Return-Path: Received: from [24.25.9.100] (HELO ms-smtp-01-eri0.southeast.rr.com) by logan.com (CommuniGate Pro SMTP 4.1.6) with ESMTP id 2712509 for flyrotary@lancaironline.net; Thu, 06 Nov 2003 11:37:03 -0500 Received: from o7y6b5 (clt78-020.carolina.rr.com [24.93.78.20]) by ms-smtp-01-eri0.southeast.rr.com (8.12.10/8.12.7) with SMTP id hA6GavN4020199 for ; Thu, 6 Nov 2003 11:36:59 -0500 (EST) Message-ID: <000e01c3a483$9c381ce0$1702a8c0@WorkGroup> From: "Ed Anderson" To: "Rotary motors in aircraft" References: Subject: Cooling Effect of Latent Heat of vaporization was Re: DIE Summary Date: Thu, 6 Nov 2003 11:32:46 -0500 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_000B_01C3A459.B31EF160" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1106 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 X-Virus-Scanned: Symantec AntiVirus Scan Engine This is a multi-part message in MIME format. ------=_NextPart_000_000B_01C3A459.B31EF160 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable =20 From: Al Gietzen=20 To: Rotary motors in aircraft=20 Sent: Thursday, November 06, 2003 9:27 AM Subject: [FlyRotary] Re: DIE Summary SNIP The key of course (more better we call it "detail devil") is coming up = with a way to bias fuel injected through each the two injectors while = maintaining the sum of the quantities injected at that needed for proper = combustion. That's one hell of a devil, but not nearly as difficult as = the original devising of a system that that figures out how much fuel to = inject and then delivers it. I would guess at this point that it would = be easier to adjust intake Temperature than intake Length. Tracy, you = listening??=20 Different folks desiring different flight regimes would need runner = lengths that bracketed the temps and rpm's they expected to be operating = at.=20 Just a theory ... Jim S.=20 Pretty cool theory, Jim. The stock Mazda injectors for turbo engines = have enough flow so either set could handle max power for a NA engine. = And the turbo guys don't need DIE anyway. Modeling an override to = modulate pulse width to allocate fuel shouldn't be that tough if you = know what the curve looks like. Al Interesting problem for sure. Since the EDDIE effect is mostly = desired at higher rpms (not much point having it at idle), the fuel flow = will already be relatively high. Latent heat of vaporization of that = much fuel will probably already have the temperature of the air near a = low point - maybe??. Pumping more fuel in might not necessarily result = in lowering temperatures if the air stream is already saturate with fuel = vapor. I don't know enough about how gasoline acts in such a situation, = so that is just a guess. The basic bottom line is that the engine will need a specific Air/Fuel = ration to produce the desired power. Anything that distrubs that ratio = is likely to disturb the power produced. Now, certainly the engine does = not know or care how that fuel quantity is delivered so long as it gets = what it needs. =20 I can sort of see a program that calculates the amount of fuel needed = to lower the air mass flow X degrees. I would think you would either = need to know the ratio of fuel needed to cool X CFM of air flow Y = degrees or would need to have a temperature sensor and then adjust the = fuel flow to get the temp desired. Then so long as that amount of fuel = injected by the upstream "Temp Control Injectors" does not exceed what = the engine needs overall for the desired Air/Fuel ratio, then the down = stream injectors could be modulated to make up any difference. =20 However, if the amount of upstream fuel needed to get the desired air = temperature exceeds that required for the desired air/fuel ratio for = engine power, then I see a problem. It looks like someone is going to = have to put some number to the problem to see what it really looks like. I found the latent heat of vaporization for gasoline and while there = are some variations (probably dependining on the exact formulation of = the gasoline), it looks like 117 Btu/lb or 900 Btu/Gallon are a rough = average. I know 6.5 lbs/gallon * 117 BTU/LB only =3D 760 Btu/gallon so there is = some inconsistency. But in any case, here are some preliminary numbers. Assuming 5500 rpm = and A/F of 13 producing 161 HP with an inlet temperature of 90F, it = appears that if all the fuel injected and evaporated before reaching the = combustion chamber, it would lower the inlet temperature by 36F down to = 54F. It would seem that regardless of where the fuel is injected, if it all = vaporizes (which you want it to in order for it to burn in the = combustion chamber), then the air temp is lowered by 36F (in this = example - assuming my calcuations are correct). So, I am not certain = how much we could materially affect the cooling and still maintain the = desired Air/Fuel ratio. It seems that if the net total of fuel is that = needed for the desired air/fuel ratio, then it must all vaporize (in = order to burn) and if it does that then regardless of where it does it, = it would seem to lower the accompanying air by 36F. =20 Now, since the velocity of the pulse is dependent on the average air = temp in the manifold, there might be something said for using fuel to = get the desired temp drop early on (immediately after the air enters the = throttle body) and then delay injection of the fuel need to achieve the = desired air/fuel ratio until the last instance (say the primary block = injection ports), then the average velocity of the pulse would likely to = be more impacted by the long average lower air temperature for most of = the runner length before that last bit of fuel is injected. Clearly = lots of things to consider. So some other folks need to run some numbers and see what they get. At 5500 rpm assuming WOT the airflow for the two rotor would be around = 255CFM which works out to around 19.51 Lbs of air/minute. For a = air/fuel ratio of 13, that give 1.50 lbs of fuel/minute. I used the=20 Te =3D Ti +Q/W*cp where Q is the latent heat of vaporation BTU = removed (per pound of fuel)from the air flow (W air in lbs/second), Cp = =3D 0.25 for air. I used 90F for Ti and calculated 54F for Te. So some = of you who understand this much better than I, jump in with your = calculations I think I will leave this approach to others, while really elegant (if = the theory holds to gether), I am going to try the more direct but = simple, brute force approach and vary the lenght of the runners.=20 Ed Anderson =20 ------=_NextPart_000_000B_01C3A459.B31EF160 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
 
From:=20 Al = Gietzen=20
Sent: Thursday, November 06, = 2003 9:27=20 AM
Subject: [FlyRotary] Re: DIE=20 Summary

  SNIP

The key of course (more better we call it = "detail=20 devil") is coming up with a way to bias fuel injected through each the = two=20 injectors while maintaining the sum of the quantities injected at that = needed=20 for proper combustion.  That's one hell of a devil, but not = nearly as=20 difficult as the original devising of a system that that figures out = how much=20 fuel to inject and then delivers it.  I would guess at this point = that it=20 would be easier to adjust intake Temperature than intake Length.  = Tracy,=20 you listening??

Different folks desiring different flight = regimes=20 would need runner lengths that bracketed the temps and rpm's they = expected to=20 be operating at.

Just a theory ... Jim S.

Pretty cool theory, Jim.  = The stock=20 Mazda injectors for turbo engines have enough flow so either set could = handle=20 max power for a NA engine.  And the turbo guys don=92t need DIE=20 anyway.  Modeling an override to modulate pulse width to allocate = fuel=20 shouldn=92t be that tough if you know what the curve looks=20 like.

Al

 

Interesting problem=20 for sure.  Since the EDDIE effect is mostly desired at = higher=20 rpms (not much point having it at idle), the fuel flow will already be = relatively high.  Latent heat of vaporization of that much fuel = will=20 probably already have the temperature of the air near a low point -=20 maybe??.  Pumping more fuel in might not necessarily result in = lowering=20 temperatures if the air stream is already saturate with fuel = vapor.  I=20 don't know enough about how gasoline acts in such a situation, so that = is just=20 a guess.

The basic = bottom=20 line is that the engine will need a specific Air/Fuel ration to = produce the=20 desired power.  Anything that distrubs that ratio is likely to = disturb=20 the power produced.  Now, certainly the engine does not know or = care how=20 that fuel quantity is delivered so long as it gets what it = needs. =20

I can = sort of see a=20 program that calculates the amount of fuel needed to lower the air = mass flow X=20 degrees.  I would think you would either need to know the ratio = of fuel=20 needed to cool X CFM of air flow Y degrees or would need to have a=20 temperature sensor and then adjust the fuel flow to get the temp=20 desired.  Then so long as that amount of fuel injected by the = upstream=20 "Temp Control Injectors" does not exceed what the engine needs overall = for the=20 desired Air/Fuel ratio, then the down stream injectors could be = modulated to=20 make up any difference. 

However, = if the=20 amount of upstream fuel needed to get the desired air temperature = exceeds that=20 required for the desired air/fuel ratio for engine power, then I see a = problem.  It looks like someone is going to have to put some = number to=20 the problem to see what it really looks like.

I found = the latent=20 heat of vaporization for gasoline and while there are some variations=20 (probably dependining on the exact formulation of the gasoline), it = looks like=20 117 Btu/lb  or 900 Btu/Gallon are a rough = average.

I know = 6.5=20 lbs/gallon * 117 BTU/LB only =3D 760 = Btu/gallon so=20 there is some inconsistency.

But in = any case,=20 here are some preliminary numbers.  Assuming 5500 rpm and A/F of = 13=20 producing 161 HP with an inlet temperature of 90F, it appears that if = all the=20 fuel injected and evaporated before reaching the combustion = chamber, it=20 would lower the inlet temperature by 36F down to = 54F.

It would seem that regardless = of=20 where the fuel is injected, if it all vaporizes = (which you=20 want it to in order for it to burn in the combustion chamber), = then the=20 air temp is lowered by 36F (in this example - assuming my calcuations = are=20 correct).  So, I am not certain how much we could materially = affect the=20 cooling and still maintain the desired Air/Fuel ratio.  It seems = that if=20 the net total of fuel is that needed for the desired air/fuel ratio, = then it=20 must all vaporize (in order to burn) and if it does that then = regardless of=20 where it does it, it would seem to lower the accompanying air by = 36F. =20

Now, = since the=20 velocity of the pulse is dependent on the average air temp in the = manifold,=20 there might be something said for using fuel to get the desired temp = drop=20 early on (immediately after the air enters the throttle body) and then = delay=20 injection of the fuel need to achieve the desired air/fuel = ratio until=20 the last instance (say the primary block injection ports), then = the=20 average velocity of the pulse would likely to be more impacted by = the=20 long average lower air temperature for most of the runner length = before that=20 last bit of fuel is injected.  Clearly lots of things to=20 consider.

So some = other folks=20 need to run some numbers and see what they get.

At 5500 = rpm=20 assuming WOT the airflow for the two rotor would be around 255CFM = which works=20 out to around 19.51 Lbs of air/minute.  For a air/fuel ratio of = 13, that=20 give 1.50 lbs of fuel/minute.  I used the

Te =3D Ti = +Q/W*cp  where Q is the latent heat of vaporation BTU removed = (per pound=20 of fuel)from the air flow (W air in lbs/second), Cp =3D 0.25 for = air. =20 I used 90F for Ti and calculated 54F for Te.  So some of you who=20 understand this much better than I, jump in with your=20 calculations

I think I = will=20 leave this approach to others, while really elegant (if the theory = holds to=20 gether), I am going to try the more direct but simple, brute force = approach=20 and vary the lenght of the runners. 

 

Ed=20 Anderson

 

 

 

 =20

------=_NextPart_000_000B_01C3A459.B31EF160--