Injector driver question...

[Hugo Villeneuve]

At 20:02 99-04-02 -0800, you wrote:
>>Hi all,
>>
>>    I have designed and tested an injector driver based on the LM1949. I
>>want to be able to run low impedance as well as high impedance type
>>injectors. I tested it with low resistance (2.4 ohms) injector and the
>>operation was as expected (at least on the scope screen!) with a peak
>>current of 4 A and hold at 1.0 A. The rise time is 0.8 mS.
>>
>>When I tested it with a high impedance injector (14.9 ohms), the peek
>>current was about 2.5 A and it took 4mS (seems very long to me...) to
>>rise to this value. After that 4 mS period, the timer fonction of the
>>LM1949 lowered the current to the hold mode of 1 A, as expected.
>>
>>My question is : why the current is so high with the high impedance
>>injector ? If we divide the battery voltage by the impedance of the
>>injector, we get 12 / 14.9 = 0.8 A, which is the value of the current in
>>the steady state.
>>
>>    Hugo Villeneuve.
>
>
>
>
>Your P&H values sound like they are right on the money for an  LM1949  with
>4/1  P&H ( did your  4-1 timer shift  take place at roughly 3.9 ms?).  On
>the other hand, your saturated values sound a bit off.  Did you change
>anything between the two experiments (except for the injector)?  I assume
>you had only one high impedance injector hooked-up during your saturated
>tests?  Seems kinda strange.  I'd be really interested in knowing more about
>your test setup.  Injectors have an equivalent circuit that can be modelled
>as an RL circuit (no capacitance).  From what I've seen/tested, the current
>response usually follows the classical "RL" equivalent circuit model?????
>
>
>Walt.




Hi Walt,

	I know that the timer shift for the low impedance injector did not occur
because the peak current was higher than the 4 A required to activate the
hold mode. I didn't change anything between the two tests. I have also
tested two differents models of high impedance injectors, with
approximatively the same results. I found some information for the values
of the injectors (general guidelines only!) :

	Low impedance : L = 2mH and R = 2.4 Ohms -> time constant = L/R = 0.83 mS

	High impedance: L = 10mH and R = 15 Ohms -> time constant = 0.66 mS

We see that the two time constants are almost the same... For the low
impedance injector, I cannot say what was the final value of the current,
because peak detection reduced the current past 4 A. But for the high
impedance type, after 4 mS it is evident from the curve I obtained that the
current have reached the steady state condition and ,since 4 mS is about
five time constant, the curve seems to be logical for a classical RL
circuit (except for the high current...). So if the current in the low
impedance injector had a chance to rise without peak detection activated,
maybe it will be more than 12V/2.4 = 5 A?

I think the circuit should work like this for high impedance injectors, but
I don't know the effect of lowering the current to 1 A for these kind of
injectors since the notion of peak and hold is never discussed for these
injectors... anyone with an idea?

For my test setup, I used a digital storage oscilloscope with a probe
directly between the 0.1 Ohms sense resistor to measure the current of the
injector, like the LM1949 is doing! . The input signal was at 50% duty
cycle, with a frequency of 83 Hz (5000 RPM).

	Hugo.