California Smog test failed

[Stevon]

95 NSX T new cats are installed but problem prevails.............
Parts installed:
4 new O2 sensors
New PCV valve
New plugs
New coils
New timing belt
New thermostat
New cats
New fuel filter
New fuel pump
Removing the old cats I was very disapointed to see that they were still good with no damage or clogging.
Cleaned and replaced the battery and engine grounds with gold terminals soldered onto new copper cables. Installed the new converters and O2 sensors and ran my car, code 64 went away and the check engine light and relay click problem went away. The grounds improved the cold running hesitation but did not eliminate the problem. When cold, you have to drive 100 yards or so then it has full power and no problems until you park and restart the car. When starting hot, the problem may last 100 feet or might not happen at all and car drives normally, bizarre eh? This emission problem is really getting to me, failed again! I’m now a certified California Gross Polluter! This car has the original 95 engine wire harness and intake manifold that was taken off the engine that was sold from the car. It’s all original parts as far as the harness and sensors are concerned, only the long block was changed. Someone mentioned that there is an internal ground that can go bad inside of the ECU itself. Not sure what to try next, have replaced a lot of parts with little results to fix the original problem. The plugs are beautiful grey and the exhaust smells sweet. Maybe the ECU is bad? should I buy a spare and try that? Will have to try other things I guess, running out of ideas. Below is today’s smog test and it is slightly better but still far away from passing.

Test date 11/17/10
Test _______ Rpm |%CO2|%O2| HC (PPM) | CO (%) |
_M1: 15 mph 2678 | 10.7 | 0.0 | MAX 85 Meas 261 | MAX 0.50 Meas 6.78|
_M2: 25 mph 2427 | 10.5 | 0.0 | MAX 50 Meas 275 | MAX 0.47 Meas 7.15|


Test date 09/13/10
Test _______ Rpm |%CO2|%O2| HC (PPM) | CO (%) |
_M1: 15 mph 2611 | 9.8 | 0.1 | MAX 85 Meas 296 | MAX 0.50 Meas 7.57|
_M2: 25 mph 2427 | 9.4 | 0.0 | MAX 50 Meas 303 | MAX 0.47 Meas 8.03|



Original thread
95T running too rich, no ECU codes

--------------------------------------------------------------------------------

UPDATE:
Finaly got a code that makes sense out of the ECU besides the engine misfire ones. Code 64 is indicating that rear secondary O2 sensor has a problem and is not working. Would that make the car run erratic power cold and put the car in failsafe or limp mode?
The exhaust is very black too which shows cats aren't working fully. They are at temps of 512 and 479 degrees after a six mile highway run, seems too cold for good running cats. The cats are from the 1990 engine installed in this car. The secondary O2 sensors are resistors and O2 simulater units. Guess it's time for Pro Speed converters with bungs installed for real secondary O2 sensors.

I have just completed my five year project 1995 NSX T and putting the car on the road. First problem was gelled gas from sitting which destroyed the original fuel pump and another replacement pump in the tank. Cleaned tank out and replaced pump and gas, put some carb cleaner in fuel rail and ran the car. The car has power and idles and accelerates fine when hot. Problem is no power when accelerating cold, super lean, pops in intake manifold. When the engine warms up the problem suddenly goes away and check engine light comes on. Stored code is 71,72,73,74,75 and 76 "cylinder misfire". If you reset the ECU (pull clock fuse) and drive the car hot it drives fine and no codes are stored in the ECU. When getting the car smoged for registration, HC and CO2 are off the scale! needless to say it fails, here are the readings:
Test ______Rpm|%CO2|%O2|___ HC (PPM)___ | CO (%) |
_M1: 15 mph 2611 9.8 | 0.1 | MAX 85 Meas 296 | MAX 0.50 Meas 7.57|
_M2: 25 mph 2427 9.4 | 0.0 | MAX 50 Meas 303 | MAX 0.47 Meas 8.03|
I was wondering what to do because there are no codes to go by, bad injectors? Any sugestions?

Did some experiments today. Measured the cluster temperature sensor cold and hot, it is within specs. When starting the car cold it turns over and starts in 3 seconds and idles smooth. When pushing the accelerator pedal the rpm drops down and if you push it all the way the car will almost stall running extremely lean. After about 3 minutes you hear a click like a relay sound, check engine light comes on at the same second, then car accelerates normal with full power and no problems other than running rich for emissions. I’ll have to get a fuel pressure gauge to see what’s going on as a baseline.

Stephen

 

[drew]

Your car is running rich.

I think you are on the right track with the temperature sender. The NSX has TWO temperature senders: one for the cluster and one for the ECU. You should be able to measure it at the ECU.

You can get a Auto Innovate LM-1 (or similar) to help you with the diagnosis. Having a data logging tool like this is invaluable to determining the problem. It logs your stoch as well as 4 (5?) other sensors...I use small safety pins through the target wires to get contact with the ECU harness.


Gross Polluter, nice. Do you still have to go to a DMV rep for that?

 

[Stevon]

The smog only shop said that I could bring the car back for a re-test. DMV rep? God no, have a 91 engine in the 95 car with 91 exhaust and headers, need to fix this myself. The ECU temp sensor sounds like it could be the problem, might explain the cold running issue as well. Will look into it today if it doesn't start raining. My 95 car has four O2 sensors but no ODBII connector, is there a way to connect to the ECU can buss for diagnostic connection? Factory manual does not show pin out wire color or connctor number for that.
Stephen

Your car is running rich.

I think you are on the right track with the temperature sender. The NSX has TWO temperature senders: one for the cluster and one for the ECU. You should be able to measure it at the ECU.

You can get a Auto Innovate LM-1 (or similar) to help you with the diagnosis. Having a data logging tool like this is invaluable to determining the problem. It logs your stoch as well as 4 (5?) other sensors...I use small safety pins through the target wires to get contact with the ECU harness.


Gross Polluter, nice. Do you still have to go to a DMV rep for that?

 

[Stevon]

Update 11/22/10
Found out that the temp sender I was testing was for gauge and not the ECT (Engine Coolant Temperature Sensor) for the ECU. It was hard to find and just as hard to get to. The harness wire colors were the clue as to which is which. Bought a new ECT only to find out the old one can't be removed? It's all plastic and the octagonal body just stripped without moving, who thought this was a good design? Guess Acura assumes in this situation the owner will just buy a new NSX and trade old one in! What now? drill it out? Any ideas?

The smog only shop said that I could bring the car back for a re-test. DMV rep? God no, have a 91 engine in the 95 car with 91 exhaust and headers, need to fix this myself. The ECU temp sensor sounds like it could be the problem, might explain the cold running issue as well. Will look into it today if it doesn't start raining. My 95 car has four O2 sensors but no ODBII connector, is there a way to connect to the ECU can buss for diagnostic connection? Factory manual does not show pin out wire color or connctor number for that.
Stephen

 

[liftnot]

Steve,
give Nick at Applied Motorsports a call: 760-401-0253.
He has been around a long time with mad skills with the NSX. He does a lot of special projects with the NSX for customers. His shop is in North Vista. He is also a lot more reasonable$$ than other shops in town.

Tim

 

[Stevon]

Today on the NSX I was able to test the ECT sensor. The cold resistance at 60 degrees F was 3K ohms and running the car up to 150 degrees resistance dropped to 512 ohms. This is consistent with the Acura listed values in the manual. Have a new ECT sensor to try out of the box and testing produced very similar readings so there is no reason to struggle drilling out old one. Unplugging the ECT connector for testing, the check engine light came on when starting car. Shorting service connector oddly brings up only code 66 (rear 2nd O2 heater circuit malfunction) which I thought was fixed with the new secondary O2 sensors installed, no other codes were stored yet. The next thing will be to buzz out the secondary O2 connection and drive the car with the injector cleaner. After that not sure what to do. Coolant thermostat was replaced when I did the timing belt and works correctly as far as I can tell. After engine is hot the radiator does get hot driving a few miles and you get heat from cabin heater controls. Only thing left is injectors and the ECU!

Your car is running rich.

I think you are on the right track with the temperature sender. The NSX has TWO temperature senders: one for the cluster and one for the ECU. You should be able to measure it at the ECU.

You can get a Auto Innovate LM-1 (or similar) to help you with the diagnosis. Having a data logging tool like this is invaluable to determining the problem. It logs your stoch as well as 4 (5?) other sensors...I use small safety pins through the target wires to get contact with the ECU harness.


Gross Polluter, nice. Do you still have to go to a DMV rep for that?

 

[liftnot]

Stevon,
FWIW, I had my injectors flowed by RC and one of them was rated "leaking" and the rest were "good". Now the flow pattern are all "excellent" and I notice the engine is a bit smoother as well. I did this AFTER I barely passed my smog last Nov.

 

[PATCHEZ]

It might be a good idea to check ur IAT sensor readings. As this would coz a rich mixture and advance timing if it's reading cold.

 

[EdinKali]

It might be a good idea to check ur IAT sensor readings. As this would coz a rich mixture and advance timing if it's reading cold.

This and I believe there is an ambient air temp sensor at the front of the car as well if I am not mistaken. Not sure if this is the same or separate sensor on the NSX.

 

[drew]

Steve

I recommend that you get a pin out for the ECU that you are using and measure the inputs/outputs <i>at the main connector</i>. The NSX engine is Speed Density (SD), there isn't a whole lot of inputs.

Also, get your secondary O2's welded in with some bungs.

At this point, I strongly recommend getting a wide band o2 sensor with data logger.

Are both sides of the engine running rich?

Did you use a 1991 engine harness with with a 1995 main harness?

Drew

/Getting some data logging will really help. There are several on Ebay and CL will have them from time to time.

//The ambient temp sensor in the front of the car is for the climate control system.

 

[Stevon]

Drew,

The new cats were made by Prospeed and he fitted bungs after the cat to install secondary O2 sensors. The ECU sees them but they don't affect mixture. Both sides seem to run the same, Smog tester tried both sides and there was no difference. Injectors could be an issue with the old jelled gas pumped through them at one point. Harness and intake manifold is all 95 only block and trans are 91. Thanks for your help!
Stephen

Steve

I recommend that you get a pin out for the ECU that you are using and measure the inputs/outputs <i>at the main connector</i>. The NSX engine is Speed Density (SD), there isn't a whole lot of inputs.

Also, get your secondary O2's welded in with some bungs.

At this point, I strongly recommend getting a wide band o2 sensor with data logger.

Are both sides of the engine running rich?

Did you use a 1991 engine harness with with a 1995 main harness?

Drew

/Getting some data logging will really help. There are several on Ebay and CL will have them from time to time.

//The ambient temp sensor in the front of the car is for the climate control system.

 

[EdinKali]

Drew,

The new cats were made by Prospeed and he fitted bungs after the cat to install secondary O2 sensors. The ECU sees them but they don't affect mixture. Both sides seem to run the same, Smog tester tried both sides and there was no difference. Injectors could be an issue with the old jelled gas pumped through them at one point. Harness and intake manifold is all 95 only block and trans are 91. Thanks for your help!
Stephen

Are those 300 or 400 cell cats? You probably won't pass CA emissions with anything less.

 

[drew]

What is the voltage of the pre-cat O2 sensor when the engine is at operating temperature?

Is there any way to put on the OEM cats?

Drew

 

[Stevon]

EdinKali,

Brian suggested and made for me 400 cell units that slightly improved emissions but it turned out the old ones weren't bad and not the problem. You can see the results below:

Test date 11/17/10 (New Prospeed Cats and 2nd O2 sensors)
Test _______ Rpm |%CO2|%O2| HC (PPM) ******| CO (%) |
_M1: 15 mph 2678 | 10.7 | 0.0 | MAX 85 Meas 261 | MAX 0.50 Meas 6.78|
_M2: 25 mph 2427 | 10.5 | 0.0 | MAX 50 Meas 275 | MAX 0.47 Meas 7.15|


Test date 09/13/10 (original 91 Cats and 2nd O2 sim circuit)
Test _______ Rpm |%CO2|%O2| HC (PPM) *****| CO (%) |
_M1: 15 mph 2611 | 9.8 | 0.1 | MAX 85 Meas 296 | MAX 0.50 Meas 7.57|
_M2: 25 mph 2427 | 9.4 | 0.0 | MAX 50 Meas 303 | MAX 0.47 Meas 8.03|

Are those 300 or 400 cell cats? You probably won't pass CA emissions with anything less.

 

[Stevon]

Drew,

Thanks for the help man, Will check the O2 hot running voltage today. Don't see any reason to put the original cats back on and the "fake" secondary O2 sensor simulator circuit, the new Cats are doing their job and performing well.

Stephen

What is the voltage of the pre-cat O2 sensor when the engine is at operating temperature?

Is there any way to put on the OEM cats?

Drew

 

[Stevon]

liftnot,

I'd rather not remove injectors to be cleaned, flowed then reinstalled. To do all that work it would make better sense to install new injectors. Considering buying Accel replacement ones part# 151370. Check out the specs:
http://www.cadvision.com/blanchas/Ranger/AccelFuelInjectors.pdf

Stevon,
FWIW, I had my injectors flowed by RC and one of them was rated "leaking" and the rest were "good". Now the flow pattern are all "excellent" and I notice the engine is a bit smoother as well. I did this AFTER I barely passed my smog last Nov.

 

[Daedalus]

EdinKali,

it turned out the old ones weren't bad and not the problem.

From the first test results alone it is pretty obvious that something other than the cats is out of whack. I would not have bothered to try a differnet set of cats. Cats reduce emissions by oxidizing hydrocarbons. But the cat HAS to have oxygen to do this.

5O2 + 4HC --> 2H2O + 4CO2 (Largely simplified)

Look at the data from the first test. There is ~zero oxygen left in the exhaust, thus the cats don't have anything left to work with. The only missing variable here is NOx, which can provide some oxygen:

2NO2 --> N2 + 2O2 (Again, simplified)

However, the amount of NOx typically present is not enough to offset the HCs shown. The x in NOx is a range, but is mostly in the small single digits, thus max available O2 can be estimated and almost certainly won't be close in this case.

O2 sensor readings from narrow-band sensors will be useless under these conditions. They will only say "RICH!".

 

[liftnot]

although I don't have a sure way to quantify the RC clean did anything to improve my smog reading but I do know a leaking injector could cause a high HC reading. My was at 80k and ran fine but I am sure it was never cleaned and flowed. My engine is fairly stock so new injectors was not necessary. It was the right thing for me to do so when next smog comes, I know its not the injectors.

liftnot,

I'd rather not remove injectors to be cleaned, flowed then reinstalled. To do all that work it would make better sense to install new injectors. Considering buying Accel replacement ones part# 151370. Check out the specs:
http://www.cadvision.com/blanchas/Ranger/AccelFuelInjectors.pdf

 

[Stevon]

Daedalus,

I found two items today working on the car. First was IAT sensor reads 2795 ohms at 60 degrees, book says it should be 500 ohms at that temperature. 2795 ohms corolates to 18 degrees F on book graph, San Diego does not get that cold! Chris from PrecisionHonda told me to make sure injectors and injector resistors are the same value, they are not! Injectors are 2.2 ohms, resistors are 12 ohms! Have the wrong resistors installed from the 91 parts car. Acura manual says resistance for the 95 injector resistors should be between 5 and 7 ohms. The injectors read 2.2 ohms and are correct for the 95. Next step is to replace IAT sensor, locate a 95/96 Injector resistor set and go from there.
Stephen

From the first test results alone it is pretty obvious that something other than the cats is out of whack. I would not have bothered to try a differnet set of cats. Cats reduce emissions by oxidizing hydrocarbons. But the cat HAS to have oxygen to do this.

5O2 + 4HC --> 2H2O + 4CO2 (Largely simplified)

Look at the data from the first test. There is ~zero oxygen left in the exhaust, thus the cats don't have anything left to work with. The only missing variable here is NOx, which can provide some oxygen:

2NO2 --> N2 + 2O2 (Again, simplified)

However, the amount of NOx typically present is not enough to offset the HCs shown. The x in NOx is a range, but is mostly in the small single digits, thus max available O2 can be estimated and almost certainly won't be close in this case.

O2 sensor readings from narrow-band sensors will be useless under these conditions. They will only say "RICH!".

 

[sr5guy]

Daedalus,

I found two items today working on the car. First was IAT sensor reads 2795 ohms at 60 degrees, book says it should be 500 ohms at that temperature. Chris from PrecisionHonda told me to make sure injectors and injector resistors are the same value, they are not! Injectors are 2.2 ohms, resistors are 12 ohms! Have the wrong resistors installed from the 91 parts car. Acura manual says resistance for the 95 injector resistors should be between 5 and 7 ohms. The injectors read 2.2 ohms and are correct for the 95. Next step is to replace IAT sensor, locate a 95/96 Injector resistor set and go from there.
Stephen

Hey, check the resistor again. Most likely you measured two of the resistors in series (6ohms + 6ohms = 12ohms). The resistor is the same part number from 1991-2005.

30700-SL0-003 RESISTOR ASSY. 91 - 05 NSX

On the other hand, the fuel injectors are different -
91-94 use this number:
06164-P0A-000 INJECTOR SET, FUEL 91 - 94 NSX

95-05 use this number:
06164-P0A-A00 INJECTOR SET, FUEL 95 - 03 NSX

I think if you change your injectors you'll be good.

-Matt

 

[Stevon]

Matt,

Thank you for the information. You were right, had measured across pins 2-6 and 4-8 instead of all pins to 5 (ground), So it appears resistors are within spec. I think Ill go with the Accel low impedance 151370 (370cc) injectors. Got four of them for 37.50 new in the box and can buy the last two for $45 clearance. That and replace the IAT sensor.

Hey, check the resistor again. Most likely you measured two of the resistors in series (6ohms + 6ohms = 12ohms). The resistor is the same part number from 1991-2005.

30700-SL0-003 RESISTOR ASSY. 91 - 05 NSX

On the other hand, the fuel injectors are different -
91-94 use this number:
06164-P0A-000 INJECTOR SET, FUEL 91 - 94 NSX

95-05 use this number:
06164-P0A-A00 INJECTOR SET, FUEL 95 - 03 NSX

I think if you change your injectors you'll be good.

-Matt

 

[sr5guy]

Matt,

Thank you for the information. You were right, had measured across pins 2-6 and 4-8 instead of all pins to 5 (ground), So it appears resistors are within spec. I think Ill go with the Accel low impedance 151379 (370cc) injectors. Got four of them for 37.50 new in the box and can buy the last two for $45 clearance. That and replace the IAT sensor.

You can't just guess at what size injector to put in. The injector matches the ECU very closely. I assume you are still running the 1995 ECU? I was hoping I could find more information on this injector but I cannot find good solid information other than this same injector was used on 94-95 Honda Accords with VTEC.

I will try to do more research and get back to you later in the morning.

-Matt

 

[EdinKali]

Make sure you follow the golden rule of not changing more than one thing at a time. Otherwise you won't know what fixed the problem or if you fixed the original problem and introduced a new one.

 

[PATCHEZ]

Matt,

Thank you for the information. You were right, had measured across pins 2-6 and 4-8 instead of all pins to 5 (ground), So it appears resistors are within spec. I think Ill go with the Accel low impedance 151379 (370cc) injectors. Got four of them for 37.50 new in the box and can buy the last two for $45 clearance. That and replace the IAT sensor.

I would just change the IAT like I said earler in the thread. Forget about injectors for now. The ecu is dumping all kinds of fuel if that IAT is reading 18degrees

 

[Stevon]

EdinKali,
That is good advise IF you have an expensive gas analyzer and better, a dyno to test the car. Don't have either so I do a few tests, replace a few parts and redo smog test after batches of repairs. Alot of the repairs are just maintenance anyways like plugs, thermostat, fuel filter etc. Others may not be neccasary like new injectors but the side affect of that repair is smoother idle and better gas mileage so it can't really hurt. It's still cheaper for me to do this my self than pay an untrusted mechanic who might just play around, replace a few parts and charge me $75 an hour weather he fixes the problem or not, had lots and lots of offers to "diagnose" my car for $65 an hour or more. None offered a refund if they wasted money on labor or parts. Have some Engineering and automotive skills myself. I see it this way, It's kind of like your health, ultimately you pay the price so if you can it's sometimes best to do your own research and work.

Stephen

Make sure you follow the golden rule of not changing more than one thing at a time. Otherwise you won't know what fixed the problem or if you fixed the original problem and introduced a new one.