Repeatedly Running On A Low Tank?

I realize all I have is antidotal evidence, but in 95% of the pumps I
replace, I always ask the customer and it is almost always the ones I am
replacing admit to running the tank low most of the time. KB

 

When my Civic tank is low, it takes a lot longer to start. I've read
different theories on why. One is that a one-way valve on the fuel
pump seals better when it is lubricated by fuel, so gas stays in the
line. Whatever the reason, keeping the tank full puts less stress on
the battery, fuel pump, and starter.

 

I have noticed a slight delay in starting when the tank is
low (say 1/8 tank and less) on my 91 Civic as well.

 

Same here. My daily driver Chrysler Concorde just turned 200k miles
last week from its 80 mile/day commute. I run it as low as I can
before filling up each time at the filling station across the street
from my house - typically with the gage anywhere from 3/16 tank to below
'E'. Original fuel pump.

The Chrysler LH car fuel filter is built into the sender/pump/regulator
assembly - tank would have to be dropped to replace it. But they are
designed to last longer than the vehicle (barring getting some really
bad/dirty gas) - no replacement interval specified - and judging from
the total lack of fuel filter problems mentioned on the forums, they did
a good job on the design.

Bill Putney
(To reply by e-mail, replace the last letter of the alphabet in my
address with the letter 'x')

 

Far be it from me to criticize Fox News, but in this case I agree with
you. I find *all* news outlets to be totally FOS when they try to
discuss technical subjects.

Bill Putney
(To reply by e-mail, replace the last letter of the alphabet in my
address with the letter 'x')

 

The overwhelming majority of the cooling (and *all* of the lubrication)
of the pump and its internal components comes from the constant flow of
fuel thru the pump and around each component. Very little cooling comes
from the mostly stangant fuel surrounding the pump.

Bill Putney
(To reply by e-mail, replace the last letter of the alphabet in my
address with the letter 'x')

 

Almost all cooling of the pump comes from fuel flowing thru it - very
little from mostly stagnant fuel around the outside of the pump.

Bill Putney
(To reply by e-mail, replace the last letter of the alphabet in my
address with the letter 'x')

 

The designs are not that thermally marginal to worry about. I am being
facetious when I say you should only drive your car in the rain to
maximize cooling of your tires, but it illustrates my point that it
isn't worth worrying about.

I can't argue with that.

Not in today's sealed (and sometimes plastic) tanks.

Wear of the bearings (which for most pumps are actually plastic holes in
pump/motor end caps (basically, plastic bushings) and wear of the brushes.

Usually the plastic end caps (that act as the bearings for the armature
shaft) are glass filled. The molded bearing (bushing) surfaces have a
microscopically thin film of plastic separating the glass fibers from
the metal shaft - an inherent result of the molding process of
glass-filled plastics. Once that thin film wears thru, the very
abrasive glass wears the metal shaft faster than the plastic itself
wears believe it or not (I learned this when I worked as a
designer/engineer/manager for a supplier of fuel pump parts to GM/Delphi
and Ford/Visteon. Once there is significant play between the shaft and
the bearings/bushings, the armature literally rattles around and
eventually crashes into the magnets (also the pumping section is mounted
on the shaft, so its close-clearance parts start rubbing against/hitting
each other, depending somewhat on the type of pumping section - gerotor,
turbine, moving vane, etc.).

Sulfur in fuel of third world countries is a big corrosion problem for
fuel pumps.

Presence of silicone compounds greatly shortens brush life.

The interconnecting wires can break from vibration - like if they have a
nick, and the car goes over a railroad track often - vibration and the
notch effect don't go well together.

IIRC, brush compounds have to be specifically designed to handle large
concentrations of ethanol (I may not be 100% correct on this point -
been out of that field for 7 years)).

I do know that sometimes metal film plastic capacitors (for EMI
suppression) will be destroyed by alcohol in the fuel (alcohol molecules
are so small, you can't keep them from leaching onto the capacitor and
etching away the metal film. When that happens, the capacitor is no
longer a capacitor - EMI problems).

I'd say the shaft/bushing wear is the biggest common cause of failure
(intermittent) - it's why a pump in the process of failing can often can
be heard rattling. Next would be worn out brushes - also intermittent
(temperature- and shock/vibration-dependent), but noiseless, but brushes
are usually over-designed as far as life (length).

Perhaps I shed some light for you with some of my comments. Certainly I
do not have anywhere near exhaustive knowledge of the subject.

Bill Putney
(To reply by e-mail, replace the last letter of the alphabet in my
address with the letter 'x')

 

No. There are no dynamic seals in fuel pumps like in a typical
automotive water pump. Running dry (not a credible situation in
general) would not affect case seals (which are static crimped seals).

Not likely since the pumping section is at the bottom of the pump, so
when you "run out of gas", there is a column of fuel extending from the
pumping section of the pump (at its very bottom), thru the pump, all the
way to the fuel rail and injectors. Granted that column of fuel is not
moving, but it's there nonetheless. And the engine dies, and the
computer turns the pump off in a matter of seconds. No real chance for
significant damage from heat or lack of lubrication.

I'll buy that.

Bill Putney
(To reply by e-mail, replace the last letter of the alphabet in my
address with the letter 'x')

 

The pump will never not see fuel inside. The pumping section is at the
very bottom of the pump, so the column of fuel will be "dancing" on the
pumping element (think of a turbine blade). The only part (in an empty
tank) filled with air will be the pickup tube.

Bill Putney
(To reply by e-mail, replace the last letter of the alphabet in my
address with the letter 'x')

 

I can visualize that happening as it starts sucking bubbles of air
rising into the pump thru the pickup tube. Kind of like cavitation,
except air instead of nothing. Froth.

Bill Putney
(To reply by e-mail, replace the last letter of the alphabet in my
address with the letter 'x')

 

Not that much of a problem these days with sealed tanks.

FWIW, Chrysler LH cars have life-of-vehicle fuel filters, and apparently
they are effective (very few posts on LH forums for fuel filter issues).

I do not agree that the article is not baloney, but I agree that it's
not a big deal.

Bill Putney
(To reply by e-mail, replace the last letter of the alphabet in my
address with the letter 'x')

 

I trust you will recall that cavitation is not bubbles
filled with nothing. It is the occurrence of bubbles filled
with the vapor of the liquid being moved. It is a
consequence of the liquid reaching a pressure low enough to
cause boiling.

 

Elle wrote:


snip

Honda Civics went to electric pumps beginning with the 1980 model year
but it was external right outside the tank. I have never heard of a
fuel pump failure with these units.

Even after setting idle for nearly 12 years, the one in my '83 works
perfectly.

JT

 

You are correct.

Bill Putney
(To reply by e-mail, replace the last letter of the alphabet in my
address with the letter 'x')

 

But...But...Ray told me I was in danger of burning out my fuel pump in my
Supra when I got it because the tank had so many holes I had to keep it
below 1/4 tank.

I trust what Ray says...

 

'Cause the OP did to begin with.

 

Bill, the article is addressing all the saps too poor to
keep their gas tanks full. You really think they're all
driving cars with a model year later than about 1998, when
fuel tanks were going plastic?

(I am just guestimating the year plastic became really
popular, based on googling on the subject.)

 

How'd you know this little factoid so quickly?! I believe
you, just saying that's way inside information.

Hm, unless you worked on Hondas of that vintage for a few
years, I am not sure I can support a hypothesis that pumps
of this vintage do not fail.

Yeabut you don't run to empty/"on fumes" (as they say), do
you?

Those of you attesting that the article's thesis is bunk may
be right. Just saying fuel pumps do fail; more authoritative
sources than Fox make the same claim the Fox article does
(which is pretty darn general; look back); and those of you
saying, 'Ain't never seen a fuel pump failure in my
guzillion years of drivin' ' just do not run the car to
empty enough or do not leave your fuel filter unchanged long
enough to have a problem.

 

As time goes on, I believe lees and less of what I read because much of
what I see reported that I have first-hand knowledge of I know to be
absolute b.s.

My '99 Concorde just rolled over 200k miles - original fuel filter (as
it is designed to last the life of the car), and I routinely run it as
low as it will go before filling up on my 80 mile daily commute - almost
always between 3/16 and below 'E'. Yeah - I know - a sample of one.
But that is lifelong habit - have never had a fuel pump fail in over 40
years of driving/DIY'ing.

Bill Putney
(To reply by e-mail, replace the last letter of the alphabet in my
address with the letter 'x')