late Pathfinder or Quest OBDII ecus on L eng?

[Tony D]

The OBD2 ECUs, ALL of them have a 'short term and long term fuel trim' feature that allows for self-adjustment of the ECU fueling over a FAR broader range than the old systems did.

 

Make no mistake: This is NOT a substitute for a proper fuel mapping in the ECU. If you have a sensor failure, you will drive home on the map you have...NOT the fuel trimmed maps!

 

What the fuel trim is intended to do is accomodate for various wear items and things that would alter the efficiency of the engine to an extent that would put it outside the 10% range of the original Bosch setup.

 

GM has this on their ECU's as well. They all do it. Even aftermarket ECUs are now employing this technology to allow for a 'self tuning' situation where the map is sent with a basic startup map, and you just go out and drive the car over varying conditions...with you able to burn the map at the end of the run, and then repeat the procedure till you have a complete and accurate map without going and individually inputting cels and going back each time to change them manually.

 

The aftermarked ECU's are getting increasingly spohisticated and CHEAP. The one I am referring to above was being sold retail for $1200, complete with Fuel Pump and the ability to run as a non-return line fueling system up to 500HP as they controlled pump speed through a PWM controller resident in the ECU working off a fuel pressure and temperature sensor in the single fuel feed line to the injectors.

[Daeron]

Okay, I went to the U-Pull and Pay on Benoist Farms today (the place that used to be called the Auto Recycle Center) and there were a few Quest minivans that I eyeballed... One of them had a loose exhaust manifold, and I pulled it forwards. The cat had been cut off, but it was a funky shaped pipe that bolted onto each bank's manifold.. it looks almost like a lopsided stethoscope. After the two banks merge, there is ONE O2 sensor bung, and then the pipe was cut off so I presume the catalytic converter was mounted roughly between the two front seats... so the Quest at least uses a single O2 sensor for both banks.

[jeffp]

I have run a Z31 box single coil system, and that unit has difficulty running power over 500Hp after 5K RPM's I also did not like the single coil with a turbo application. I tried all of the common stuff to stop misfires, and I am sure there is an answer, but I got sick of spending money on ignition components. I do think the MSD 6A and the MALLORY Hyfire coil is the best setup.

I also have installed an run the VE30DE twin cam unit. I liked that unit and still do, but it is a ytpe III board and there is no eprom to reprogram like on the Z31 box. The variable cam timing output is really cool, because you can adjust for RPM and for LOAD, when the output is enabled. Well that workes out VERY well for a secondary pump for HIGH boost applications. Very nice option. The CAS will fit into the stock 280ZXT distributor, and if you can get the ( A ) build maxima distributior (OR CAS) it has a steel lid that you can use on the 280ZXT distributor. That was the mitsibishi built unit for the maxima, wonder why they changed?

The maxima has ONE O2 sensor and it is the Titania type O2. That was also cool for the L engine.

There are some tuning issues and lost tables that have made me go to the next setup the Z32 box. if the tables I need to have acess to are found, then this box is ideal for the L engine.

The Z32 uses the same CAS, turning the same direction, and after you have your wire harness built, the maxima and the Z32 boxes are both plug and play. so you can swap them out with no additional wiring changes.

As you know the Z32 box has been hacked to greatest extent, and there are guys running 100Hp with the unit. I am now running a 1990 N/A Manual FED box I got off of ebay.

The Z32 does two O2 sensors, both the Titania type sensors. The wiring from the maxima is the same for the left bank O2 and the wiring for the right bank obviously is not wired in. That is the big difference in the two. So there are two options I think that are the best ones. Since the O2 sensor is a resistance based output, then it would not be very difficult to use a decade resistor, with the help of consultant, to set the right bank O2 at a FIXED 14.7 AFR with a resistor. Then you are free to use the single O2? I have some people I can email about that question. Then there is the other option of just turning the O2 off in software all together, this is the option I have used. All of the other fuel and timing related tables work, and it is very possible to get an excellent tune and run without O2 sencing all together. I do want to look into building a parallel input circuit and use the O2 in the future, but for now it is off.

I still have the variable cam timing output that is the same output as the 92-94 maxima VE30DE box, so I still have the option of a secondary pump enable based on RPM, and LOAD, now that is slick.

The single coils will give me any specified length of fire at the plug (2 milliseconds or less) that will still fire, with the right coils under HIGH cylinder pressure.

NOTE: with single coil nissan systems like the M30 and Z31 boxes, at 7500 RPM, your effective spark duration is about .9 milliseconds or 900 microseconds, and that is very near the failing point of the ignition under high cylinder pressures like a turbo application. I know this to be fact, that was one of the problems with the Z31 boxes, and I verified that on the bench.

 

Example, I was trying to get the car going the first time. I ended up flooding the car so bad that I filled the bottom EGR portion of my stock manifold with fuel, enough to RUN for 10 MINUTES with no fuel pump running. The car was running 10:1 and 11:1 AFR's the whole time @ 750 RPM. The plugs did not foul out, and I am still running them right now. I am not saying they are good, but they run the car because I have a hefty coil feeding each plug.

 

I wired the maxima and the Z32 box into my car using all of the stock connectors to the 280ZXT that the ECCS harness plugged into and built my own EFI harness. The system looks bone stock Nissan equipment and it runs very well. My problems with fuel management I believe are solved finally. I am in the tuning stages, and by the time I am done this will be a fully intergrated Nissan system for the L engine, works for me.

The last thing about the Z32 ECU is that it is a type 2 box, so it has a removable eprom, if you can solder good, so when you are done tuning, pop the reprogramed chip back in and you are done.

I did do a couple things with both of the boxes. I put a 500M resistor across the Knock sensor input to stop the ECU from faulting the knock. I also put a 1K resistor on the fuel temperature sensor and fixed that value to 44 degrees C. I dont have ANY trouble codes now from the boxes. Good to go, tune away and be happy.

Also, after looking at all the BS ford and GM have done to their EFI systems, you could not get me to run one if my life depended on it. My opinion is that they all are JUNK, and the engineers that developed them have bandaided them to make them run soso, but pass smog! I think of the two, the ford box is much better then the GM unit. Any design engineer that believes he needs to have MAP based sencing, AND MAF based sencing all in the same ECM is out of his mind! and the MAF is a frequency based output, totally STUPID!

anyway, the whole thing cost me less then megasquirt. The cost was getting all of the wires for the wireharness and the blue connector terminals, I got racked on those parts, but it sure is nice.

[HowlerMonkey]

I tested, repaired, and engineered ecus for standard motor products as well as supported them and my findings are that the GM and chrysler ecus are much more advanced......in operation than any other brand but they suffer from cheap components.

 

They have learning abilities such that a customer who's replaced a sensor complains the car runs worse with the new sensor or iac.....etc.

 

The reason for this is the moron mechanic didn't reset it and the computer was using the learned values of the skewed sensor readings of the failed sensor.

 

A simple memory reset sends them down the road happy because they weren't saavy enough to realize that a few drive cycles or a reset are what they needed.

 

A computer expecting said skewed signal will just not run right until a few drive cycles or a reset but these guys who call up saying standard sold them 4 bad sensors just lack the skill requiored for his chosen line of work.

 

These same guys will put 4 ecus in a car hoping the ecu replacement will be the magic bullet and are befuddled when all 4 exhibit the same symptoms.........then they call up saying we sold them 4 bad ecus.......morons.

 

That said.....the gm and chrysler ecus are smarter than the techs working on them....in many cases.

 

These ecus will run on signals that would choke a ford, nissan. or toyota ecu.

 

Believe me, I spent the last 3 years fixing up to 20 cars a day over the phone as well as managing the rebuild facility, pioneering new test regimens in the shop and defending our products to technicians who don't belong in the industry.

 

One guy called me weekly for 11 months straight on the same car through multiple ecus and never did fix the car.

 

I faxed him every thing possible until I finally had enough and told him I would fax him something that would fix all of his problems.

 

So......I went to burger king, picked up an application for employment, and faxed it to him.

 

I now suspect that fax may be related to my firing last month.

[jeffp]

I will agree with you that MOST of the techs working on a newer car really trouble guesses the problem, that along with the shot gun approach to repairs. I have seen this many times, perts that were replaced, charged to the customer, and it did not fix the problem, ESPECIALLY WITH electronic EFI systems.

Nissan boxes are advanced with all of the functions such as memory and check summing, or a learn function. These features came out in the early 90's with most of the manufacturers.

Nissan and BMW now have variable cam timing adjustments for performance and fuel economy. I haven't seen any fords or gm's with these types of features in their systems.

I am use to the Nissan method of calculating fuel and to me it seems very strait forward.

I just got a bad taste in my mouth with the later 80's ECU's from them. This was the transitional time between OBD I and OBD II systems and boy what a mess, at least for the few cars I worked on.

So after looking at all of the sensors involved with a few systems, I decided this was not an option for my car. The hardest part of the whole thing was getting the harness built correctly and mounting the CAS, but other then that, all of the sensors I had stock still work on the newer boxes. There were added improvements like the TPS resistor, along with the TPS switch that indicates off idle. The coil on plug is excellent and drives just about any coil you can find to fit the application.

So for me, I will run with that for a while. I have done a few installes of various ECU's testing them with the L28 and I have found the Nissan solution to be the best fit to the engine, and the car.

anyway

[Daeron]

One guy called me weekly for 11 months straight on the same car through multiple ecus and never did fix the car.

 

I faxed him every thing possible until I finally had enough and told him I would fax him something that would fix all of his problems.

 

So......I went to burger king, picked up an application for employment, and faxed it to him.

 

I now suspect that fax may be related to my firing last month.

 

HybridZ Pulitzer nominee!!

[HowlerMonkey]

I was able to check out a 1997 quest with a VG30e single cam engine and it had the post cat 02 sensor.

 

When money permits, I will experiment with this ecu on the L28et.

[Tony D]

Sounds like you and I run in the same circles. "I checked everything, and it's all checking good."

 

Book a flight, fly for 18 hours. Stay in a hotel, drive 4 hours to a site in the middle of nowhere...start at square one of the diagnostic regimen...and usually within a couple of easy steps I nail the issue.

 

I then usually get to spend three more days 'proving' why I fixed it, because 'there is no way you could have fixed it that quickly, Qwak is a Masters Degree in Electrical Engineering it must be something different as he was telling us!' Finally it's accepted that I DO know what I'm doing, and I can go home. A week (or more) at exorbitant rates, a roud trip plane ticket, and 36 hours in the air all because a guy can't use a VOM or follow precise testing suggestions.

 

On the way out of the jobsite, I have to restrain myself from yelling at the top of my lungs as the guy repeats AGAIN for the millionth time how hystersis looks just like these symptoms: "Yes, theoretically hystersis in the input filter circuit could cause a problem, but not in this case----this time (and most times) it was condensate filling up the transmitter and causing it to fail intermittently and give bad output. If you had more time behind a digital VOM and less time behind his master's thesis, perhaps you would have understood this when I tried to explain it lo these past several months. If I tell you to WIGGLE the damn thing while standing on one foot balancing a plate on your head, you don't ASK WHY, you just damn well DO IT and tell me what the f-ing meter reads!"

 

***************

 

Thankfully, I deal with foreign customers for the most part now. Outside of the USA. I don't have to fight with the ludicrous arguments I used to do when I only handled domestic 'distributors'.

 

I feel for ya when you say "tech's that should not be in the business"...

[eec564]

Tony, you're absolutely right. People all over can't preform proper diagnostics. I see it constantly at automotive shops where their only solution is to replace a part and see what happens.

 

I worked on a CNC lathe that wouldn't run, and upon checking the registers, the bit for door closed was a 0. Checked the door, and it wasn't fully closed. Had to slam it to get the sensor to catch. A readjustment, and everything worked smoothly, no slamming shut access doors any more. That was a machine shop that had a strict policy of outsourcing all their diagnostic, repair, and maintenance work. They'd had employees try to fix things in the past that cost them far more than paying a guy to come in a shut the door a couple times a year.

[HowlerMonkey]

The saddest part is that the dealership pay structure reward parts changers and penalizes the true diagnostic technicians.

[Tony D]

I almost went apopletic on a national service manager who offhandedly said "that's the cost of troubleshooting" when I asked about the customer paying for a $2800 Printed Circuit Board alone (which we called in the first place---it was a broken trace with a capacitor bleeding down) but the FACTORY said we FIRST had to replace two different power supplies and a motor transition timer to get to that point!

 

Customer was NOT happy.

 

Neither was I, my reply was "That's not the cost of troubleshooting, that's the cost of replacing every part in the system from one end to the other until the most expensive part is the only thing left!"

 

They then decided it was time to review all my warranty claims. Which I was fine with...found out my boss was spoofing them up crazily and raping them for extra money.

 

About that time...I quit!

 

I hate unethical people, but incompetent unethical people really set me off!

[Pharaohabq]

So anyway, back to the ECU's. did any of you get to try one on an L28 yet? I know I could make it work with a GM ECU, but that's not the point. One of the Nissan ECU's shouldn't be too much more difficult as they require basically the same inputs, MAP Temp, etc. though you'll need to duplicate the timing signals for the CPS. Usually this is done off from the flywheel, and unfortunately isn't often the same 36:1 wheels that could be used for MS or EDIS. though, On here, Derek has made a bolt on timing wheel for the L28 that could be modified to show the appropo timing signal. Some systems also use a Cam sensor to detect where the distributor is, but on a newer engine with coilpacks you wouldn't need it. NDIS runs off the CPS. So setting up the Crank Timing wheel, and CPS will be the hardest part of this project.

 

OBD-I or OBD-II are really very similar, mainly it's just that 2nd O2 sensor in the exhaust, which any shop can put in for you. tho you may need a tube to fool the comp to say there's a cat in place. OBD-1 doesn't have that, though Nissan used Consult and thier own standed prior to 96.

 

Phar

[HowlerMonkey]

You could modify a 1981 280zx turbo pulley to duplicate the z31 style signal (wider slot in cylinder 1 than the rest) but then you would have to rely on a part that hasn't been made in many years.

 

Next is to check the distributor disc on the obd2 ecu vg30s single cams and hunt around the engine for any type of crank reference sensor.......maybe even today.

 

If the waveform is the same as the Z31 distributor waveform then it should be gravy.............unless there is a true crank position sensor on engine.

 

Off to the junkyard.

[HowlerMonkey]

Ok........back from the junkyard recon. mission.

 

The wheel in the distributor has the same slotting as earlier distributors and the same connector so it seems the distributor signal will be identical.......However........

 

The OBD2 quests do indeed have a crank angle sensor on the bellhousing just behind the front exhaust manifold and covered with a heat shielding material.

 

This is because the obd2 standards and other EPA regs require a car to go a damn long way with zero maintenance.

 

Since the distributor relies on a good timing belt, you just cannot meet the regulations concerning mileage traveled with no maintenance and still have good crank reference because of belt slop.

 

They get around this with the very exact reference from a bellhousing mounted crankshaft position sensor and they also get a degree of redundancy in signalling.

 

My main reason to try this project is to be able to flash the unit with calibration files simply by plugging into the OBD2 port as well as get bettter control on emissions since I may very well be moving to a state that has tight emissions standards.

 

Money prevents me from going further with this other than trying to see the orientation the "tone wheel" (probably on the flex plate) and how to get this same signalling out to the ecu on a L28 or earlier VG30 with as little effort as possible.

 

If I come up with anything, I will add to this thread.

[Daeron]

So basically you are saying that unless you find an easy way to transfer the signalling apparatus to the L-flywheel, your efforts on this front are pretty much stalled?

 

Don't forget that Jeeps use a crank sensor in the bellhousing as well.. Certainly wouldn't be as likely to adapt as the Nissan piece but you never know.

[slown280z]

If it is for redundancy, can it be reprogramed to ignore the signal, or does it even matter?

I know some SR20DE Sentras (OBDII 200SXs and SE-Rs) throw a CEL if the wrong toothed flywheel is used, but as far as I know it doesn't hinder performance (the ECU "counts" flywheel ring gear teath).

[HowlerMonkey]

Thanks for the "counting ring gear teeth" information.

 

I will spend some time counting teeth or at least look it up somewhere to see what can be done.

[Pharaohabq]

Yeah, GM counts the teeth on the flywheel, but theres kits to swap the CPS to the front of the crank on the harmonic balancer. I did this on my Jeep, www.projectjeep.com

 

There's no reason you shouldn't be able to set something similar up on an L28 engine. Again Derek, here on Hybridz, already made a Harmonic balancer adapter for EDIS / NDIS, it should be as easy as modifying the tabs on a wheel to match the Nissan pattern on the flywheel.

 

The CAS is what I was worried about, but if the Z31 distributor, or the tone wheel can be made to work on a L28, then you won't have any trouble. I'm not entirely clear on what's holding you up? the CAS or the CPS?

 

Phar

[HowlerMonkey]

What's holding me up?

 

Cash...........I have none........at least none to spend on anything other than food lodging and other necessities.

 

This is why my infiniti M30 with L28et sits needing nothing more than a driveshaft to become mobile.

 

I lost my job as general manager of a ECU rebuilder and have zero cash to spend.

 

The saddest part was losing all of the resources I had since we could test any and all function of most any ecu and scope/graph any signal.

 

About 70% of my business was rebuilding and repairing Jeep and chrysler ecus (same thing) because, while quite advanced, they are made with inferior parts.

 

If there's any thing I can give to the world from my experiences there it's telling everybody not to buy a Cirrus/Stratus or another chrysler with the 2.5 liter V-6 and that 90% of perceived ECU troubles are not the ecu.........unless it is a jeep/chrysler or mitsubishi ecu and even then it's still near 75%.

 

I did compare the flexplates from both a 1991 infiniti M30 and the L24e from a first generation maxima and both have 120 teeth.

 

If the nissan quest has 180 teeth on the ring gear, it's signal should be identical to the distributor or the 1991 280zx turbo crank angle sensor signal.

 

It it has 120 teeth...............then things will be a bit more complicated and require physical fittment of that sensor to a L engine bellhousing.

 

What I need to do tomorrow check how many teeth the are on the flexplate of the OBD2 quests as well as count pole pieces on the sensor which will tell me what the waveform will look like.

 

I wish I had access to the waveforms that the OBD2 ecu expects to see from the crank angle sensor.

 

I had all that information available to me at my former job.

[Pharaohabq]

ooh, I think you're being a little misled. It's not the teeth on the ring gear. there's a timing ring that's usually around the perimeter of the flywheel. The number of teeth vary, but usually aren't more than 36.

 

It would be nice to have that kind of diagnostic equipment. it's too bad about the job. there ARE other jobs out there, just likely not the same type. anyhow it's the project we're talking about.

 

There's a ton of OBD-II information on the net.

 

Phar