Mr. Midas

I've worked out a lot of bugs with my 1988 SE automatic N327 EFI system, but it is still running way too rich.



I'm going to do another fuel pressure /injector leak test, and do a couple other things to make sure it is really running in closed loop mode. I think the fuel and even timing maps are going to need tweaking for best economy and performance.





Mods on my fully rebuilt 88' NA include:



S5 rotor housings and RX-8 stat gears.



Fidanza light weight flywheel.



Pineapple Racing port sleeve inserts.



2-cycle OMP mod.



Converted to manual S5 5spd tranny.



PS deleted per Flying miata's How-to.



Custom long tube 2-1/2" collected header-- into resonator(formerly an aftermarket cat)-- into Borla dual exhaust.



S5 port-matched intake and fuel rail swap with S4 TB(modded). Custom separate rpm switching for AUX ports and VDI.



3-wire heated O[sub]2[/sub] sensor.



S5 AFM swap /w cone filter.





I was having difficulty getting it to idle smoothly in the beginning(newly rebuilt) and went through a few flooding incidents. I started diagnosing the problems and found and corrected the following:



Faulty fuel pump check valve-- Replaced fuel pump with new unit.



Clogged fuel filter. Tank had some sediment from sitting 5+ years, even after flushing.-- Went through a couple big ford fuel filters to filter out remaining sediment.



Flooding issues, violent backfires once unflooded. Traced to excessive 2-cycle oil in the fuel. After rebuild I put a 100/1 mix in the gas tank. Once the OMP was primed it began to smoke and flood at different times.-- Stopped mixing oil into the fuel, problem solved.(or so I thought)



Idle and backfiring probs.-- Faulty TPS replaced, throttle plates adjusted. Still idled high?



High idle.-- Found aftermarket catalytic converter destroyed and blocking exhaust flow.(Probably overheated and hammered to death by flooding and backfires.) Gutted cat and rechecked /adjusted throttle plates; was finally able to set Idle and timing to stock specks but...



Engine seemed not to run as smooth as it should.-- With a 5-7[sup]o[/sup] timing advance it idled and ran much better. Because i have S5 housings, or?

The Idle was still a little finicky at times and was never spot-on steady. I always attributed that to the lightweight flywheel.

I had driven around for a while after all of that and discovered I was getting very poor mpg's, but No check engine lights. If I'm working around my car while it idles, my eyes readily burn from the fumes. I figured that was normal, but the oil also appears to be getting a healthy dose of fuel dilution as well, even though the motor has great compression all around...







Round 2:



Adjusting the variable resistor did nothing noticeable. I've turned it several revolutions in the "L" direction.(check con. jumpered)



Checked the(new since rebuild) O[sub]2[/sub] sensor's output and found it reading sluggishly and when pulled found the tip very dirty.-- Retrofitted a heated 3-wire sensor which is giving very responsive readings in less than two minutes after startup. It bounces around .985v on acceleration in neutral and under deccel it hits zero and bounces back.



For the heck of it, I swapped on the S5 AFM. Lo and behold, the car has a rock solid steady idle! I have a feeling it was because the slide type MAF has a smoother output because the cone sits inside the air stream; whereas, the flapper door oscillates a little from the turbulence it creates behind it. (Some rudimentary bench testing has also brought me to this conclusion.)



After having it that way for about week, I still don't see a difference in fuel consumption. The fumes are still bad and a the oil is already getting a gas smell after a fresh oil change about week before. I'm about to go fill up and get an accurate figure on recent mpg's, but have a feeling it will be in the low teens again.



I apologize for the long-winded post. I figured it would help to get everything out on the table before seeking input. Since this was an automatic car have I just missed some connection that needs to be bypassed in order to go into closed loop? I've patched into the auto harness and have the 5spd neutral switch acting as the auto's neutral starter interlock. According to the ECU pin out page in the FSM, the auto's OD switch is the equivalent to the 5'th switch but their outputs are reversed. I've left the 5'th switch disconnected, but plan on using a relay to invert the switch's function when I do hook it up.



As far as I can tell, all critical sensors are working properly now, so it should be running in closed loop mode. Which means I need some way to lean this thing out. And since I have S5 housings with slightly different plug placement, being able to adjust the S4's timing maps would be a good thing too.



The motor's ports could still be considered "stock". Since I only ground out the exhaust diffusers and smoothed the intake ports out I figured my other add-ons would be fine.

I've been looking into Rtek's stuff. Anyone have good or bad experience with them? I'd like to have a setup with a couple different maps. One for normal driving to maximize mpg's and another set for spirited driving/autoxing.



Is there anything else out there that won't take days of rewiring?





Thanks

j9fd3s

did you put the voltmeter on the 02 sensor, while you're driving? the o2 voltage should oscilate between 1500-3500rpms.



the rtek stuff is supposed to be very good, and the s5 housings will take a little more advance, although if you have no air pump/acv then it will run rich at low rpm say under 2000.

Mr. Midas

Well half a tank has yielded 13.5mpg. And I've tried so hard to be good with the pedal(I'm glad I was tonight, cops and troopers everywhere!)



I'll check the O2 sensor while driving tomorrow. It's funny that you should mention that. The FSM calls for using the monitor lamp of the diagnostic checker to test the feed back system. The other day I disconnected the neutral switch, got the motor up to that speed, and hooked my LED code checker to the GL(yellow wire) connector of the diagnostic connector. It flashed kind of sporadically. Much lower than the specified 8 flashes in 10 seconds. At idle it flashed more often. Didn't know what to make of it.



I still have the air pump and acv... *thinking*... could there be some way it's injecting too much air into the exhaust and throwing the O2 sensor off? I know that 5'th switch does something with the emissions system. I've got to wire that up, and double check the radiator thermo-switch wire continuities back to the ECU.



I'll test the O2 sensor before and after to see if there is a change, and post back tomorrow some time.





I might be working on a race-prepped aprilia tuono 1000r tomorrow as well. Could make the 7 a little jealous



Stock image

Mr. Midas

Ok,

Got some base O2 readings.



1500-2000 rpms. Cruising in 2'nd and 3rd



Idle: average .033-.047v



Up gentle hill: .86v <->.2v



Flat road: .85v <-> .5-4v



Down gentle hill: .003-.001v--blips around a lot(hard to read) avg maybe .023v?





Will be back later(maybe much later) with some readings after I check some wires and hook the 5'th switch up properly.









I just sold my first bike like three days ago, so I have a little moneys I can set aside for an Rtek7 stage 2 chip and cheap WBO[sub]2[/sub] setup.

j9fd3s

13mpg is way low....



the air pump air should all be going in front of the o2 sensor (throwing it off) well it depends on load, if you're just in the driveway, it'll go up to 2500ish rpms, the more throttle you have the lower rpm the acv opens up.



oh you dont have the radiator switch hooked up? thats it right there, it turns the emissions system "on" if you're lazy just make it on all the time

Mr. Midas

I've tested the radiator thermo switch(works fine), but I wanted to check the wires because the alt. lamp wire went bad. And I thought I'd rebuilt the wire harness and checked it with an ohm meter before putting it back on...







Right now I'm leaning toward the idea some part of the system is causing excess air ahead of the O2 sensor. This dialog is really helping to put the all the pieces together now; I know it has to be something simple.





Didn't have time to do much more with it today. I'll be working on it tomorrow for sure!

Mr. Midas

Have run another half tank of gas through it after doing some things... This time it got 16.24mpg.





O2 readings(getting better at watching the meter while driving):



Idle: .132v - .033v[sub]AVG[/sub] -.022





1500-2000 rpms. Cruising in 2'nd and 3rd:



Up hill: .872v -.545v[sub]AVG[/sub] - .383v (Under full power load up a hill highest I saw on the gauge was .878)



Flat road: .352v - .036v[sub]AVG[/sub] - .023



Down hill: .000v - .002v[sub]AVG[/sub] - .006







What I've done:



Double checked thermo-switch wiring and function... working just fine.



When I hooked the DVM up to the S5 tranny's 5'th switch, I was surprised to see it acting like the '88 FSM says the OD oil pressure switch operates. (is the S5 switch different?) It is closed in every gear but 5'th, so all I had to do was plug one wire into the OD pressure switch connector(E-12) and grounded the other wire out



Disconnected battery - to reset the ECU.



Here's the real kicker. I feel really stupid for not catching this sooner. This whole time I had left the Split Air outlet capped off! I was very careful to cap all openings on the motor while rebuilding it(have had bad luck with mud wasps getting into the weirdest places) and it never crossed my mind to make sure the split air could vent after I put the motor in. I hope the rotary gods don't punish me to much for that. I completely overcomplicated a simple problem that I had initiated in the first place. Maybe all this time it has peen pushing the back pressure from the split air ahead of the O2 sensor giving it a false lean reading? Not having the 5th switch hooked up probably made it worse too. Since I have no split air pipe running to the cat I made up a hose elbow with a screen on the end to allow it to vent and still keep the damned wasps out.



Although i haven't had it running in the garage for long, I think the eye burning fumes have been cut back a little and the exhaust pulses are more even sounding as well.



16.25 mpg is pretty close to stock now, considering my setup... I forgot to mention I have 185/60 tires on the car right now, so I've been converting recorded miles into actual miles. A strait reading from the odometer says I'm getting 17.22mpg.



Well maybe now the ECU upgrade can wait a little longer until after I install some gauges and a stereo that works.