[Diy_efi] Re: peak-and-hold injector driver ic's

[Klaus Allmendinger]

Hi Marcell,

1. The 40106 runs at 5V, not 14V, but according to specs it can run as high
as 18V Vcc (MC14106B).
BTW, a jumper over C1 would make it work for saturation type injectors.

2. The emitter follower design does work. I use it on a PWM controlled fan
motor in my car. The motor (max 10A) is switched at 20kHz with a logic
level MosFet. The npn/pnp transistors are simple 2N3904/2N3906.

3. The problem with PWM limited current is the same as in all designs that
don't have a serial resistor for the flyback dissipation: The switchoff
behaviour is slow because it takes time for the stored energy to dissipate
with a small resistance. It is actually worse because the max current
during the ON time (at the trailing edge of the last PWM pulse) is higher
than the averaged hold current. If you put a resistor in series with the
shottky in your design you can improve the switchoff time at the cost of
lower efficiency, higher voltage load in the Mosfet and faster required PWM
frequency. Also the current sense resistor of 0.47Ohm increases the turn-on
time of the injector. The voltage across Rcurrent is essentially subtracted
from the available voltage to bring the current up to snuff (I = U*t/L).
(There just ain't no such thing as a free lunch) ;-).

Regards,
Klaus


At 11:41 PM 3/10/03 +0100, you wrote:
>Hi, 
>
>Klaus Allmendinger writes:
>> be wrong in something as simple as the attached design?
>
>simplicity: exellent 
>
>I also thought about the emitter-follower design
>driven by a CMOS gate.
>But would CD40106 not burn at 14 V supply? Maybe a diode
>at GND and another at VCC can effectively decrease its
>supply without having much effect on FET operation.
>(if it worx, my circuit can use the same and get rid of a few
>components). 
>
>> Example: 1.2 Ohm Injector, 12V battery voltage, 1A hold current. R-Limit =
>> 10.8 Ohm. 
>
>So in hold mode efficiency is about 10%. Much worse than
>with pulse-mode flyback current limiting.
>Adding those 8 simple components (sum ~ $1) worth
>the efficiency, I think: 
>
>http://x-dsl.hu/genboard/injector/ 
>
>I uploaded injdrive2.pdf
>(thanx to Alexei Pavlov for his valuable comments) 
>
>injdrive and injdrive2 are very similar.
>There is a small difference around Q4.
>Some will like one, others the other. injdrive2 has a little higher 
>dissipation, but
>setting values might be simpler since margins are higher. The same PCB can 
>be designed that can be populated
>either way. Look at injdrive.txt for more info. 
>
>Marcell
>-- 
>+36 70 2 654 789 
>
>_______________________________________________
>Diy_efi mailing list
>Diy_efi at diy-efi.org
>http://www.diy-efi.org/mailman/listinfo/diy_efi
>
>


_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi