FW: [Diy_efi] 555 timers

Mon Jan 24 16:34:03 GMT 2005

We had some off-line discussion while the list was down for a few days.  I
will post all of those exchanges incase others want to jump in.

Thanks for getting us back online Steve!

Dan Nicoson

-----Original Message-----
From: Daniel Nicoson [mailto:A6intruder at adelphia.net]
Sent: Wednesday, January 19, 2005 10:08 AM
To: A list for Do-It-Yourself EFI
Subject: RE: [Diy_efi] 555 timers

James,

I've been thinking about your project of a bare-bones EFI system.  I was one
of the ones that suggested a 555 timer based system.

First, I wouldn't totally rule out your original route of a blow-through
system.  Check in on the Yahoo groups Jyturbo.  Some of those guys have some
blow-through experience and may be able to help you troubleshoot.  In fact I
think there is another turbo Yugo there.

I say don't rule out the blow-through carb setup because the 555 based EFI
system will take some real time to work out well enough to run a turbo.  Not
that it can't be done, but it is a whole project of its own. (A VERY cool
project!)

If you want to pursue the 555 EFI here are some thoughts:

First, how much fuel do you need to deliver?  Will the injector you have
selected even start to deliver that much fuel?  figure if your engine makes
90 Hp stock (I know NOTHING about Yugos), 7-8 psi typically gives 40-50%
increase, so you would make roughly 130 Hp on boost.  So if you use
brake-specific-fuel-consumption (BSFC) of .55(lbs/hphr) then you will use
.55*130 = 71.5 lbs per hour.  That's a pretty good rate for a single
injector.  A typical early 1990's Ford V-8 injector was 14#/hr, Mustangs
used 19#/hr.  These are cheap and plentiful in JY's.  Use several if needed.

Pulse Widths etc.  I realize you are trying to set up a throttle body style
injector (like a carb).  But pulse width limitations will still apply
regardless of your final injector configuration.  For each revolution of the
crank you must achieve 1/2 of your fueling.  At 6,000 RPM a 4 cycle engine
has 20ms (micro seconds) to inject fuel at 100% duty cycle, 4,000 RPM brings
that max pulse width to 30ms.  Most injectors take from .5 to 2.0 ms to
fully open (this takes away from useable max pulse width).

Google is your friend learning about the 555 timer.  Here are some links I
have found very helpful:

http://casemods.pointofnoreturn.org/pwm/circuit2.html

http://www.williamson-labs.com/480_555.htm

http://home.cogeco.ca/~rpaisley4/LM555.html

http://www.robietherobot.com/storm/mapsensor.htm

http://www.digikey.com/

You started asking specific questions about current, voltages etc on the
555.  Go to Digi-Key and pull down the product data sheet for the specific
chip you have.  Although most 555 chips function the same (logically)
different models have different specifications.  These data sheets ARE your
friend.  Also, some contain good application notes with example circuits.
Pull down the data sheets from several different manufacturers.

I have been thinking about the overall function of what you are trying to
do.  I would set this up as a "load-based" TB EFI.  As someone else pointed
out you can't do this purely on RPM.  I would consider adding a MAP sensor
such as the MPXA4250A series (find it in Digi-Key, $18.00) or find a MAP
sensor out of a junk car that puts out 0 - 5 volts.

Think of LOAD as follows:  Drive your car in second gear, 3,000 RPM steady
speed.  That is a very light load.  Now from that steady state, floor it and
you will experience high LOAD.  Same RPM very different fuel requirements.
On my 1994 Mustang GT, the light load situation would require 2.5ms pulse
widths, full load would require 12.5ms pulse widths.

You'll hear us discussing pulse width modulation a lot with regard to this
project.  The typical way of doing this on a 555 is by changing the
resistance and/or capacitance.  There are some real limitations to this
method, especially achieving full range 0-100% duty cycle.  The first link
above has a very interesting way to fully achieve 0 - 100% duty cycle.  They
use a few more components but get a well controlled circuit.

Here's how I would approach this project.  TB injectors sufficient to flow
the fuel needed.  Batch fire these injectors.  On the 555 control circuit,
set the 555 up as a "one-shot" or mono-stable circuit with about a 45ms
pulse length (120K or so resistance & .1uF capacitor) trigger it off the
primary side of the ignition coil (good luck dealing with the inductive
spike - another whole subject...) then modulate the pulse width via the MAP
signal (using the ideas in the first link). You'll probably need an opamp or
two to set gain of the MAP signal etc.

So basically if you open the throttle, the MAP sensor will notice the change
in manifold pressure and output a different signal which changes the fuel
pulse width.

For the injector drive part of the circuit take a look at the LM1949 chip
(Digi-Key).  It has a nice layout of an injector drive circuit for a low
impedance injector.  This is a good starting point.

Final point.  This project will not be easy to develop without an
oscilloscope.  I started my project without one and found very quickly I HAD
to have one.  I wouldn't consider this EFI project without one.  This EFI
project is more involved than my own which I am 1.5 years into (a secondary
sequential injector driver).  I'm currently re-working my circuit to make it
more simple (www.myo-p.com).

Good luck with this project.

Dan Nicoson

-----Original Message-----
From: diy_efi-bounces at diy-efi.org [mailto:diy_efi-bounces at diy-efi.org]On
Behalf Of James Seabolt
Sent: Tuesday, January 18, 2005 8:01 AM
To: diy_efi at diy-efi.org
Subject: [Diy_efi] 555 timers

You guys got me thinking over the weekend and I think I'm going to give
this 555 timer a shot.
I've had some experience with this chip. I built a timer that would turn an
LED on and off and when I was looking through an old shoe box, found the
circuit still hooked up to the breadboard. That was almost 15 years ago!

It's been so long since my Tech School days so I'm hoping someone will
refresh my memory. If I can put anything to use from my Digital Circuits
class, I'm hoping this will be it. Since I was unable to use my computer
degree. I ended up working at a chemical plant instead after school.

What I want to do is take the tach signal from the - side of the coil which
will be the trigger, feed that into the 555 chip which will pulse the
injector. And use a potentiometer to adjust the pulse rate.

I want to start out using a Bosch cold start valve since it would be easier
to mount. Perhaps use the throttle position switch to trigger a second
injector under wide open throttle and boost.

I downloaded some pages on the 555 timer but can't figure out what size
capacitors I need. And what range the potentiometer needs to be.

The timer chip I have is stamped NE555N. Is this the chip that will
tolerate 5 to 18 volts? Or do I need a different chip?

Also I'm a bit confused on the reset (pin #4). One schematic shows it tied
to pin #8 (Vcc). Another one shows something else.

If someone could draw me up a schematic with what components I need that
would be really cool!

I'm off for the next three days (I work 12 hour shifts) and would like to
spend time on my Yugo since it's too cold to do anything else this time of
year.

Is it possible to build a working fuel injection system by simply
controlling the injector using just a tach signal? I know it probably
wouldn't be able to adapt any better to temperature and altitude changes
than a carburetor but I'm thinking it just might work.

I'm having my doubts whether I am going to successful at getting my Yugo to
accept blowing through a carburetor so I'm looking toward fuel injection.

James Seabolt
Tennessee, United States
Homepage --->>> http://users.chartertn.net/jseabolt/

2003 Subaru Baja
1987 Yugo GV
1980 FIAT 2000 "TURBO" Spider
1968 Ford Fairlane 500 (Not a Ford Galaxie!!)

_______________________________________________
diy_efi mailing list
diy_efi at diy-efi.org
http://lists.diy-efi.org/mailman/listinfo/diy_efi

_______________________________________________
diy_efi mailing list
diy_efi at diy-efi.org
http://lists.diy-efi.org/mailman/listinfo/diy_efi