Hi all,
A month or so ago I emailed a rant about the issues I was having with
the RetrotekSpeed EFI system I'm running on my Tiger. Since then I did
some testing and determined that most if not all of the problems were
ignition related. I had consistent misfiring at about 1500-1800 RPM, at
3000 RPM, and above 4000 RPM, with the system set up as suggested by
Retrotek - except that I was using the Ford TFI module to drive an
MSD6-AL.
I had confirmed that the problem at 3000 RPM was definitely due to the
Retrotek missing one of the eight distributor timing pulses... Ford
makes one of the reluctor teeth in the TFI distributor narrower so that
it can recognize the #1 cylinder and do port-timed injection in the OEM
application. For whatever reason, this pulse was coming in at slightly
lower voltage than the others, and the Retrotek computer wasn't always
recognizing it, and then it wouldn't send out a corresponding pulse to
fire the ignition.
I didn't have this problem at SUNI, because I wasn't using the TFI
module there... I'd done my own driver circuit to interface the PIP
module in the distributor to the EFI input, with a manual switch to
bypass the computer for starting. So I went back to that driver circuit,
and used a small relay triggered by the 'start' position of the ignition
switch to handle the switchover duties. This got rid of the 3000 RPM
miss but still left the 1500-1800 RPM problem and later I discovered
that it also developed a miss above 4500 RPM (I don't often play up
there right now).
I'd been very careful to route high-power wires for the injectors and
for the coil well away from the sensor wires; this is a common problem
in aftermarket EFI installations, but I was pretty sure I wasn't having
a problem there.
I had noticed that when I was using a timing light to check the timing,
occasionally I'd get indications that were way out of whack, then the
timing would jump back to where it was supposed to be. This can be
caused by crossfires, but after I replaced the plugs and plug wires I
still had the miss.
I finally decided to bite the bullet. Took the old Frankenstein
distributor (1969 casting with matching lower distributor shaft,
first-generation Duraspark reluctor upper shaft, welded and ground
advance plate, and vacuum advance unit) off the shelf, pulled off the
broken distributor gear, pressed on a new steel MSD gear, plugged the
distributor into the MSD, rewired the computer to take its RPM signal
from the MSD tach output, and bam! Instant success. No misfires, no
hesitation, and it pulls smooth from 1000 RPM in fifth if I want... too
bad I had to go all the way back to a mechanical distributor to get it
there.
And yes, I do recognize that it's possible there was a problem with my
earlier setup that may have contributed to the miss. I am running a
"tiny" 55 amp alternator, and the computer and MSD are both wired to a
bus bar that is 10 feet (cable-length) from the actual battery. These
things put together make the 12V supply rail a little softer than some
computers will like. But an automotive control should be robust enough
to handle some amount of garbage on the 12V circuit... for now I'm happy
to have a smooth running Tiger and we'll see if the winter brings me a
different engine computer.
Theo
_______________________________________________
Support Team.Net http://www.team.net/donate.html
Tigers@autox.team.net
http://autox.team.net/mailman/listinfo/tigers
http://www.team.net/archive
|