late Pathfinder or Quest OBDII ecus on L eng?

[HowlerMonkey]

I've seen threads on using Z31 or a hard to find M30 ecu on the L engine but none on anybody who tried an ecu out of the last couple model year cars that came with the single cam VG engine such as the late 1990s pathfinders, pick-ups, and quest vans.

 

If one would go through the trouble for early consult capable ecus, then whatever capability they seek would be much better fulfilled with the above mentioned ecus......right?

 

Just wondering.

 

I'll try it myself and report my findings, I guess.

[Tony D]

JeffP and Bernard have been using Maxima ECUs from the VG30DET---which is coil on plug (I want to say 92-94).

 

Thing is you need to find an ECU that used only ONE O-2 sensor on a V-Engine, and that's pre OBD2.

 

If you want to put two sensors in a divorced set of header collectors, and then run a dual exhaust with four catalysts for full OBD capabilities....feel free, but I'm not up on the configuration of all that mid-level engineered stuff. I could tolerate the divorced headers and two O2 sensors, but the full on OBD quad sensor dual cat systems really won't be giving you anything more than what the earlier systems give you in terms of flexibility.

[HowlerMonkey]

In the test profiles we use at standard motor products with our ecu testing equipment, we usually just spoof the post cat sensor signal to the ecu in that the ecu only wants to see the very slightest of changes to prove to the ecu that the sensor is indeed working. Many of the tests we do have it at a steady value and the ecu doesn't complain at all.

 

We also send the post cat signal matching the upstream sensor signal so we can whether the catalyst moniter in the ecu catches that the cat is not doing it's job.

 

So it seems many ecus only moniter the resistence of the heater circuit and whether the cat is working.

 

You can probably just spoof the heater coil resistence and get away with sending a steady signal in the range one would see from a sensor behind a working cat.......which normally is steady with near zero variance.

 

I'll venture a bet it can be done with a resistor with the expected value on each circuit.

 

I'll look into it more since it would be a lot easier than having to spoof all the signals one would need to spoof on a DOHC ecu to keep it happy.

[woldson]

The guts in my saab's cat are awall.

I was told to get a 5m resistor to clear the check engine code.

I do believe you are correct in your assumtions.

[Tony D]

Oh, I'm sure spoofing the rear O2's is the way to go...

the real question is about the pre-cat O2 sensors. It's fine for a header system if someone wanted to separate the L engine into 'two banks of three' but for a turbo application, short of Twins... it may be difficult to reconcile the twin O2's in the circuitry if they get a bleedover pulse that doesn't jibe with the rpm indicated for operation---false spike for cylinder firing---am I making any sense in what I'm trying to say?

 

At least with JP and Bernard, the single O2 box seems to be more (er...) in-line with the requirements of the L-Engine...I don't believe there is as much spoofing of signals as you would think. Drive the cam position sensor off the old dizzy drive and you're most of the way there with a pre-made housing from an old ZXT. Jeff mentioned several models he looked into where their camshaft or crankshaft sensors rotated the same way as the OEM 280ZXT dizzy arrangement, so they were 'the easy ones to use' (his definition of easy, not mine!)

 

BWAHAHAHAHA!

 

Seriously, though...

[jeffer949]

They make O2 sensor eliminators that you can buy. My friend has them on his twin turbo Audi S4 that has no cats. and he isnt getting any check engine codes from it.

[Tony D]

Like mentioned, the post catalyst O2 sensors aren't the issue.

It's the dual O2's BEFORE the catalysts that adjust fuel mixture, and how the ECU reconciles the signals given from each bank and adjusts accordingly that may be the issue.

 

This is why JeffP selected the single O2 sensor version. Sure, it's Consult and not OBD2, but will it really make that much difference in practical application?

 

Now, if you were fitting a GM ECU to the inline Nissan, I'd go for the OBD2 boxes, they have very sophisticated self-tuning subroutines and are already thoroughly aftermarket cracked for any adjustments you would care to want to make!

[HowlerMonkey]

I wish I had spent more time with the simulators so I could reverse engineer things that would benefit us here but the president of my company needed a scapegoat to sacrifice to appease angry non-technical managers at carquest who couldn't figure out that six ecus sold to the same tech for the same car were indeed not bad since all six exhibited the same symptoms.

 

Amazing how you can get in trouble for a customers lack of understanding but I am taking small consolation in the fact that they had to hire four people to do what I did for them.

 

Rant over.....back to discussion.

 

Lexus divides the 2jz into two banks so it shouldn't be too much a stretch.

 

If not for variable valve timing and changing of indexing of the cams causing skewing of timing signals sent by sensors driven by said cams, it would be easy to use newer systems on older engines.

 

This is why I mention the late single cam car models as a good place to start when looking for engine management systems that could help us with our non-variable valve timing engines.

 

Any vg30 or vg33 car made later than 1997 should easily be OBDII so I will ztart there with my research now that I have lots of free time.

[jeffer949]

Why not just put both O2 sensors in the dp and let it run that way? what would be a down fall of this? Granted both banks would be running the same but it would still work.

[HowlerMonkey]

It's possible the ecu checks whether any fueling changes made to a certain bank is actually affecting said bank.

Having the sensors of both banks together in mergred exhaust on an engine the ecu divides into multiple banks could possibly "dilute" the changes the ecu expects when making an adjustment and you could run into the ecu "hunting" the proper adjustment or just throwing a light and possibly falling back on an algorithm that is not performance friendly.

 

If one could get over the hurdles of running an ecu that expects variable valve timing, the 2jz OBDII ecu might be the ticket if you made sure your exhaust hardware coincided with the logical bank.

 

Wait a minute......did toyota make a non vvti OBDII ecu for the 2jz?

 

If so, that would be a good starting point.

 

Done now....surfing on a blackberry hurts my head.

[Tony D]

Yeah, that's what I was getting at: the way it was explained to me the ECU will watch firing events in each bank, and look for the corresponding spikes in O2 in the exhaust. If it read 6 spikes when it is registered to only be reading three...then I'm betting it throws a code due to a disparity to it's RPM -vs- Sensed Input. The rpm input shows say 1000 rpms, and due to the pulses of the O2 sensor being exposed to all six pulses in the collector would be returning spikes corresponding to 2000 rpms (2X what is 'should' be seeing.)

 

I'm sure it's just some BIN or HEX hacking like JeffP and Bernard have been doing....but if you change that, it's just one thing that's reading it. What else will have to be redone or tweaked to keep working?

 

I know what you are saying HM, I had to go audit a distributor where there were 13 Element Failures in a year. All the same frame machine, all the same element. All failed by some idiot over-using RTV and having it exude out the oil injector nozzle that sprayed on the internal gears and bearings.

 

Curiously, the failures corresponded to a former Direct Technician getting canned and going to work....guess where? We always called him 'RTV-Dan'... Birds don't change their spots.

 

Another time, a distributor 'failed' 13 elements in short order, but on four different units. Each failed just inside the warranty period, but because they were tracked separately nobody ever picked it up but me...

 

When I went in to Audit the Distributor, I found 5# of rust inside each of the machines intercoolers. The rust was passing through the second stage and failing them prematurely. Simple failure to follow the standard work procedures was directly to blame---their tech (same guy each time) was flat-rating the job and never doing the simple cleaning I did.

 

When I quit there, they lost 2.2 Million in service work the first year. But I was the 'bad guy' for leaving and taking the work with me!

 

Seems customers knew who did the work the right way...the FIRST time!

 

Each of my subsequent 'replacements' has been a resounding bozo which the customers hated. Poor them. Too bad, huh?

[Bernardd]

I don't see any benefit to doing this unless you can tune the ecu. While the obdII ecu might be able to keep the afr's ok in an NA application, I doubt it would be able to for a setup with forced induction, and then there's the timing aspect which is far more critical. The only reason for consult (in my case) is to allow nistune access to datalogging the maps to make tuning easier. I switched to the M30 ecu to get away from the prototype nistune board I was using and to have accel enrichment for crisper response.

[woldson]

I don't see any benefit to doing this unless you can tune the ecu. While the obdII ecu might be able to keep the afr's ok in an NA application, I doubt it would be able to for a setup with forced induction, and then there's the timing aspect which is far more critical. The only reason for consult (in my case) is to allow nistune access to datalogging the maps to make tuning easier. I switched to the M30 ecu to get away from the prototype nistune board I was using and to have accel enrichment for crisper response.

 

 

My brother is useing a obd1 or 2, (I"ll have to ask), and a osterage to tune his turbo honda engine he built.

He is dynoed at 305hp. I want to say he is running a b16a with ls pistons, but again I will have to confrim this.

Pretty simple/inexpensive setup to do, at least in his application.

[HowlerMonkey]

True......maybe the consult port yields enough information to.make the hassle of setrting up an OBDII ecu more hassle than it's worth.

 

I'll still check out the late nissan single cam units and possibly explore the GM units mentioned above.....or maybe even the jeep OBDII units for the 4.0 liter.

 

I will definately report any findings either positive or negative since it seems you guys here at hybridz are on the sharper end of the knowledge/motivation end of the spectrum.

 

Now it's off to the junkyard to check some stuff out.

[Pharaohabq]

Yeah I've always wanted to try out Jeep. 4.0 EFI on a L28 engine. There's a few things to figure, but really it's pretty simple when you get into the swap bacause so many people have done it to fix the carbs on the old 4.2L using 4.0 parts. OBD1 is much easier to swap, but with ODB2 you just put the 2nd O2 sensor furthor down the pipe and it looks like it's on the other side of the Cat, tho a resistor is needed to change the value. It's not as tuneable, but it's a big step up from the bosch EFI.

 

http://www.projectjeep.com/Projects/EFI.html

 

Phar

[woldson]

My brother is running OBD1.

Said it is perfered because it has be hacked up the wazzu.

[HowlerMonkey]

I'm not sure the jeep ecu needs any tuning.

 

My co-worker put a kenne bell supercharger kit with much larger injectors on his year 2000 cherokee and nothing else.

 

It ran pig rich for a few drive cycles and cleaned right up with his LM-1 showing normal mixture while driving.......Hmmmmmmm.

[Daeron]

Hey, I was going to the junkyards tomorrow.. I'll keep my eyes peeled for the applicable models! the U-Pull on Benoist Farms Road shows a 98 Pathy in the online inventory (hardly guaranteed) and my oldest brother needs a headlight assembly from that vehicle anyhow, so if it is there I will be A-pickin' at it. I need parts for my 91 CRX that are pretty model/year specific, so I will be walking both yards on Benoist.. Not sure if I am going to hit the one on Dyer Road or not, even though it is on the way.. that place seems to be falling slowly into the crapper...

 

PM me with your telephone number if you want; I'll probly be leaving the house around lunchtime.

[Pharaohabq]

I'm not sure the jeep ecu needs any tuning.

 

My co-worker put a kenne bell supercharger kit with much larger injectors on his year 2000 cherokee and nothing else.

 

It ran pig rich for a few drive cycles and cleaned right up with his LM-1 showing normal mixture while driving.......Hmmmmmmm.

 

that's the same as it did on my jeep, I think it learns your engine and holds a better functionality. that's why I want to try it on an L28. (not to mention it's made for a 6cyl, and is MPI non-batch fired too... )

[HowlerMonkey]

One possible hurdle with a later ecu regardless of manufacturer would be security in that many late ecus have immobilzer circuitry that expects a certain key. Some are wholly contained within an ecu and some communicate with security modules on either ccd or pci data lines. It would suck to put together a system and have no way to run the car.

 

I do know a bit about it and will rerport back soon on which ones are easiest to work around or spoof but be advised that many are unbeatable requiring a new ecu and master key to be purchased together as I have many lexus ecus that are supposed bricks because the customer lost his keys.