DFI revisited

[Howard Chu]

I've had some time to re-think my previous approach. One of the problems I'm
trying solve by going to direct-fire ignition is insufficient dwell time at
high RPM. My previous design doesn't solve that problem. I started looking
at how I could rig up a timer with my switching logic, but I think I'm going
to give up on the switches altogether and just use a PIC16F84 or something.

And, having made that decision, I now realize I don't need to buy any new
sensors at all; the stock setup already has two CKP sensors and a CID
signal. I just need to tap off the CID and one of the CKP sensors to keep
track of which cylinder needs to fire next. The control logic will be
something like this: run a countdown timer set at a 5ms interval. If the
timer runs down before the next ignition pulse from the ECU, just activate
the current coil by echoing the ECU spark signal to the coil. If the pulse
comes before the counter runs down, then we have to start charging the next
coil while the current coil is operating.

So, the questions - has anyone done something like this before? It looks
like I need 3 inputs - CID, CKP, and SPARK and 6 +5V coil outputs (These are
LS1 "smart coils", remember). Any recommendations on how to do this with the
absolute minimum number of components? As another alternative - how smart
are these coils anyway? If I always tell them to dwell for two cylinder's
worth of time, will they operate correctly? (Someone please correct these
specs if I'm wrong: I believe the LS1 coils are 1Ohm resistance, charge time
of 4.4ms. I'm working with a maximum engine speed of 7500rpm, or 2.667ms per
cylinder. I guess the practical limit is 7000rpm, or 2.857ms per cylinder.
And with idle at 650rpm, 30.769ms per cylinder.)
  -- Howard

----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org