Injector Driver Module

[Jose Carlos Rublescki]

On many cars you'll find a temperature sensor under the hood that's
connected to nothing. It'll turn on the cooling fan if the under the hood
ambient temperature reaches unacceptable high values. This way I believe
they can keep the temperature at a maximum of 95 degrees Celsius or so (and
not 125 as you predicted). In this case you have two benefits: first you get
a much better delta-temp to dissipate heat. Second, you'll get forced
ventilation of the heat-sinks if you can put them close enough to the
cooling fan.

PS: I have a peak-and-hold injector driver board that's currently doing a
good job under the hood of my car. I use rather small heat-sinks on each of
the power transistors (bipolar darlington, by the way) and they do not get
very hot. When I designed the system I didn't make any calculations to see
what total power dissipation would be. Giving it a new thought now, I assume
that my system works well because most of the time the injectors open on a
duty-cycle that's much less than the maximum possible and only when you need
full power they'll open the maximum time (in my case, if too much fuel is
needed they'll just remain open and not close at all)

Hope this helps.
Jose Rublescki


>  I am designing an injector driver module. I would like your input, to see
if my calculations are correct, particularly in my assumption that all eight
injectors could be on at the same time, for up to 85% duty cycle. Thanks in
advance for your answers!
>
>  The current specs I have are:
>
>  Eight injectors.
>  Injector coil resistance: 1.5 ohms (it could also be 2.3 ohms)
>  Injector current: 4A peak, 1A hold.
>  Duty cycle: up to 80..85%
>  Operating ambient temp: -40C..125C
>  Operating voltage: 4.5V to 16V.
>
>  I did some calculations, to estimate how much power would I need to
dissipate, worst case, ASSUMING ALL EIGHT INJECTORS COULD BE ON, FOR 85% OF
THE TIME (is this assumption correct?)
>
>  First, I set Vbatt=14V. In order to have 4A circulating, the driver's
output voltage would be (14-4*1.5)=8V. Injector's coil voltage = (14-8) =
6V.
>  Instantaneous power at the driver: (8V)*(4A) = 32W. Instantaneous power
thru injector's coil: (6V)*(4A) = 24W.
>  The 4A will only circulate for a brief period of time, so average power
over one cycle will be much less.
>
>  One amp, sustaining, automatically means (14*1)=14W have to be dissipated
between the injector driver and the injector coil (100% duty cycle).
>  In order to have the 1A sustaining, the driver's output voltage would be
(14-1*1.5)=12.5V. Injector's voltage =(14-12.5) = 1.5V.
>  Inst. power at driver: (12.5V)*(1A) = 12.5W. Inst power at injector:
(1.5V)*(1A) = 1.5W. (total=14W, as predicted).
>
>  Thus, since the Duty Cycle can be up to 85%, and disregarding the initial
4A power peak, and the switching power losses, then the average power for
each injector driver is (12.5W)*(0.85) = 10.625W. The average power for each
injector coil is (1.5W)*(0.85)=1.275W.
>
>  If all 8 injectors could be on 85% of the time, then the maximum average
power at Vbatt=14V (again, disregarding some power losses) would be
(8)*(10.625W) = 85W at the injector driver side. Obviously, at 16V the power
dissipation would be higher (98.6W!).
>
>  As you can imagine, dissipating 98.6W+ into an under-the-hood ambient of
125C, with no expected air flow, is a formidable task. The heatsink's
surface temperature should not be allowed to go above 150C; that would mean
the thermal resistance from the heatsink to ambient would be (150-125)/98.6
= 0.25C/W. I can not see any feasible heatsink doing that, without the need
for forced cooling.
>
>
>