Wednesday, December 31, 2008

Friday, December 19, 2008

New intake manifold with throttle-body injection

Look what I got as a christmas present today. Ole Mobeck has finished
the machining of my intake manifold. The manifold is supposed to replace the original shown to the right. It has a fixture for a Bosch fuel injector which will make throttle-body injection (Central Fuel Injection/Single-point injection) possible. Now I only need to drill a few mounting holes and two holes for the map sensor and vacuum diaphragm.

Wednesday, December 17, 2008

Home made electronic ignition and fuel injection for my 1961 Fiat 600 - part 2

I finally got some extra time to figure out how to build the actuating part of the home made ignition + injection. I got the final bricks to build the circuit after reading about how to interface high current loads (up to 10A) with a microcontroller from this really nice web page.

In the last post I figured out how to get sparks with a 12V input directly from a battery and a subaru impreza igniter. This time I wanted to use a microcontroller to create controlled sparks. As a bonus I wanted to check if the same circuit could switch the two Bosch injection coils that I got.

To make a quick setup for the microcontroller I set up the arduino NG that I got to read the input from a potmeter (0-5V) and adjust the on time for the digital output from 0 to 2.048ms. I know from literature that 1.2ms usually is sufficient to make a nice spark. I was a little bit uncertain to wheter this would be sufficient for a waste spark system that I am using.

I set up the circuit as shown below to the left:

Ignition + Injection circuit test


The physical test - lots of wires

After some fumbling around, double checking that all the wires were correctly installed, I turned the knob on the potmeter. Voila - The sparks were exactly as I wanted them - nice and blue. The sparks were steady at approximately 1/3 adjustment for two spark plugs, which indicates that a minimum of 0.7ms would be necessary. I am not sure what turn on time would be necessary while the cylinders are with pressure.

After successfully testing the prototype for the ignition i removed the ignitor+the spark plug pack from the circuit. I installed a snubber diode, as shown on the reference page above, to avoid high current back emf.


Connected to a injection coil with a snubber diode

The setup produced a nice clicking sound when connected to the orange injector shown above. The fully black injector had no reaction when adjusted to maximum coil charging time. A multimeter showed that the orange coil had 16ohm and the fully black 2.5ohm resistance. It might be that the black injector needs more charing time for the solenoid to open.

To summarize: I have figured out how to actuate the spark plugs and one ignition coil with a 5V output from a microcontroller. Now I need to start measuring some input from the sensors to correctly adjust spark advance/retard and injection time/pulse width. I allready have the old mechanical settings for the spark diagrams from a workshop manual.
Posted by Picasa