helps to cool a hot = spot.=20 We are talking about sub cooled boiling here where the bulk liquid is = cooler=20 than the boiling point. Locally the liquid boils and transfers all the = heat the=20 metal can move. The limiting factor is actually the metal conduction for = this=20 case. The bubbles of vapor are cooled by the surrounding coolant and=20 collapse. Put a pot on the stove and watch as you transfer from = sub-cooled=20 to nucleate and finally bulk boiling. You can see the process=20 happen.

Both bulk = and nucleate boiling are=20 to be avoided. Sub cooled boiling, a thin boundary layer = and turbulence are=20 all good things. NPG strikes out on all three. In addition it = requires more=20 power to pump and the pressure drop through the evaporator core type = coolers at=20 low temps is suspect.

In short:

The 13b was developed to use = water/glycol as a=20 coolant.

To = properly validate NPG you need a=20 dyno and a lot of thermocouples, plus a way to measure the mass flow of = the=20 coolant, pressure drops, pump power, and the heat transfer = coefficient.=20

Anybody got that laying around in their = hangar?

Do you want to be a guinea = pig?

I would not use NPG.

Monty

------=_NextPart_000_0015_01C62766.A6E857F0--