X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Received: from mail-qy0-f187.google.com ([209.85.221.187] verified) by logan.com (CommuniGate Pro SMTP 5.3.4) with ESMTP id 4164634 for flyrotary@lancaironline.net; Sat, 13 Mar 2010 13:35:40 -0500 Received-SPF: pass receiver=logan.com; client-ip=209.85.221.187; envelope-from=rwstracy@gmail.com Received: by qyk17 with SMTP id 17so1833794qyk.11 for ; Sat, 13 Mar 2010 10:35:05 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:sender:received:in-reply-to :references:date:x-google-sender-auth:message-id:subject:from:to :content-type; bh=GsAgYdD7ukfGCa/TM+L7RR1ATBiKtdAI+/YAip6la3g=; b=r/R+IS95Bi3dnjX6PsIHFL2+7OSfFskKQJVvXItca2v7aMvtS/SkcZq9kZBMTMsrbf xfuZ0bghwzHRZAuEra0rltj6TwFxZoXkuopAMm5vFvQz93tDIO7yyIDzWba7UJgh6jG9 aw5eYLhjiCcQsTEZfjpEGi/zoUdLDx9Uwjv8Y= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type; b=beQanU8Ok0XKrXeNmiaub+Lo+4GzFa8gU7e+nZxPUP9+o2W7CaJCyFLI2OcYF84/fs wTpE0KBIVWhAz5ptL1IjL3XcZ5Qx8QTXwKyIBtA/DwRMW0iZqqaR2Ce6tqQP6pBjC88a YWAxIK6r8vUiW+gNdbzj7xn+A8aE/s/aGvj4Q= MIME-Version: 1.0 Sender: rwstracy@gmail.com Received: by 10.224.105.30 with SMTP id r30mr950662qao.162.1268505305517; Sat, 13 Mar 2010 10:35:05 -0800 (PST) In-Reply-To: References: Date: Sat, 13 Mar 2010 13:35:05 -0500 X-Google-Sender-Auth: b17c5e226fedeead Message-ID: <1b4b137c1003131035n7cca639dk4da5738bfc03631f@mail.gmail.com> Subject: Re: [FlyRotary] Re: fuel supply problem? From: Tracy Crook To: Rotary motors in aircraft Content-Type: multipart/alternative; boundary=00c09f905bd8218fff0481b2e8c5 --00c09f905bd8218fff0481b2e8c5 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable I had a similar thing happen but the engine would recover when I reduced throttle. Problem was clogged filter between tank and pump. At low flow rate all was OK but pressure drop across filter at high flow rate caused pump to cavitate and fuel pressure to drop. Did you try high throttle on ground test later? But you didn't mention a filter between tank & pump. That has potential problems as well since it doesn't take much trash to foul up an EFI pump & might have cleared up after it passed. Do you have only the slots in Van's suggested tank pickup between tank & pump? Problem might have been trash going through pump. I guess Ed already mentioned the loose tank pickup problem he had. Tracy On Sat, Mar 13, 2010 at 11:29 AM, Mike Wills wrote: > I'm pretty sure the turn was coordinated. Knowing that both tanks were > approaching (if not below) 5 gallons, it was the first thing I thought of > and I looked at the ball. It was centered. And I'm sure there is going to > turn out to be about 5 gallons left in the tank so its hard to understand > how this could have been an issue with the pickup uncovering. I just don= =92t > have a better explanation. If the pump failed I would expect it to stay > failed, not work fine on the ground. > > I didn=92t copy it, but my system I believe is the same that Ian Beadle u= sed. > I have 1 pump for each tank. The outputs are T'eed together - check valve= s > built into the pumps prevents crossfeed. Both tanks have return lines. Th= e > return line from the fuel rail goes through an industrial grade electric > valve. > > I chose this design for a couple of reasons: > 1) Operational simplicity. I reasoned that most fuel related problems in > flight are due to stupid pilot tricks so wanted a system that was as simp= le > to operate as possible. In its original configuration all that was requir= ed > to switch tanks was to flip a single toggle switch on the panel which wou= ld > energize the appropriate pump and configure the return valve to return fu= el > to the tank it came from. This has since been modified slightly to have > individual switches for each pump so both can be on at the same time if > needed. > 2) I had previous experience (bad) with EFI in some project cars before I > built the airplane. I had a couple of fuel pump failures and in researchi= ng > found strong recommendations against putting the fuel pump too far from t= he > fuel tank. These pumps are designed to push fuel, not draw fuel. A design > where both pumps have access to both tanks requires the pumps to be > downstream of the fuel selector with several feet of fuel line ahead of t= he > pump. I wanted to avoid this type of design although it apparently is > working fine for Ed. My pumps are just inboard of the wing roots literall= y a > couple of inches from the pickups in the tanks. > > Tracy's system is mechanically simple but has the potential for pilot err= or > resulting in pumping fuel overboard if the transfer pump is forgotten and > left on. Yes, I know there are ways to address that. And clearly it works > for him. > > I think (thought?) I pretty well understand the pros and cons in my setup= . > All things being equal I think before yesterday's incident I would do it > this way again. If it turns out that the issue was due to uncovering the > pickup (not real sure how to prove that) I'd consider changing to include= a > header tank with a deep sump to prevent re-occurance. But I don=92t think= I'd > rely on suction to fill the header, think I'd use a low pressure pump lik= e > the typical Facet pump. > > One thing is for certain. Unlike a carbureted engine which has a little > cushion due to fuel in a float bowl, the instant fuel pressure drops this > engine quits. It's an attention getter. Particularly when you are at the > opposite corner of your test box from your home base. I ended up flying b= ack > about 70 miles hopscotching from field to field. > > Mike > > *From:* Ed Anderson > *Sent:* Saturday, March 13, 2010 1:53 AM > *To:* Rotary motors in aircraft > *Subject:* [FlyRotary] Re: fuel supply problem? > > There is certainly more than one way to design a reliable fuel system = =96 > each with its pros and cons. I too did not want a six way value for > switching fuel between two tanks and the engine with the ugh return line. > Besides I had already constructed my fuel system pretty much according to > Van=92s instructions. So to preclude a return line I came up with my =93= almost > returnless=94 system. It uses a =BD pint capacity small header tank to r= eturn > the injector fuel =96 the fuel injected to the engine comes from this hea= der > tank thereby creating a =93vacuum=94 in the tank which pulls fresh fuel f= rom the > wing tanks. It has worked fine for over 10 years. > > > > Both fuel pumps draw from this header tank and either tank can feed it an= d > I have no return lines going back to the tanks. > > > > But, Tracy=92s approach has shown to work just fine =96 not knocking it b= y any > means. > > > > Ed > > > > Ed Anderson > > Rv-6A N494BW Rotary Powered > > Matthews, NC > > eanderson@carolina.rr.com > > http://www.andersonee.com > > http://www.dmack.net/mazda/index.html > > http://www.flyrotary.com/ > > http://members.cox.net/rogersda/rotary/configs.htm#N494BW > > http://www.rotaryaviation.com/Rotorhead%20Truth.htm > ------------------------------ > > *From:* Rotary motors in aircraft [mailto:flyrotary@lancaironline.net] *O= n > Behalf Of *Bktrub@aol.com > *Sent:* Friday, March 12, 2010 11:43 PM > *To:* Rotary motors in aircraft > *Subject:* [FlyRotary] Re: fuel supply problem? > > > > I just copied Tracy's setup- all fuel is pumped from the right tank, and > fuel is transferred from the left into the right by a Facet pump. I didn'= t > want to get into having a six port fuel valve in order to get the fuel > injection to return to the tank I was using at the time. > > It's really simple, and hopefully that means reliable. We'll have to see, > as this plane is looking for it's airworthiness inspection in the next fe= w > months. > > > > Brian Trubee > > > > __________ Information from ESET NOD32 Antivirus, version of virus > signature database 3267 (20080714) __________ > > The message was checked by ESET NOD32 Antivirus. > > http://www.eset.com > --00c09f905bd8218fff0481b2e8c5 Content-Type: text/html; charset=windows-1252 Content-Transfer-Encoding: quoted-printable I had a similar thing happen but the engine would recover when I reduced th= rottle.=A0 Problem was clogged filter between tank and pump.=A0 At low flow= rate all was OK but pressure drop across filter at high flow rate caused p= ump to cavitate and fuel pressure to drop.=A0 Did you try high throttle on = ground test later?

But you didn't mention a filter between tank & pump.=A0 That ha= s potential problems as well since it doesn't take much trash to foul u= p an EFI pump & might have cleared up after it passed.=A0 Do you have o= nly the slots in Van's suggested tank pickup between tank & pump?= =A0 Problem might have been trash going through pump.

I guess Ed already mentioned the loose tank pickup problem he had.
T= racy

<= div class=3D"gmail_quote"> On Sat, Mar 13, 2010 at 11:29 AM, Mike Wills <rv-4mike@cox.net> wrote:
I'm pretty sure the turn was coordinated.= =A0Knowing=20 that both tanks were approaching (if not below) 5 gallons, it was the first= =20 thing I thought of and I looked at the ball. It was centered. And I'm s= ure there=20 is going to turn out to be about 5 gallons left in the tank so its hard to= =20 understand how this could have been an issue with the pickup uncovering. I = just=20 don=92t have a better explanation. If the pump failed I would expect it to = stay=20 failed, not work fine on the ground.
=A0
I didn=92t copy it, but my system I believe is = the same=20 that Ian Beadle used. I have 1 pump for each tank. The outputs are T'ee= d=20 together - check valves built into the pumps prevents crossfeed. Both tanks= have=20 return lines. The return line from the fuel rail goes through an industrial= =20 grade electric valve.
=A0
I chose this design for a couple of=20 reasons:
1)=A0Operational simplicity. I reasoned that mo= st=20 fuel related problems in flight are due to stupid pilot tricks so wanted a= =20 system that was as simple to operate as possible. In its original configura= tion=20 all that was required to switch tanks was to flip a single toggle switch on= the=20 panel which would energize the appropriate pump and configure the return va= lve=20 to return fuel to the tank it came from. This has since been modified sligh= tly=20 to have individual switches for each pump so both can be on at the same tim= e if=20 needed.
2) I had previous experience (bad) with EFI in = some=20 project cars before I built the airplane. I had a couple of fuel pump failu= res=20 and in researching found strong recommendations against putting the fuel pu= mp=20 too far from the fuel tank. These pumps are designed to push fuel, not draw= =20 fuel. A design where both pumps have access to both tanks requires the pump= s to=20 be downstream of the fuel selector with several feet of fuel line ahead of = the=20 pump. I wanted to avoid this type of design although it apparently is worki= ng=20 fine for Ed. My pumps are just inboard of the wing roots literally a couple= of=20 inches from the pickups in the tanks.
=A0
Tracy's system is mechanically simple but h= as the=20 potential for pilot error resulting in pumping fuel overboard if the transf= er=20 pump is forgotten and left on. Yes, I know there are ways to address that. = And=20 clearly it works for him.
=A0
I think (thought?) I pretty well understand the= pros=20 and=A0cons in my setup. All things being equal I think before yesterday'= ;s=20 incident I would do it this way again.=A0If it turns out that the issue was= =20 due to uncovering the pickup (not real sure how to prove that) I'd cons= ider=20 changing to include a header tank with a deep sump to prevent re-occurance.= But=20 I don=92t think I'd rely on suction to fill the header, think I'd u= se a low=20 pressure pump like the typical Facet pump.
=A0
One thing is for certain. Unlike a carbureted e= ngine=20 which has a little cushion due to fuel in a float bowl, the instant fuel=20 pressure drops this engine quits. It's an attention getter. Particularl= y when=20 you are at the opposite corner of your test box from your home base. I ende= d up=20 flying back about 70 miles hopscotching from field to field.
=A0
Mike=A0

Sent: Saturday, March 13, 2010 1:53 AM
To: Rotary motors in aircraft <= /div>
Subject: [FlyRotary] Re: fuel supply problem?

There is certa= inly more=20 than one way to design a reliable fuel system =96 each with its pros and=20 cons.=A0 I too did not want a six way value for switching fuel between two= =20 tanks and the engine with the ugh return line.=A0 Besides I had already=20 constructed my fuel system pretty much according to Van=92s instructions.= =A0 So=20 to preclude a return line I came up with my =93almost returnless=94 system.= =A0 It=20 uses a =BD pint capacity small header tank to return the injector fuel =96 = the fuel=20 injected to the engine comes from this header tank thereby creating a =93va= cuum=94=20 in the tank which pulls fresh fuel from the wing tanks.=A0 It has worked fi= ne=20 for over 10 years.

=A0

Both fuel pump= s draw=20 from this header tank and either tank can feed it and I have no return line= s=20 going back to the tanks.

=A0

But, Tracy=92s= approach has shown=20 to work just fine =96 not knocking it by any means.

=A0

Ed

=A0

From:=20 Rotary motors in aircraft=20 [mailto:fl= yrotary@lancaironline.net] On=20 Behalf Of Bk= trub@aol.com
Sent: = Friday, March 12, 2010 11:43=20 PM
To: Rotary motors in= aircraft
Subject: [Fly= Rotary] Re: fuel supply=20 problem?

=A0

I just copie= d=20 Tracy's setup-=20 all fuel is pumped from the right tank, and fuel is transferred from the le= ft=20 into the right by a Facet pump. I didn't want to get into having a six = port fuel=20 valve in order to get the fuel injection to return to the tank I was using = at=20 the time.=A0=A0

It's rea= lly simple,=20 and hopefully that means reliable. We'll have to see, as this plane is = looking=20 for it's airworthiness inspection in the next few months.=20

=A0

Brian=20 Trubee



____= ______=20 Information from ESET NOD32 Antivirus, version of virus signature database = 3267=20 (20080714) __________

The message was checked by ESET NOD32=20 Antivirus.

http://= www.eset.com


--00c09f905bd8218fff0481b2e8c5--