X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Sender: To: lml@lancaironline.net Date: Wed, 10 Jun 2009 13:35:07 -0400 Message-ID: X-Original-Return-Path: Received: from web81207.mail.mud.yahoo.com ([68.142.199.111] verified) by logan.com (CommuniGate Pro SMTP 5.2.14) with SMTP id 3677044 for lml@lancaironline.net; Tue, 09 Jun 2009 15:39:13 -0400 Received-SPF: none receiver=logan.com; client-ip=68.142.199.111; envelope-from=cwfmd@yahoo.com Received: (qmail 5240 invoked by uid 60001); 9 Jun 2009 19:38:27 -0000 DomainKey-Signature:a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Reply-To:Subject:To:MIME-Version:Content-Type; b=ByvBZ3O3k94u+An77QEpdy0zK4ZAkt+8JT8Fv5unKrGpMepkjisJwcpOySkcFqJ9OojxAcM1WRymNxPFXr3AWbpgb4zxofODR+dAnHKTTFzef9zGJ6OzWa+sSXrvSFW5JI03VN6J63eCQez698dP8gZSaNYBZ2KOwNP1U2oc90s=; X-Original-Message-ID: <490777.4030.qm@web81207.mail.mud.yahoo.com> X-YMail-OSG: dBbJRJUVM1kSZnYSuRWdLnB9na7rEH223flgozEjNR49jC2yIzqr9O8.tuCtQVr76Jht3aFfR1lkosf2384yzVUVxORB.eSWrsP8.2NkHpoW0g6hf5J4YG2hB2jY.leYF2mQ3Ws7h6Bcw7XBXw.W4fVzs.e.mvh1bgEk9JWovmBCkf3a9hkty8cdj1ag77f801wV2OANr8YHELm6hKNpBt6r_xoP7dsJfF1CUg4KgwQw1CDJsiFkAu.EDlBHWzZZWpdSLRnOOrxNsJuB6N_ikcYss0BIRE0iW.sjVfI42v23viego68ZbIE- Received: from [98.201.146.11] by web81207.mail.mud.yahoo.com via HTTP; Tue, 09 Jun 2009 12:38:27 PDT X-Mailer: YahooMailClassic/5.3.9 YahooMailWebService/0.7.289.10 X-Original-Date: Tue, 9 Jun 2009 12:38:27 -0700 (PDT) From: cwfmd@yahoo.com Reply-To: cwfmd@yahoo.com Subject: Aux Tank plumbing explains R vapor return to L fuel line X-Original-To: lml@lancaironline.net MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="0-657592525-1244576307=:4030" --0-657592525-1244576307=:4030 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Executive=A0 Summary: We DID find a small potential leak from the Right amb= ient vapor return pressure (in the upper R tank) through a small transfer p= ump and partially open shutoff valve to the left 1/2" fuel line halfway bet= ween the Left tank and the selector valve. =A0We believe that hot soaking the fuel on a Gulf Coast afternoon combined = with a rapid climb to altitude and this small source of vapor at the lowest= pressure part of the line introduced bubbles into the flow and cavitated t= he boost pump (set on low boost). We doubt that a switch to high boost woul= d have corrected the problem. It did clear after 5 minutes, after the tank = was switched and the small air leak removed. There is no connection between= the pressurized cabin and this source. It comes from the right vapor retur= n line to the left fuel=0A line. =A0We were able to duplicate on the ground only after 5-6 high power hot af= ternoon runs simulating takeoff at full power. Though ambient pressure was = higher at sea level than at FL220, the rumbling down the runway probably sh= ook the "coke can" a bit and allowed us to duplicate the Fuel Flow interrup= tion. Thanks for all the help from the many responders. I plan to review the Reid= Vapor Pressure (RVP=3D44 KPA for 100LL) implications here and get some dat= a on what's really happening in that very low pressure part of the system. Detail: =A0On Friday, while waiting for parts and the inspected engine driven fuel = pump, =A0we examined the filters and fuel flow again. Since the flow was not routed through the FF sensor, we could only estimate, but flow seemed higher out of the left than the right tank, with = intermittent flow and possible air bubbles. Looking further, I recalled a modification. There was a (unused) fuel transfer system for an "aux=0A tank" that had been "capped off". We wondered if the combination of a slightly open valve and a leaky cap might explain the problems. We found the valve slightly open and there = was no cap at all. We discovered the flow path was not capped, as we had all been informed, but was set up as a transfer from the left to the right tank. Fuel from the left=A0 tank 1/2" feed line to=0A the selector valve also supplies fuel through a T into the transfer pump,=20 then through a shutoff valve to the right tank's vapor return line!=20 Question: Anybody see this config before? Why would the earlier builder set= it up this way? It seems backward for the fuel transfer scenario from an aux tank= in the back seat, because of the T into the left fuel line. In any case, this open path from the right vapor return to the left gravity feed line to the boost pump solves the mystery of intermittent fuel flow. W= hy no other occurence in almost 200 hrs?=A0=A0 Usually it was cooler, and t= he stepped climbs took much longer because of the baffle seal leaks. We had= to allowing engine cooling=A0 in stepped climbs, which likely cooled off t= he fuel too and helped avoid this event. (Remember this all started on a da= y when we were very happy to have achieved far better CHT cooling on climb!= ) We had experienced slight roughness usually at night or in weather or ove= r water, but thought it was corrected by the low boost on above 10k procedu= re --0-657592525-1244576307=:4030 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable
Executive  Summary: We DID find a small= potential leak from the Right ambient vapor return pressure (in the upper = R tank) through a small transfer pump and partially open shutoff valve to t= he left 1/2" fuel line halfway between the Left tank and the selector valv= e.

 We believe that hot soaking the fuel on a Gulf Coast aftern= oon combined with a rapid climb to altitude and this small source of vapor = at the lowest pressure part of the line introduced bubbles into the flow an= d cavitated the boost pump (set on low boost). We doubt that a switch to hi= gh boost would have corrected the problem. It did clear after 5 minutes, af= ter the tank was switched and the small air leak removed. There is no conne= ction between the pressurized cabin and this source. It comes from the righ= t vapor return line to the left fuel=0A line.
 We were able to duplicate o= n the ground only after 5-6 high power hot afternoon runs simulating takeof= f at full power. Though ambient pressure was higher at sea level than at FL= 220, the rumbling down the runway probably shook the "coke can" a bit and a= llowed us to duplicate the Fuel Flow interruption.
Thanks for all the he= lp from the many responders. I plan to review the Reid Vapor Pressure (RVP= =3D44 KPA for 100LL) implications here and get some data on what's really h= appening in that very low pressure part of the system.

Detail:
&n= bsp;On Friday, while waiting for parts and the inspected engine driven fuel= pump,
 we examined the filters and fuel flow
again. Since the f= low was not routed through the FF sensor, we could only
estimate, but fl= ow seemed higher out of the left than the right tank, with intermittent flo= w and possible air bubbles. Looking
further, I recalled a modification. = There was a (unused)
fuel transfer system for an "aux=0A tank" that had been "capped
off". We won= dered if the combination of a slightly open valve and a leaky
cap might = explain the problems. We found the valve slightly open and there was no cap= at all.

We discovered the flow path was not capped, as we
had al= l been informed, but was set up as a transfer from the left
to the right= tank. Fuel from the left  tank 1/2" feed line to=0A the
selector v= alve also supplies fuel through a T into the transfer pump,
then throug= h a shutoff valve to the right tank's vapor return line!

Question: = Anybody see this config before? Why would the earlier builder set it up
= this way? It seems backward for the fuel transfer scenario from an aux tank= in
the back seat, because of the T into the left fuel line.

In a= ny case, this open path from the right vapor return to the left gravity
= feed line to the boost pump solves the mystery of intermittent fuel flow. W= hy no other occurence in almost 200 hrs?   Usually it was cooler,= and the stepped climbs took much longer because of the baffle seal leaks. = We had to allowing engine cooling  in stepped climbs, which likely coo= led off the fuel too and helped avoid this event. (Remember this all starte= d on a day when we were very happy to have achieved far better CHT cooling = on climb!) We had experienced slight roughness usually at night or in weather or over water, but thought it was corrected= by the low boost on above 10k procedure




--0-657592525-1244576307=:4030--