SpeedPro Tuning

[Jörgen Karlsson]

> I wonder if anyone remembers like I do from past posts, about a guy that
> was using a SpeedPro, but noticed it getting thrown off by bogus
> situations like, a plug wire breaks, that cylinder stops firing, the HC's
> in the exhaust header rise, fooling the SEFI8LO into thinking it is too
> rich, so it trys to compensate, leading to a dangerous leanment for the
> engine overall?

I have had my chare of broken plug wires, it haven’t caused any big problems
for the speedpro. That is strange since the revlimiter makes a mess.

Detonation on the other hand is a big problem, the reading are all over the
place when that happens. If the o2correction is big when this happens there
can be a nasty oscillation. The biggest problems occur when the next shift
occurs before the oscillation has ended.

When we had a problem with overrun on first gear there was often a
detonation when the second gear dropped in (what do I expect with tree
trigger pulses per rpm?). The overrun often caused the wrong mixture, then
the detonation came and messed things up further. The oscillation continued
for a second or so, but third gear came in before the oscillation was over
most of the time. If the mixture was lean then there could be a detonation
right after the shift of the same reason as there often was a detonation at
the first to second gear shift.

For the o2 to work the rev limiter must be soft touch fuel, since sequential
injection is needed for this and I don’t think that even the sequential
speedpro can do this there is a problem. The revlimiter must not be tripped
at any cost. If it is you must run open loop. We did some experimenting with
an antilag system before we installed the speedpro, when we first activated
it after intalling the speedpro the engine didn’t respond as planned ;) We
just weren’t thinking.

One other problem is that the speedpro assume that the current o2 correction
is correct after the shift too. Lets say that we blindly follow the
+-7%rule. At 4000rpm the correcton add 7% fuel and at the shift at 5500 it
subtract 7%. After the shift the ecu will make the mixture 14% to lean, that
is not good. As long as there is not a detonation because of this the
mixture will be back where it belong a quarter of a second later, but if it
detonates we could have a problem.

I think that the engine acceleration is the problem, when the car shift more
then once a second closed loop does not work that well. I will probably tune
a turbo civic next month, I wonder how it works with an engine that is
extremely sensitive to mixture variations. A manual transmission will also
be interesting.

Jörgen

----------------------------------------------------------------------------
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