Paging NSX OEM electronic experts - Code 12 - Yes the EGR crap

[integrals]

Ok guys, brief backstory... please read to the end!

Yes I have searched and read dozens of threads on here regarding this, but most do not post a resolution that has not been tried already.

Car started showing intermittent code 12 about 2 years ago (it would go on, then go off on Key Off and stay off for 7 months or so, so no cause for alarm). Starting end of last year it started to become more semi permanent (coming on weekly). Now it actually will stay come on daily after about 1 hour of driving (car fully warmed up).

I started with the most obvious culprit and did the somewhat lazy 'replace egr valve' with a factory unit. Obviously problem was not solved so we are here.

So I opened up the FSM and went through the full section from TOP to bottom. The system passes the entire diag tree with flying colors (has 8" vacuum on hose 5 when EGR solenoid is activated. 12 V and ground at the EGR solenoid wiring. EGR stalls car when vacuum applied. The power wire at EGR has 4.9V Key On (not running) and the center wire has 1.2V Key on (not running) and smoothly moves from 1.2 to 4V as vacuum is applied. The pins at the ECU both show correct voltage as well according to the FSM. I swapped ECU's with my other 91's ECU. Still no change . CEL code 12 after driving for 1 hour or so.

I swapped the entire ECU control box with 2 other units (known working good). Same exact symptoms.

I swapped intake manifolds with a spare that I just thoroughly cleaned the EGR ports from the inside of the runners out to the external pipe with solvent and a pipe cleaner.

I did the reset procedure between each test run as well.

SO. I have run out of ideas. The only things I can think of is the computer is getting incorrect data from either the IAT or O2 sensors which is causing a conflict on the EGR lift tables. (Aka o2 or IAT sensor is within voltage range but stuck or something like that). The O2 sensors are NTK sensors. There is no other codes in the car except 36 as my TCS is removed.

Due to lack of information in the FSM, I am unsure how the other sensors such as O2 TW and IAT affect the EGR tables but that is ALL I can think of.

If you have information please let me know. I am not one to be stumped by electrical usually but this one has got me running in circles.

Cliff notes to avoid asking questions already answered:
Used FSM for full diag (passes all flowchart tests and is deemed intermittent by FSM)
Replaced EGR no change
Replaced Intake manifold with port cleaned unit no change
Replaced EGR Solenoid, and CVC with 2x known working units and no change
Replaced ECU with known working ECU no change
Need information how the o2 iat and tw sensors determine EGR lift (has NTK o2 sensors and OEM IAT TW with no codes relevant to these sensors).

Thank you in advanced.

BTW 1991 US Model, 224k miles on it. Stock minus Comptech headers and K and N air filter.

 

[Old Guy]

The service manual is a little thin on details so I am going to engage in some speculation. From the little information that I can extract from my '96-2000 service manual, I believe the #12 error code is driven purely by the voltage that the ECU reads off of the EGR lift potentiometer. Under specific engine operating conditions the ECU is looking for a specific voltage from the lift potentiometer to confirm that the EGR is in the correct position. You did the vacuum test on the EGR valve and confirmed that under test conditions the wiper voltage on the lift pot is moving from 1 to 4 volts so it seems like the EGR valve is working correctly (plus this is the new valve that we are talking about?).

Now this is the part where I am really weak. My recollection from a discussion of EGR valve operation and emission testing issues was that the EGR operation by the ECU is only a function of the EGR map. The EGR map as I recall was purely a function of manifold pressure (MAP) and engine RPM and the ECU opens the EGR in a certain range of those cells in order to control NOx. I remember this because there was carping about the change in California emission testing procedures which resulted in the engine not operating in an RPM / MAP range that caused the EGR to open (or completely open) which resulted in failing the NOx standard. Somebody actually posted the map showing the cells in which the EGR was operated and I seem to remember that the map axis were only MAP and RPM. As I recall the trick discussed was to externally apply vacuum to the EGR valve to get it to open in the new test range which would then allow the engine to pass the NOx component of the test. The only reason this is important is that I don't think (as opposed to actually know) that the operation of the O2, IAT or coolant sensor is material to the generation of the error code for the EGR. I 'think' that when the engine is operating in that particular MAP/RPM range in the EGR map it is looking for a particular voltage from the lift potentiometer to confirm operation and if it doesn't get that voltage it generates the error code.

From the service manual, I interpret that as vacuum is applied to the EGR and it opens the voltage increases and the voltage drops as the valve closes. The later OBDII cars have two codes related to the EGR and one of those is voltage too high so I think the later codes allow you to differentiate between the valve being open (voltage high) when it should not be open or not open when it should be open. I don't know whether the 12 error comes up just with low voltage (EGR no open) or low and high voltage. The other item I do not know about EGR operation is whether it is either fully open or fully closed or whether the ECU PWM modulates the EGR solenoid to create partial openings. I think it is fully open or fully closed; but, again that is not something that I know for sure.

It seems like you have addressed all the likely candidates for the error code. I take it that the problem is still 'technically' intermittent - not showing up until things have come up to temperature after an hour of driving? If that is correct, then at the risk of stating the obvious the factory test procedures will probably fail to diagnose the problem if they are not carried out under exactly the same conditions as when the error code emerges - which is probably going to be really difficult to arrange. Has the error code ever shown up after a cold start or does it only emerge after a period of operation? If it only shows up after the vehicle is up to temperature I am inclined to think a thermally induced electrical connection failure. The obvious candidate would be to check the electrical connectors on the EGR valve and the EGR solenoid, looking for damaged pins and frayed / broken wiring where the wires come out of the connectors. Is there any place in the wiring harness where there could be an intermittent short to ground on the wiring which could cause a loss of voltage on the EGR lift pot leading to an error code? Also, check the condition of your vacuum lines. If they are getting hard and have developed cracks there might be a leak which shows up when hot which results in the EGR valve failing to open or open completely.


Are you by any chance from a jurisdiction that requires emission testing? If so, are you passing the emission tests, in particular the NOx component? If you are passing the NOx test then your EGR valve is probably doing its thing correctly and its a problem with the lift pot voltage signal getting back to the ECU that is causing the error code.

If inspection of the wiring and vacuum lines yields nothing, a final rather drastic step involves the use of a data logger. There are small USB based and stand alone data loggers that can be set up to monitor a voltage (typically 0 - 5 volts). The most basic versions with a couple of channels can be had for around $100. I would be inclined to hook one of these data loggers up to the EGR wiper terminal at the ECU and just record the voltage and wait for the 12 error code to emerge. Then dump the data from the data logger and check to see if you can spot a sudden change in the wiper voltage which could be a sign of a short or open circuit condition in the wiring.

 

[integrals]

I think I saw that thread someplace too regarding RPM vs MAP. Which would kinda put me back at square 1.

And correct, the car DEFINITELY has to be in closed loop / warmed up for the code 12 to surface.

I have checked the wiring to the best of my ability (continuity is correct on all wires in question) as well as vacuum lines are all intact.

I do not have emissions where I reside fortunately and unfortunately lol.

I am about ready to just get an EMS lol

 

[Larry Bastanza]

I am thinking you want to throughly clean the throttle body. Could be a minor vacuum issue when the EGR valve should open.

JMO,

Regards,
LarryB

 

[integrals]

I am thinking you want to throughly clean the throttle body. Could be a minor vacuum issue when the EGR valve should open.

JMO,

Regards,
LarryB

I did overlook the TB itself but did clean the ports inside the IM side. Will check this possibility out as well!

Thanks Larry!