Date: Sat, 28 May 2005 10:09:11 -0400
From: Finn Lassen <finn.lassen@verizon.net>
Subject: Bubbles in fuel lines to pumps. (Was: Why insist on a sump?)
In-reply-to: <list-966119@logan.com>
To: Rotary motors in aircraft <flyrotary@lancaironline.net>
Message-id: <42987B87.3040507@verizon.net>
MIME-version: 1.0
Content-type: text/plain; charset=us-ascii; format=flowed
Content-transfer-encoding: 7bit
References: <list-966119@logan.com>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4)
 Gecko/20030624 Netscape/7.1 (ax; PROMO)

atlasyts@bellsouth.net wrote:

>All pumps, filter, fuel lines etc. are below the lowest point of my tanks and in such a way that any bubbles before the pumps can go only up in to the tanks.
>
Ha! You'd think so, but...

In actual fact, while you have any kind of flow of fuel in one 
direction, bubbles and pockets of air are very reluctant to move in the 
opposite direction (back up to the tank).

After a lot of experimentation I have given up getting rid of bubbles 
and vapor pockets in the line(s) to my pumps.
I expanded my lines to my two facet pumps to 3/8" ID, with mininum ID in 
fittings 5/16"+.
With a continuous downslope from tank to pumps (about 2 feet of gravity 
head pressure), I still see a vapor pocket forming in the line when I 
turn off one of the pumps. And it stays where it is until the flow drops 
to maybe 3 to 4 GPH! The fuel runs around it, following the inside walls 
of the line.

I finally redesigned the fuel system: straight lines from tank to pumps. 
(Need to put shutoff valve back in.)
Pumps truly in parallel. Pumps, then filter, then fuel flow sensor. Then 
line splits: one line to two outer carbs, and the other line, via 
shut-off valve to the center carb (used for leaning mixture at altitude).

Now I can have both pumps on as long as I want - truly redundant.

The disadvantage (which made me reluctant to make the change) is that 
the fuel flow sensor is sensitive to the pulses from the pumps. Reads 
about 0.2 GPH with no actual flow. That is why my original configuration 
was: valve, filter, sensor, pumps (one for each set of carbs).

As to the origin of the bubbles/vapor pocket, still uncertain. One 
suggestion is that the pulsing/shaking action of the facet pumps are 
generating then, like when shaking a small container of fluid?

Finn