> anybody know about DIY ecu with basic stamp??
>
> thanks..
>
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject
> and Body of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>
ecu-engine control unit. small computer used to control air/fuel
mixture for car engine.
--- In basicstamps@y..., veewee77@a... wrote:
> ecu?
>
> What's this?
>
> Doug
>
> On 12 Nov 2001, at 15:25, rusram@p... wrote:
>
> > anybody know about DIY ecu with basic stamp??
> >
> > thanks..
> >
> >
> > To UNSUBSCRIBE, just send mail to:
> > basicstamps-unsubscribe@y...
> > from the same email address that you subscribed. Text in the
Subject
> > and Body of the message will be ignored.
> >
> >
> > Your use of Yahoo! Groups is subject to
> > http://docs.yahoo.com/info/terms/
> >
> >
I would be interested in this, but the ecu control much more than just A/F
ratio. Please keep us informed on anything you find out.
Personally, I think it would be beyond the capabilities of the stamp based
upon I/O count alone. I think you might be able to get around all the
memory locations with additional eeproms.
Anyway, this would be a very interesting project and I would like to help if
anyone wants to start on something.
ecu-engine control unit. small computer used to control air/fuel
mixture for car engine.
--- In basicstamps@y..., veewee77@a... wrote:
> ecu?
>
> What's this?
>
> Doug
>
> On 12 Nov 2001, at 15:25, rusram@p... wrote:
>
> > anybody know about DIY ecu with basic stamp??
> >
> > thanks..
> >
> >
> > To UNSUBSCRIBE, just send mail to:
> > basicstamps-unsubscribe@y...
> > from the same email address that you subscribed. Text in the
Subject
> > and Body of the message will be ignored.
> >
> >
> > Your use of Yahoo! Groups is subject to
> > http://docs.yahoo.com/info/terms/
> >
> >
To UNSUBSCRIBE, just send mail to: basicstamps-unsubscribe@yahoogroups.com
from the same email address that you subscribed. Text in the Subject and
Body of the message will be ignored.
this is not for a car, but he touch a lot of detail from his ecu
project..
http://www.5bears.com/index.htm
--- In basicstamps@y..., "Mark Adams" <madams@a...> wrote:
> I would be interested in this, but the ecu control much more than
just A/F
> ratio. Please keep us informed on anything you find out.
>
> Personally, I think it would be beyond the capabilities of the
stamp based
> upon I/O count alone. I think you might be able to get around all
the
> memory locations with additional eeproms.
>
> Anyway, this would be a very interesting project and I would like
to help if
> anyone wants to start on something.
>
> Mark
>
>
Original Message
> From: rusram@p... [noparse][[/noparse]mailto:rusram@p...]
> Sent: Monday, November 12, 2001 9:38 AM
> To: basicstamps@y...
> Subject: [noparse][[/noparse]basicstamps] Re: ecu
>
>
> ecu-engine control unit. small computer used to control air/fuel
> mixture for car engine.
>
> --- In basicstamps@y..., veewee77@a... wrote:
>
> > ecu?
> >
> > What's this?
> >
> > Doug
> >
> > On 12 Nov 2001, at 15:25, rusram@p... wrote:
> >
> > > anybody know about DIY ecu with basic stamp??
> > >
> > > thanks..
> > >
> > >
> > > To UNSUBSCRIBE, just send mail to:
> > > basicstamps-unsubscribe@y...
> > > from the same email address that you subscribed. Text in the
> Subject
> > > and Body of the message will be ignored.
> > >
> > >
> > > Your use of Yahoo! Groups is subject to
> > > http://docs.yahoo.com/info/terms/
> > >
> > >
>
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@y...
> from the same email address that you subscribed. Text in the
Subject and
> Body of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
--- In basicstamps@y..., rusram@p... wrote:
> anybody know about DIY ecu with basic stamp??
>
> thanks..
You will Have A few problems. 1st, the stamp can only do things one
at a time, while the ECU (Electronic Control Unit) can do meny things
at one time. 2nd, there are a lot of sensors that tell the ECU what
the motor is doing. 3rd, to program a fuel map and/or a Ignition map
would take months or eaven years to do. But if you got the time and
the money, I wish you good luck, and if you need any help just ask.
I think it is doable for the following reasons. ECU's do not need to
track every revolution of the engine for control purposes. Sampling
should be sufficient. Throttle position can be eliminated with close
monitoring of engine vacuum or a unique sensor fabricated that combines
the two functions. Fuel injection can be done with a singular injector
similar to early GM's. Reference maps can be dispensed with if you have
a starting value and the last learned figure for the manifold vacuum/rpm
blocks or ballparked with a formula adjusted by ox sensor readings.
Some tasks can be offloaded to dedicated circuits or devices (for
example, keep the distributor, servo advance).
aconti@n... wrote:
>
> --- In basicstamps@y..., rusram@p... wrote:
> > anybody know about DIY ecu with basic stamp??
> >
> > thanks..
>
> You will Have A few problems. 1st, the stamp can only do things one
> at a time, while the ECU (Electronic Control Unit) can do meny things
> at one time. 2nd, there are a lot of sensors that tell the ECU what
> the motor is doing. 3rd, to program a fuel map and/or a Ignition map
> would take months or eaven years to do. But if you got the time and
> the money, I wish you good luck, and if you need any help just ask.
>
> TC
> Conti Custom Dash (CCD)
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject and Body
of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
M> I think it is doable for the following reasons.
Doable yes, easy no.
M> ECU's do not need to
M> track every revolution of the engine for control purposes. Sampling
M> should be sufficient.
Injectors have to be opened in each revolution. The microcontroller itself
(inside the ECU) perhaps doesn't need to track each rev, but the complete
ECU certainly will.
M> Throttle position can be eliminated with close
M> monitoring of engine vacuum or a unique sensor fabricated that combines
M> the two functions.
The throttle position sensor on some engines is used for accelerator pump
function. Planting the throttle can lead to a lean condition if this
accelerator pump function isn't operating, however a number of economy
engine do not have this part of the TPS.
The idle switch in the TPS puts the ECU into an idle mode. You can't sense
that using a vaccuum sensor as the vaccuum changes depending on a lot of
factors, one of which is cold idle up. (Unless idle quality isn't an
issue).
M> Fuel injection can be done with a singular injector
M> similar to early GM's.
It can be, but why? If you're replacing a carby setup with injection then
fair enough. If you wish to use an EFI existing setup then there is no
advantage to change your multiple injectors into a single injector. You can
run all the injectors in parallel so electrically they will look like one
injector - from a fuel delivery viewpoint, that's a hell of a lot better
then single point injection.
Newer ECUs control each injector individually.
M> Reference maps can be dispensed with if you have
M> a starting value and the last learned figure for the manifold vacuum/rpm
M> blocks or ballparked with a formula adjusted by ox sensor readings.
Uhm.. if you don't have maps, then where will you store your last learned
figure? (I'm not completely sure what you're getting at).
It is very difficult to adjust fuel delivery based on a single narrow band
oxygen sensor on a real-time basis. OEM ECUs only use the narrow band
oxygen sensor to tweak the fuel delivery while cruising (closed loop mode).
When you apply the throttle the ECU generally goes into open loop mode and
delivers fuel based solely on the stored maps.
The narrow band of the oxygen sensor means it (usefully) indicates only a
lean or a rich condition. The ECUs that I have seen use an oscillating
approach in closed loop mode. Ox shows lean, add fuel, when ox shows rich,
remove fuel and so on.
A wide band oxygen sensor can tell you how lean or rich the engine is
running. Those sensor are very expensive and not what you find on your
typical EFI engine. AFAIK there is only one manufacturer that equips an
engine with a wide band sensor - Honda (Civic). Aftermarket support for a
wideband sensor is also very expensive (relatively speaking compared to a
BS).
M> Some tasks can be offloaded to dedicated circuits or devices (for
M> example, keep the distributor, servo advance).
Ignition advance kits are available, and would be easy to duplicate.
Most ignition advance units only base advance on engine rpm - they do not
sense the load on the engine, hence cannot correct for it. That's why it's
beneficial to have the ignition advance part built into the same device
that knows exactly what the engine is doing.
It really depends on what you want to do. If you are trying to replace your
factory ECU, and make your engine run "better", then budget a lot of blood,
sweat and tears (and time and money too). OEM's spend millions of dollars
and years developing their ECUs.
If all you are doing is trying to get an engine to run, then you can run
one with a 555 timer IC. It wont run very well, but it will run.
In such a case a BS can probably do the job quite nicely.
You could build a data logger to see what your current ECU is doing and get
an idea of what goes on.
I highly recommend you have a browse of the http://www.diy-efi.org/
website. It deals specifically with the topic of making your own ecu.
This post is not to be construed as a discouragement. I also don't mean to
be hostile, but if it comes out like that I apologise in advance.
There are many things to consider to make a good ECU and these are just
touching the tip of the iceberg. I haven't build a diy ecu yet, however
have been researching it on and off for a few years.
TC, What's your CCD about? [noparse]:)[/noparse]
Luke.
M> aconti@n... wrote:
M> >
M> > --- In basicstamps@y..., rusram@p... wrote:
M> > > anybody know about DIY ecu with basic stamp??
M> > >
M> > > thanks..
M> >
M> > You will Have A few problems. 1st, the stamp can only do things one
M> > at a time, while the ECU (Electronic Control Unit) can do meny things
M> > at one time. 2nd, there are a lot of sensors that tell the ECU what
M> > the motor is doing. 3rd, to program a fuel map and/or a Ignition map
M> > would take months or eaven years to do. But if you got the time and
M> > the money, I wish you good luck, and if you need any help just ask.
M> >
M> > TC
M> > Conti Custom Dash (CCD)
--- In basicstamps@y..., Luke Szymanski <lsz@c...> wrote:
> (Longish).
>
> On Mon, 12 Nov 2001, Don Denhardt wrote:
>
> M> I think it is doable for the following reasons.
>
> Doable yes, easy no.
>
> M> ECU's do not need to
> M> track every revolution of the engine for control purposes.
Sampling
> M> should be sufficient.
>
> Injectors have to be opened in each revolution. The microcontroller
itself
> (inside the ECU) perhaps doesn't need to track each rev, but the
complete
> ECU certainly will.
>
> M> Throttle position can be eliminated with close
> M> monitoring of engine vacuum or a unique sensor fabricated that
combines
> M> the two functions.
>
> The throttle position sensor on some engines is used for
accelerator pump
> function. Planting the throttle can lead to a lean condition if this
> accelerator pump function isn't operating, however a number of
economy
> engine do not have this part of the TPS.
>
> The idle switch in the TPS puts the ECU into an idle mode. You
can't sense
> that using a vaccuum sensor as the vaccuum changes depending on a
lot of
> factors, one of which is cold idle up. (Unless idle quality isn't an
> issue).
>
> M> Fuel injection can be done with a singular injector
> M> similar to early GM's.
>
> It can be, but why? If you're replacing a carby setup with
injection then
> fair enough. If you wish to use an EFI existing setup then there is
no
> advantage to change your multiple injectors into a single injector.
You can
> run all the injectors in parallel so electrically they will look
like one
> injector - from a fuel delivery viewpoint, that's a hell of a lot
better
> then single point injection.
> Newer ECUs control each injector individually.
>
> M> Reference maps can be dispensed with if you have
> M> a starting value and the last learned figure for the manifold
vacuum/rpm
> M> blocks or ballparked with a formula adjusted by ox sensor
readings.
>
> Uhm.. if you don't have maps, then where will you store your last
learned
> figure? (I'm not completely sure what you're getting at).
>
> It is very difficult to adjust fuel delivery based on a single
narrow band
> oxygen sensor on a real-time basis. OEM ECUs only use the narrow
band
> oxygen sensor to tweak the fuel delivery while cruising (closed
loop mode).
> When you apply the throttle the ECU generally goes into open loop
mode and
> delivers fuel based solely on the stored maps.
>
> The narrow band of the oxygen sensor means it (usefully) indicates
only a
> lean or a rich condition. The ECUs that I have seen use an
oscillating
> approach in closed loop mode. Ox shows lean, add fuel, when ox
shows rich,
> remove fuel and so on.
>
> A wide band oxygen sensor can tell you how lean or rich the engine
is
> running. Those sensor are very expensive and not what you find on
your
> typical EFI engine. AFAIK there is only one manufacturer that
equips an
> engine with a wide band sensor - Honda (Civic). Aftermarket support
for a
> wideband sensor is also very expensive (relatively speaking
compared to a
> BS).
>
> M> Some tasks can be offloaded to dedicated circuits or devices (for
> M> example, keep the distributor, servo advance).
>
> Ignition advance kits are available, and would be easy to duplicate.
> Most ignition advance units only base advance on engine rpm - they
do not
> sense the load on the engine, hence cannot correct for it. That's
why it's
> beneficial to have the ignition advance part built into the same
device
> that knows exactly what the engine is doing.
>
>
> It really depends on what you want to do. If you are trying to
replace your
> factory ECU, and make your engine run "better", then budget a lot
of blood,
> sweat and tears (and time and money too). OEM's spend millions of
dollars
> and years developing their ECUs.
>
> If all you are doing is trying to get an engine to run, then you
can run
> one with a 555 timer IC. It wont run very well, but it will run.
> In such a case a BS can probably do the job quite nicely.
>
> You could build a data logger to see what your current ECU is doing
and get
> an idea of what goes on.
>
> I highly recommend you have a browse of the http://www.diy-efi.org/
> website. It deals specifically with the topic of making your own
ecu.
>
> This post is not to be construed as a discouragement. I also don't
mean to
> be hostile, but if it comes out like that I apologise in advance.
>
> There are many things to consider to make a good ECU and these are
just
> touching the tip of the iceberg. I haven't build a diy ecu yet,
however
> have been researching it on and off for a few years.
>
> TC, What's your CCD about? [noparse]:)[/noparse]
>
> Luke.
>
>
>
> M> aconti@n... wrote:
> M> >
> M> > --- In basicstamps@y..., rusram@p... wrote:
> M> > > anybody know about DIY ecu with basic stamp??
> M> > >
> M> > > thanks..
> M> >
> M> > You will Have A few problems. 1st, the stamp can only do
things one
> M> > at a time, while the ECU (Electronic Control Unit) can do meny
things
> M> > at one time. 2nd, there are a lot of sensors that tell the ECU
what
> M> > the motor is doing. 3rd, to program a fuel map and/or a
Ignition map
> M> > would take months or eaven years to do. But if you got the
time and
> M> > the money, I wish you good luck, and if you need any help just
ask.
> M> >
> M> > TC
> M> > Conti Custom Dash (CCD)
>
> --
> '80 (ex)KE38 '86 ST141 '84 AE86 '90 ST185
Luke it appears that you lack confidence to accomplish this task you are
interested in. Go for it. It can be done. You will be successful.
No new clean sheet task is ever easy. Never made that claim.
Just that I believe it is doable. The project would call for some
creativity and perhaps some reinvention of the wheel. You do not have
to take what others have done as the standard or the only way to
accomplish a task. Never chain yourself to conventional thinking. Else
all you will do is attempt to copy what others have done.
The homemade unit has an advantage over car manufacturers requirements.
It does not have to meet any emissions standards, unless you live in a
state like California and most of them will be underwater when the big
quake hits anyway. It merely has to make the engine run smoothly while
getting reasonable mileage. OEM's spend millions of dollars and years
developing their ECUs because they do have to meet emissions
requirements. Additionally, the OEM units manage more than the engine
(air conditioning, radiator fan, etc...).
Injectors do not have to be opened by the Stamp in each revolution if
you use a singular modulated feed and merely vary the drip rate. Or if
you prefer port injection, that is one of the tasks that could be
offloaded to a dedicated circuit that maintains a set pulse width to
each injector unless instructed otherwise by the controller. The
dedicated circuit can feed the injector(s) the same pulse widths for
quite a few revolutions before needing an update. How about a master
timing pulse triggering a 555 whose resistive elements are
programmable. Remember the Controller does not have to micromanage
every revolution.
Planting the throttle will result in an incredibly low vacuum reading or
rate of change. One of the tricks I use in programming devices is to
sprinkle tests for critical real time events throughout the program. If
the program is sensitive to critical items it can respond quickly
enough.
An engine running at 6000 RPM makes 10 revolutions per second or a
revolution in 1/10th of a second. The BS2-sx will process 1000
instructions in that time frame. 6000 RPM could be construed as really
sticking your foot into it and not much finesse is needed. If you were
to assume an engine would average about 2000 RPM in normal driving
conditions the sx can process 3000 instruction per revolution. If the
controller is mostly sampling 1 out of 3 revolutions, 9,000 instructions
can be processed. You do not have to micromanage each revolution.
Reference mapping is only required for open loop operation where
environmental and engine conditions are unknown or too much of a wild
card to deal with. All engines that I've worked with while on the road
with diagnostic computer connected, never left closed loop operation
once the engine came up to temperature. Once in closed loop operation
the engine is in a predefined state. I.E. the inlet air temperature is
regulated, the oil is of a predetermined viscosity and engine rotating
friction is within a narrow band. This narrow operating band is what
makes computer control to a fine degree possible.
If you have formulas (adjusting for real time conditions by sensor
readings) the compensation for fuel quality (the fudge factor) can be
written to the eprom.
Oooops got to run.
Luke is this is something you are interested in doing, give it a shot.
There will always be plenty of help available if you run into a
stumbling block.
Take small steps - get it to start and idle first. Concern yourself with
road conditions later. You'll have a lot of fun.
Good luck,
Don
Luke Szymanski wrote:
>
> (Longish).
>
> On Mon, 12 Nov 2001, Don Denhardt wrote:
>
> M> I think it is doable for the following reasons.
>
> Doable yes, easy no.
>
> M> ECU's do not need to
> M> track every revolution of the engine for control purposes. Sampling
> M> should be sufficient.
>
> Injectors have to be opened in each revolution. The microcontroller itself
> (inside the ECU) perhaps doesn't need to track each rev, but the complete
> ECU certainly will.
>
> M> Throttle position can be eliminated with close
> M> monitoring of engine vacuum or a unique sensor fabricated that combines
> M> the two functions.
>
> The throttle position sensor on some engines is used for accelerator pump
> function. Planting the throttle can lead to a lean condition if this
> accelerator pump function isn't operating, however a number of economy
> engine do not have this part of the TPS.
>
> The idle switch in the TPS puts the ECU into an idle mode. You can't sense
> that using a vaccuum sensor as the vaccuum changes depending on a lot of
> factors, one of which is cold idle up. (Unless idle quality isn't an
> issue).
>
> M> Fuel injection can be done with a singular injector
> M> similar to early GM's.
>
> It can be, but why? If you're replacing a carby setup with injection then
> fair enough. If you wish to use an EFI existing setup then there is no
> advantage to change your multiple injectors into a single injector. You can
> run all the injectors in parallel so electrically they will look like one
> injector - from a fuel delivery viewpoint, that's a hell of a lot better
> then single point injection.
> Newer ECUs control each injector individually.
>
> M> Reference maps can be dispensed with if you have
> M> a starting value and the last learned figure for the manifold vacuum/rpm
> M> blocks or ballparked with a formula adjusted by ox sensor readings.
>
> Uhm.. if you don't have maps, then where will you store your last learned
> figure? (I'm not completely sure what you're getting at).
>
> It is very difficult to adjust fuel delivery based on a single narrow band
> oxygen sensor on a real-time basis. OEM ECUs only use the narrow band
> oxygen sensor to tweak the fuel delivery while cruising (closed loop mode).
> When you apply the throttle the ECU generally goes into open loop mode and
> delivers fuel based solely on the stored maps.
>
> The narrow band of the oxygen sensor means it (usefully) indicates only a
> lean or a rich condition. The ECUs that I have seen use an oscillating
> approach in closed loop mode. Ox shows lean, add fuel, when ox shows rich,
> remove fuel and so on.
>
> A wide band oxygen sensor can tell you how lean or rich the engine is
> running. Those sensor are very expensive and not what you find on your
> typical EFI engine. AFAIK there is only one manufacturer that equips an
> engine with a wide band sensor - Honda (Civic). Aftermarket support for a
> wideband sensor is also very expensive (relatively speaking compared to a
> BS).
>
> M> Some tasks can be offloaded to dedicated circuits or devices (for
> M> example, keep the distributor, servo advance).
>
> Ignition advance kits are available, and would be easy to duplicate.
> Most ignition advance units only base advance on engine rpm - they do not
> sense the load on the engine, hence cannot correct for it. That's why it's
> beneficial to have the ignition advance part built into the same device
> that knows exactly what the engine is doing.
>
> It really depends on what you want to do. If you are trying to replace your
> factory ECU, and make your engine run "better", then budget a lot of blood,
> sweat and tears (and time and money too). OEM's spend millions of dollars
> and years developing their ECUs.
>
> If all you are doing is trying to get an engine to run, then you can run
> one with a 555 timer IC. It wont run very well, but it will run.
> In such a case a BS can probably do the job quite nicely.
>
> You could build a data logger to see what your current ECU is doing and get
> an idea of what goes on.
>
> I highly recommend you have a browse of the http://www.diy-efi.org/
> website. It deals specifically with the topic of making your own ecu.
>
> This post is not to be construed as a discouragement. I also don't mean to
> be hostile, but if it comes out like that I apologise in advance.
>
> There are many things to consider to make a good ECU and these are just
> touching the tip of the iceberg. I haven't build a diy ecu yet, however
> have been researching it on and off for a few years.
>
> TC, What's your CCD about? [noparse]:)[/noparse]
>
> Luke.
>
> M> aconti@n... wrote:
> M> >
> M> > --- In basicstamps@y..., rusram@p... wrote:
> M> > > anybody know about DIY ecu with basic stamp??
> M> > >
> M> > > thanks..
> M> >
> M> > You will Have A few problems. 1st, the stamp can only do things one
> M> > at a time, while the ECU (Electronic Control Unit) can do meny things
> M> > at one time. 2nd, there are a lot of sensors that tell the ECU what
> M> > the motor is doing. 3rd, to program a fuel map and/or a Ignition map
> M> > would take months or eaven years to do. But if you got the time and
> M> > the money, I wish you good luck, and if you need any help just ask.
> M> >
> M> > TC
> M> > Conti Custom Dash (CCD)
>
> --
> '80 (ex)KE38 '86 ST141 '84 AE86 '90 ST185
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject and Body
of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
Oooops was sleepy and math was a tad off [noparse]:)[/noparse]
After a short nap I remember my name clearly and that an engine running
at 6000 RPM effectively has an ignition speed of 3000 cycles per minute
or one cycle in 1/50th of a second. the BS2sx will process 200
instructions in that time frame. At 1000 cycles per minute (2000RPM)
the sx can process 600 instructions. If the controller is sampling 1
out of 3 cycles 1800 instructions can be processed.
Don Denhardt wrote:
>
> Luke it appears that you lack confidence to accomplish this task you are
> interested in. Go for it. It can be done. You will be successful.
>
> No new clean sheet task is ever easy. Never made that claim.
>
> Just that I believe it is doable. The project would call for some
> creativity and perhaps some reinvention of the wheel. You do not have
> to take what others have done as the standard or the only way to
> accomplish a task. Never chain yourself to conventional thinking. Else
> all you will do is attempt to copy what others have done.
>
> The homemade unit has an advantage over car manufacturers requirements.
> It does not have to meet any emissions standards, unless you live in a
> state like California and most of them will be underwater when the big
> quake hits anyway. It merely has to make the engine run smoothly while
> getting reasonable mileage. OEM's spend millions of dollars and years
> developing their ECUs because they do have to meet emissions
> requirements. Additionally, the OEM units manage more than the engine
> (air conditioning, radiator fan, etc...).
>
> Injectors do not have to be opened by the Stamp in each revolution if
> you use a singular modulated feed and merely vary the drip rate. Or if
> you prefer port injection, that is one of the tasks that could be
> offloaded to a dedicated circuit that maintains a set pulse width to
> each injector unless instructed otherwise by the controller. The
> dedicated circuit can feed the injector(s) the same pulse widths for
> quite a few revolutions before needing an update. How about a master
> timing pulse triggering a 555 whose resistive elements are
> programmable. Remember the Controller does not have to micromanage
> every revolution.
>
> Planting the throttle will result in an incredibly low vacuum reading or
> rate of change. One of the tricks I use in programming devices is to
> sprinkle tests for critical real time events throughout the program. If
> the program is sensitive to critical items it can respond quickly
> enough.
>
> An engine running at 6000 RPM makes 10 revolutions per second or a
> revolution in 1/10th of a second. The BS2-sx will process 1000
> instructions in that time frame. 6000 RPM could be construed as really
> sticking your foot into it and not much finesse is needed. If you were
> to assume an engine would average about 2000 RPM in normal driving
> conditions the sx can process 3000 instruction per revolution. If the
> controller is mostly sampling 1 out of 3 revolutions, 9,000 instructions
> can be processed. You do not have to micromanage each revolution.
>
> Reference mapping is only required for open loop operation where
> environmental and engine conditions are unknown or too much of a wild
> card to deal with. All engines that I've worked with while on the road
> with diagnostic computer connected, never left closed loop operation
> once the engine came up to temperature. Once in closed loop operation
> the engine is in a predefined state. I.E. the inlet air temperature is
> regulated, the oil is of a predetermined viscosity and engine rotating
> friction is within a narrow band. This narrow operating band is what
> makes computer control to a fine degree possible.
>
> If you have formulas (adjusting for real time conditions by sensor
> readings) the compensation for fuel quality (the fudge factor) can be
> written to the eprom.
>
> Oooops got to run.
>
> Luke is this is something you are interested in doing, give it a shot.
> There will always be plenty of help available if you run into a
> stumbling block.
>
> Take small steps - get it to start and idle first. Concern yourself with
> road conditions later. You'll have a lot of fun.
>
> Good luck,
>
> Don
>
> Luke Szymanski wrote:
> >
> > (Longish).
> >
> > On Mon, 12 Nov 2001, Don Denhardt wrote:
> >
> > M> I think it is doable for the following reasons.
> >
> > Doable yes, easy no.
> >
> > M> ECU's do not need to
> > M> track every revolution of the engine for control purposes. Sampling
> > M> should be sufficient.
> >
> > Injectors have to be opened in each revolution. The microcontroller itself
> > (inside the ECU) perhaps doesn't need to track each rev, but the complete
> > ECU certainly will.
> >
> > M> Throttle position can be eliminated with close
> > M> monitoring of engine vacuum or a unique sensor fabricated that combines
> > M> the two functions.
> >
> > The throttle position sensor on some engines is used for accelerator pump
> > function. Planting the throttle can lead to a lean condition if this
> > accelerator pump function isn't operating, however a number of economy
> > engine do not have this part of the TPS.
> >
> > The idle switch in the TPS puts the ECU into an idle mode. You can't sense
> > that using a vaccuum sensor as the vaccuum changes depending on a lot of
> > factors, one of which is cold idle up. (Unless idle quality isn't an
> > issue).
> >
> > M> Fuel injection can be done with a singular injector
> > M> similar to early GM's.
> >
> > It can be, but why? If you're replacing a carby setup with injection then
> > fair enough. If you wish to use an EFI existing setup then there is no
> > advantage to change your multiple injectors into a single injector. You can
> > run all the injectors in parallel so electrically they will look like one
> > injector - from a fuel delivery viewpoint, that's a hell of a lot better
> > then single point injection.
> > Newer ECUs control each injector individually.
> >
> > M> Reference maps can be dispensed with if you have
> > M> a starting value and the last learned figure for the manifold vacuum/rpm
> > M> blocks or ballparked with a formula adjusted by ox sensor readings.
> >
> > Uhm.. if you don't have maps, then where will you store your last learned
> > figure? (I'm not completely sure what you're getting at).
> >
> > It is very difficult to adjust fuel delivery based on a single narrow band
> > oxygen sensor on a real-time basis. OEM ECUs only use the narrow band
> > oxygen sensor to tweak the fuel delivery while cruising (closed loop mode).
> > When you apply the throttle the ECU generally goes into open loop mode and
> > delivers fuel based solely on the stored maps.
> >
> > The narrow band of the oxygen sensor means it (usefully) indicates only a
> > lean or a rich condition. The ECUs that I have seen use an oscillating
> > approach in closed loop mode. Ox shows lean, add fuel, when ox shows rich,
> > remove fuel and so on.
> >
> > A wide band oxygen sensor can tell you how lean or rich the engine is
> > running. Those sensor are very expensive and not what you find on your
> > typical EFI engine. AFAIK there is only one manufacturer that equips an
> > engine with a wide band sensor - Honda (Civic). Aftermarket support for a
> > wideband sensor is also very expensive (relatively speaking compared to a
> > BS).
> >
> > M> Some tasks can be offloaded to dedicated circuits or devices (for
> > M> example, keep the distributor, servo advance).
> >
> > Ignition advance kits are available, and would be easy to duplicate.
> > Most ignition advance units only base advance on engine rpm - they do not
> > sense the load on the engine, hence cannot correct for it. That's why it's
> > beneficial to have the ignition advance part built into the same device
> > that knows exactly what the engine is doing.
> >
> > It really depends on what you want to do. If you are trying to replace your
> > factory ECU, and make your engine run "better", then budget a lot of blood,
> > sweat and tears (and time and money too). OEM's spend millions of dollars
> > and years developing their ECUs.
> >
> > If all you are doing is trying to get an engine to run, then you can run
> > one with a 555 timer IC. It wont run very well, but it will run.
> > In such a case a BS can probably do the job quite nicely.
> >
> > You could build a data logger to see what your current ECU is doing and get
> > an idea of what goes on.
> >
> > I highly recommend you have a browse of the http://www.diy-efi.org/
> > website. It deals specifically with the topic of making your own ecu.
> >
> > This post is not to be construed as a discouragement. I also don't mean to
> > be hostile, but if it comes out like that I apologise in advance.
> >
> > There are many things to consider to make a good ECU and these are just
> > touching the tip of the iceberg. I haven't build a diy ecu yet, however
> > have been researching it on and off for a few years.
> >
> > TC, What's your CCD about? [noparse]:)[/noparse]
> >
> > Luke.
> >
> > M> aconti@n... wrote:
> > M> >
> > M> > --- In basicstamps@y..., rusram@p... wrote:
> > M> > > anybody know about DIY ecu with basic stamp??
> > M> > >
> > M> > > thanks..
> > M> >
> > M> > You will Have A few problems. 1st, the stamp can only do things one
> > M> > at a time, while the ECU (Electronic Control Unit) can do meny things
> > M> > at one time. 2nd, there are a lot of sensors that tell the ECU what
> > M> > the motor is doing. 3rd, to program a fuel map and/or a Ignition map
> > M> > would take months or eaven years to do. But if you got the time and
> > M> > the money, I wish you good luck, and if you need any help just ask.
> > M> >
> > M> > TC
> > M> > Conti Custom Dash (CCD)
> >
> > --
> > '80 (ex)KE38 '86 ST141 '84 AE86 '90 ST185
> >
> > To UNSUBSCRIBE, just send mail to:
> > basicstamps-unsubscribe@yahoogroups.com
> > from the same email address that you subscribed. Text in the Subject and
Body of the message will be ignored.
> >
> >
> > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
> _________________________________________________________
> Do You Yahoo!?
> Get your free @yahoo.com address at http://mail.yahoo.com
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject and Body
of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
Archiver
Posts: 46,084
Comments
What's this?
Doug
On 12 Nov 2001, at 15:25, rusram@p... wrote:
> anybody know about DIY ecu with basic stamp??
>
> thanks..
>
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject
> and Body of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>
mixture for car engine.
--- In basicstamps@y..., veewee77@a... wrote:
> ecu?
>
> What's this?
>
> Doug
>
> On 12 Nov 2001, at 15:25, rusram@p... wrote:
>
> > anybody know about DIY ecu with basic stamp??
> >
> > thanks..
> >
> >
> > To UNSUBSCRIBE, just send mail to:
> > basicstamps-unsubscribe@y...
> > from the same email address that you subscribed. Text in the
Subject
> > and Body of the message will be ignored.
> >
> >
> > Your use of Yahoo! Groups is subject to
> > http://docs.yahoo.com/info/terms/
> >
> >
ratio. Please keep us informed on anything you find out.
Personally, I think it would be beyond the capabilities of the stamp based
upon I/O count alone. I think you might be able to get around all the
memory locations with additional eeproms.
Anyway, this would be a very interesting project and I would like to help if
anyone wants to start on something.
Mark
Original Message
From: rusram@p... [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=MqZEcJ5kHODGORxSNFVrEIH8nDBPjbPAwdrkZc815oYVXiI3fDX4MXHZiaqyl4rJw3PxhPE9__nWG1M]rusram@p...[/url
Sent: Monday, November 12, 2001 9:38 AM
To: basicstamps@yahoogroups.com
Subject: [noparse][[/noparse]basicstamps] Re: ecu
ecu-engine control unit. small computer used to control air/fuel
mixture for car engine.
--- In basicstamps@y..., veewee77@a... wrote:
> ecu?
>
> What's this?
>
> Doug
>
> On 12 Nov 2001, at 15:25, rusram@p... wrote:
>
> > anybody know about DIY ecu with basic stamp??
> >
> > thanks..
> >
> >
> > To UNSUBSCRIBE, just send mail to:
> > basicstamps-unsubscribe@y...
> > from the same email address that you subscribed. Text in the
Subject
> > and Body of the message will be ignored.
> >
> >
> > Your use of Yahoo! Groups is subject to
> > http://docs.yahoo.com/info/terms/
> >
> >
To UNSUBSCRIBE, just send mail to:
basicstamps-unsubscribe@yahoogroups.com
from the same email address that you subscribed. Text in the Subject and
Body of the message will be ignored.
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
project..
http://www.5bears.com/index.htm
--- In basicstamps@y..., "Mark Adams" <madams@a...> wrote:
> I would be interested in this, but the ecu control much more than
just A/F
> ratio. Please keep us informed on anything you find out.
>
> Personally, I think it would be beyond the capabilities of the
stamp based
> upon I/O count alone. I think you might be able to get around all
the
> memory locations with additional eeproms.
>
> Anyway, this would be a very interesting project and I would like
to help if
> anyone wants to start on something.
>
> Mark
>
>
Original Message
> From: rusram@p... [noparse][[/noparse]mailto:rusram@p...]
> Sent: Monday, November 12, 2001 9:38 AM
> To: basicstamps@y...
> Subject: [noparse][[/noparse]basicstamps] Re: ecu
>
>
> ecu-engine control unit. small computer used to control air/fuel
> mixture for car engine.
>
> --- In basicstamps@y..., veewee77@a... wrote:
>
> > ecu?
> >
> > What's this?
> >
> > Doug
> >
> > On 12 Nov 2001, at 15:25, rusram@p... wrote:
> >
> > > anybody know about DIY ecu with basic stamp??
> > >
> > > thanks..
> > >
> > >
> > > To UNSUBSCRIBE, just send mail to:
> > > basicstamps-unsubscribe@y...
> > > from the same email address that you subscribed. Text in the
> Subject
> > > and Body of the message will be ignored.
> > >
> > >
> > > Your use of Yahoo! Groups is subject to
> > > http://docs.yahoo.com/info/terms/
> > >
> > >
>
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@y...
> from the same email address that you subscribed. Text in the
Subject and
> Body of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to
http://docs.yahoo.com/info/terms/
> anybody know about DIY ecu with basic stamp??
>
> thanks..
You will Have A few problems. 1st, the stamp can only do things one
at a time, while the ECU (Electronic Control Unit) can do meny things
at one time. 2nd, there are a lot of sensors that tell the ECU what
the motor is doing. 3rd, to program a fuel map and/or a Ignition map
would take months or eaven years to do. But if you got the time and
the money, I wish you good luck, and if you need any help just ask.
TC
Conti Custom Dash (CCD)
track every revolution of the engine for control purposes. Sampling
should be sufficient. Throttle position can be eliminated with close
monitoring of engine vacuum or a unique sensor fabricated that combines
the two functions. Fuel injection can be done with a singular injector
similar to early GM's. Reference maps can be dispensed with if you have
a starting value and the last learned figure for the manifold vacuum/rpm
blocks or ballparked with a formula adjusted by ox sensor readings.
Some tasks can be offloaded to dedicated circuits or devices (for
example, keep the distributor, servo advance).
aconti@n... wrote:
>
> --- In basicstamps@y..., rusram@p... wrote:
> > anybody know about DIY ecu with basic stamp??
> >
> > thanks..
>
> You will Have A few problems. 1st, the stamp can only do things one
> at a time, while the ECU (Electronic Control Unit) can do meny things
> at one time. 2nd, there are a lot of sensors that tell the ECU what
> the motor is doing. 3rd, to program a fuel map and/or a Ignition map
> would take months or eaven years to do. But if you got the time and
> the money, I wish you good luck, and if you need any help just ask.
>
> TC
> Conti Custom Dash (CCD)
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject and Body
of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
On Mon, 12 Nov 2001, Don Denhardt wrote:
M> I think it is doable for the following reasons.
Doable yes, easy no.
M> ECU's do not need to
M> track every revolution of the engine for control purposes. Sampling
M> should be sufficient.
Injectors have to be opened in each revolution. The microcontroller itself
(inside the ECU) perhaps doesn't need to track each rev, but the complete
ECU certainly will.
M> Throttle position can be eliminated with close
M> monitoring of engine vacuum or a unique sensor fabricated that combines
M> the two functions.
The throttle position sensor on some engines is used for accelerator pump
function. Planting the throttle can lead to a lean condition if this
accelerator pump function isn't operating, however a number of economy
engine do not have this part of the TPS.
The idle switch in the TPS puts the ECU into an idle mode. You can't sense
that using a vaccuum sensor as the vaccuum changes depending on a lot of
factors, one of which is cold idle up. (Unless idle quality isn't an
issue).
M> Fuel injection can be done with a singular injector
M> similar to early GM's.
It can be, but why? If you're replacing a carby setup with injection then
fair enough. If you wish to use an EFI existing setup then there is no
advantage to change your multiple injectors into a single injector. You can
run all the injectors in parallel so electrically they will look like one
injector - from a fuel delivery viewpoint, that's a hell of a lot better
then single point injection.
Newer ECUs control each injector individually.
M> Reference maps can be dispensed with if you have
M> a starting value and the last learned figure for the manifold vacuum/rpm
M> blocks or ballparked with a formula adjusted by ox sensor readings.
Uhm.. if you don't have maps, then where will you store your last learned
figure? (I'm not completely sure what you're getting at).
It is very difficult to adjust fuel delivery based on a single narrow band
oxygen sensor on a real-time basis. OEM ECUs only use the narrow band
oxygen sensor to tweak the fuel delivery while cruising (closed loop mode).
When you apply the throttle the ECU generally goes into open loop mode and
delivers fuel based solely on the stored maps.
The narrow band of the oxygen sensor means it (usefully) indicates only a
lean or a rich condition. The ECUs that I have seen use an oscillating
approach in closed loop mode. Ox shows lean, add fuel, when ox shows rich,
remove fuel and so on.
A wide band oxygen sensor can tell you how lean or rich the engine is
running. Those sensor are very expensive and not what you find on your
typical EFI engine. AFAIK there is only one manufacturer that equips an
engine with a wide band sensor - Honda (Civic). Aftermarket support for a
wideband sensor is also very expensive (relatively speaking compared to a
BS).
M> Some tasks can be offloaded to dedicated circuits or devices (for
M> example, keep the distributor, servo advance).
Ignition advance kits are available, and would be easy to duplicate.
Most ignition advance units only base advance on engine rpm - they do not
sense the load on the engine, hence cannot correct for it. That's why it's
beneficial to have the ignition advance part built into the same device
that knows exactly what the engine is doing.
It really depends on what you want to do. If you are trying to replace your
factory ECU, and make your engine run "better", then budget a lot of blood,
sweat and tears (and time and money too). OEM's spend millions of dollars
and years developing their ECUs.
If all you are doing is trying to get an engine to run, then you can run
one with a 555 timer IC. It wont run very well, but it will run.
In such a case a BS can probably do the job quite nicely.
You could build a data logger to see what your current ECU is doing and get
an idea of what goes on.
I highly recommend you have a browse of the http://www.diy-efi.org/
website. It deals specifically with the topic of making your own ecu.
This post is not to be construed as a discouragement. I also don't mean to
be hostile, but if it comes out like that I apologise in advance.
There are many things to consider to make a good ECU and these are just
touching the tip of the iceberg. I haven't build a diy ecu yet, however
have been researching it on and off for a few years.
TC, What's your CCD about? [noparse]:)[/noparse]
Luke.
M> aconti@n... wrote:
M> >
M> > --- In basicstamps@y..., rusram@p... wrote:
M> > > anybody know about DIY ecu with basic stamp??
M> > >
M> > > thanks..
M> >
M> > You will Have A few problems. 1st, the stamp can only do things one
M> > at a time, while the ECU (Electronic Control Unit) can do meny things
M> > at one time. 2nd, there are a lot of sensors that tell the ECU what
M> > the motor is doing. 3rd, to program a fuel map and/or a Ignition map
M> > would take months or eaven years to do. But if you got the time and
M> > the money, I wish you good luck, and if you need any help just ask.
M> >
M> > TC
M> > Conti Custom Dash (CCD)
--
'80 (ex)KE38 '86 ST141 '84 AE86 '90 ST185
--- In basicstamps@y..., Luke Szymanski <lsz@c...> wrote:
> (Longish).
>
> On Mon, 12 Nov 2001, Don Denhardt wrote:
>
> M> I think it is doable for the following reasons.
>
> Doable yes, easy no.
>
> M> ECU's do not need to
> M> track every revolution of the engine for control purposes.
Sampling
> M> should be sufficient.
>
> Injectors have to be opened in each revolution. The microcontroller
itself
> (inside the ECU) perhaps doesn't need to track each rev, but the
complete
> ECU certainly will.
>
> M> Throttle position can be eliminated with close
> M> monitoring of engine vacuum or a unique sensor fabricated that
combines
> M> the two functions.
>
> The throttle position sensor on some engines is used for
accelerator pump
> function. Planting the throttle can lead to a lean condition if this
> accelerator pump function isn't operating, however a number of
economy
> engine do not have this part of the TPS.
>
> The idle switch in the TPS puts the ECU into an idle mode. You
can't sense
> that using a vaccuum sensor as the vaccuum changes depending on a
lot of
> factors, one of which is cold idle up. (Unless idle quality isn't an
> issue).
>
> M> Fuel injection can be done with a singular injector
> M> similar to early GM's.
>
> It can be, but why? If you're replacing a carby setup with
injection then
> fair enough. If you wish to use an EFI existing setup then there is
no
> advantage to change your multiple injectors into a single injector.
You can
> run all the injectors in parallel so electrically they will look
like one
> injector - from a fuel delivery viewpoint, that's a hell of a lot
better
> then single point injection.
> Newer ECUs control each injector individually.
>
> M> Reference maps can be dispensed with if you have
> M> a starting value and the last learned figure for the manifold
vacuum/rpm
> M> blocks or ballparked with a formula adjusted by ox sensor
readings.
>
> Uhm.. if you don't have maps, then where will you store your last
learned
> figure? (I'm not completely sure what you're getting at).
>
> It is very difficult to adjust fuel delivery based on a single
narrow band
> oxygen sensor on a real-time basis. OEM ECUs only use the narrow
band
> oxygen sensor to tweak the fuel delivery while cruising (closed
loop mode).
> When you apply the throttle the ECU generally goes into open loop
mode and
> delivers fuel based solely on the stored maps.
>
> The narrow band of the oxygen sensor means it (usefully) indicates
only a
> lean or a rich condition. The ECUs that I have seen use an
oscillating
> approach in closed loop mode. Ox shows lean, add fuel, when ox
shows rich,
> remove fuel and so on.
>
> A wide band oxygen sensor can tell you how lean or rich the engine
is
> running. Those sensor are very expensive and not what you find on
your
> typical EFI engine. AFAIK there is only one manufacturer that
equips an
> engine with a wide band sensor - Honda (Civic). Aftermarket support
for a
> wideband sensor is also very expensive (relatively speaking
compared to a
> BS).
>
> M> Some tasks can be offloaded to dedicated circuits or devices (for
> M> example, keep the distributor, servo advance).
>
> Ignition advance kits are available, and would be easy to duplicate.
> Most ignition advance units only base advance on engine rpm - they
do not
> sense the load on the engine, hence cannot correct for it. That's
why it's
> beneficial to have the ignition advance part built into the same
device
> that knows exactly what the engine is doing.
>
>
> It really depends on what you want to do. If you are trying to
replace your
> factory ECU, and make your engine run "better", then budget a lot
of blood,
> sweat and tears (and time and money too). OEM's spend millions of
dollars
> and years developing their ECUs.
>
> If all you are doing is trying to get an engine to run, then you
can run
> one with a 555 timer IC. It wont run very well, but it will run.
> In such a case a BS can probably do the job quite nicely.
>
> You could build a data logger to see what your current ECU is doing
and get
> an idea of what goes on.
>
> I highly recommend you have a browse of the http://www.diy-efi.org/
> website. It deals specifically with the topic of making your own
ecu.
>
> This post is not to be construed as a discouragement. I also don't
mean to
> be hostile, but if it comes out like that I apologise in advance.
>
> There are many things to consider to make a good ECU and these are
just
> touching the tip of the iceberg. I haven't build a diy ecu yet,
however
> have been researching it on and off for a few years.
>
> TC, What's your CCD about? [noparse]:)[/noparse]
>
> Luke.
>
>
>
> M> aconti@n... wrote:
> M> >
> M> > --- In basicstamps@y..., rusram@p... wrote:
> M> > > anybody know about DIY ecu with basic stamp??
> M> > >
> M> > > thanks..
> M> >
> M> > You will Have A few problems. 1st, the stamp can only do
things one
> M> > at a time, while the ECU (Electronic Control Unit) can do meny
things
> M> > at one time. 2nd, there are a lot of sensors that tell the ECU
what
> M> > the motor is doing. 3rd, to program a fuel map and/or a
Ignition map
> M> > would take months or eaven years to do. But if you got the
time and
> M> > the money, I wish you good luck, and if you need any help just
ask.
> M> >
> M> > TC
> M> > Conti Custom Dash (CCD)
>
> --
> '80 (ex)KE38 '86 ST141 '84 AE86 '90 ST185
interested in. Go for it. It can be done. You will be successful.
No new clean sheet task is ever easy. Never made that claim.
Just that I believe it is doable. The project would call for some
creativity and perhaps some reinvention of the wheel. You do not have
to take what others have done as the standard or the only way to
accomplish a task. Never chain yourself to conventional thinking. Else
all you will do is attempt to copy what others have done.
The homemade unit has an advantage over car manufacturers requirements.
It does not have to meet any emissions standards, unless you live in a
state like California and most of them will be underwater when the big
quake hits anyway. It merely has to make the engine run smoothly while
getting reasonable mileage. OEM's spend millions of dollars and years
developing their ECUs because they do have to meet emissions
requirements. Additionally, the OEM units manage more than the engine
(air conditioning, radiator fan, etc...).
Injectors do not have to be opened by the Stamp in each revolution if
you use a singular modulated feed and merely vary the drip rate. Or if
you prefer port injection, that is one of the tasks that could be
offloaded to a dedicated circuit that maintains a set pulse width to
each injector unless instructed otherwise by the controller. The
dedicated circuit can feed the injector(s) the same pulse widths for
quite a few revolutions before needing an update. How about a master
timing pulse triggering a 555 whose resistive elements are
programmable. Remember the Controller does not have to micromanage
every revolution.
Planting the throttle will result in an incredibly low vacuum reading or
rate of change. One of the tricks I use in programming devices is to
sprinkle tests for critical real time events throughout the program. If
the program is sensitive to critical items it can respond quickly
enough.
An engine running at 6000 RPM makes 10 revolutions per second or a
revolution in 1/10th of a second. The BS2-sx will process 1000
instructions in that time frame. 6000 RPM could be construed as really
sticking your foot into it and not much finesse is needed. If you were
to assume an engine would average about 2000 RPM in normal driving
conditions the sx can process 3000 instruction per revolution. If the
controller is mostly sampling 1 out of 3 revolutions, 9,000 instructions
can be processed. You do not have to micromanage each revolution.
Reference mapping is only required for open loop operation where
environmental and engine conditions are unknown or too much of a wild
card to deal with. All engines that I've worked with while on the road
with diagnostic computer connected, never left closed loop operation
once the engine came up to temperature. Once in closed loop operation
the engine is in a predefined state. I.E. the inlet air temperature is
regulated, the oil is of a predetermined viscosity and engine rotating
friction is within a narrow band. This narrow operating band is what
makes computer control to a fine degree possible.
If you have formulas (adjusting for real time conditions by sensor
readings) the compensation for fuel quality (the fudge factor) can be
written to the eprom.
Oooops got to run.
Luke is this is something you are interested in doing, give it a shot.
There will always be plenty of help available if you run into a
stumbling block.
Take small steps - get it to start and idle first. Concern yourself with
road conditions later. You'll have a lot of fun.
Good luck,
Don
Luke Szymanski wrote:
>
> (Longish).
>
> On Mon, 12 Nov 2001, Don Denhardt wrote:
>
> M> I think it is doable for the following reasons.
>
> Doable yes, easy no.
>
> M> ECU's do not need to
> M> track every revolution of the engine for control purposes. Sampling
> M> should be sufficient.
>
> Injectors have to be opened in each revolution. The microcontroller itself
> (inside the ECU) perhaps doesn't need to track each rev, but the complete
> ECU certainly will.
>
> M> Throttle position can be eliminated with close
> M> monitoring of engine vacuum or a unique sensor fabricated that combines
> M> the two functions.
>
> The throttle position sensor on some engines is used for accelerator pump
> function. Planting the throttle can lead to a lean condition if this
> accelerator pump function isn't operating, however a number of economy
> engine do not have this part of the TPS.
>
> The idle switch in the TPS puts the ECU into an idle mode. You can't sense
> that using a vaccuum sensor as the vaccuum changes depending on a lot of
> factors, one of which is cold idle up. (Unless idle quality isn't an
> issue).
>
> M> Fuel injection can be done with a singular injector
> M> similar to early GM's.
>
> It can be, but why? If you're replacing a carby setup with injection then
> fair enough. If you wish to use an EFI existing setup then there is no
> advantage to change your multiple injectors into a single injector. You can
> run all the injectors in parallel so electrically they will look like one
> injector - from a fuel delivery viewpoint, that's a hell of a lot better
> then single point injection.
> Newer ECUs control each injector individually.
>
> M> Reference maps can be dispensed with if you have
> M> a starting value and the last learned figure for the manifold vacuum/rpm
> M> blocks or ballparked with a formula adjusted by ox sensor readings.
>
> Uhm.. if you don't have maps, then where will you store your last learned
> figure? (I'm not completely sure what you're getting at).
>
> It is very difficult to adjust fuel delivery based on a single narrow band
> oxygen sensor on a real-time basis. OEM ECUs only use the narrow band
> oxygen sensor to tweak the fuel delivery while cruising (closed loop mode).
> When you apply the throttle the ECU generally goes into open loop mode and
> delivers fuel based solely on the stored maps.
>
> The narrow band of the oxygen sensor means it (usefully) indicates only a
> lean or a rich condition. The ECUs that I have seen use an oscillating
> approach in closed loop mode. Ox shows lean, add fuel, when ox shows rich,
> remove fuel and so on.
>
> A wide band oxygen sensor can tell you how lean or rich the engine is
> running. Those sensor are very expensive and not what you find on your
> typical EFI engine. AFAIK there is only one manufacturer that equips an
> engine with a wide band sensor - Honda (Civic). Aftermarket support for a
> wideband sensor is also very expensive (relatively speaking compared to a
> BS).
>
> M> Some tasks can be offloaded to dedicated circuits or devices (for
> M> example, keep the distributor, servo advance).
>
> Ignition advance kits are available, and would be easy to duplicate.
> Most ignition advance units only base advance on engine rpm - they do not
> sense the load on the engine, hence cannot correct for it. That's why it's
> beneficial to have the ignition advance part built into the same device
> that knows exactly what the engine is doing.
>
> It really depends on what you want to do. If you are trying to replace your
> factory ECU, and make your engine run "better", then budget a lot of blood,
> sweat and tears (and time and money too). OEM's spend millions of dollars
> and years developing their ECUs.
>
> If all you are doing is trying to get an engine to run, then you can run
> one with a 555 timer IC. It wont run very well, but it will run.
> In such a case a BS can probably do the job quite nicely.
>
> You could build a data logger to see what your current ECU is doing and get
> an idea of what goes on.
>
> I highly recommend you have a browse of the http://www.diy-efi.org/
> website. It deals specifically with the topic of making your own ecu.
>
> This post is not to be construed as a discouragement. I also don't mean to
> be hostile, but if it comes out like that I apologise in advance.
>
> There are many things to consider to make a good ECU and these are just
> touching the tip of the iceberg. I haven't build a diy ecu yet, however
> have been researching it on and off for a few years.
>
> TC, What's your CCD about? [noparse]:)[/noparse]
>
> Luke.
>
> M> aconti@n... wrote:
> M> >
> M> > --- In basicstamps@y..., rusram@p... wrote:
> M> > > anybody know about DIY ecu with basic stamp??
> M> > >
> M> > > thanks..
> M> >
> M> > You will Have A few problems. 1st, the stamp can only do things one
> M> > at a time, while the ECU (Electronic Control Unit) can do meny things
> M> > at one time. 2nd, there are a lot of sensors that tell the ECU what
> M> > the motor is doing. 3rd, to program a fuel map and/or a Ignition map
> M> > would take months or eaven years to do. But if you got the time and
> M> > the money, I wish you good luck, and if you need any help just ask.
> M> >
> M> > TC
> M> > Conti Custom Dash (CCD)
>
> --
> '80 (ex)KE38 '86 ST141 '84 AE86 '90 ST185
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject and Body
of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
After a short nap I remember my name clearly and that an engine running
at 6000 RPM effectively has an ignition speed of 3000 cycles per minute
or one cycle in 1/50th of a second. the BS2sx will process 200
instructions in that time frame. At 1000 cycles per minute (2000RPM)
the sx can process 600 instructions. If the controller is sampling 1
out of 3 cycles 1800 instructions can be processed.
Don Denhardt wrote:
>
> Luke it appears that you lack confidence to accomplish this task you are
> interested in. Go for it. It can be done. You will be successful.
>
> No new clean sheet task is ever easy. Never made that claim.
>
> Just that I believe it is doable. The project would call for some
> creativity and perhaps some reinvention of the wheel. You do not have
> to take what others have done as the standard or the only way to
> accomplish a task. Never chain yourself to conventional thinking. Else
> all you will do is attempt to copy what others have done.
>
> The homemade unit has an advantage over car manufacturers requirements.
> It does not have to meet any emissions standards, unless you live in a
> state like California and most of them will be underwater when the big
> quake hits anyway. It merely has to make the engine run smoothly while
> getting reasonable mileage. OEM's spend millions of dollars and years
> developing their ECUs because they do have to meet emissions
> requirements. Additionally, the OEM units manage more than the engine
> (air conditioning, radiator fan, etc...).
>
> Injectors do not have to be opened by the Stamp in each revolution if
> you use a singular modulated feed and merely vary the drip rate. Or if
> you prefer port injection, that is one of the tasks that could be
> offloaded to a dedicated circuit that maintains a set pulse width to
> each injector unless instructed otherwise by the controller. The
> dedicated circuit can feed the injector(s) the same pulse widths for
> quite a few revolutions before needing an update. How about a master
> timing pulse triggering a 555 whose resistive elements are
> programmable. Remember the Controller does not have to micromanage
> every revolution.
>
> Planting the throttle will result in an incredibly low vacuum reading or
> rate of change. One of the tricks I use in programming devices is to
> sprinkle tests for critical real time events throughout the program. If
> the program is sensitive to critical items it can respond quickly
> enough.
>
> An engine running at 6000 RPM makes 10 revolutions per second or a
> revolution in 1/10th of a second. The BS2-sx will process 1000
> instructions in that time frame. 6000 RPM could be construed as really
> sticking your foot into it and not much finesse is needed. If you were
> to assume an engine would average about 2000 RPM in normal driving
> conditions the sx can process 3000 instruction per revolution. If the
> controller is mostly sampling 1 out of 3 revolutions, 9,000 instructions
> can be processed. You do not have to micromanage each revolution.
>
> Reference mapping is only required for open loop operation where
> environmental and engine conditions are unknown or too much of a wild
> card to deal with. All engines that I've worked with while on the road
> with diagnostic computer connected, never left closed loop operation
> once the engine came up to temperature. Once in closed loop operation
> the engine is in a predefined state. I.E. the inlet air temperature is
> regulated, the oil is of a predetermined viscosity and engine rotating
> friction is within a narrow band. This narrow operating band is what
> makes computer control to a fine degree possible.
>
> If you have formulas (adjusting for real time conditions by sensor
> readings) the compensation for fuel quality (the fudge factor) can be
> written to the eprom.
>
> Oooops got to run.
>
> Luke is this is something you are interested in doing, give it a shot.
> There will always be plenty of help available if you run into a
> stumbling block.
>
> Take small steps - get it to start and idle first. Concern yourself with
> road conditions later. You'll have a lot of fun.
>
> Good luck,
>
> Don
>
> Luke Szymanski wrote:
> >
> > (Longish).
> >
> > On Mon, 12 Nov 2001, Don Denhardt wrote:
> >
> > M> I think it is doable for the following reasons.
> >
> > Doable yes, easy no.
> >
> > M> ECU's do not need to
> > M> track every revolution of the engine for control purposes. Sampling
> > M> should be sufficient.
> >
> > Injectors have to be opened in each revolution. The microcontroller itself
> > (inside the ECU) perhaps doesn't need to track each rev, but the complete
> > ECU certainly will.
> >
> > M> Throttle position can be eliminated with close
> > M> monitoring of engine vacuum or a unique sensor fabricated that combines
> > M> the two functions.
> >
> > The throttle position sensor on some engines is used for accelerator pump
> > function. Planting the throttle can lead to a lean condition if this
> > accelerator pump function isn't operating, however a number of economy
> > engine do not have this part of the TPS.
> >
> > The idle switch in the TPS puts the ECU into an idle mode. You can't sense
> > that using a vaccuum sensor as the vaccuum changes depending on a lot of
> > factors, one of which is cold idle up. (Unless idle quality isn't an
> > issue).
> >
> > M> Fuel injection can be done with a singular injector
> > M> similar to early GM's.
> >
> > It can be, but why? If you're replacing a carby setup with injection then
> > fair enough. If you wish to use an EFI existing setup then there is no
> > advantage to change your multiple injectors into a single injector. You can
> > run all the injectors in parallel so electrically they will look like one
> > injector - from a fuel delivery viewpoint, that's a hell of a lot better
> > then single point injection.
> > Newer ECUs control each injector individually.
> >
> > M> Reference maps can be dispensed with if you have
> > M> a starting value and the last learned figure for the manifold vacuum/rpm
> > M> blocks or ballparked with a formula adjusted by ox sensor readings.
> >
> > Uhm.. if you don't have maps, then where will you store your last learned
> > figure? (I'm not completely sure what you're getting at).
> >
> > It is very difficult to adjust fuel delivery based on a single narrow band
> > oxygen sensor on a real-time basis. OEM ECUs only use the narrow band
> > oxygen sensor to tweak the fuel delivery while cruising (closed loop mode).
> > When you apply the throttle the ECU generally goes into open loop mode and
> > delivers fuel based solely on the stored maps.
> >
> > The narrow band of the oxygen sensor means it (usefully) indicates only a
> > lean or a rich condition. The ECUs that I have seen use an oscillating
> > approach in closed loop mode. Ox shows lean, add fuel, when ox shows rich,
> > remove fuel and so on.
> >
> > A wide band oxygen sensor can tell you how lean or rich the engine is
> > running. Those sensor are very expensive and not what you find on your
> > typical EFI engine. AFAIK there is only one manufacturer that equips an
> > engine with a wide band sensor - Honda (Civic). Aftermarket support for a
> > wideband sensor is also very expensive (relatively speaking compared to a
> > BS).
> >
> > M> Some tasks can be offloaded to dedicated circuits or devices (for
> > M> example, keep the distributor, servo advance).
> >
> > Ignition advance kits are available, and would be easy to duplicate.
> > Most ignition advance units only base advance on engine rpm - they do not
> > sense the load on the engine, hence cannot correct for it. That's why it's
> > beneficial to have the ignition advance part built into the same device
> > that knows exactly what the engine is doing.
> >
> > It really depends on what you want to do. If you are trying to replace your
> > factory ECU, and make your engine run "better", then budget a lot of blood,
> > sweat and tears (and time and money too). OEM's spend millions of dollars
> > and years developing their ECUs.
> >
> > If all you are doing is trying to get an engine to run, then you can run
> > one with a 555 timer IC. It wont run very well, but it will run.
> > In such a case a BS can probably do the job quite nicely.
> >
> > You could build a data logger to see what your current ECU is doing and get
> > an idea of what goes on.
> >
> > I highly recommend you have a browse of the http://www.diy-efi.org/
> > website. It deals specifically with the topic of making your own ecu.
> >
> > This post is not to be construed as a discouragement. I also don't mean to
> > be hostile, but if it comes out like that I apologise in advance.
> >
> > There are many things to consider to make a good ECU and these are just
> > touching the tip of the iceberg. I haven't build a diy ecu yet, however
> > have been researching it on and off for a few years.
> >
> > TC, What's your CCD about? [noparse]:)[/noparse]
> >
> > Luke.
> >
> > M> aconti@n... wrote:
> > M> >
> > M> > --- In basicstamps@y..., rusram@p... wrote:
> > M> > > anybody know about DIY ecu with basic stamp??
> > M> > >
> > M> > > thanks..
> > M> >
> > M> > You will Have A few problems. 1st, the stamp can only do things one
> > M> > at a time, while the ECU (Electronic Control Unit) can do meny things
> > M> > at one time. 2nd, there are a lot of sensors that tell the ECU what
> > M> > the motor is doing. 3rd, to program a fuel map and/or a Ignition map
> > M> > would take months or eaven years to do. But if you got the time and
> > M> > the money, I wish you good luck, and if you need any help just ask.
> > M> >
> > M> > TC
> > M> > Conti Custom Dash (CCD)
> >
> > --
> > '80 (ex)KE38 '86 ST141 '84 AE86 '90 ST185
> >
> > To UNSUBSCRIBE, just send mail to:
> > basicstamps-unsubscribe@yahoogroups.com
> > from the same email address that you subscribed. Text in the Subject and
Body of the message will be ignored.
> >
> >
> > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
> _________________________________________________________
> Do You Yahoo!?
> Get your free @yahoo.com address at http://mail.yahoo.com
>
> To UNSUBSCRIBE, just send mail to:
> basicstamps-unsubscribe@yahoogroups.com
> from the same email address that you subscribed. Text in the Subject and Body
of the message will be ignored.
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com