Searching \ for 'Instantaneous Fuel Consuption Measurement' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=instantaneous+fuel
Search entire site for: 'Instantaneous Fuel Consuption Measurement'.

Truncated match.
PICList Thread
'Instantaneous Fuel Consuption Measurement'
1998\06\25@042818 by Nuno Pedrosa

flavicon
face
Hi there!

 WITHOUT going way OT like in that injection discussion, I would like
to know if someone has used some kind of sensor to measure fuel flow.
 The idea is to measure the instantaneous fuel consuption. Problem is
that it's a small flow. If cars spent more fuel, it would be a lot
easier! 8)
 Why am I here?
 Well, I want to build a small car-computer, based on PIC. I really
would like to include fuel consuption indication on its features.

 Thanx,

Nuno Pedrosa.


--

                                            _/^\_            __
_____/\_________________________________oOOo__O-O__oOOo______/oo\____
 /\/ /  Nuno Filipe Freitas Pedrosa                   __ /..\__/
 \ \/\  spam_OUTNuno.PedrosaTakeThisOuTspamoen.siemens.de           /`'\\__/-\ __
  \/\ \ Tel. : 0049-89-72222342                      \__/  \__//@@\
___/ /\/_______________________________________________/OO\_..__\__/_
  \/                                                  \__/

1998\06\25@045626 by David BALDWIN

flavicon
face
I found a project for 16C84 of full consumption. Already done, just
have a look on the net. If you nedd it, I got the files here.

David

1998\06\25@053327 by Nuno Pedrosa

flavicon
face
Well David, could you please send me the files?

I can't find that project through yahoo.

Thanks,

 Nuno.

David BALDWIN wrote:
>
> I found a project for 16C84 of full consumption. Already done, just
> have a look on the net. If you nedd it, I got the files here.
>
> David

--

                                            _/^\_            __
_____/\_________________________________oOOo__O-O__oOOo______/oo\____
 /\/ /  Nuno Filipe Freitas Pedrosa                   __ /..\__/
 \ \/\  .....Nuno.PedrosaKILLspamspam@spam@oen.siemens.de           /`'\\__/-\ __
  \/\ \ Tel. : 0049-89-72222342                      \__/  \__//@@\
___/ /\/_______________________________________________/OO\_..__\__/_
  \/                                                  \__/

1998\06\25@083129 by WF AUTOMACAO

flavicon
face
Nuno Pedrosa wrote:
{Quote hidden}

RS Components has!

Miguel.

1998\06\25@084546 by Nuno Pedrosa

flavicon
face
Yeah, that sounds easy enough. Problem is the flow. Too small for such
an "homemade sensor". Comercial sensors of this kind requires flows
larger than 1liter/min, I think... That's a lot of fuel!

 Nuno.
--

                                            _/^\_            __
_____/\_________________________________oOOo__O-O__oOOo______/oo\____
 /\/ /  Nuno Filipe Freitas Pedrosa                   __ /..\__/
 \ \/\  .....Nuno.PedrosaKILLspamspam.....oen.siemens.de           /`'\\__/-\ __
  \/\ \ Tel. : 0049-89-72222342                      \__/  \__//@@\
___/ /\/_______________________________________________/OO\_..__\__/_
  \/                                                  \__/

1998\06\25@085006 by Timothy D. Gray

flavicon
face
not if you make it tiny... 1/4 inch dia. I've gotten 10ml per minute
readings that way before (water not fuel) It took me about 2 weeks to
build it and it did leak :-/ but it could be done.
If you could ensure a clean fuel stream a positive displacement system
would work (Measure 1ml at a time is possible that way.) at work we use a
laser to measure minute flow and particulates, but that is beyond the
scope of a home-brewer.


On Thu, 25 Jun 1998, Nuno Pedrosa wrote:

{Quote hidden}

1998\06\25@125322 by tjaart

flavicon
face
Timothy D. Gray wrote:

> not if you make it tiny... 1/4 inch dia. I've gotten 10ml per minute
> readings that way before (water not fuel) It took me about 2 weeks to
> build it and it did leak :-/ but it could be done.
> If you could ensure a clean fuel stream a positive displacement system
> would work (Measure 1ml at a time is possible that way.) at work we use a
> laser to measure minute flow and particulates, but that is beyond the
> scope of a home-brewer.
>
> On Thu, 25 Jun 1998, Nuno Pedrosa wrote:
>
> > Yeah, that sounds easy enough. Problem is the flow. Too small for such
> > an "homemade sensor". Comercial sensors of this kind requires flows
> > larger than 1liter/min, I think... That's a lot of fuel!

I read about a technique where you run the petrol pipe
through a magnetic field and then past a hall sensor. I
can't vouch for this because I have never seen it work.

--
Friendly Regards

Tjaart van der Walt
tjaartspamspam_OUTwasp.co.za

Add your voice !!
Vote at the Great G Com Public Vote. Go to :
http://www.wasp.co.za/~tjaart/gcomvote.html

|--------------------------------------------------|
|                WASP International                |
|R&D Engineer : GSM peripheral services development|
|--------------------------------------------------|
|SMS @spam@0832123443KILLspamspamwasp.co.za  (160 chars max)|
|     http://www.wasp.co.za/~tjaart/index.html     |
|Voice: +27-(0)11-622-8686  Fax: +27-(0)11-622-8973|
|          WGS-84 : 26¡10.52'S 28¡06.19'E          |
|--------------------------------------------------|

1998\06\25@142839 by Nuno Pedrosa

flavicon
face
Well, ok. You're right about the size. But would you like your car to
leak fuel through the fuel meter?
 You've tried it, and you know it ain't easy to make such a sensor at
home without leakages...

 I'll try to have a look in RS components, like someone (Marcelo, I
think... Sorry! Lost that mail.8(  ) suggested. I should have a catalog
somewhere in this building... 8)

 Nuno.

Timothy D. Gray wrote:
>
> not if you make it tiny... 1/4 inch dia. I've gotten 10ml per minute
> readings that way before (water not fuel) It took me about 2 weeks to
> build it and it did leak :-/ but it could be done.
> If you could ensure a clean fuel stream a positive displacement system
> would work (Measure 1ml at a time is possible that way.) at work we use a
> laser to measure minute flow and particulates, but that is beyond the
> scope of a home-brewer.

--

                                            _/^\_            __
_____/\_________________________________oOOo__O-O__oOOo______/oo\____
 /\/ /  Nuno Filipe Freitas Pedrosa                   __ /..\__/
 \ \/\  KILLspamNuno.PedrosaKILLspamspamoen.siemens.de           /`'\\__/-\ __
  \/\ \ Tel. : 0049-89-72222342                      \__/  \__//@@\
___/ /\/_______________________________________________/OO\_..__\__/_
  \/                                                  \__/

1998\06\25@153145 by David Tait

flavicon
picon face
David Baldwin wrote:

> found a project for 16C84 of full consumption. Already done, just
> have a look on the net. If you nedd it, I got the files here.

Perhaps a URL would be useful:

http://indigo.ie/~grabe/PICfuelRate/PICmpg.html

David

1998\06\25@154618 by Martin Green

flavicon
face
    We had a long thread on this many months ago. The short answer is that
    if your vehicle has fuel injection, then the trick is to measure the
    pulse rate to the injector solenoids. Multiply this by the average
    fuel quantity injected per pulse, and you have fuel flow.

    The only problem is I don't know where you would get the injection
    quantity per pulse value, but it might be in the shop manual for the
    vehicle in question.


    CIAO - Martin.


______________________________ Reply Separator _________________________________
Subject: Instantaneous Fuel Consuption Measurement
Author:  pic microcontroller discussion list <RemoveMEPICLISTTakeThisOuTspamMITVMA.MIT.EDU> at
Internet
Date:    6/25/98 10:08 AM


Hi there!

 WITHOUT going way OT like in that injection discussion, I would like
to know if someone has used some kind of sensor to measure fuel flow.
 The idea is to measure the instantaneous fuel consuption. Problem is
that it's a small flow. If cars spent more fuel, it would be a lot
easier! 8)
 Why am I here?
 Well, I want to build a small car-computer, based on PIC. I really
would like to include fuel consuption indication on its features.

 Thanx,

Nuno Pedrosa.


--

                                            _/^\_            __
_____/\_________________________________oOOo__O-O__oOOo______/oo\____
 /\/ /  Nuno Filipe Freitas Pedrosa                   __ /..\__/
 \ \/\  spamBeGoneNuno.PedrosaspamBeGonespamoen.siemens.de           /`'\\__/-\ __
  \/\ \ Tel. : 0049-89-72222342                      \__/  \__//@@\
___/ /\/_______________________________________________/OO\_..__\__/_
  \/                                                  \__/

1998\06\25@155747 by David VanHorn

flavicon
face
The way I've always seen this done is either to tap an injector, or
to pick up current pulses from the fuel pump.  After a calibration
run to establish the number of pulses per gallon, you have a
fairly accurate indication of fuel flow.

1998\06\25@162400 by David VanHorn

flavicon
face
>     The only problem is I don't know where you would get the injection
>     quantity per pulse value, but it might be in the shop manual for the
>     vehicle in question.



Fill the tank, zero the counter, drive for a while (the longer the better)
refill the tank, and enter the fuel quantity used to refill.

During this process, we've been counting injector pulses, so we now
know that there are X pulses per gallon. (liter or whatever)

1998\06\26@021719 by Oyvind Kaurstad

flavicon
face
>>     The only problem is I don't know where you would get the injection
>>     quantity per pulse value, but it might be in the shop manual for the
>>     vehicle in question.


>Fill the tank, zero the counter, drive for a while (the longer the better)
>refill the tank, and enter the fuel quantity used to refill.

>During this process, we've been counting injector pulses, so we now
>know that there are X pulses per gallon. (liter or whatever)

Hmm...
How can this be correct?
It will probably get you in the ballpark, but all you get is an average.
To measure instantaneous fuel consumption you will have to both time and count
the injector pulses for a very short period of time. The amount of fuel flowing
in
the injector is constant while it is open, and therefore you have to measure
for how
long it stays open. This is typically in the ms range. When you know this you
can calculate
fuel consumption.

-Oyvind

1998\06\26@041302 by Nuno Pedrosa

flavicon
face
Why do I want to use a flow meter?

Gasoline Engine:
 Situation 1: Carburator
    Usually there's a fuel pump, electrical most of the times. But it
works always with the same speed. It's fuel flow is constant. All the
excess fuel returns to the tank. So, I cannot use this to measure flow.
In mechanical pumps I don't even have the chance, naturally. There's a
vacuum meter that's related to the fuel usage, but not very accurate.

 Situation 2: Injection
    Like it has been said, I would have to count the injections, and
measure the time each one lasts. It looks feasible. Calibration would be
possible, but difficult without official data. But it could be possible
to obtain this anyway.

Diesel Engine:
 Situation 1: Pressure Injection
    There's no electric relevant parts in most engines circulating. The
pump works with a constant flow. Excess fuel is returned to the tank. I
would have to buid a piezo detector (don't know of any available), but
it's very difficult to measure injection duration.
 Situation 2: Electric Injection
    Still seldom used in my "target cars". No chance of using it.

  So, since my "targeted" vehicles (namely mine!) work with Diesel
Engines, with pressure injectors, the only feasable way I see is with a
fuel flow sensor. Worse than that, TWO sensors. One for infuel, another
for outfuel.
  The bonus with this approach: it's universal! Measure the fuel that
goes out of the tank. Subtract the amount that returns, and voila!
Instaneous (almost) fuel consuption.

 Sorry for the long mail, but it seemed to be necessary a more accurate
description of the situation.

 Thanx for all the answers! I find amazing the amount of experience
contained in this list!

 Bye,
  Nuno.


--

                                            _/^\_            __
_____/\_________________________________oOOo__O-O__oOOo______/oo\____
 /\/ /  Nuno Filipe Freitas Pedrosa                   __ /..\__/
 \ \/\  TakeThisOuTNuno.PedrosaEraseMEspamspam_OUToen.siemens.de           /`'\\__/-\ __
  \/\ \ Tel. : 0049-89-72222342                      \__/  \__//@@\
___/ /\/_______________________________________________/OO\_..__\__/_
  \/                                                  \__/

1998\06\26@043320 by tjaart

flavicon
face
Nuno Pedrosa wrote:

>  So, since my "targeted" vehicles (namely mine!) work with Diesel
> Engines, with pressure injectors, the only feasable way I see is with a
> fuel flow sensor. Worse than that, TWO sensors. One for infuel, another
> for outfuel.
>    The bonus with this approach: it's universal! Measure the fuel that
> goes out of the tank. Subtract the amount that returns, and voila!
> Instaneous (almost) fuel consuption.
>
>   Sorry for the long mail, but it seemed to be necessary a more accurate
> description of the situation.

I figure out how the hall effect flow sensor works.When the fuel flows through the field, the
(weakly
magnetic) molecules line up magnetically. As soon
as they leave the field, they probably tend to misalign
again. The faster it flows, the further it goes before
misaligning, and the higher magnetic field will be
detected by the sensor. You wouldn't even have to
cut a line to measure.

--
Friendly Regards

Tjaart van der Walt
RemoveMEtjaartspamTakeThisOuTwasp.co.za

Add your voice !!
Vote at the Great G Com Public Vote. Go to :
http://www.wasp.co.za/~tjaart/gcomvote.html

|--------------------------------------------------|
|                WASP International                |
|R&D Engineer : GSM peripheral services development|
|--------------------------------------------------|
|SMS 0832123443EraseMEspam.....wasp.co.za  (160 chars max)|
|     http://www.wasp.co.za/~tjaart/index.html     |
|Voice: +27-(0)11-622-8686  Fax: +27-(0)11-622-8973|
|          WGS-84 : 26¡10.52'S 28¡06.19'E          |
|--------------------------------------------------|

1998\06\26@153956 by Mike Keitz

picon face
On Fri, 26 Jun 1998 10:01:35 +0200 Nuno Pedrosa
<EraseMEnuno.pedrosaspamOEN.SIEMENS.DE> writes:

>   So, since my "targeted" vehicles (namely mine!) work with Diesel
>Engines, with pressure injectors, the only feasable way I see is with
>a
>fuel flow sensor. Worse than that, TWO sensors. One for infuel,
>another
>for outfuel.

Doing that will work very poorly.  You'll be subtracting two large
measurements to get a small measurement.  Slight errors in the
measurements will cause a large error in the result.

Maybe the system could be rearranged so the recirculation occurs after
the flow sensor.  For example, add a small "day tank" fuel reservoir,
then measure how much fuel needs to be transferred from the main tank to
keep it full.  The measurement could be done indirectly by using a
calibrated fuel injector or a positive-displacement pump to do the
transfer.

A simple approach that doesn't involve the possibility of leaks is to
just measure the position of the linkage that controls the fuel injector
pump and engine RPM.  For every position of the throttle, the pump will
inject a definite amount of fuel per revolution.  Integrate the flows
over time.


_____________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com
Or call Juno at (800) 654-JUNO [654-5866]

1998\06\26@204847 by Igorp

flavicon
face
Hi Nuno:

If you have a car with a injection, the best sensor is a car computer
itself. I am close to finish the computer that you mentioned  and all
signals that I needed I have found on a car computer. I do the computer
for my Ford Escort with the central injection. The frequency of pulses
to the injection valve is more or less (it depends of temperature,
pressure of petrol, etc.) equal a small quantity of petrol. I did some
measurement and I am certain that it would be enough accurate for your
purpose.

Let me know if you are interested about my solution. I built a car
computer with PIC16F84.

The best sensor, used with car computers for measurment a consumption is
a heated wire with two thermistors (very small of course) in a way of
petrol. You heat a wire with pulses and measure the resistance of both
thermistor - basicaly you measure how much time a hotter petrol needs
for a distance between thermistors. I think that one thermistor only
could serve well too.

Regards

Igor

{Quote hidden}

1998\06\26@205537 by David Peterson

flavicon
face
Two Australian magazines(Electronics Australia & Silicon Chip) have
published projects for car computers in the last few years and both had
sources for two types of fuel flow sensors.
The sensor was positioned between the fuel return line & the injector
assembly (or carby)
Another thought, is the return line really nesessary or can it be blocked,
early model vehicals didn't have them. Maybe the flow pressure wasn't as
high ? My old jalopy has an electric fuel pump with no reticulation.
hth David

1998\06\26@212544 by Clyde Smith-Stubbs

flavicon
face
On Sat, Jun 27, 1998 at 09:05:50AM +1000, David Peterson wrote:

> Another thought, is the return line really nesessary or can it be blocked,
> early model vehicals didn't have them. Maybe the flow pressure wasn't as
> high ? My old jalopy has an electric fuel pump with no reticulation.

It is necessary; the fuel pumps used on injected engines are different
to those used on carburetted engines. The pump for an injection system operates
at a high pressure (it provides the injection pressure). The electric pumps
used on carburettors basically provide a low pressure via a spring-loaded
diaphragm with a solenoid to reload the spring as required; with no flow the
pump simply stops. An injection type pump is quite different and the return
line is required so that the pressure can be regulated (the injectors depend on
a constant pressure).

I suspect that cars with injection systems and trip computers get the fuel
flow data from the engine controller, which calculates it from the injection
parameters. To measure this externally would require monitoring of the pulse
rate and duration to the injectors and knowing the parameters to convert this
to flow rate.


--
Clyde Smith-Stubbs               |            HI-TECH Software
Email: RemoveMEclydeEraseMEspamEraseMEhtsoft.com          |          Phone            Fax
WWW:   http://www.htsoft.com/    | USA: (408) 490 2885  (408) 490 2885
PGP:   finger RemoveMEclydespam_OUTspamKILLspamhtsoft.com   | AUS: +61 7 3354 2411 +61 7 3354 2422
---------------------------------------------------------------------------
HI-TECH C: compiling the real world.

1998\06\26@213641 by Ron Fial

flavicon
face
Perhaps s little off topic, but when the Rutan airplane the 'voyager' flew aroun
d the world, they had an accurate fuel flow system measuring flow to the engine.
 As I recall, the completely forgot that the engine had a return line from the
carb/fuel injectors, so they thought they were out of fuel quite early.  Testing
by measuring the speed at which the airplane stalled (related to the airplanes
weight) reassured them that the fuel was still (or again) in the tanks!

 Regards,
    Ron Fial

======================================
At 11:08 AM 6/27/98 +1000, you wrote:
{Quote hidden}

1998\06\27@064300 by Morgan Olsson

picon face
At 18:28 1998-06-26 -0700, you wrote:
>Perhaps s little off topic, but when the Rutan airplane the 'voyager' flew
around the world, they had an accurate fuel flow system measuring flow to
the engine.  As I recall, the completely forgot that the engine had a
return line from the carb/fuel injectors, so they thought they were out of
fuel quite early.  Testing by measuring the speed at which the airplane
stalled (related to the airplanes weight) reassured them that the fuel was
still (or again) in the tanks!
>
>  Regards,
>     Ron Fial
>
I think I read about it.
As I understood it, the problem was that in bad weather due to the
airplane«s movement, the fuel occasionally flowed backwards from some kind
of collecting tank back to the source tank.  Problem was that the sensor
they used could not differ between forward and backward...

/Morgan

/  Morgan Olsson, MORGANS REGLERTEKNIK, SE-277 35 KIVIK, Sweden \
\  RemoveMEmrtKILLspamspaminame.com, ph: +46 (0)414 70741; fax +46 (0)414 70331    /

1998\06\29@064943 by Nuno Pedrosa

flavicon
face
> Doing that will work very poorly.  You'll be subtracting two large
> measurements to get a small measurement.  Slight errors in the
> measurements will cause a large error in the result.

I'm afraid you could be very right about that... Things can look in
theory, when you forget real-world constraints, and then, on practice,
it just doesn't work...

> Maybe the system could be rearranged so the recirculation occurs after
> the flow sensor.  For example, add a small "day tank" fuel reservoir,
> then measure how much fuel needs to be transferred from the main tank to
> keep it full.  The measurement could be done indirectly by using a
> calibrated fuel injector or a positive-displacement pump to do the
> transfer.

Like a dosing system, I think... I had thought of that... Fill a small
tank, and dump it, fill and dump, fill and dump... Resolution would
depend on the size of the tank. But I don't know where I could put it...
After the pump is not possible. There's the pressure. Before the pump,
I had to find some system to avoid lack of fuel in the system... It's
air-tight, which is a problem... Mechanical, but still a problem.

> A simple approach that doesn't involve the possibility of leaks is to
> just measure the position of the linkage that controls the fuel injector
> pump and engine RPM.  For every position of the throttle, the pump will
> inject a definite amount of fuel per revolution.  Integrate the flows
> over time.

I have to study this option... Although not as generic as I would like,
it seems the easiest one! I already intend to have an RPM sensor, so
the only extra component it's the position of the linkage. I need to
get a hold of data-sheets and talk to my father (he's the expert in
diesel mechanics). Problems that jump to my mind... Diesel pumps have
internal fuel regulation systems... The older ones are mechanical. This
could be a problem when you have the engine in high effort. The RPM is
low, but you are spending a lot of fuel... But the accelator pedal is
down, also, so it could be used to calculate... Seems promising.

Next weekend I'll talk to my father about it.

Nuno.
--

                                            _/^\_            __
_____/\_________________________________oOOo__O-O__oOOo______/oo\____
 /\/ /  Nuno Filipe Freitas Pedrosa                   __ /..\__/
 \ \/\  Nuno.PedrosaSTOPspamspamspam_OUToen.siemens.de           /`'\\__/-\ __
  \/\ \ Tel. : 0049-89-72222342                      \__/  \__//@@\
___/ /\/_______________________________________________/OO\_..__\__/_
  \/                                                  \__/


'Instantaneous Fuel Consuption Measurement'
1998\07\16@024906 by Chuck Rice
flavicon
face
At 1:08 AM -0700 6/25/98, Nuno Pedrosa wrote:


> Hi there!
>
>   WITHOUT going way OT like in that injection discussion, I would like
> to know if someone has used some kind of sensor to measure fuel flow.
>   The idea is to measure the instantaneous fuel consuption. Problem is
> that it's a small flow. If cars spent more fuel, it would be a lot
> easier! 8)
>   Why am I here?
>   Well, I want to build a small car-computer, based on PIC. I really
> would like to include fuel consuption indication on its features.

You might try using a diaphragm to measure the fuel. Create a small chamber
with a inflow port and an outflow port. Divide the chamber in half with
the diaphragm. One side would contain gas and the two ports, the other
side would be sealed and contain air and a non-contact switch (If the
diaphragm breaks, would not want to spark a fire).

Close the output port and allow the fuel pump to fill the gas side of
the chamber expanding the diaphragm. Close the input port and open the
output port. When enough fuel has been used to relax the diaphragm,
you count that as one unit. Then your fuel flow is the the number of
units times the volume of the diaphragm as related to time. The PIC
would use one output pin to control the inflow and outflow ports, and
one input pin connected to the switch on the diaphragm. -Chuck-

__________________________________________________________________________
Chuck Rice                                     <spamBeGoneChuckSTOPspamspamEraseMEWildRice.com>

1998\07\16@034555 by Nuno Pedrosa

flavicon
face
Chuck Rice wrote:
> You might try using a diaphragm to measure the fuel. Create a small chamber
> with a inflow port and an outflow port. Divide the chamber in half with
> the diaphragm. One side would contain gas and the two ports, the other
> side would be sealed and contain air and a non-contact switch (If the
> diaphragm breaks, would not want to spark a fire).

I had thought of a chamber measurement, but never thought of the
diaphragm. That would seem a real nice idea, but...

> Close the output port and allow the fuel pump to fill the gas side of
> the chamber expanding the diaphragm.

...I can not close the output port. The fuel system is air tight. If I
close the output port, I close the fuel to the engine and it stops. I
would need some kind of parallel chambers: close one, use the other.
Invert. Repeat.
So, I would need four electro-valves, 2 diaphragms and 2 switches,
besides the chambers. And all of this for one single device...

>Close the input port and open the
> output port. When enough fuel has been used to relax the diaphragm,
> you count that as one unit. Then your fuel flow is the the number of
> units times the volume of the diaphragm as related to time. The PIC
> would use one output pin to control the inflow and outflow ports, and
> one input pin connected to the switch on the diaphragm. -Chuck-

Maybe I'm seeing it all wrong... What do you think?

Nuno.

__________________________________________________________________________
> Chuck Rice                                     <KILLspamChuckspamBeGonespamWildRice.com>

--
   .^.                                              _,^,_         /\
___( | )_____ Nuno Filipe Freitas Pedrosa __________ o(`} | , __/\/ /__
/*\\|//*\    SIEMENS S.A. Portugal                  (]|  /'    \ \/\
\(\\V//)/    EraseMENuno.PedrosaspamEraseMEoen.siemens.de     (]|`%   (") \/\ \
 ` -=- '     Tel.  :00351-1-3502454         (")      /`/        / /\/
__B//|\\P___________________________________________\' / `/______\/____
   `-' "Try and leave this world a little better than you found it..."

1998\07\16@035427 by Ron Fial

flavicon
face
Most continuous fuel meas. system use a closed circular body with a paddle that
goes around and around as the fuel flows through it.  Each revolution is counted
by a hall (magnetic) sensor, because a tiny magnet is inside each paddle, so th
e fuel system is completely closed.  I am familiar with this type of sensor beca
use of its use in small airplane engines.  It is extremely reliable.   You shoul
d be able to find a company that makes such a device on the WEB.  Perhaps if you
find a picture, you can manufacture one of your own.
  Regards,
    Ron Fial.

At 08:35 AM 7/16/98 +0000, you wrote:
{Quote hidden}

1998\07\16@040303 by

flavicon
face
{Quote hidden}

1998\07\16@113126 by Peter L. Peres

picon face
 You could try to use an ultrasonic differential Doppler method, but I
suspect that this will be expensive, in development costs and transducers.
This method is metrology instrumentation grade, once you have a tube that
does not vary in dimensions and a constant density liquid. I know that
something like this is used to measure blood flow in artificial kidneys
and other delicate places. If you want to get really picky, also measure
the temperature and pressure in the tube to correct for compressibility
and density variation ;) The differential Doppler method requires an
ultrasonic transmitter and two receivers, all immersed in a tube section
of constant cross-section.  It can measure flow in both directions, and
has a lower speed limit under which no measurement is possible (due to the
phase resolution of the detection equipment). There are no S/N and
preamplifier problems as the transducer output signal is huge and clean.
imho this method is excellent for a PIC, used both to generate the
ultrasonic frequency directly (by bit-bashing) and for data aquisition.

 Another method that does not use moving parts, uses the thermal transfer
properties of the liquid. You have a piece of tubing that is thermally
insulated, and inside it, a heated thermistor, and at some distance
downstream, another one that is not heated. Knowing the thermal
conductvity of the liquid and the temperature read-out at the 2nd
thermistor, and tube diameter etc you can determine flow rate with uncanny
accuracy. A third unheated thermistor is used upflow of the heated one to
give a temperature reference if the liquid temperature is not constant.
This method requires serious calibration, unlike the differential Doppler
one, which needs none. However, it measures flow down to very, very near
zero, in both directions.

hope this helps,

       Peter

1998\07\16@125819 by Bruce Turrentine

picon face
The problem with measuring fuel flow in a fuel system such as fuel injection
is the fuel pressure regulator bleeds fuel back to the tank to control the
pressure. Therefore, you would have to measure the fuel returning as well as
the total flow and compute the difference.

1998\07\18@020611 by Chuck Rice

flavicon
face
At 1:35 AM -0700 7/16/98, Nuno Pedrosa wrote:


> Chuck Rice wrote:
> > You might try using a diaphragm to measure the fuel. Create a small chamber
> > with a inflow port and an outflow port. Divide the chamber in half with
> > the diaphragm. One side would contain gas and the two ports, the other
> > side would be sealed and contain air and a non-contact switch (If the
> > diaphragm breaks, would not want to spark a fire).
>
> I had thought of a chamber measurement, but never thought of the
> diaphragm. That would seem a real nice idea, but...
>
> > Close the output port and allow the fuel pump to fill the gas side of
> > the chamber expanding the diaphragm.
>
> ...I can not close the output port. The fuel system is air tight. If I
> close the output port, I close the fuel to the engine and it stops.

I think that that would depend on how long it takes to pressurize the
diaphragm with respect to how long depresurization of the downstream system
takes. You could add some reserve by including another chamber downstream
with another diaphram. No valves would be needed as it would just provide
pressure during the time that the measure chamber fills, but I think that
it might not be needed. Depends on how close the system tolerances are.
If there were a problem, it would only show up at the maximum flow rate.
-Chuck-

__________________________________________________________________________
Chuck Rice                                     <.....ChuckspamRemoveMEWildRice.com>

1998\07\19@133236 by Mark Willis

flavicon
face
What would happen if you read the fuel flow after the bleedback unit
drains fuel off to regulate pressure?

 I.E. Tank --> Pump --> Regulator --> Flow Measurement --> Injectors
        ^                   |
        +-------------------+

 (I've not torn a car fuel injection system apart, just Carbs (Holly's
& Quadrajets mostly <G>) - perhaps the pressure regulator is after all
the injectors, but seems to me that if it's inline, adding a small flow
measurement body inline AFTER the return line, would give the desired
flow rate!

 Mark Willis, RemoveMEmwillisspamspamBeGonenwlink.com

Bruce Turrentine wrote:
>
> The problem with measuring fuel flow in a fuel system such as fuel injection
> is the fuel pressure regulator bleeds fuel back to the tank to control the
> pressure. Therefore, you would have to measure the fuel returning as well as
> the total flow and compute the difference.

1998\07\19@133240 by Mark Willis

flavicon
face
Chuck Rice wrote:
> > <snipped>
> I think that that would depend on how long it takes to pressurize the
> diaphragm with respect to how long depresurization of the downstream system
> takes. You could add some reserve by including another chamber downstream
> with another diaphram. No valves would be needed as it would just provide
> pressure during the time that the measure chamber fills, but I think that
> it might not be needed. Depends on how close the system tolerances are.
> If there were a problem, it would only show up at the maximum flow rate.
> -Chuck-
> __________________________________________________________________________
> Chuck Rice                                     <spamBeGoneChuck@spam@spamspam_OUTWildRice.com>

In hydraulics they use an "accumulator", consisting of a spring-loaded
piston in a cylinder;  Push fluid in, increase volume of stored fluid by
pushing the piston up; stop pumping fuel in, & you have a reserve of
pressurized fuel & can use that fuel.  (Small systems such as fuel
injection could use a diaphragm instead of a piston, so long as you
tried to make pressure fairly constant so the injectors work right!)

 You'd want a fairly constant pressure spring;  Also, a leak-free
system would be good!

 If I were completely replacing the whole system from scratch, I'd
think about a bi-directional piston pump with an accumulator, you could
eliminate the need for fuel re-circulation, I suspect.  (Same deal if
you use two or more identical pumps, in parallel.)  Stroke the piston a
known stroke, back & forth, say 60cc each direction - use a stepper
motor to move the piston - I get a "Wild Guess" of around 3.5 ml /
second for fuel consumption on an 8 Cylinder truck, at 60 MPH at 2200
RPMs, getting 18 miles per gallon - this is about the sort of mileage
you get in a "Jeep" type truck - so you could cobble this together as
"hackware" from a pair or three 60ml Horse Syringes, stepping it at any
rate from motionless up to 5 seconds for a one-way stroke should provide
plenty of fuel for a V-8.  I think (The accelerator pumps on carbs might
be harder to keep up with, I'm sure the peak flow rates on fuel
injectors go UP during heavy acceleration!  Maybe make that 1 second
full stroke, should be do-able.  Or run several {2 of 4?} pumps in
parallel if heavy fuel demand?)

 Only problems I can think of would be, "don't run out of syringes",
and "I wonder how well the seals on a standard syringe last in benzene &
the other solvents in unleaded supreme fuel?"!  {Probably want a
different pump than a medical syringe, in other words!}  Obviously a set
of 3 or 4 60-cc {or larger} pumps, cycled with one active at a time,
would WORK for precision fuel measurement, but there might be a lots
better & cheaper way (radial pump with an odd number of pistons?  I'm
NOT a fluids expert!)

 I know the medical folks use a syringe on their "PCA" pain control
apparatus & can meter a TINY dose of pain-killer with a linear stepper
this way, they can make a 5ml syringe of morphine or dilaudid or
whatever last a LONG time (my girlfriend's disabled & has had TOO MANY
surgeries, fortunately all over with, hopefully forever!)

 Just a stupid idea to be shot down as dumb <G>

 Mark Willis, TakeThisOuTmwillisspamspamnwlink.com

1998\07\19@185212 by Bruce Turrentine

picon face
Most automotive fuel injection systems have the pressure regulator mounted on
the fuel rail where the injectors are mounted. If the pressure regulator was
removed from the fuel rail and placed upstream of the injectors, it would seem
reasonable that a flow meter could be inserted between the two.
Bruce

1998\07\20@115201 by

picon face
I have not followed this thread completely, but it seems to me your making
this way to complicated. Fuel Injectors utilize "choked flow" their flow rate
is dependent on pulse time and upstream pressure. Injectors have fairly
precise flow curves. Just measure the pulse width of an injector and do some
math.

Jon

More... (looser matching)
- Last day of these posts
- In 1998 , 1999 only
- Today
- New search...