mx5cartalk.com

i *think* the NB6 ecu controls the alternator, you might need to swap in an NA8 alternator.

bpt4w wrote:update , the car has been wired up with the reluctors as needed but with no reference on timing points for the computer it doesn't look like it will be easy as i thought. If anyone has some info for sync events ect on aftermarket computers please post it up.

If you haven't already, it might be worth asking the question on the UK MX5 forum as they got the NB6 there. Good to hear it coming along - best of luck!

bpt4w wrote:oops almost forgot , the 1989 1.6 cam/crank angle sensor puts out two signals one for the crank and one for the cam (or sync signal) , which is a white wire and a blue/yellow stripe wire respectively (you can find them easily looking on the back of the sensor plug and then into the computer) .

I've been finding it hard to find good info on this - so the na6 are all an optical sensor putting out a 5v square wave? and there is definitely a reference and a sync wire (ie: reference will be rising or falling edge of the square wave, 4 times a cycle / sync will be the once per cycle just before cyl 1 reference)? assuming all this is correct, does anyone know what the stock ecu listens to, rising or falling edge?

cheers.

to answer a couple of my own questions:

from http://www.aus-cartalk.com/tech/MiataFuelandEngineElectronicSystemsInterchangeGuide.htm

" Cam Angle Sensor

The cam angle sensor was upgraded in 3/95 for the new ODBII electronics. The cam angle sensor mount and cam drive is the same from 1990-1997, what this means is that a 1.6 sensor can be used on a 94-97 1.8 engine for the purposes of a engine swap using the 90-3/95 ECU. Starting in 1999 the cam angle sensor is gone and replaced with a crank angle sensor mounted behind the crank pulley. The sensor versions are 90-3/95, 3/95-97, 99+

The '90-95.5 CAS differs in construction to the 95.5-97 CAS. Both are Hall Effect but the earlier CAS is an optical unit that reads holes punched in a disk while the later unit is magnetic based. In theory the later style is more accurate and offers potentially less spark scatter. In the Miata the ignition trigger signal rises 70 deg before each TDC (the 4 equal holes) and uses the other two to determine cylinders #1/4 and #2/3 (the primary and waste spark). "


It looks like being a wasted spark setup, and having two injector banks, the computer doesn't actually know which cylinder is firing out of the two that are getting spark. looking at the disc in the optical cas (na6) it gives the ecu 2 sync signals. if you didn't mind not being able to run a stock ecu, covering (filling or whatever works) one of those reference points and adding two wires to make all the injectors independent would give you sequential injection.

picture borrowed from http://www.miatapower.com/cas.htm

I'm surprised that a vendor hasn't released a 10" timing advance replacement crank plate. Isn't it just a laser cut disc with some magnets?

I've plugged in the wolf today and i'm having trouble getting good signals from the cas to the wolf... seems the pick up the sync signal when the signal is set to 'very low' - i measured the voltage and it looks like it's gonig from 0v to 1v - not what i expected.

Checking the plug going to the CAS there are 4 wires - a ground, a 12v (gets this with ignition on, not from ecu), then the two signal wires. those two wires read 0v when the wolf ecu is connected (obviously, as they are inputs) - but reading some cas tests a way of testing to make sure the wiring is all correct with factory ecu, is to check those two wires at cas plug, and they should both have 5v.

so the factory ecu supplies 5v to both those wires and the CAS pulls them to ground? I'm guessing this is all taken care of with the plug and play wolf, but i'm wiring mine in from scratch, does that mean i have to supply 5v from the wolf (or from a regulator or seomthing) to those wires?

any help appreciated. I don't think i have any hair left to work this out on my own after spending the day chasing what turned out to be dry joints in a lpg controller on a slant 6 valiant. arghh!

edit: found some info.

http://rivercityroad.com/garage/cas_testing.htm

bpt-4w wrote:now this mod will work for all the 1.8's up to 2002 which is where the vvt kicks in and you will have to sort that one out (i will try maybe in a few years), people with the variable intakes will require the computer to use one of its outputs to activate the valve and tune accordingly.

will take a few photo's of the wiring i have rigged for the conversion of the old wiring to the new sensors when i get it back.

thanks for sharing & look fwd to the pics - good to hear you're making progress