Shop OBEX P1 Docs P2 Docs Learn Events
Fuel Injection Conversion — Parallax Forums

Fuel Injection Conversion

Guy KGuy K Posts: 6
edited 2010-09-01 18:26 in Propeller 1
I currently own a 69 simplicity 3112 garden tractor equiped with a 12 horse-power Cast Iron, Briggs and Stratton carborated engine.

My plan thus far is to convert this carborated engine to a PWM fuel injected engine, utillizing the Propeller Chip in a dip 40 package. I will first start by gathering some basic info from the engine while running on its normal carborator( I.E. Control Run) such as head temp, cyclinder temp, exhuast gas temp, rpm, fuel comsumption, there are a few more.
Since this carborator is a bowl and float type, I will drain it of gas, plug up the holes and remove the choak plate. I will be controlling the throttle plate with a servo so I can get positional feed back. I plan on mounting an automotive fuel injector in the new airway tube I will machine. This will make it more of a throttle body design rather than direct port injection style.
I am in favor of using multiple Propeller chips but I will see if one can do it for now. I will be using dual 4x20 serial LCD for looking at engine data like rpm ans temperatures.
I will mount a sensor to detect the magnet on the flywheel, at some point right before the intake cycle begains to use as a input signal to active the fuel injector then tweeking it untill operational temperatures become close to the control run, that should bring it back to operational status.

I am still in the planning phase of this project and would like to hear what everyone thinks, I do expect this to take some time as I am on a very limited budget. I will post pictures, notes, and code as the project progresses.

Comments

  • Graham StablerGraham Stabler Posts: 2,510
    edited 2010-08-31 07:25
    Despite pulling my hair out over the weekend with a pair of carburettors on a motorcycle I can't really comment on the mechanical side of things but as far as sequencing pulses to an injector goes it should not be any problem for the Propeller, a single Propeller.

    Graham
  • RinksCustomsRinksCustoms Posts: 531
    edited 2010-08-31 08:58
    one of those $25 electronic fuel pumps will do nicely, they pump ~25psi, perfect for what your doing. "right before the intake cycle" may not give nough headroom as most fuel injection systems today start firing the injector as far back as half way through the previous exhaust stroke, remembering that it takes a finite ammount of time for the air/fuel mix to travel from the throttle body to the cylinder depending on intake runner length x volume. For what your doing, simply varying the frequency of the pulses to the injector based off RPM will do, while varying the duty cycle based on engine load will yield good results and will be easy to program.
    And a thermocouple in the exhaust will be much faster response to rich/lean condition over CHT. CHT is goot for indicating overheating/general op-temp but not so good for adjusting the pulse width/duty cycle for the injector. Better yet, a heated oxygen sensor is about $35 from the parts store, even cheaper if you yank it off a car from the local "u-pull-it".
    Also, you need a way to determine engine loading (unless you just want to sit there and play with the RPM's all day), your gonna need at least a 1 bar MAP sensor. but do take accurate readings of EGT,RPM, & MAP after they stabilize under various load conditions.. you may want to make a kind of data logger with the prop to capture these conditions, a link to a laptop with Prop serial terminal on it should be the simplest solution over an SD card idea.

    If you use a large enough injector, something like a single 240cc injector, that should give plenty enough headroom, since a 240cc injector comes stock on a 1.5L honda civic.

    Good luck
  • markaericmarkaeric Posts: 282
    edited 2010-09-01 13:18
    Guy K,

    While a MAP sensor might be easier to incorporate in hardware, the software side would probably much easier to tune if you were to use a mass airflow meter instead.
  • Bob Lawrence (VE1RLL)Bob Lawrence (VE1RLL) Posts: 1,720
    edited 2010-09-01 14:36
  • RinksCustomsRinksCustoms Posts: 531
    edited 2010-09-01 18:26
    markaeric wrote: »
    Guy K,

    While a MAP sensor might be easier to incorporate in hardware, the software side would probably much easier to tune if you were to use a mass airflow meter instead.

    Trust me on this one, its not! i have a MAF from a LT1 and some excel data somewhere for a few differnt similar p/n's i found on the internet.. data from manufacturers and aftermarket companies is proprietary and guarded better than fort knox, also it seems there is somewhat sloppy tolerances between units of the same p/n. i was going to use two MAF's on a twin turbo setup i was working on for my 4.3 Vortech. I'm still planing on doing it but need to make a test rig to accurately measure airflow first and then get my mass air flow readings vs frequency output. It's a task to say the least. I'm not sure if MAF's are humidity compensated, as water grains affect oxygen content negatively.
Sign In or Register to comment.