Fuel flow meter for cars .. ?

[Geir Holmavatn]

Hi,

Have all those good ol' 2" fuel flow instruments with separate sensor
been wept off the market and replaced by cruise control computers..?

Anyone knowing where to find a fuel-flow meter with sensor for gasoline
with display in European units: milliliters per minute or liters per
hour..?

Thanks for tips and comments

geir

 

[Dave Plowman (News)]

Have all those good ol' 2" fuel flow instruments with separate sensor
been wept off the market and replaced by cruise control computers..?

Anyone knowing where to find a fuel-flow meter with sensor for gasoline
with display in European units: milliliters per minute or liters per
hour..?

Thanks for tips and comments

They're not any use with injection as the fuel runs in a loop sending
unused back to the tank. And at high pressure. The normal way these days
is to count injection pulses and duration.

 

[Trevor Wilson]

Hi,

Have all those good ol' 2" fuel flow instruments with separate sensor been
wept off the market and replaced by cruise control computers..?

Anyone knowing where to find a fuel-flow meter with sensor for gasoline
with display in European units: milliliters per minute or liters per
hour..?

Thanks for tips and comments

**Here is an article published in and Australian magazine which describes a
car computer:

http://www.siliconchip.com.au/cms/A_102510/article.html

None of the local suppliers supports the kit any longer. However, as I
recall, the kit was designed by a contractor to the magazine. As a
consequence, that person probably can assist with stuff like sensors, PCBs,
programmed microcontrollers, et al. You will probably need to purchase the
article to ascertain the details. Or perhaps a polite email to the publisher
may elicit the contact details of the author.

Good luck. I almost built one, but found a (second hand) commercial device
at a very good price first. I do recall that the design was most impressive.

Trevor Wilson

 

[Allodoxaphobia]

They're not any use with injection as the fuel runs in a loop sending
unused back to the tank. And at high pressure. The normal way these days
is to count injection pulses and duration.

I suppose you could employ two (2) in-line sensors -- one on the supply
side, and one on the return side and subtract the return flow from the
supply flow in the u-processor. (I'm assuming the project used a
MicroChip [or similar uP) thingy.)

Jonesy

 

[Dave Plowman (News)]

I suppose you could employ two (2) in-line sensors -- one on the supply
side, and one on the return side and subtract the return flow from the
supply flow in the u-processor. (I'm assuming the project used a
MicroChip [or similar uP) thingy.)

The site Trevor gave in a later post says it can work by either counting
injector pulses or a flow device.

 

[Trevor Wilson]

I suppose you could employ two (2) in-line sensors -- one on the supply
side, and one on the return side and subtract the return flow from the
supply flow in the u-processor. (I'm assuming the project used a
MicroChip [or similar uP) thingy.)

The site Trevor gave in a later post says it can work by either counting
injector pulses or a flow device.

**Correct. It's a bit fiddly, but more accurate actually measuring the fuel
flow. I ended using the pulse counting method myself (I didn't care much for
mucking about with fuel lines), but I did have two fuel flow meters, which
came with the computer. One was plastic, the other a substantial die-cast
affair. Accuracy was pretty impressive, according to the manufacturer's
data.

Trevor Wilson

 

[Dave Plowman (News)]

**Correct. It's a bit fiddly, but more accurate actually measuring the
fuel flow. I ended using the pulse counting method myself (I didn't
care much for mucking about with fuel lines), but I did have two fuel
flow meters, which came with the computer. One was plastic, the other a
substantial die-cast affair. Accuracy was pretty impressive, according
to the manufacturer's data.

I must admit to being surprised. Both my cars are injection and both use
the pulse measuring system to calculate MPG and both are very accurate.
Years ago I did have an aftermarket one that measure fuel flow and it
wasn't.

 

[Chuck]

Hi,
Have all those good ol' 2" fuel flow instruments with separate sensor
been wept off the market and replaced by cruise control computers..?

Anyone knowing where to find a fuel-flow meter with sensor for gasoline
with display in European units: milliliters per minute or liters per
hour..?

Thanks for tips and comments

geir

Hi,

Fuel flow is porportional to the Duty Cycle of the injector signal. A volt
meter across across an injector will give you uncalibrated fuel flow,
Assuming the time delay to open is = to the closing delay and the waveform
is square. Web search indicates typical wave form is sq. with a 70v
inductive kick. Current rise time ~1.5ms so I guess the on time is long
resulting in a positive offset. Anyone know how = the times are? The
injectors take >.6A so I'd start with a 470 ohm R and a zener or LM431 to
filter battery changes. For relative calibration: connect across battery =
100%. Good enough for my needs: check effect of tire change, syn oil, wax
job, etc..

Check http://www.robietherobot.com/storm/fuelinjectorguide.htm for injector
flow rates and discussion of on/off delay..

Chuck

 

[James Sweet]

Fuel flow is porportional to the Duty Cycle of the injector signal. A
volt
meter across across an injector will give you uncalibrated fuel flow,
Assuming the time delay to open is = to the closing delay and the waveform
is square. Web search indicates typical wave form is sq. with a 70v
inductive kick. Current rise time ~1.5ms so I guess the on time is long
resulting in a positive offset. Anyone know how = the times are? The
injectors take >.6A so I'd start with a 470 ohm R and a zener or LM431 to
filter battery changes. For relative calibration: connect across battery
=
100%. Good enough for my needs: check effect of tire change, syn oil, wax
job, etc..

Check http://www.robietherobot.com/storm/fuelinjectorguide.htm for
injector
flow rates and discussion of on/off delay..

Look up Megasquirt, it's a DIY fuel injection system and related sites have
all the information you could want about injector pulsewidths.

 

[Trevor Wilson]

I must admit to being surprised. Both my cars are injection and both use
the pulse measuring system to calculate MPG and both are very accurate.
Years ago I did have an aftermarket one that measure fuel flow and it
wasn't.

**As I understand it (please note, I am not speaking from any real authority
here) no injector is perfect. Some are clogged a little. Some are clogged a
lot. Some are not clogged at all. Pumps are subject to wear. As a
consequence, the BEST way to measure actual fuel consumption, is to measure
the amount of fuel used, rather than inferring how much fuel SHOULD be going
into the cylinders.

Having said all that: My present car probably uses pulse measurement
techniques and the computer appears to be respectably accurate (though I've
never actually checked it).

Trevor Wilson

 

[HapticZ]

i used to just coast down hills to gain MPG advantage ;-))

 

[Michael A. Terrell]

i used to just coast down hills to gain MPG advantage ;-))

Anyone can do that. OTOH, if you can do it while going uphill... ;-)

--
Service to my country? Been there, Done that, and I've got my DD214 to
prove it.
Member of DAV #85.

Michael A. Terrell
Central Florida

 

[James Sweet]

**As I understand it (please note, I am not speaking from any real
authority here) no injector is perfect. Some are clogged a little. Some
are clogged a lot. Some are not clogged at all. Pumps are subject to wear.
As a consequence, the BEST way to measure actual fuel consumption, is to
measure the amount of fuel used, rather than inferring how much fuel
SHOULD be going into the cylinders.

Having said all that: My present car probably uses pulse measurement
techniques and the computer appears to be respectably accurate (though
I've never actually checked it).

All injectors should be within 5% of rated flow, or engine performance and
fuel economy will suffer. When new, they're within a percent or so, and
usually hold up pretty well. Measuring the pulsewidth should in most cases
be at least as accurate as typical flow sensors.

 

[Dave Plowman (News)]

**As I understand it (please note, I am not speaking from any real
authority here) no injector is perfect. Some are clogged a little. Some
are clogged a lot. Some are not clogged at all. Pumps are subject to
wear.

Yehbut EFI systems are closed loop and if the injectors are slightly out
of spec, the ECU will compensate. If they're wildly out an inaccurate OBC
will be the least of your worries...

As a consequence, the BEST way to measure actual fuel
consumption, is to measure the amount of fuel used, rather than
inferring how much fuel SHOULD be going into the cylinders.

That assumes you can make the perfect flow meter which will maintain its
performance.

Having said all that: My present car probably uses pulse measurement
techniques and the computer appears to be respectably accurate (though
I've never actually checked it).

I have to log all fuel bought for tax purposes so I can do a pretty
accurate check of the overall MPG - and when I have, the OBC has been as
near as dammit spot on.

 

[Trevor Wilson]

Yehbut EFI systems are closed loop and if the injectors are slightly out
of spec, the ECU will compensate. If they're wildly out an inaccurate OBC
will be the least of your worries...

**OK, here's where my ignorance begins to show. EFI systems are sort of
closed loop. However, under certain conditions, that may not be the case. At
startup, for instance. Under very heavy acceleration, for another. There are
probably other conditions where the system is not closed loop. So-called
'limp-home mode' would be another. I'm not well versed in automotive issues
to comment with great authority. I do understand a little, however.

That assumes you can make the perfect flow meter which will maintain its
performance.
**Indeed.


I have to log all fuel bought for tax purposes so I can do a pretty
accurate check of the overall MPG - and when I have, the OBC has been as
near as dammit spot on.

**As has mine.

Trevor Wilson

 

[Dave Plowman (News)]

**OK, here's where my ignorance begins to show. EFI systems are sort of
closed loop. However, under certain conditions, that may not be the
case. At startup, for instance. Under very heavy acceleration, for
another. There are probably other conditions where the system is not
closed loop. So-called 'limp-home mode' would be another. I'm not well
versed in automotive issues to comment with great authority. I do
understand a little, however.

Yes - but a 'dribble' from an injector or injectors would have far more of
a percentage effect at cruise or low throttle conditions. Remember the
combination of depression and fuel pressure acting on the injector is a
constant. If you think of a dripping tap (faucet?) the drip is irrelevant
when the tap is fully on.

 

[mike]

Hi,

Fuel flow is porportional to the Duty Cycle of the injector signal. A volt
meter across across an injector will give you uncalibrated fuel flow,
Assuming the time delay to open is = to the closing delay and the waveform
is square. Web search indicates typical wave form is sq. with a 70v
inductive kick. Current rise time ~1.5ms so I guess the on time is long
resulting in a positive offset. Anyone know how = the times are? The
injectors take >.6A so I'd start with a 470 ohm R and a zener or LM431 to
filter battery changes. For relative calibration: connect across battery =
100%. Good enough for my needs: check effect of tire change, syn oil, wax
job, etc..

Check http://www.robietherobot.com/storm/fuelinjectorguide.htm for injector
flow rates and discussion of on/off delay..

Chuck

What's an injector? Some of us don't have 'em and still want a diy fuel
metering system.

 

[Trevor Wilson]

What's an injector? Some of us don't have 'em and still want a diy fuel
metering system.

**Then you need a fuel flow sensor.

Trevor Wilson

 

[Dave Plowman (News)]

What's an injector? Some of us don't have 'em and still want a diy fuel
metering system.

Then I'd ask why. Wanting fuel metering suggests you're concerned with MPG.
Converting to injection will improve this.

 

[Clint Sharp]

Anyone can do that. OTOH, if you can do it while going uphill... ;-)

LOL, of course a modern EFI vehicle uses less fuel if you coast with it
in the highest gear and the throttle closed. Drop it out of gear or use
the clutch and it uses fuel to keep the engine spinning.