Engine injector control ICs

what are you trying to do?

there's no magic to driving a fuel injector, turn it on for X milliseconds and you get and you an ~X*flowrate milliliter squirt of fuel (at a specific fuel pressure and battery voltage)

To run an engine you need the engine management to figure out how much and when to inject

-Lasse

Have you looked into the 'Megasquirt' ECM? It's open source hardware and firmware.

Cheers

Gentlemen,

Does anyone know of a readily available chip to control the timing and duty cycle of engine fuel injectors? Or a PWM chip that could do the job equally well? Most of the ICs I'm finding seem to be entire engine management chips that do far too much other stuff whereas I only want injector control.

Thanks. ===================================================

If you just want to drive the injector, if it is high impedance (over 12 ohms) just use a bjt or fet like it is a relay coil. If it is low impedance, generally 1.5-3 ohms, you need a peak and hold driver. Usually they ramp from zero to 4 amps in 1 msec then drop to 1 amp and hold. Cherry Semi used to make the CS452 and CS453 (different current profiles) but I'm not sure what is on the market these days. If you need the smarts that decides when to inject and for how long, that's a complete ecu :-). The Megasquirt is very popular and can be had as kits or complete. Even cheaper, there are lots of arduino based projects, just google. One such is discussed on this forum:

----- Regards, Carl Ijames

LM1949 from TI. but a 555 timer can do this.... have it start with a short delay before pusling. drive a darlington which has a diode in it that can be used for the flyback.

a flyback diode will make it close far too slow

Well, it doesn't matter because the diode in the Darlington is in the wrong position to act as a flyback diode.

--sp

How about a discrete, ultra-fast recovery diode, then?

Don't think so, I believe the diode keeps current flowing longer in the coil, so the solenoid plunger stays in the "open" position longer.

if you clamps the flyback voltage with a diode the injector closes very slow, you could use a zener

or just do like they do in cars, use a clamped FET like a VND7NV04

-Lasse

The issue with a flyback diode is that it holds the voltage across the coil to a small value, which slows down the current decay.

A flyback diode in series with a resistor, sized so that the transistor doesn't go "poof", is the way to go. Or a flyback in series with a Zener, or a Zener to the transistor emitter. The higher-voltage transistor you're willing to buy (assuming matching snubbers), the faster you can make the current decay.

The app note for the chip Mr. Philbrook pointed us to uses a 33V 5W zener for snubbing purposes:

Probably the simplest solution!

how is that simpler than a single automotive FET made for the purpose with build in clamping at the right voltage, short circuit and temperature protection?

if you absolutely have to use low a impedance injector add a 10R series resistor

I have yet to obtain the full data for the injectors so have no idea what their impedance is as yet. What's the 10R resistor for and in series with what - the injector coil or the snubber?

gasoline fuel injectors are generally high impedance ~12R or low impedance ~2R-3R

high impedance you just drive drectly with ~12V, With low impedance you can put a ~10R resistor in series with the injector to limit current.

You don't need a snubber if the FET is rated for clamping

how so? if it's pulling one side to low from the coil via the collector, how does the diode not help? for that matter, a mosfet can do the same in the same position. jamie

Den onsdag den 7. december 2016 kl. 23.51.55 UTC+1 skrev M Philbrook:

The flyback is positive so the diode is in the wrong direction to do anything and if it was in the other direction it would be always on

A mosfet would work assuming it is rated for it but that would be from avalanching at Vds max

again please! diodes in darlingtons are in same positon as mosfets. think, open collector/ open drain.

Den torsdag den 8. december 2016 kl. 00.52.31 UTC+1 skrev M Philbrook:

the collector/drain is pulled positive, above the supply during flyback

and the diode in a mosfet wouldn't either if it didn't act like a zener and avalached at ~Vds max

-Lasse

yes, you are absolutely correct. Which is why we don't run fast raise and fall pulses where they are not needed.

DC r in the coilds tend to reduce the effects greatly . 101 E but for those times when flyback is high in cases of design or shorts in coils which do happen.... we use zener boost via tranny for the flyback clamp