X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Received: from mail-fx0-f52.google.com ([209.85.161.52] verified) by logan.com (CommuniGate Pro SMTP 5.4.1) with ESMTPS id 5097807 for flyrotary@lancaironline.net; Fri, 19 Aug 2011 15:47:02 -0400 Received-SPF: pass receiver=logan.com; client-ip=209.85.161.52; envelope-from=msteitle@gmail.com Received: by fxd18 with SMTP id 18so2111614fxd.25 for ; Fri, 19 Aug 2011 12:46:26 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; bh=45iqXMgKn4lxl2sT4CKgp6YUTIwy7xzYS56D0JOCoqM=; b=Yeh6SpNMSyicmxKK9JaB67UV/impeQnviAvTknwLg/b3YBMts9K0SNVCn68lbt54QR 2rXi9mdrSlfCbXfwW4zEMUHzhQv4ZLj3NU8KZG/CBbVm6Nz6hSxqH92t6R8rC6XOEnEA gMF4QflV0UGqLdx5mHhkFj7zEECf05zSkQ1aU= MIME-Version: 1.0 Received: by 10.223.9.6 with SMTP id j6mr150417faj.75.1313783186546; Fri, 19 Aug 2011 12:46:26 -0700 (PDT) Received: by 10.223.74.129 with HTTP; Fri, 19 Aug 2011 12:46:26 -0700 (PDT) In-Reply-To: References: Date: Fri, 19 Aug 2011 14:46:26 -0500 Message-ID: Subject: Re: [FlyRotary] Re: Dennis Haverlah Fuel System...or any others, for that matter. From: Mark Steitle To: Rotary motors in aircraft Content-Type: multipart/alternative; boundary=0015174485882561c204aae0fced --0015174485882561c204aae0fced Content-Type: text/plain; charset=ISO-8859-1 Al, Yes, the inlet filters are coarse screen, one per tank. Mark On Fri, Aug 19, 2011 at 2:13 PM, Al Wick wrote: > ** > Mark said: > Way to be. Glad you went to the effort to test it. Others might have been > tempted to respond just by cracking the fitting on fuel rail. Fuel starts to > flow: "Took care that problem". You pursued further so that running one tank > dry is not catastrophic. > > Top view of your diagram looks good. It's the fine details that make a > difference. Some plumbing oversight prevented your pumps from getting wet. > I do hope those two inlet filters are coarse ones with low pressure drop > and high surface area. You can't beat the oem inlet filter bags. > > -al wick > > ----- Original Message ----- > *From:* Mark Steitle > *To:* Rotary motors in aircraft > *Sent:* Wednesday, August 17, 2011 11:46 AM > *Subject:* [FlyRotary] Re: Dennis Haverlah Fuel System...or any others, > for that matter. > > Al, > > You wrote,* "You also test it without the bleed line?"** * > > Yes, the test led to the search for a solution. You can see the "bypass > circuit" on Egg's fuel system schematic at > http://www.eggenfellneraircraft.com/Manual-Updates/EarlyManualFuel.html. > This is what I incorporated into my fuel system, and it solved the problem. > Its been quite a few years now, but I may still have some of the posts on > the Eggenfellner list that occurred in search for this as the final > solution. > > You wrote,* "FWIW, everything you describe is not vapor lock, has nothing > to do with vapor lock." * > > As I said, I'm not sure if my condition fits the definition of "vapor > lock", but that's what someone labeled it, and it stuck. The problem is > that EFI pumps don't draw or push air very well. So, if the flowing fuel is > replaced by air due to an empty tank, the pump keeps running, but > essentially stops pumping fuel. With a 40# spring in the downstream > circuit, it is virtually impossible for the pump to pass the air even > though the selector valve has been swiched to the full tank. I would agree > that the fuel injectors should be able to pass the air, provided the prop is > still turning. (This was not the case with my ground test.) The bypass > circuit ensures that flow will be restored as soon as the bubble passes > through the pump and the fuel pump regains its prime. > > Mark S. > > On Wed, Aug 17, 2011 at 12:06 PM, Al Wick wrote: > >> ** >> > as low as possible >> >> Instead of speculations, you can actually measure how good it is. A pump, >> clear fuel lines, and water instead of fuel. Pinch fuel line with pliers to >> increase pressure drop. Watch bubbles come out of solution. Hear the rattle >> from pump due to it's inability to flow air. >> If you also put a pressure gage on pump inlet, you can see how fine fuel >> filter increases risk. You can see how small diameter tubing increases risk, >> heat, head pressure, etc etc. >> >> > Egg had one forced landing. It was caused by vapor lock. Absolutely >> NOTHING to do with the lack of bleed line. A day later, Jan made it to the >> crash site. Pump would not flow fuel until he cracked the line loose. This >> is simply because the inlet to pumps was not wet. He did not understand >> this. He just reacted to symptoms. Nothing to do with vapor lock. Remember, >> this was crash site. Who knows what attitude plane was at. As I described, >> if you screwed up your plumbing from tank to pump, then pump is not self >> priming. >> >> > You also test it without the bleed line? >> >> FWIW, everything you describe is not vapor lock, has nothing to do with >> vapor lock. >> >> -al wick >> >> >> ----- Original Message ----- >> *From:* Mark Steitle >> *To:* Rotary motors in aircraft >> *Sent:* Wednesday, August 17, 2011 9:19 AM >> *Subject:* [FlyRotary] Re: Dennis Haverlah Fuel System...or any others, >> for that matter. >> >> Al, >> >> I would think that a well-designed fuel system with the pumps located as >> low as possible, in a cool area, and a return system should not be troubled >> with vapor locking. Although I guess it could happen in Death Valley >> in August if the a/c is left out in the sun for 8 hours, and running auto >> gas. So far, I haven't had any problem here in Texas (temps 105* today). >> >> >> My purpose for installing the bypass circuit was more critical to safe >> flight. (The solution was the result of a forced landing, or two, in an >> Eggenfellner Subaru installation.) During ground runs, using a 5-gallon >> can, I found that the efi fuel pumps could not restore pressure once the >> tank had been allowed to run dry. When this occurred, the EFI pump would >> suck a big slug of air into the inlet side of the pump, loose pressure, and >> being unable to build psi equal to the pressure regulator setting, it would >> stop pumping fuel (vapor lock). The only way to restore operation was to >> crack open the system downstream of the pump until the slug of air could be >> passed on through the fuel pump. Not sure this meets your definiton of >> "vapor lock", but I think it does, but for a different reason than fuel >> vapor pressure. >> >> I have tested this system on the ground and it works as advertized. If >> the tank runs dry, switched to a full tank, the efi pump will restore >> pressure to the system within a few seconds. I urge all builders to run >> this test on their fuel system to determine whether or not their pumps are >> able to restore pressure after running a tank dry. Or, just don't ever run >> a tank dry. >> >> Mark S. >> >> >> >> On Wed, Aug 17, 2011 at 10:22 AM, Ernest Christley wrote: >> >>> Al Wick wrote: >>> >>>> I'm really concerned for some of these fuel designs. The fuel bleed has >>>> nothing to do with vapor lock. Virtually no effect at all. >>>> >>>> >>> >>> I don't know why others are doing it, but for me, the bleed has nothing >>> at all to do with vapor lock. Some conversations have been mixed together, >>> so I can see how that could be the impression. The point of the pressure >>> bleed is to bleed off the pressure after shutdown. >>> >>> I have a strong, positive head pressure going into my pumps. They, and >>> the regulator, are about 8" directly below the tank. Excess fuel goes back >>> to the opposite side of the tank from the pickup, and a single line goes >>> forward to feed the injectors. The fuel lines are arranged such that heat >>> soaking the lines to the point of boiling the gas will push liquid fuel down >>> hill and behind the firewall, isolating the gaseous gas with its heat at the >>> top of the line. Turning the pumps on will pressurize the line to 55psi, >>> returning most of the gaseous fuel back to a liquid state. The ECM is >>> programmed for a longer clearing pulse on hot start. >>> >>> The point of the bleed is to allow fuel to move back to the tank. I had >>> the issue of a the pressurized lines being perfectly sealed. The >>> pressurized fuel was finding the path of least resistance out, which just >>> happened to be out the injector and into the intake manifold where it sat as >>> a little puddle. Heat soaking the lines would not push liquid fuel downhill >>> and back behind the firewall. It would push more fuel into the manifold. A >>> puddle of gas sitting in a composite manifold, just above a hot exhaust >>> stack is just bad mojo. A poorly sealed regulator allows the pressure to >>> bleed off in about 5 seconds (give or take), isolating the hot fuel in front >>> of the firewall, and keeping the rest cool and out of the intake manifold. >>> >>> Got nuthin' to do with vapor lock. >>> >>> -- >>> Homepage: http://www.flyrotary.com/ >>> Archive and UnSub: http://mail.lancaironline.net:** >>> 81/lists/flyrotary/List.html >>> >> >> > ------------------------------ > > -- > Homepage: http://www.flyrotary.com/ > Archive and UnSub: > http://mail.lancaironline.net:81/lists/flyrotary/List.html > > --0015174485882561c204aae0fced Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Al,=A0

Yes, the inlet filters are coarse screen, one per= tank. =A0

Mark

= On Fri, Aug 19, 2011 at 2:13 PM, Al Wick <alwick@juno.com> wrote:
Mark said:
<Yes, the test led to the search for a solution
Way to be. Glad you went to the effort to test it. Others might have b= een=20 tempted to respond just by cracking the fitting on fuel rail. Fuel starts t= o=20 flow: "Took care that problem". You pursued further so that runni= ng one tank dry=20 is not catastrophic.
=A0
Top view of your diagram looks good. I= t's the fine=20 details that make a difference. Some plumbing oversight prevented your pump= s=20 from getting wet.
I do hope those two inlet filters are = coarse ones=20 with low pressure drop and high surface area. You can't beat the oem in= let=20 filter bags.
=A0
-al wick
----- Original Message -----
From:=20 Mark Steitle=20
Sent: Wednesday, August 17, 2011 11= :46=20 AM
Subject: [FlyRotary] Re: Dennis Hav= erlah=20 Fuel System...or any others, for that matter.

Al,
=A0
You wrote, "You also test it without the bleed=20 line?"=A0
=A0
Yes, the test led to the search for a solution.=A0 You can see the= =20 "bypass circuit" on Egg's fuel system schematic at http://www.eggenfellneraircraft.com/Manual-Updates/Earl= yManualFuel.html.=A0=20 This is=A0what I=A0incorporated into my=A0fuel system, and it solved=20 the problem.=A0 Its been quite a few years now, but I=A0may still=20 have=A0some of the posts on the Eggenfellner=A0list=A0that occurred=20 in search for this as the final solution.
=A0
You wrote,=20 "FWIW, everything you describe is not vapor lock, has nothin= g to do=20 with vapor lock."=A0
=A0
As I said, I'm not sure if my condition fits the definition of &= quot;vapor=20 lock", but that's what someone labeled it, and it stuck.=A0 The = problem is=20 that EFI pumps don't=A0draw or push air very well.=A0 So, if the flow= ing=20 fuel is replaced by air due to an empty tank, the pump keeps running, but= =20 essentially=A0stops pumping fuel.=A0 With a 40# spring in the downstream= =20 circuit, it is virtually impossible for the pump to pass the air=A0even= =20 though=A0the selector valve has been swiched to the full tank.=A0 I=20 would agree that the fuel injectors should be able to pass the air, provi= ded=20 the prop is still turning.=A0 (This was not the case with my ground test.= )=20 The bypass circuit ensures that flow will=A0be restored as soon as the=20 bubble=A0passes through the pump and=A0the fuel pump regains its=20 prime.=A0
=A0
Mark S.

On Wed, Aug 17, 2011 at 12:06 PM, Al Wick <alwi= ck@juno.com>=20 wrote:
<I would think that a well-designed fuel system with the pumps= =20 located as low as possible
=A0
Instead of speculations, you can a= ctually=20 measure how good it is. A pump, clear fuel lines, and water instead of = fuel.=20 Pinch fuel line with pliers to increase pressure drop.=A0 Watch bubbles= =20 come out of solution. Hear the rattle from pump due to it's inabili= ty to=20 flow air.
If you also put a pressure gage on= pump inlet,=20 you can see how fine fuel filter increases risk. You can see how small= =20 diameter tubing increases risk, heat, head pressure, etc etc.
=A0
<the result of a forced landing
Egg had one forced landing. It was caused by vapor lock. Absolutel= y=20 NOTHING to do with the lack of bleed line. A day later, Jan made it to = the=20 crash site. Pump would not flow fuel until he cracked the line loose. T= his=20 is simply because the inlet to pumps was not wet. He did not understand= =20 this. He just reacted to symptoms. Nothing to do with vapor lock. Remem= ber,=20 this was crash site. Who knows what attitude plane was at. As I describ= ed,=20 if you screwed up your plumbing from tank to pump, then pump is not sel= f=20 priming.
=A0
<tested this system on the ground and it works as=20 advertized
You also test it without the bleed line?
=A0
FWIW, everything you describe is n= ot vapor=20 lock, has nothing to do with vapor lock.
=A0
-al wick
=A0
----- Original Message -----
Sent: Wednesday, August 17, 201= 1 9:19=20 AM
Subject: [FlyRotary] Re: Dennis= =20 Haverlah Fuel System...or any others, for that matter.

Al,
=A0
I would think that a well-designed=A0fuel system=A0with the=20 pumps located as low as possible, in a=A0cool area,=A0and=A0a=20 return system=A0should not be troubled with vapor=20 locking.=A0=A0Although I guess it could happen in Death Valley=20 in=A0August if the a/c is left out in the sun for 8 hours, and runnin= g=20 auto gas.=A0 So far, I haven't had any problem here in Texas (tem= ps=20 105* today).=A0 =A0
=A0
My purpose for installing the bypass circuit was more critical t= o=20 safe flight.=A0 (The solution was=A0the result of a forced landing,= =20 or two,=A0in an Eggenfellner Subaru installation.)=A0 During ground= =20 runs, using a 5-gallon can, I found that the efi fuel pumps could not= =20 restore pressure once the tank had been allowed to run=20 dry.=A0=A0When this occurred, the EFI pump=A0would suck a big=20 slug of air into the inlet side of the pump, loose pressure,=A0and=20 being unable to build psi equal to the pressure regulator setting, it= =20 would stop pumping fuel (vapor lock).=A0 The only way to restore=20 operation was to crack open the system downstream of the pump until t= he=20 slug of air could be passed on through the fuel pump.=A0=A0Not sure= =20 this meets your definiton of "vapor lock", but I think it d= oes, but for a=20 different reason than fuel vapor pressure.
=A0
I have tested this system on the ground and it works as=20 advertized.=A0 If the tank runs dry, switched to a full tank, the efi= =20 pump will restore pressure to the system within a few seconds.=A0 I= =20 urge all builders to run this test on their fuel system to determine= =20 whether or not their pumps are able to restore pressure after running= a=20 tank dry.=A0 Or, just don't ever run a tank dry.
=A0
Mark S.


=A0
On Wed, Aug 17, 2011 at 10:22 AM, Ernest C= hristley=20 <echristley@att.net> wrote:
Al Wick wrote:
I'm really concerned for = some of these fuel designs.=20 The fuel bleed has nothing to do with vapor lock. Virtually no ef= fect=20 at all.
=A0

I don't know why oth= ers are=20 doing it, but for me, the bleed has nothing at all to do with vapor= =20 lock. =A0Some conversations have been mixed together, so I can see= =20 how that could be the impression. =A0The point of the pressure blee= d=20 is to bleed off the pressure after shutdown.

I have a strong= ,=20 positive head pressure going into my pumps. =A0They, and the=20 regulator, are about 8" directly below the tank. =A0Excess fue= l goes=20 back to the opposite side of the tank from the pickup, and a single= line=20 goes forward to feed the injectors. =A0The fuel lines are arranged= =20 such that heat soaking the lines to the point of boiling the gas wi= ll=20 push liquid fuel down hill and behind the firewall, isolating the= =20 gaseous gas with its heat at the top of the line. =A0Turning the=20 pumps on will pressurize the line to 55psi, returning most of the= =20 gaseous fuel back to a liquid state. =A0The ECM is programmed for a= =20 longer clearing pulse on hot start.

The point of the bleed i= s to=20 allow fuel to move back to the tank. =A0I had the issue of a the=20 pressurized lines being perfectly sealed. =A0The pressurized fuel w= as=20 finding the path of least resistance out, which just happened to be= out=20 the injector and into the intake manifold where it sat as a little= =20 puddle. =A0Heat soaking the lines would not push liquid fuel downhi= ll=20 and back behind the firewall. =A0It would push more fuel into the= =20 manifold. =A0A puddle of gas sitting in a composite manifold, just= =20 above a hot exhaust stack is just bad mojo. =A0A poorly sealed=20 regulator allows the pressure to bleed off in about 5 seconds (give= or=20 take), isolating the hot fuel in front of the firewall, and keeping= the=20 rest cool and out of the intake manifold.

Got nuthin' to= do with=20 vapor lock.

--
Homepage: =A0http://www.flyrotary.com= /
Archive and UnSub: =A0=20 http://mail.lancaironline.net:81/lists/flyrota= ry/List.html


--
Homepage:=A0 http://www.flyrotary.com/
Archive and=20 UnSub:=A0=A0=20 http://mail.lancaironline.net:81/lists/flyrotary/List.html<= /a>


--0015174485882561c204aae0fced--