Homebrew EFI box

[Alex Samson]

Alex samson wrote:
>After more than 4 years I finally found this site that I have been
>looking for!  I build... <snip>

Ken, I was referring to this reflector at ohio state.  I am a newbee
here and I am glad to have found it.

Sorry Ken, havent done work on the 13B or so.  Had several invitations
in the past but just never got started to try one out. I hear that the
Weber fired 13B's output a LOT of power. 

Building boxes is a career by itself.  You will figure how to make one
yourself in the long run if you are determined enough. This reflector is
perfect as a means for that goal. An MCU based box is better than a
hard-wired logic one because the MCU based system runs a software that
allows the author to explore all possibilities without ever picking up a
soldering iron.  You will have to become adept at a specific MCU or as
an alternative perhaps acquire one of those ready made "universal" EFI
boxes.  

regards,
alex samson


> -----Original Message-----
> From:	Kenneth Bailey [SMTP:iflyrc at bellsouth.net]
> Sent:	Wednesday, August 26, 1998 10:33 PM
> To:	diy_efi at efi332.eng.ohio-state.edu
> Subject:	Re: Homebrew EFI box
> 
> Alex, Where is this web site you seek and found. Also have you done
> any work
> with a mazda RZ-7 rotary ?
> 
> I would like to learn to build hobby boxes ...
> 
> 
> Ken
> 
> 
> 
> 
> 
> 
> 
> 
> 
> Alex Samson wrote:
> 
> > Hi to All,
> > After more than 4 years I finally found this site that I have been
> > looking for!  I build EFI boxes for a hobby and every time I start a
> > project, it is always one big massive guessing game.  I always
> wondered
> > if there exists a website or reflector for this activity to help
> make
> > the guesses a little less wilder.
> >
> > My last project was a complete engine management box for a Subaru
> 2.2L
> > flat-4.  This engine went into an early 80's VW Vanagon which had
> that
> > famous engine commonly known as a "grenade with a short fuse".  The
> box
> > I made was based on an 80C31 for the EFI, another 80C31 for the
> > distributorless ignition system, and a PIC for the idle stabilizer.
> Man
> > this was a wild project for about 6 months! I did a lot of guessing
> > because we never had access to a Subaru with the same engine. I had
> 8k
> > bytes of scratchpad RAM to log sensor readings to help in developing
> the
> > software written in assembler. At present, the van has done over 30K
> > miles already and I havent made any more changes the last 2 years.
> Its
> > a daily driver owned by a shop mechanic who simply loves the novelty
> of
> > his engine setup.   Exhaust has the usual stuff and an O2 sensor
> which
> > the box reads for mixture control.  Softwares and look-up tables for
> the
> > 80C31's are stored in individual SRAMs with  a small backup battery
> to
> > allow easy upgrade downloading from a laptop. Naturally, a default
> setup
> > is in Eprom.  I used the original Subaru air flow sensor, Cam &
> Crank
> > sensors and other peripherals.  Fuel injectors were quite rare in
> that
> > they didnt need a step-start scheme, they were merely turned
> completey
> > ON or OFF with a straight FET switch (no modulation). The power
> output
> > of this new engine is way outside the league of the original
> aircooled
> > VW.  The owner once had damaged a pinion gear bearing, then
> obliterated
> > a clutch, and the last repair was the transmission getting stuck in
> 1st
> > gear.  Unfortunately, the engine has proven to be too strong for the
> > tranny but otherwise it is very smooth and more appropriate for the
> van.
> > One just has to learn to be easy on the throttle to preserve the
> drive
> > train.  Would I do it again?  YES!  It was a lot of fun. Kenedy
> > Engineering in Palmdale made the adapter plate.
> >
> > I have started a new project, this time for an old Bosch D-jetronic
> in a
> > 73 Mercedes 450SLC.  This one is for myself.  At the present, I just
> > finished installation of an O2 sensor in the exhaust, and I finished
> a
> > small box based on the AT89C2051 that processes the EFI points from
> the
> > distributor.  The problem with the original system design is that
> the 4
> > EFI trigger points are located at the bottom of the distributor and
> the
> > lower pair of points get fouled by oil sling.   So my small box
> > "synthesizes" the action of the lower pair of points by just reading
> the
> > upper pair of points.  I will start on the main box next.  The main
> box
> > will allow mixture correction with the O2 sensor and will implement
> a
> > more positive step-start injector control instead of the weak
> original
> > Bosch circuit.  This should hopefully improve emissions, power, and
> fuel
> > mileage when I am done.  We use the Atmel AVR at work so I am
> tempted to
> > use the AVR for this project this time.  I want to keep the
> distributor
> > original with regards to suggestions from others in the past to
> install
> > a GM magnetic trigger scheme to the Mercedes distributor.  I want to
> > maintain full compatibility with the original box so its easy to go
> back
> > to stock.
> >
> > By the way, I was recently at a junque yard because I needed a
> connector
> > for a D-Jetronic computer.  I found an old 1972 (or older?) Mercedes
> 280
> > SE 3.5 thats been sitting out in the open for probably many years.
> The
> > D-jetronic box was still there although exposed to the rain etc in
> the
> > exposed engine compartment.  I picked up the box (and connectors)
> and
> > ran it in my bench simulator back home (for which I needed the
> connector
> > for in the first place).  I wanted to see what kind of failure this
> > junque yard box  had acquired after all the exposure.  Would you
> > believe, this box still works?  I was amazed. This was the first
> > generation of Bosch EFI and they certainly must have made sure that
> the
> > boxes are dependable!
> >
> > I can go on and on with this subject but I would first like to get
> > feedback from the members of the list if my topics are appropriate.
> >
> > regards to all,
> > alex samson