Stamp 2 and Fuel Injection
Archiver
Posts: 46,084
FYI
Fuel injection is a set of complex variables, I know as
I program and calibrate EFI systems on peoples toys
(sports cars, trucks, off-road 4x4's, etc...). At the
low RPM's as you have experienced you may have some
control capability, however the stamp does not have
enough speed to do "all the calculations" needed for
good control. At for example 6000 rpm you have 10ms to
read all sensors, gather the data from look up tables,
do all the calculations and setup a hardware timer to
control the injector on-time.
Also the data needed to map the fuel delivery is a 3-D
curved surface, at least that is what will be graphed
when you plot a Speed-Density system ( x & y data) with
fuel ( being z data ).
Included in most all EFI systems is the fact that the
data from the 3-D look up table is interpolated,
meaning the exact point in space the engine is running
is not on top of one of these data points, rather it
exists between them. Thus all nearby data points are
used to calculate the best data on the slope existing
between four fuel data points.
Thus a minimum data table consists of 16 columns x 16
rows with a byte value at each intersection. That byte
value is the fuel to be injected under the conditions
defined by the column (RPM) and the row (manifold air
density). An Alpha-N system replaces the row data with
throttle position. This means you now also have to do
barometric pressure correction
In the end the BS2 does not have enough speed to do
"all the calculations". If you wish to use a BS2 to
monitor and report the current data like RPM or
injector pulse width to an LCD that is not a problem as
the data will come at about the same speed as a human
can contemplate the data.
There are ready made fuel injection systems available
for install on engine applications with access to the
complete calibration data tables, which are adjustable
via a PC.
www.jegs.com
The above web site sells these systems by ACCEL and for
the price they are cheap compared to the effort to roll
you own system.
Before trying to roll your own, research, buying and
testing everyone else's EFI system is a good way to get
a handle on the state of the art.
I have worked for those who rolled there own EFI
system. Being involved with these projects they spent
millions of dollars in order to be successful.
Attempting to do this is a wonderful learning project,
so feel free to press forward and learn, just be aware
that what you learn may also be "the limitations" too.
Sincerely,
Ron
_______________________________________________________
_________________
>
>
> Message: 10
> Date: Mon, 06 Oct 2003 22:10:36 -0000
> From: "djrevolution99" <djrevolution99@y...>
> Subject: Is a stamp 2 quick enough?....
>
> in this program below, i have it setup to inject fuel
from a fuel
> injector that is triggered by input 15 going high
which is a magnetic
> coil pick up with a 10k resistor from the input pin
going to 5+ ...
> when i have this program running on my engine it will
idle and pull
> up to around 1500~2000 rpm's (25 ~ 33 times a second
input15 is being
> pulled high) but once i give it full thottle or
3/4''s thottle it
> will start to rev up then die and rev up and die, but
i tried fooling
> around with the pulseout command to have it open up
more to alow more
> fuel into the engine but for some reason still acts
the same, am
> starting to think that ither the stamp isnt keeping
up with magnetic
> pick up coil, or could i be over driveing the input
coil with the
> engien rev'ing up so high it pulses to much
current/voltage into the
> input15...... or is my stamp not being able to be
triggered quickley
> enoguh so it can open injector and look back to it
self once
> more .... ive heard from other people stamp 2 can
only go so fast but
> i figure if im runing my engine with in 5000~6000 rpm
range that it
> should be able to see the input pin, trigger the sub
routine, read
> the thottle postion then do the lookup command for
the correct pulse
> for the injector and loop back to its self agian....
>
> another question would be , in dealing with magnetic
pick up coils
> and interfaceing them into the stamp, i probley need
a better way to
> interface the two, in low rpm's it works fine but i
have no idea how
> to tell if the stamp is realling reading every time
the input is high
> then low agian or if its "skipping" sometimes across
the coil
> pickup....
>
> also befor i forget... on the injector, i have a NPN
power swithcing
> transistor to switch the state of the injector off/on
, i have it
> setup where the base is pluged directly into the
stamps output pin,
> the collector pluged into the ground and the emitter
pluged into the
> injector and the other wire for injector is pluged
into 12+ dc...
>
>
> hopefully someone can help me with this it sucks to
see it my
> invention work but still have bugs in it... let me
know what u guys
> think...
>
> Sean...
>
>
>
>
> '{$STAMP BS2}
> TB_POS VAR BYTE
> X VAR WORD
> x0 CON 1300
> x1 CON 1700
> x2 CON 2000
> x3 CON 3000
> x4 CON 4000
> x5 CON 4000
> x6 CON 5000
> x7 CON 6000
> x8 CON 8000
> x9 CON 10000
> x10 CON 12000
> x11 CON 13000
> x12 CON 14000
>
> MAIN:
>
> input 15
>
> IF IN15 = 1 THEN tb
> goto main
>
> TB:
>
> PULSIN 8, 1, TB_POS
> lookup TB_POS min 5 max 18 - 5,
> [noparse][[/noparse]x0,x1,x2,x3,x4,x5,x6,x7,x8,x9,x10,x11,x12], x
> pulsout 7,x
> GOTO MAIN
>
>
> **********end of programm
>
>
>
Fuel injection is a set of complex variables, I know as
I program and calibrate EFI systems on peoples toys
(sports cars, trucks, off-road 4x4's, etc...). At the
low RPM's as you have experienced you may have some
control capability, however the stamp does not have
enough speed to do "all the calculations" needed for
good control. At for example 6000 rpm you have 10ms to
read all sensors, gather the data from look up tables,
do all the calculations and setup a hardware timer to
control the injector on-time.
Also the data needed to map the fuel delivery is a 3-D
curved surface, at least that is what will be graphed
when you plot a Speed-Density system ( x & y data) with
fuel ( being z data ).
Included in most all EFI systems is the fact that the
data from the 3-D look up table is interpolated,
meaning the exact point in space the engine is running
is not on top of one of these data points, rather it
exists between them. Thus all nearby data points are
used to calculate the best data on the slope existing
between four fuel data points.
Thus a minimum data table consists of 16 columns x 16
rows with a byte value at each intersection. That byte
value is the fuel to be injected under the conditions
defined by the column (RPM) and the row (manifold air
density). An Alpha-N system replaces the row data with
throttle position. This means you now also have to do
barometric pressure correction
In the end the BS2 does not have enough speed to do
"all the calculations". If you wish to use a BS2 to
monitor and report the current data like RPM or
injector pulse width to an LCD that is not a problem as
the data will come at about the same speed as a human
can contemplate the data.
There are ready made fuel injection systems available
for install on engine applications with access to the
complete calibration data tables, which are adjustable
via a PC.
www.jegs.com
The above web site sells these systems by ACCEL and for
the price they are cheap compared to the effort to roll
you own system.
Before trying to roll your own, research, buying and
testing everyone else's EFI system is a good way to get
a handle on the state of the art.
I have worked for those who rolled there own EFI
system. Being involved with these projects they spent
millions of dollars in order to be successful.
Attempting to do this is a wonderful learning project,
so feel free to press forward and learn, just be aware
that what you learn may also be "the limitations" too.
Sincerely,
Ron
_______________________________________________________
_________________
>
>
> Message: 10
> Date: Mon, 06 Oct 2003 22:10:36 -0000
> From: "djrevolution99" <djrevolution99@y...>
> Subject: Is a stamp 2 quick enough?....
>
> in this program below, i have it setup to inject fuel
from a fuel
> injector that is triggered by input 15 going high
which is a magnetic
> coil pick up with a 10k resistor from the input pin
going to 5+ ...
> when i have this program running on my engine it will
idle and pull
> up to around 1500~2000 rpm's (25 ~ 33 times a second
input15 is being
> pulled high) but once i give it full thottle or
3/4''s thottle it
> will start to rev up then die and rev up and die, but
i tried fooling
> around with the pulseout command to have it open up
more to alow more
> fuel into the engine but for some reason still acts
the same, am
> starting to think that ither the stamp isnt keeping
up with magnetic
> pick up coil, or could i be over driveing the input
coil with the
> engien rev'ing up so high it pulses to much
current/voltage into the
> input15...... or is my stamp not being able to be
triggered quickley
> enoguh so it can open injector and look back to it
self once
> more .... ive heard from other people stamp 2 can
only go so fast but
> i figure if im runing my engine with in 5000~6000 rpm
range that it
> should be able to see the input pin, trigger the sub
routine, read
> the thottle postion then do the lookup command for
the correct pulse
> for the injector and loop back to its self agian....
>
> another question would be , in dealing with magnetic
pick up coils
> and interfaceing them into the stamp, i probley need
a better way to
> interface the two, in low rpm's it works fine but i
have no idea how
> to tell if the stamp is realling reading every time
the input is high
> then low agian or if its "skipping" sometimes across
the coil
> pickup....
>
> also befor i forget... on the injector, i have a NPN
power swithcing
> transistor to switch the state of the injector off/on
, i have it
> setup where the base is pluged directly into the
stamps output pin,
> the collector pluged into the ground and the emitter
pluged into the
> injector and the other wire for injector is pluged
into 12+ dc...
>
>
> hopefully someone can help me with this it sucks to
see it my
> invention work but still have bugs in it... let me
know what u guys
> think...
>
> Sean...
>
>
>
>
> '{$STAMP BS2}
> TB_POS VAR BYTE
> X VAR WORD
> x0 CON 1300
> x1 CON 1700
> x2 CON 2000
> x3 CON 3000
> x4 CON 4000
> x5 CON 4000
> x6 CON 5000
> x7 CON 6000
> x8 CON 8000
> x9 CON 10000
> x10 CON 12000
> x11 CON 13000
> x12 CON 14000
>
> MAIN:
>
> input 15
>
> IF IN15 = 1 THEN tb
> goto main
>
> TB:
>
> PULSIN 8, 1, TB_POS
> lookup TB_POS min 5 max 18 - 5,
> [noparse][[/noparse]x0,x1,x2,x3,x4,x5,x6,x7,x8,x9,x10,x11,x12], x
> pulsout 7,x
> GOTO MAIN
>
>
> **********end of programm
>
>
>