[xj40] Fuel Line Check Valve Placement

XJ40
1

My March 7, 2013 post detailing my installation of a fuel
line check valve prompted several off-line conversations
regarding valve placement. The most common opinion was that
placing the valve at the fuel rail was not only acceptable,
but preferable to placing it close to the fuel pump/tank,
because �All that matters is the pressure in the fuel rail.�

For convenience, I first installed my after-market valve in
the flexible hose at the rail, but fellow member
aerostarflyr told me that a check valve vendor informed him
that the valve should be installed as close as possible to
the fuel pump/tank. Since I happened to have the IRS out of
our car for a differential overhaul, I moved the valve to
the flexible line beneath the tank on our �93, which has the
in-tank pump/housing/check valve assembly.

When members began emailing me to say that the valve would
work just as well at the fuel rail, the only response I
could muster was, �So why do so most auto makers place the
check valve just ahead of the fuel pump?� Only one member
had an answer: �It�s less expensive for the manufacture to
install a single pump/check valve assembly than separate
components,� which may be true, but doesn�t explain why, on
earlier XJ40s, the check valve is a separate part from the
pump, yet is not mounted at the fuel rail but rather
connects directly to the outlet of the pump.

My extensive online searches have found no authoritative
explanation. However, in pondering this dilemma, I believe
I may have deduced the answer. For your consideration, I
offer my theories of operation of the two configurations,
assuming all other parts/systems are working properly (FPR,
coolant temp sensor, carbon canister control valve, ignition
system, battery and starter circuit, ECU, etc.), and all
cylinders have good compression.

Scenario 1, Check Valve at the Fuel Rail: While the engine
is off, fuel line pressure dissipates back through the pump
from the check valve to the tank. When the engine is
cranked and pressure in the rail drops as the injectors
operate, the pump must re-pressurize the entire fuel line
from tank to check valve/fuel rail before pressure at the
rail is restored. Since the pump�s rate of flow is
insufficient to re-pressurize the entire fuel line
instantaneously, it seems logical that there will be a lapse
of pressure as the engine tries to start, contributing to
extended cranking times.

Scenario 2, Check Valve at the Fuel Pump: While the engine
is off, pressure is maintained between the check valve and
fuel rail, which is to say, in nearly the entire fuel feed
line. When the engine is cranked, the pump needs only to
re-pressurize/prime the short distance between the pump
outlet and check valve before it is already restoring fuel
rail pressure lost through the injectors. Theoretically,
this should minimize any pressure lapse at the fuel rail and
shorten cranking times.

That�s my hypothesis. I welcome all thoughts of assent or
dissent!

Cheers,

Don–
Don B : '93 VDP Flamenco Red 189K : (ex-'88 Sovereign)
Franklin, TN, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

2

In reply to a message from Don B sent Tue 2 Apr 2013:

The external Bosch fuel pump is rated at 220 liters per hour
(70+ gallons) the short priming burst as you turn the key to
‘start’ is sufficient to repressurize the fuel line and rail.
On internal pumps, the check valve is built into the lid of
the plastic canister housing the pump. The Walbro pump also
is high capacity – 240 lph – but about half the output is
directed to an eductor that keeps the canister filled, even
with an almost empty tank. I have photos of the canister in
my albums, too.
In either case, the fuel lines stay full of fuel because
it’s a tightly closed system and both feed and return hoses
go into the tank below the fuel level.–
The original message included these comments:

rail is restored. Since the pump�s rate of flow is
insufficient to re-pressurize the entire fuel line
instantaneously, it seems logical that there will be a lapse
of pressure as the engine tries to start, contributing to
extended cranking times.

–
Pete Peterson 70E(193K) XJ40s(88-270K,89-97K, 94-122K)
Severna Park, Maryland, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

3

In reply to a message from Jaguarpete sent Tue 2 Apr 2013:

Pete,

So, based on your info, what would be the purpose of the
check valve? Your comments make it seem that there would be
no reason to even have one, yet nearly all autos do…

Is it your understanding that the flow rate of the pumps is
sufficient to fully pressurize the fuel system within a
couple of seconds without the presence of a check valve?

220 liters per hour equals 0.06 liter per second, or 61 cc
per second, or 0.016 gallons/2.15 ounces per second. Does
this flow rate decrease as the pressure in the system increases?–
The original message included these comments:

The external Bosch fuel pump is rated at 220 liters per hour
(70+ gallons) the short priming burst as you turn the key to
‘start’ is sufficient to repressurize the fuel line and rail.
On internal pumps, the check valve is built into the lid of
the plastic canister housing the pump. The Walbro pump also
is high capacity – 240 lph – but about half the output is
directed to an eductor that keeps the canister filled, even
with an almost empty tank. I have photos of the canister in
my albums, too.
In either case, the fuel lines stay full of fuel because
it’s a tightly closed system and both feed and return hoses

–
Don B : '93 VDP Flamenco Red 189K : (ex-'88 Sovereign)
Franklin, TN, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

4

In reply to a message from Don B sent Tue 2 Apr 2013:

My experience with a drop in pressure in the fuel line is that
the sudden loss of pressure causes the fuel in the line to
‘boil’ resulting in a sort of ‘vapour lock’ which, after the
engine has been shut down for an hour or more, takes a number
of ‘priming burst’ from the fuel pump before the fuel rail
achieves its operating pressure of 40 psi to facilitate engine
starting.

I suspect that is why the fuel feed line check valve is fitted
as close as possible to the pump itself so that the fuel in
the feed line and the fuel rail upstream of the fuel pressure
regulator remains a ‘non-compressable’ fluid.–
The original message included these comments:

So, based on your info, what would be the purpose of the
check valve? Your comments make it seem that there would be
no reason to even have one, yet nearly all autos do…

–
Bryan N, '91 Sovereign 4.0 L, RHD
Cambridgeshire, United Kingdom
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

5

In reply to a message from Don B sent Tue 2 Apr 2013:

The pumps are positive-displacement roller pumps so

capacity is proportional to pump speed and should not vary
with back pressure (although high pressure would make it
slow down a tad) As the commutator wears (see photos) the
motor spins slower because the brushes contact more
commutator segments at one time until it finally quits.
I don’t know why one would need a check valve since the
pump suction is always flooded and the return is almost
always submerged (except for a very low level in the tank).
Assuming the feed line is 3/8 inch tubing, it would hold
one cubic inch of fuel per foot (a gallon is 232 cubic
inches). A short priming burst should fill the line, and
repressurize quickly as soon as you start to crank the engine.
It would certainly be interesting to see how a car runs
without a check valve in the line.–
The original message included these comments:

So, based on your info, what would be the purpose of the
check valve? Your comments make it seem that there would be
no reason to even have one, yet nearly all autos do…
Is it your understanding that the flow rate of the pumps is
sufficient to fully pressurize the fuel system within a
couple of seconds without the presence of a check valve?
220 liters per hour equals 0.06 liter per second, or 61 cc
per second, or 0.016 gallons/2.15 ounces per second. Does
this flow rate decrease as the pressure in the system increases?

–
Pete Peterson 70E(193K) XJ40s(88-270K,89-97K, 94-122K)
Severna Park, Maryland, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

6

In reply to a message from Jaguarpete sent Tue 2 Apr 2013:

Pete,

Since the check valve in my car failed I can report that the
car runs fine without it, it just takes a lot more cranking
before the engine starts. I don’t think the issue is that
there isn’t fuel in the line, it’s that the fuel isn’t under
sufficient pressure for the injectors to spray proper
charges (or anything) into the intake manifold.

Your description of the fuel pump operation was enlightening
and prompted me to look up the description in the diagnostic
manual, which states, �When the (fuel pump) motor is
energized, centrifugal force acting on the rollers forces
them outward so that they act as seals. The fuel between
the rollers is then forced to the outlet side of the pump.�
This clearly implies that when the pump is not energized
and the rollers are not forced outward by centrifugal force,
the pump is not sealed and will allow pressure in the fuel
line to dissipate through the pump. This seems to explain
at least one necessity of the check valve.

According to the diagnostic manual�s description of the
in-tank fuel pump module for the �91 MY on, �Both the outlet
and return ports through the pump module have check valves.
The outlet check valve reduces back flow from the fuel rail
when the pump is off. The return check valve holds fuel
pressure in the return line from the fuel rail and prevents
siphoning if a fuel line is disconnected.� This is the
first I�ve heard of a check valve in the return line, and
this alleged second valve is not illustrated in the cutaway
view of the pump module, so I wonder if it actually exists.

Pump flow rate and pressure build-up may or may not be
proportionally related � the latter may also be related to
pump power/torque or other factors. But given the
real-world experience of many XJ40 owners it does seem that
when the check valve has failed it can take the fuel pump
several seconds to pressurize the fuel in the line to 40psi
or so for proper starting (the diagnostic manual description
of the fuel pressure regulator states that the fuel pressure
will vary from 32 psi at overrun to 44 psi at full load).

Bryan’s comments are also interesting regarding what happens
to the fuel when the pressure drops suddenly, as when the
engine is shut off, and possibly represent yet another
reason the check valve is important.

By the way, on our car the fuel lines have an inside
diameter of 8 mm, or just over 5/16 in., so your estimate of
one foot of line holding approximately one cubic inch of
fuel was very close!

Don–
The original message included these comments:

I don't know why one would need a check valve since the

pump suction is always flooded and the return is almost
always submerged (except for a very low level in the tank).
Assuming the feed line is 3/8 inch tubing, it would hold
one cubic inch of fuel per foot (a gallon is 232 cubic
inches). A short priming burst should fill the line, and
repressurize quickly as soon as you start to crank the engine.
It would certainly be interesting to see how a car runs
without a check valve in the line.

–
Don B : '93 VDP Flamenco Red 189K : (ex-'88 Sovereign)
Franklin, TN, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

7

In reply to a message from Don B sent Tue 2 Apr 2013:

I installed my check valve years ago a few inches before the
fuel filter. Never had any problem to start the car ever
since, starts right up every single time and it’s easy to
get to it.–
Joseph '94 XJ40 (British Steel, Maiden England) Naples, FL
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

8

In reply to a message from Jaguarpete sent Tue 2 Apr 2013:

Pete,

I was just checking your photo album on the fuel pump module
housing and in the photo at the link below you identify a
‘‘red rubber check valve for eductor suction.’’ This must be
the mystery ‘‘return port check valve’’ mentioned in the
diagnostic manual on the page that describes the Fuel Pump
Module:

http://www.jag-lovers.org/include/iv3.php3?zx=SdAEzBAIiM8ED8ZK%2FZnkxwTKDgSemMgJygcYtq2j3L1HN6OZvET8PwvO2gpCvAwPi5jc98kfRd3dxDz4QPvS1Qw3CUz81dj998kfQdTHBkW8CxTZ1PdC9kk21doFMQo%2BMdrPBUD2QjzKyw4FxUk21pm7Bd1TRouZ2iX7Ghfe3uZJyRI8urbmOtgqBsy%2F2Dz6JR7Nnd4D6CIPi5jLBNlENZ3Q4z3ZKx%2B3vdg1%2FD9H3bLLF9g6J9LN7x0RKxOu1LsEzAsQqrbZCv4lD8Cc7hb%2BQxmpzMwpBCY14NvoI7wLA5mqvDUCTDujnfcLzAkwnZ7IAgcRBJ6fz0zQEv%2BYmc4DDw8BlpbGBMk68KE%3D--
Don B : '93 VDP Flamenco Red 189K : (ex-'88 Sovereign)
Franklin, TN, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

9

In reply to a message from Don B sent Tue 2 Apr 2013:

Trying again to include Pete’s photo of the red rubber check
valve for eductor suction:

www.jag-lovers.org/include/iv3.php3?zx=SdAEzBAIiM8ED8ZK%2FZnkxwTKDgSemMgJygcYtq2j3L1HN6OZvET8PwvO2gpCvAwPi5jc98kfRd3dxDz4QPvS1Qw3CUz81dj998kfQdTHBkW8CxTZ1PdC9kk21doFMQo%2BMdrPBUD2QjzKyw4FxUk21pm7Bd1TRouZ2iX7Ghfe3uZJyRI8urbmOtgqBsy%2F2Dz6JR7Nnd4D6CIPi5jLBNlENZ3Q4z3ZKx%2B3vdg1%2FD9H3bLLF9g6J9LN7x0RKxOu1LsEzAsQqrbZCv4lD8Cc7hb%2BQxmpzMwpBCY14NvoI7wLA5mqvDUCTDujnfcLzAkwnZ7IAgcRBJ6fz0zQEv%2BYmc4DDw8BlpbGBMk68KE%3D--
Don B : '93 VDP Flamenco Red 189K : (ex-'88 Sovereign)
Franklin, TN, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

10

In reply to a message from Don B sent Tue 2 Apr 2013:

The red rubber flap valve in the photo is at the suction to
the eductor. It’s function is to ensure that the canister
housing the fuel pump stays full of fuel at all times, even
if there is very little fuel in the tank.–
The original message included these comments:

I was just checking your photo album on the fuel pump module
housing and in the photo at the link below you identify a
‘‘red rubber check valve for eductor suction.’’ This must be
the mystery ‘‘return port check valve’’ mentioned in the
diagnostic manual on the page that describes the Fuel Pump

–
Pete Peterson 70E(193K) XJ40s(88-270K,89-97K, 94-122K)
Severna Park, Maryland, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

11

In reply to a message from Jaguarpete sent Wed 3 Apr 2013:

Does anyone know the size of the o-ring that is attached to
the tube inside of the in tank fuel pump can?

I had pulled out my gas tank last fall on a whim. I have a
problem where if the tank gets below 1/2 full, I run out of
gas. I was hoping that there was something that only needed
to be tightened. I cleaned everything in the tank but did
not replace any parts.

The interesting thing was that I could go below a 1/2 tank
for a few days without running out of gas. The problem came
back and I haven’t tested it since.

I researched a new pump, but I do not see the o-ring listed
with any pumps.

Sorry to hijack the thread - but it seemed the perfect time
to ask.–
1991 XJ6 (XJ40) Sovereign - 153,000 miles
Oyster Bay, NY, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

12

In reply to a message from Don B sent Tue 2 Apr 2013:

Don B,

Late in on posting to your thread, but the math intrigued me.

Using info from other replies…

The fuel pump flows 61cc/second. The line from the pump to rail is
appx 15 feet long…(my guess). The fuel line over that length
holds 15 cubic inches, or +/- 244 cc. If the line is completely
empty of fuel, then it will take 244/61 to refill the line…appx 4
seconds.

A check valve should NOT leak. The best are s/s ball valve against
a s/s seat. Use that type on my private well plumbing. None of my
personal vehicles ever have a long cranking period. One or two
seconds with the key in crank position, and they fire right up.

SD Faircloth–
www.jaguarfuelinjectorservice.com
Jacksonville, Florida, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

13

In reply to a message from SD Faircloth sent Thu 4 Apr 2013:

Hi SD,

Interesting thoughts!

I think Pete has convinced us that the fuel feed line from
pump to rail is never actually empty, so the pump doesn’t
have to fill an empty line on startup so much as it has to
replenish a partially- or nearly-full line and then
pressurize it to around 40psi. The pressurization is the
process I don’t know how to calculate since a pump’s flow
rate and ability to build pressure are no doubt related but
not the same thing. Two engines may be capable of spinning
at the same rpm, but the one with greater torque will climb
a steep hill faster.

Based purely on the difference between how long our car had
to crank to start before and after I installed a new check
valve, my conclusion is that it can take at least two or
three seconds for the pump to replenish the fuel in the line
and then pressurize it sufficiently to be forced through the
injectors in proper doses.

On an XJ40, whether an earlier car with the in-line fuel
pump or a later car with an in-tank pump, the pump sits
higher than the section of fuel line from the rear wheel
well the front wheel well, so I tend to agree with Pete that
little or no fuel ‘‘drains’’ back through the pump when it is
de-energized. However, without a working check valve, it
seems logical that all the pressure in the system is
released back through the pump when it is de-energized.

I am also pondering the thought that without a working check
valve, the pump must work even harder to maintain proper
pressure since, if the rollers do not seal 100% when in
operation (and what’s the likelihood that they do in a
20-year-old pump?), pressure can continuously leak back
through the pump, like water slipping past the blades of a
steam boat’s paddle wheel. That leads me to wonder if one
of the purposes of the check valve may be to lengthen the
life of the pump, which could be another reason automakers
put the check valve as close to the pump as possible, to
minimize the length of pressurized line the pump must
constantly fight against.

Thoughts?–
The original message included these comments:

The fuel pump flows 61cc/second. The line from the pump to rail is
appx 15 feet long…(my guess). The fuel line over that length
holds 15 cubic inches, or +/- 244 cc. If the line is completely
empty of fuel, then it will take 244/61 to refill the line…appx 4
seconds.

–
Don B : '93 VDP Flamenco Red 189K : (ex-'88 Sovereign)
Franklin, TN, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

14

In reply to a message from Don B sent Thu 4 Apr 2013:

I looked at my photos in my album again. I put the Bosch
pump parts on a paper-cutting board with half-inch grid to
show relative sizes of components. The five small rollers
are held against the outer periphery by centrifugal force
and the notched rotor ‘pushes’ the rollers around. Except
for the side clearances on the rollers, it’s pretty tight,
with little slippage. The capacity takes that slippage into
account. Ans, although gasoline has little lubrication for
sliding surfaces – evidenced by the erosion of the
commutator – the rollers and rotor show little wear ,if
any, over time. The eccentricity of the rotor to the body
causes the pumping action. I now wonder whether polarity of
the electrical connections actually makes a difference in
pump performance, given the symmetry.
Gasoline is essentially incompressible, so pressure rise is
going to be virtually instantaneous in a flooded set of
lines as soon as the pump starts to turn.–
The original message included these comments:

Based purely on the difference between how long our car had
to crank to start before and after I installed a new check
valve, my conclusion is that it can take at least two or
three seconds for the pump to replenish the fuel in the line
and then pressurize it sufficiently to be forced through the
injectors in proper doses.
On an XJ40, whether an earlier car with the in-line fuel
pump or a later car with an in-tank pump, the pump sits
higher than the section of fuel line from the rear wheel
well the front wheel well, so I tend to agree with Pete that
little or no fuel ‘‘drains’’ back through the pump when it is

–
Pete Peterson 70E(193K) XJ40s(88-270K,89-97K, 94-122K)
Severna Park, Maryland, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

15

In reply to a message from Jaguarpete sent Thu 4 Apr 2013:

I think you are correct that a rise in pressure will occur nearly
instantaneously, but isn’t the real question how soon will full
pressurization occur? It would occur nearly instantaneously if the
pump has unlimited torque and 100% efficiency and gasoline is truly
incompressible, but we have to question whether the tiny motor in
the fuel pump has sufficient torque and efficiency to pressurize an
incompressible liquid without rotating a relatively significant
number of revolutions (which takes at least some time so is not
instantaneous).

I thought it would be worth checking to see just how incompressible
gasoline is. According to the website below, in the SI unit of the
Pascal(One Pa equals one Newton-per-Square Meter), gasoline has a
Bulk Modulus Elasticity of 1.3, compared to 2.15 for Water, 4.35
for Glycerin and 28.5 for Mercury (all times 10 to the 9th power).
The higher the BME the more incompressible the fluid.
Compressibility is the inverse of Bulk Modulus, so for gasoline
1/1.3=0.769, for water 1/2.15=0.465, and for mercury 1/28.5=0.0351
(all times 10 to the -9th power). For comparison, the
BME/Compressibility of glass is about 55/0.0182 and of steel
160/0.00625 (both times 10 to the 9th/-9th powers). The
relationship of BMEs for various materials appears to be linear,
since one reference website states that ‘‘glass is three times as
compressible as steel’’ (BMEs of 55 and 160 respectively). If the
relationship really is linear, then gasoline is 65% more
compressible than water ((0.769-0.465)/0.465=0.6537), which is not
insignificant.

http://www.engineeringtoolbox.com/bulk-modulus-elasticity-
d_585.html

I also found a paper on fuel pump performance by a student at MIT
that contains some interesting information for anyone who is
interested. Some of the possibly relevant issues addressed are
pump efficiency as the rollers wear and a pump’s difficulty in
handling bubbles and vapor which are inevitably created in the fuel
system:

http://www.google.com/url?
sa=t&rct=j&q=&esrc=s&frm=1&source=web&cd=1&cad=rja&ved=0CDIQFjAA&url
=http%3A%2F%2Fdspace.mit.edu%2Fbitstream%2Fhandle%2F1721.1%2F37737%
2F31369689.pdf%
3F…&ei=sh1eUbjqE5L08ATWnIGQCw&usg=AFQjCNG4PhUqzv5Pw7VbMDzJbMql5_x8h
A&sig2=febyxyExTG_6P31hlWm71Q&bvm=bv.44770516,d.eWU

I’m still trying to gain a full understanding of this issue,
particularly as it relates to the purpose of the check valves and
reasoning for their placement near the pump. I found a Wikipedia
entry on mechanical and electric fuel pumps that I hadn’t read
before. Regarding check valves, it states: A check valve is
located at both the inlet and outlet ports of the pump chamber to
force the fuel to flow in one direction only.‘’ Perhaps that is the
principal reason for their placement near the pump?–
The original message included these comments:

Gasoline is essentially incompressible, so pressure rise is
going to be virtually instantaneous in a flooded set of
lines as soon as the pump starts to turn.

–
Don B : '93 VDP Flamenco Red 189K : (ex-'88 Sovereign)
Franklin, TN, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

16

In reply to a message from Don B sent Thu 4 Apr 2013:

Don,

I believe I can answer that question from practical
experience - refer to these pictures from my photo album:-

http://www.jag-lovers.org/snaps/snap_view.php3?id=1357201999

On a car with check valves at the pump and the fuel
pressure regulator working as designed (which means that
(near) incompressible liquid fuel remains in the full length
of the fuel line from the pump outlet to the far end of the
fuel rail) a single half-second priming burst from the fuel
pump when the ignition is first turned on will
instantaneously raise the fuel pressure to the required
level to allow the engine to start. See the third picture on
the top row of that sequence.

Do not forget, that priming burst lasts only about half a
second. I do not know the rpm of the pump at full chat, but
it can’t turn far in half a second! :slight_smile: Also, do not forget
that the fuel line remains pressurised (the ‘hold’ pressure)
for a goodly length of time after an engine shut down - in
fact, with both check valves working correctly, the pressure
in the fuel feed line actually increases (as shown in the
later pictures in that sequence) after engine shut-down due
to latent heat.

The starting problems arise when one or both of those check
valves (at pump and/or FPR) fail. As I mentioned in an
earlier post in this thread, the sudden loss of pressure in
the fuel line, particularly after a hot engine is shut down,
probably causes the fuel to ‘boil’ (vaporise) and that means
that you end up with very compressible fuel vapour for the
whole length of the fuel line. That requires repeated
half-second priming bursts from the fuel pump (by using the
‘key-cycling’ trick) until sufficient pressure is built up
to allow the engine to fire. Of course, the same effect can
be achieved by repeated cranking so that the CPS signal
allows the fuel pump to work in addition to the half-second
priming burst.

Clearly, if the check valve were to be installed at the
entrance to the fuel rail instead of at the outlet of the
pump, you are still left with the length of fuel line back
to the pump which will immediately lose its pressure when
the fuel pump stops working on engine shut down. Whether or
not that also causes fuel vaporisation it will still require
the fuel pump to do much more work when you next try to
start the engine.

I would think that is a good enough reason to position the
check valve at the outlet from the fuel pump.–
The original message included these comments:

I think you are correct that a rise in pressure will occur nearly
instantaneously, but isn’t the real question how soon will full
pressurization occur? It would occur nearly instantaneously if the
pump has unlimited torque and 100% efficiency and gasoline is truly
incompressible, but we have to question whether the tiny motor in
the fuel pump has sufficient torque and efficiency to pressurize an
incompressible liquid without rotating a relatively significant
number of revolutions (which takes at least some time so is not
instantaneous).

–
Bryan N, '91 Sovereign 4.0 L, RHD
Cambridgeshire, United Kingdom
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

17

In reply to a message from Don B sent Thu 4 Apr 2013:

Re: purpose of the check valve.

I got to thinking about the Wiki entry and explanation. To force
fuel to flow in one direction only. I suggest it has another
function, which I think may have already been pointed out, but
can’t remember. A check valve should also not allow a liquid to
flow back upstream of the valve. In other words, it keeps the line
downstream of the valve full of whatever liquid is in the line.

If the valve is placed near the fuel rail, then it should keep fuel
downstream of that location…which is the regulator and fuel rail.
That placement would/may allow the line upstream of the valve to
drain back thru the pump and into the tank. Or drain back to a
point where the fuel level is located…(a liquid seeking its’ own
level due to gravity). This could potentially allow the entire fuel
line upstream of the valve near the rail to empty.

If the check valve is placed near the pump outlet, then the valve
should maintain fuel in everything downstream of the valve…which
includes the fuel line to the rail, the filter, and the fuel rail.
If the pump and valve are in the tank and submerged in fuel, then
everything should theoretically always be full of fuel.

Therefore, if my above summation is accurate, the best location for
the check valve is just beyond the pump outlet.

I think I indicated earlier that I have a private well which I
maintain. The well pump sits above ground maybe 18’‘. I have a check
valve located about 24’’ upstream of the pump. This in theory should
keep the pump always full of water, and water not draining back
thru the upstream piping, causing the pump to empty. Seems to work
fine.

SD Faircloth–
The original message included these comments:

I’m still trying to gain a full understanding of this issue,
particularly as it relates to the purpose of the check valves and
reasoning for their placement near the pump. I found a Wikipedia
entry on mechanical and electric fuel pumps that I hadn’t read
before. Regarding check valves, it states: A check valve is
located at both the inlet and outlet ports of the pump chamber to
force the fuel to flow in one direction only.‘’ Perhaps that is the
principal reason for their placement near the pump?

–

Jacksonville, Florida, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

18

In reply to a message from Bryan N sent Fri 5 Apr 2013:

Bryan,

An excellent summation! Your reminder about fuel
vaporization due to heat coincides directly with statements
in the MIT paper I linked in my previous post.

Thanks for making sense of it all, as usual!

Don–
The original message included these comments:

I believe I can answer that question from practical
experience - refer to these pictures from my photo album:-
http://www.jag-lovers.org/snaps/snap_view.php3?id=1357201999
fact, with both check valves working correctly, the pressure
in the fuel feed line actually increases (as shown in the
later pictures in that sequence) after engine shut-down due
to latent heat.
The starting problems arise when one or both of those check
valves (at pump and/or FPR) fail. As I mentioned in an
earlier post in this thread, the sudden loss of pressure in
the fuel line, particularly after a hot engine is shut down,

–
Don B : '93 VDP Flamenco Red 189K : (ex-'88 Sovereign)
Franklin, TN, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

19

In reply to a message from SD Faircloth sent Fri 5 Apr 2013:

SD,

Your well pump illustration raise another excellent point,
that positioning the two check valves a closely as possible
to the inlet and outlet of the pump has the benefit of
keeping the pump full of fuel at all times and therefore
always fully primed.–
The original message included these comments:

maintain. The well pump sits above ground maybe 18’‘. I have a check
valve located about 24’’ upstream of the pump. This in theory should
keep the pump always full of water, and water not draining back
thru the upstream piping, causing the pump to empty. Seems to work

–
Don B : '93 VDP Flamenco Red 189K : (ex-'88 Sovereign)
Franklin, TN, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !

20

In reply to a message from Don B sent Fri 5 Apr 2013:

Don,

I am not familiar with where the fuel pump is located on an XJ40?
However, I suspect it is in the fuel tank. If so, then the pump is
fully submerged in fuel and should therefore always be full of fuel.
If that is true, then there should be no need for a check valve on
the suction/upstream side of the pump in order to keep the pump
full of fuel. So then, only requiring a check valve on the
outlet/discharge side, to keep everything downstream full of
fuel…the fuel line to the engine, the filter and the rail.

SD–
The original message included these comments:

Your well pump illustration raise another excellent point,
that positioning the two check valves a closely as possible
to the inlet and outlet of the pump has the benefit of
keeping the pump full of fuel at all times and therefore
always fully primed.

–

Jacksonville, Florida, United States
–Posted using Jag-lovers JagFORUM [forums.jag-lovers.org]–
–Support Jag-lovers - Donate at http://www.jag-lovers.org/donate04.php –

Visit the Jag Lovers homepage at http://www.jag-lovers.org for exciting services and resources including Photo Albums, Event Diary / Calendar, On Line Books and more !