EFI and Ignition Systems
Archiver
Posts: 46,084
Bill,
Remember I mentioned with the RIGHT HARDWARE one may
have success, well the additional is true too, if you
do not understand the hardware, meaning the standard
components used out in the EFI and Engine field you
will kill a lot of components trying to learn. What
research have you done to educate yourself on what is
"standard" in the auto engine electronics control
industry?
Both ignition systems and injector coils generate huge
back emf pulses which must be dealt with correctly.
Also these emf pulse are necessary for these items to
work properly.
In order to make the injector close the magnetic field
must be collapsed. The only way to do this fast is to
make the back emf pulse to a high voltage and clamp it
with a high voltage zener across the transistor/switch.
This allows more precise control of fuel delivery.
Randy below is correct. I have been involved with
designing EFI systems before and the characteristics of
the components involved need to be thoroughly
understood to build a successful system. I have not as
yet seen anyone not involved in that industry whip a
system together on the cheap. I have watched systems
being developed with million dollar budgets, over
several years.
You will learn a lot in the process, however you need
to be prepared so spend several thousand dollars to
make a reasonable working system, not including your
time and all the fried parts along the way, that you
may need to purchase several times over.
Anytime I want to fuel inject an engine I will buy and
off the shelf ACCEL fuel injection system which has all
the bells and whistles needed for reliable control of
any engine from a lawn mower to a V12 Allison, the
right hardware in terms of fuel injectors, pumps, fuel
pressure regulation still need to matched to the
engines need.
An ACCEL system can be had for about 900-1200 dollars
through JEGS catalog.
Using pc software all calibration info can be loaded
and adjusted on the fly (with the engine running). In
terms of having a successful project, this system
cannot be beat by a home brew setup. Sorry, that's just
the way it is.
If you have money and stamps to burn, carry on. If not
you might wish to evaluate your project again.
Ron
*******************************************************
**********
> Message: 17
> Date: Sun, 26 Jan 2003 00:53:45 -0000
> From: "billakay <billak@f...>"
<billak@f...>
> Subject: Stamp crapped out?
>
> Hey guys, I got that injector driver circuit working
last night, I
> was playing with it all last evening, I started
messing with it
> tonight, and I could hear the injector clicking for a
couple
> seconds, like it was working correctly, then it
stopped. Now the
> program recognizes the stamp as connected, but can't
download any
> programs to it.
>
> What could have happened? I didn't do anything wrong
I don't think,
> I was just messing with a circuit I had working just
fine last night.
>
> Do they have a warranty on these things?
>
> Bill
> Message: 20
> Date: Sun, 26 Jan 2003 03:23:56 -0000
> From: "billakay <billak@f...>"
<billak@f...>
> Subject: Re: Stamp crapped out?
>
> No I didn't have any diode like that...could
something have hurt the
> stamp even through the transistor?
>
> I guess I'll have to order another one, live and
learn I guess...
>
>
> Bill
> Message: 23
> Date: Sat, 25 Jan 2003 20:30:41 -0800
> From: "Randy Jones" <randyjones@w...>
> Subject: Re: Stamp crapped out?
>
> Automotive injector driver circuits do not use a
spike suppression diode
> connected across the injector coil, as is often done
with relay coils. If
> you watch an injector waveform with a scope, you will
see that the spike is
> allowed to go fairly high -- 60 to 80 volts is
common. A zener is often
> included in the injector driver to clip the spikes at
that level.
>
> If you connect a diode across the injector winding,
you will find that the
> injector stays open MUCH longer after the driver
turns off. Some years ago
> I had a "discussion" about this with an acquaintance
who had written an
> injector theory article which was published in one of
the automotive trade
> magazines. This spike and injector closing stuff was
erroneously explained,
> and I ended up cooking up a little experiment to show
what happens...
>
> I programmed a PIC (Stamp would have been fine too)
to switch a MOSFET which
> drove an injector for a specific number of 2mS
pulses. Using fuel (er... I
> mean non-flammable test fluid) at a constant
pressure, I was able to measure
> the total injected quantity in a large graduated
cylinder with repeatability
> better than 1%. The MOSFET was rated for 100 volts
or more so it handled
> the raw injector spikes just fine. Then I installed
a spike suppression
> diode across the injector and ran the test again.
MUCH more fuel was
> collected.... and a correction was eventually
published.
>
> So, if you want tight control of your injector, you
have to do what the
> automotive industry does - let the spike go fairly
high, and use a driver
> transistor that can handle it.
>
> FYI, unless Parallax has changed their policy
"cancelled" Stamps can be
> exchanged for a replacement at about 1/2 the cost of
a new one.
>
> Randy
> www.glitchbuster.com
Remember I mentioned with the RIGHT HARDWARE one may
have success, well the additional is true too, if you
do not understand the hardware, meaning the standard
components used out in the EFI and Engine field you
will kill a lot of components trying to learn. What
research have you done to educate yourself on what is
"standard" in the auto engine electronics control
industry?
Both ignition systems and injector coils generate huge
back emf pulses which must be dealt with correctly.
Also these emf pulse are necessary for these items to
work properly.
In order to make the injector close the magnetic field
must be collapsed. The only way to do this fast is to
make the back emf pulse to a high voltage and clamp it
with a high voltage zener across the transistor/switch.
This allows more precise control of fuel delivery.
Randy below is correct. I have been involved with
designing EFI systems before and the characteristics of
the components involved need to be thoroughly
understood to build a successful system. I have not as
yet seen anyone not involved in that industry whip a
system together on the cheap. I have watched systems
being developed with million dollar budgets, over
several years.
You will learn a lot in the process, however you need
to be prepared so spend several thousand dollars to
make a reasonable working system, not including your
time and all the fried parts along the way, that you
may need to purchase several times over.
Anytime I want to fuel inject an engine I will buy and
off the shelf ACCEL fuel injection system which has all
the bells and whistles needed for reliable control of
any engine from a lawn mower to a V12 Allison, the
right hardware in terms of fuel injectors, pumps, fuel
pressure regulation still need to matched to the
engines need.
An ACCEL system can be had for about 900-1200 dollars
through JEGS catalog.
Using pc software all calibration info can be loaded
and adjusted on the fly (with the engine running). In
terms of having a successful project, this system
cannot be beat by a home brew setup. Sorry, that's just
the way it is.
If you have money and stamps to burn, carry on. If not
you might wish to evaluate your project again.
Ron
*******************************************************
**********
> Message: 17
> Date: Sun, 26 Jan 2003 00:53:45 -0000
> From: "billakay <billak@f...>"
<billak@f...>
> Subject: Stamp crapped out?
>
> Hey guys, I got that injector driver circuit working
last night, I
> was playing with it all last evening, I started
messing with it
> tonight, and I could hear the injector clicking for a
couple
> seconds, like it was working correctly, then it
stopped. Now the
> program recognizes the stamp as connected, but can't
download any
> programs to it.
>
> What could have happened? I didn't do anything wrong
I don't think,
> I was just messing with a circuit I had working just
fine last night.
>
> Do they have a warranty on these things?
>
> Bill
> Message: 20
> Date: Sun, 26 Jan 2003 03:23:56 -0000
> From: "billakay <billak@f...>"
<billak@f...>
> Subject: Re: Stamp crapped out?
>
> No I didn't have any diode like that...could
something have hurt the
> stamp even through the transistor?
>
> I guess I'll have to order another one, live and
learn I guess...
>
>
> Bill
> Message: 23
> Date: Sat, 25 Jan 2003 20:30:41 -0800
> From: "Randy Jones" <randyjones@w...>
> Subject: Re: Stamp crapped out?
>
> Automotive injector driver circuits do not use a
spike suppression diode
> connected across the injector coil, as is often done
with relay coils. If
> you watch an injector waveform with a scope, you will
see that the spike is
> allowed to go fairly high -- 60 to 80 volts is
common. A zener is often
> included in the injector driver to clip the spikes at
that level.
>
> If you connect a diode across the injector winding,
you will find that the
> injector stays open MUCH longer after the driver
turns off. Some years ago
> I had a "discussion" about this with an acquaintance
who had written an
> injector theory article which was published in one of
the automotive trade
> magazines. This spike and injector closing stuff was
erroneously explained,
> and I ended up cooking up a little experiment to show
what happens...
>
> I programmed a PIC (Stamp would have been fine too)
to switch a MOSFET which
> drove an injector for a specific number of 2mS
pulses. Using fuel (er... I
> mean non-flammable test fluid) at a constant
pressure, I was able to measure
> the total injected quantity in a large graduated
cylinder with repeatability
> better than 1%. The MOSFET was rated for 100 volts
or more so it handled
> the raw injector spikes just fine. Then I installed
a spike suppression
> diode across the injector and ran the test again.
MUCH more fuel was
> collected.... and a correction was eventually
published.
>
> So, if you want tight control of your injector, you
have to do what the
> automotive industry does - let the spike go fairly
high, and use a driver
> transistor that can handle it.
>
> FYI, unless Parallax has changed their policy
"cancelled" Stamps can be
> exchanged for a replacement at about 1/2 the cost of
a new one.
>
> Randy
> www.glitchbuster.com
Comments
You should take a look here if you haven't already:
http://www.bgsoflex.com/megasquirt.html
and
http://autos.groups.yahoo.com/group/megasquirt/
Bucky