Multi 555 chip controlling a single fuel injector


Joined Nov 6, 2012
An Engine can still "run" in an absolutely atrocious state of Tune,
this is the only reason that there is any reason to even consider the 555 as any part
of a Fuel-Injection/Ignition-Control-System.

If You are simply "wanting to see if You can make it run", more power to ya,
but if You expect the Engine to provide reasonable performance when you're done,
you'll find that the original Carburetor will work much better.

Most Late-Model-Dirt-Bikes have a very simple Fuel-Injection-System if you're interested,
even some of the larger Lawn-Mower/Industrial, Single, and Twin-Cylinder, Engines are
now coming equipped with crude forms of Fuel-Injection.

A MAP-Sensor won't work very well on a single-cylinder-Engine, ( if that's what you're working on ),
so you'll be limited to what you can make work with just a TPS-Sensor, and a Temperature-Sensor.
And when using only a TPS-Sensor,
it will be very important to provide a rather radical "Accelerator-Pump" function.

The fact that You mentioned "Barometric-Correction", indicates to me that You really
don't have a complete enough understanding of the Fueling requirements of a Gasoline-Engine.
"Barometric-Correction" is only useful for maintaining tight control of the Fuel-Mixture when
radically changing the Altitude that the Engine will be operating in,
such as driving from the bottom, to the top, of a large Mountain-Range, and then back down again.
The Barometric-Changes that occur because of changes in the Local-Weather
are almost completely inconsequential, and may be completely ignored.

You must also have a method for creating 4 separate
"Response-Curves" which will then be mixed together to produce a
total Injector-Pulsewidth in milliseconds,
these factors will be ........
Throttle-Position, and
If You think You can pull this off using 555-Timers, I want to see it,
that would be a really unique Circuit.

A Fuel-Injection-Computer can certainly be done fully Analog,
and can be designed and Tuned to work surprisingly well,
but I can't think of a single reason to use a 555 in this project,
and I think the requirements are much more complex than You may realize.


Joined Aug 31, 2022
but I can't think of a single reason to use a 555 in this project,
I'd could rely on the argument that it's been done before but another reason is that it can be used as an extremely reliable and simple monostable where the pulse width can be controlled with an analogue voltage.

I'm actually using a Seeeduino XIAO fed from variable reluctance sensor and a 36/1 trigger wheel to have a go at replacing a distributor, (with a wasted spark double coil with built in drivers) and eventually perhaps the Stromberg carburettor for a single injector EFI for my 1969 Triumph Herald. Getting the spark timing wrong can be catastrophic but a flaky EFI is more forgiving.

The fact that the Stromberg works as well as it does, relying mostly on the venturi vacuum to control the needle position and hence the amount of fuel is a tribute to the elegantly simple design which was probably the best of its day. So I really can't agree that something as efficient as this can't be done with relatively simple analogue electronics. Even non-linearity can be adjusted with diodes in an Op Amp feedback loop, as was done years ago in digital panel meters using a variety of very non-linear temperature sensors.

But having a processor on board is much easier for fine tuning, adding the influence of more sensors with A/D conversion, and stored maps. One final word on the 555! My original idea was to use a 555 to control the firing of the injectors (for a time determined by the singe DAC output on the XIAO) as it's an event whilst controlling the on/off timing of the firing requires timer interrupts which, when the interrupt comes, other processing stops - but the speed of the XIAO is so fast it's easy enough to work around this.

Thanks for your thought all the same - maybe we can just agree to differ, although your post has made me think about how I would actually make an analogue EFI at least as good as the Stromberg. Triumph was an early adopter of fuel injection to replace carburettors.



Joined Jan 6, 2022

I reply only because I've seen both approaches used to control a fuel injector in a small engine.

This was about 25 years ago, when my school entered a fuel efficient car into the "Shell Mileage Marathon" in Australia.

A friend had already achieved over 3000mpg using a 28cc engined single seater vehicle. It used a 556 timer, thumbwheel rotary switch adjustment to select different dwell and different timing widths etc. The catch is that it was a very simple set up, designed to accelerate up to and run at only one particular RPM, for about 7 seconds for each 700m or something like that (I forget the actual length but anyone could find out... It was the lower loop of Amaroo Park in Sydney which was fairly flat but ever so slightly downhill for the most part) So while it achieved an extreme fuel efficiency, it was not required to be fast, or to accelerate fast, it only had to hold the vehicle at a set speed and there weren't many inputs (if any) as it could be tuned on the day by the mechanic/engineer to suit the conditions of the weather etc.

Later we built another couple of designs for similar engines (36cc this time, completely hand built with SOHC) and one of the student's parents built a rudimentary micro-controller based design, again, very few sensor inputs (I don't think it even had an airflow meter) but could again be tweaked to suit the conditions of the engine, and the day and any other external factors.

A lot of fun, but for the most part I was an observer on the electronics side... Though I did build the electronic ignition system for the first attempt in the years prior to our efi efforts (which was a carby engine) That was simply triggered by an adjustable hall effect switch, worked just fine.

Have fun, I certainly did!