[Diy_efi] RE More ECU progress

[Ian Molton]

Adam Wade wrote:
> --- Ian Molton <spyro at f2s.com> wrote:
> 
>> It will take CAS and cam angle inputs, and use them
>> to compute the crank angle to better than one
>> degree. how much better will depend on how good 
>> our interpolation algorithms work).
> 
> One thing to keep in mind if we want to get the rotary
> engine crowd.  No cam, rotors have 3 "combustion
> chambers" per, and they rotate at 1/3 "crank"
> (actually called an eccentric shaft) speed.  So in
> practice, the fuel and spark events are equivalent to
> a 2 cylinder two-stroke engine (one spark and one fuel
> event per "crank" rotation).

My design can handle this as one of the chip registers is a cycle angle 
register. This sets the maximum that the crank angle value can become 
before it is reset to zero.

obvious values are 720 for a four stroke and 360 for a 2 stroke :)

 > To complicate matters a
> bit more, there is a set of trailing-edge spark plugs
> that fire substantially later.

Not a problem. simply hook up three more of the chips outputs to spark 
driver MOSFETs and tell the CPU to add three more spark events to the 
event table at the deisred advance.

> What's the status of closed-loop spark control?  Is
> this do-able under the current ion sensing setup?

Well, the timing controller can certainly handle the programming of any 
timing for spark.

the practical implementation where the CPU reads the ion data and 
processes it in order to recalculate the timings it programs into the 
timing controller will have to wait until the system is on a runnable 
engine.

The point is - the hardware design _can_ do this already.

 > I have data
> showing that it's quite possible for a typical
> gasoline-powered engine to make the same power with
> lower fuel consumption with a combination of water
> injection, spark advance over base, and a leaner
> mixture -- even under full throttle).

Nice :-)

the timing controller also would allow one to hook up injectors for gas 
/ nos / water / anything else without any trouble at all. the controller 
doesnt care what the event is for, it just triggers it.

Hm, actually, Im going to increase the size of the fuelling table to 
allow  other types of injector to be PWM'd too. Thanks for getting me 
thinking on that one. I think 256 entries should be ample...

(sheesh, here was me thinking 32 independant PWM'd outputs would be 
enough... I think I'm going to have to use staggered timing on the PWM 
clock to prevent silly loads being applied to the power source at high 
frequencies...)