1991 NSX just doesnt want to run right/pass smog...

[akeyzoo]

Alrighty - not so brief history - am restoring a 91 that has sat for the last 8 years - a high mileage A/T (212K miles). "Car ran fine when parked...lol"

Tons of restoration work - but here is what I did with engine....

New timing set/water pump.
Rebuilt alternator
new AC compressor
seals, plugs, vtec seals and filters and pretty much all maintenance you would do with engine out.

Got car running easily - but didn't do an amazing job of holding an idle... Would stall
Cleaned out idle air control valve - wasn't horrific but it helped...
Car now holds idle but has three symptoms. Shifts relatively hard and wont downshift at WOT AND the idle when warm hunts around (550-1300). Engine also had a stutter at mid RPM....
So I replaced injectors with rebuild and put new coils on. This solved the stutter completely but idle still hunts...

So I took apart and thoroughly cleaned throttle body. I tried setting idle per the manual and could not really get idle above 550 RPM with the IACV unplugged. No amount of idle screw setting would get me there. Also, is all my trials here now I have a new CEL that will not reset... code 7 which manual says is the TPS...

But... I thought, lets check her out at the smog station... Well, I live in CA and it wasn't even close. Hydrocarbons and CO2 elevated.

Since I am the proud owner of a parts cannon I grabbed 2 new CATS from science of speed and 2 new 02 sensors.... Smog tests improved but wont pass. Holy crap were those O2 sensors brutal to remove....

Purchased a TPS and awaiting its delivery. I did check TPS with voltmeter and it seemed within spec (.55V - 4.8V), but I cannot do the thorough test via manual as I do not have the ECU service jumper cable thingy... Oh, and thanks manual, I don't have a known good ECU to swap...

Haven't touched the EGR or the AIT (air intake temp) sensor.

What on earth do I try next? Need this idling smooth when warm and THEN I need it to pass smog lol.....

Chris

Guessing the engine hunting at idle is first order of business as I presume that would make engine run rich at smog test...

 

[akeyzoo]

Adding that TCS and SRS are lit (havent got to diagnosing that yet)

 

[Honcho]

The warm idle is a clue. It means the car is likely not getting enough idle air and is surging the EACV to try and hit the desired warm idle number. When cold, the FITV is probably adding enough air to compensate.

You said you cleaned the throttle body- did you also clean the idle air passage in it? It's a small hole in the throttle bore that leads to the idle air screw port. It's really tough to get to.



Also, I would check/clean the idle air passage in the intake manifold. This is where the FITV and EACV get their idle air from. It's a pretty big passage, but if obstructed it could be the cause.



Finally, make sure you have cleaned out the four vacuum line ports on the top of the intake manifold.




Once you have completed the above, you can be certain that there is no mechanical blockage that could be obstructing idle air. After that, it's testing and making sure the EACV is working properly. If faulty, it won't hold a stable idle.

The EACV idle logic is pretty simplistic. The ECU will set a computed desired idle under different conditions and the EACV will try to hit it by adjusting solenoid pulsewidth. The warm idle default setting is 800 RPM. However, if the base idle airflow is too far outside what the ECU considers to be "normal" parameters (either because the idle screw is not set properly or there is a blockage in the idle air pathway), it will cause idle problems and the EACV will chase the idle setting without being able to hit it, causing surging, high idle, or low idle.

Assuming the above all checks out, you may have a fueling issue- most likely clogged or dripping injectors. This could be the cause of your rich condition and also affect the warm idle. The car's narrowband O2 feedback isn't going to matter if the injector isn't spraying fuel properly. We had another member here who recently had this problem- it turns out a couple of his fuel injectors were constantly dripping fuel into the cylinder regardless of firing cycle. Something to look at too.

 

[akeyzoo]

Rock on. I absolutely thoroghly cleaned the idle scre port when I had the TB off. Pulled idle screw all the way out and blasted the hole out with carb cleaner. Will clean the vacuum ports. I did blast some carb cleaner in the intake manifold
idle air port (the larger one) and it did not seem obstructed...

I did replace injectors... Will work on this, ty!

 

[Old Guy]

First off, I have the later drive by wire throttle so I have no direct experience with the earlier idle control system. So, take that into consideration in terms of any of my advice.

First off, how is your ignition timing? If the ignition timing is significantly off at idle that can make it hard to set idle speed. If the ignition timing is really retarded the idle speed will be low and may be out of the range that the EACV can adjust for.

@Honcho has addressed the intake / throttle body 'issues'. I have a few more suggestions.

1) Since you have an automatic, is your warm idle problem with the transmission in gear or in neutral? If it won't idle in gear then check to make sure that the ECU is getting the shift position signal from the transmission.

2) Make sure that the coolant temperature sensor that the ECU uses (not the dash gauge sensor) is in spec. The ECU raises idle speed when the engine is cold. Normally an error from the sensor would do the opposite of what you are experiencing; but, make sure that it is in spec because if you have multiple problems they can have a cross influence which causes unexpected responses.

3) You cleaned the EACV; but, that does not guarantee that it operates smoothly. If you have, or have access to a variable voltage bench power supply, connect up the EACV and starting at 0 volts slowly increase the voltage up to 12 volts and listen and look in the port to try to try and determine if the valve is sticking. If the valve had barbed ports on it you could attach a tube and blow through it to confirm operation; but, it doesn't so you are pretty much limited to visual and auditory inspection. If the valve has 'stiction' that can lead to wild swings in the idle control. However, the fact that you cannot set the warm base idle speed with the EACV disconnected is in my opinion a bigger issue that you should address first - refer to point #6.

4) You talked about measuring the TPS voltages. Is that the first part of the TPS test where you measure the voltages on the wire harness side of the connector? If so, that is important; but, it does not really tell you anything about the actual TPS. You do not absolutely need the factory test harness to do the second part of the TPS voltage measurements. You can fabricate a makeshift interposing harness to allow you to access the TPS voltages right at the TPS or you can just back-probe the TPS connection right at the ECU. D11 is the terminal for the TPS 'wiper'. D22 is the TPS ground terminal. You can measure the voltage at those two points and the voltage should swing smoothly from around 0.5 volts to 4.8 volts (wide open). If the closed voltage is higher than 0.5 volts that could be a problem because the ECU measures the voltage to determine when the throttle is closed and its time to go into idle control. If the voltage is too high then the ECU does not go into idle control. Note the comment in the service manual about the influence of the auto transmission on TPS voltage/ @Honcho or @MotorMouth93 may be able to advise on the idle control threshold voltage. I don't think this is your problem because your hunting idle sounds like it is going into idle control, its just that the EACV doesn't seem to be able to do its thing; but, it is worth checking just to cross it off the list as a confounding variable.

5) Normally the EGR passages plug up with age which causes NOX compliance issues due to no EGR. If the EGR were stuck open at idle this might be elevating the manifold pressure causing addition fuel to be added and screwing up the HC / CO. I also believe that EGR flow into the intake at idle will cause idle control problems and stalling. In the service manual there are tests including measuring the voltage on the EGR lift potentiometer (need to back-probe at the ECU). If the lift pot voltage is 1.2 volts at idle I think that means that the EGR should be closed.

6) If your ignition timing is correct, the fact that you are not able to achieve the warm base idle speed with the idle adjusting screw really does sound like something is still plugging the idle air passage. Increasing air flow through the idle air passage is just like opening the throttle a tiny bit. If opening the throttle a bit increases RPM; but, opening the idle screw passage some more does nothing, something has to be blocking the idle air passage. You may need to pull off the throttle body for closer inspection and cleaning.

With respect to your high HC and CO levels. It is not obvious to me why the idle speed issue would cause this. However, the problem causing the HC / CO non compliance may be contributing to your idle control problems.
- make sure the engine coolant temperature sensor is in spec because if the ECU reads a cold engine it will apply fuel enrichment and will block closed loop operation (and possible screw up idle setting) and probably guarantee non compliance on a HC / CO test.
- the IAT sensor can mess with idle HC / CO; but, should have a minor to no effect on idle control. Check the resistance to make sure it is in spec and clean it to scratch it off the list as a confounding variable.

 

[akeyzoo]

Traveling for a week and can’t peek at car or workshop manual, but how do I check or even adjust the ignition timing? I do have a variable voltage bench power supply so can test that as needed. Really appreciate the input yall, like I said, I won’t be able to touch car for a bit but will be back!

 

[Honcho]

Traveling for a week and can’t peek at car or workshop manual, but how do I check or even adjust the ignition timing? I do have a variable voltage bench power supply so can test that as needed. Really appreciate the input yall, like I said, I won’t be able to touch car for a bit but will be back!

Ignition timing isn't really mechanically adjustable on the car. There is an adjustment screw, but it is locked down and the Honda literature is VERY clear to not touch it! The only other way to adjust timing is in the ECU, but you need a hacked one to do that.

While not directly related, the crank position sensor could be failing, causing the sparks to misfire out of order. Though, the engine might not run at all under those conditions...

My bet is still on the idle air passage in the throttle body. I know you said you cleaned the screw opening, but did you clean the tiny port going to the screw opening from the throttle passage? When I did mine, black soot kept pouring out for quite a while. If this port is partially blocked the car will surge and it would also explain why turning the idle screw makes no difference.

 

[Old Guy]

There is an ignition timing adjuster and as Honcho notes, Honda does discourage adjustment with a tamper proof cover; but, the service manual does set out a procedure for checking and adjusting the timing.

Since the ignition curve is coded in the ECU memory, the normal way you would adjust digitally controlled ignition is to write the new timing settings into memory. Given that the timing values are digitally coded, it is not absolutely clear what the ignition timing adjuster does. My guess is that because the crank sensor is a reluctance device (rather than Hall or optical device) the adjustment pot might allow the factory techs to adjust the trigger threshold for the crank sensor. This would provide a mechanism to get the specified 15 deg BTDC at 750 RPM idle on an auto rather than mechanically move the crank sensor to get the correct value. Page 23-91 provides instructions on checking the ignition timing and the procedure for altering the ignition timing. At present, I don't advocate for adjusting the timing, merely checking the timing to confirm that it is where it should be.

Like Honcho, I think the idle culprit is most likely something still blocking the idle air control passage .

 

[akeyzoo]

OK... Some answers - back from my trip to Kentucky and have acquired a shit-ton of bourbon ;-)

I took the TB off again (getting better at minimizing coolant loss lol) and re-cleaned the idle circuit on TB and sprayed a few shots in intake manifold idle passage... While taking off the TB I noticed that I hadn't completely reseated the plug for the TPS. The pedal position sensor was, but not the TPS. SHIT. Plugged that in. I then took out the IAT (needed impact, woofah) and cleaned IAT sensor.

Started car and hot-damn it runs and idles sooooo much better. I still couldnt meet the spec for setting idle with the idle control screw to a T, as it idled very low with the IACV unplugged.... couldn't get idle high enough.... But got up to 520RPM or so and then when i plugged IACV back in and reset CPU it idles spot on. Drives very good.

Will try to see how the smog test results look, hope improved!

I did notice one thing tho... MY Tach and Speedometer do not read correctly. Tach reads too high compared to external tach conected to tach sense port) Additionally my speedo reads too low at speeds above 40 MPH. I may create a new thread on this...

Thanks for all help and pointers, I am moving in the right direction!!!

Chris

 

[akeyzoo]

Found the thread on gauge cluster repair ;-)

 

[akeyzoo]

UPDATE - Car seems to run perfectly. Soooo, I took her to my local smog shop for a pre-check. Here is the net:

HC ppm are very high at idle (500ppm), but ok at higher RPM (20 ppm). There are two components to smog test, checking at idle, and checking at higher RPM and it would fail miserably at idle, but pass at higher RPM.

I have completely unloaded the parts cannon here - thoughts?

Rebuilt injectors
New plugs
New coils
New timing set
cleaned IACV
Cleaned TB
NEW o2 sensors
NEW Cats from SoS
Cleaned IAT sensor
New coolant temp sensor

Currently no CEL, but SRS and ABS lights lit...

Any recommendations on what to do next????

 

[Honcho]

UPDATE - Car seems to run perfectly. Soooo, I took her to my local smog shop for a pre-check. Here is the net:

HC ppm are very high at idle (500ppm), but ok at higher RPM (20 ppm). There are two components to smog test, checking at idle, and checking at higher RPM and it would fail miserably at idle, but pass at higher RPM.

I have completely unloaded the parts cannon here - thoughts?

Rebuilt injectors
New plugs
New coils
New timing set
cleaned IACV
Cleaned TB
NEW o2 sensors
NEW Cats from SoS
Cleaned IAT sensor
New coolant temp sensor

Currently no CEL, but SRS and ABS lights lit...

Any recommendations on what to do next????

When you sprayed cleaner into the idle screw port, did you see it drain out of the inside bore of the throttle body?

 

[Old Guy]

Is an SoS cat an OEM cat from SoS or is it a cat they market? If it is a cat that they market do they guarantee emission compliance?

Since you don't have an OBDII car you can't plug in and measure fuel trims. The next best thing might be to get a wide band O2 sensor and controller and one of those tail pipe clamp things and take some tailpipe measurements at idle to see if you are running at 14.7 AFR. If it is below 14.7 then you might want to spend some time trying to figure out why - perhaps it is not going in to closed loop operation in which case you would need to figure out why it is not in closed loop. Moutermouth93 may be able to help you with the parameters in the NA1 ECU for entering closed loop control.

Which coolant temperature sensor did you replace? There are two.

 

[akeyzoo]

@Honcho - Yep, sprayed nice and free.

@Old Guy - The SoS cat's are aftermarket. Will look into a wide band 02 sensor. Also, I replaced the temp sensor near the thermostat housing, it seems to be the one for the dash gauge. Perhaps I replace the other? ((where is it?)

 

[Honcho]

Also, I replaced the temp sensor near the thermostat housing, it seems to be the one for the dash gauge. Perhaps I replace the other? ((where is it?)

The temp sensor on the thermo cover is for the radiator fan control. The little one on the water jacket is for the dash gauge. The one you want is the ECT sensor, which is on the front head below the VTEC spool valve. It is the sensor that the ECU uses to measure coolant temp for all sorts of things, including closed/open loop idle mode.

 

[akeyzoo]

@Honcho Coolant temp sensor in the mail... ty.

(37870-PJ7-003)​

 

[akeyzoo]

Well shit, this is embarrassing. Seriously embarrassing. @Honcho @Old Guy .

So, I bought a gauge set and confirmed fuel pressure was ok - it was within spec.

I then decided to hook up the timing light and BINGO. Timing at idle was at TDC. Drilled out the rivets on the timing adjuster and I could not get it in spec.... What could it be? Well, it was my timing belt install. Pulled the engine apart and sure enough she was off ONE TOOTH +/-. Sweet Jesus. After 5 failed attempts to reinstall timing belt (after I rotate a few times it would be off) I fabbed some 5mm pins for the camshafts (I hadn't been pinning the cams this whole time) and bingo. TIming set. Put everything back together and she is right on the money on the red mark (15 degrees) with a little adjustment of timing adjustment screw. I now have confidence that she will pass smog once i get a chance to take her in. She has a little more power and the idle could be more reliably set.

Seriously, I am better than this. I installed the belt with engine out of car, and after 6 attempts in the car I have no more skin on my knuckles..... That really sucked but I am confident she will pass smog.

I did not do a compression check post reinstall of timing belt. I cant fathom a tooth would have wiped a valve out, and she really does run smooth and strong now.

Wish me luck ;-)
Chris

 

[drew]

Which sprocket was off?

I say this because my timing belt had both the exhaust cams off by one tooth. It idled fine, compression in spec, and passed CA smog without issue. There is a caveat that CARB continually tightens up the passing specs by specific vehicle models as the field test results are posted.

 

[Old Guy]

Well that is an interesting way to detect a timing belt error. It is not quite what I was expecting when I suggested checking the ignition timing; but, if it works don't look a gift horse in the mouth.

Delayed ignition timing will reduce the cylinder burn time which may be responsible for your elevated HC numbers. Hopefully, this correction will bring you into emission compliance.

 

[akeyzoo]

@drew It appeared the front exhaust was a full tooth ahead of the intake AND the rear exhaust was line up with intake (not 1/2 tooth advanced).....

So, check your timing folks. I cant wait to get her to a smog show so may stay home from work tomorrow....

 

[akeyzoo]

BTW, Had the "shower of shame" this am, man my hands hurt like hell, nicks and cuts everywhere! Must be these soft programmer hands lol.

 

[Honcho]

Well shit, this is embarrassing. Seriously embarrassing. @Honcho @Old Guy .

So, I bought a gauge set and confirmed fuel pressure was ok - it was within spec.

I then decided to hook up the timing light and BINGO. Timing at idle was at TDC. Drilled out the rivets on the timing adjuster and I could not get it in spec.... What could it be? Well, it was my timing belt install. Pulled the engine apart and sure enough she was off ONE TOOTH +/-. Sweet Jesus. After 5 failed attempts to reinstall timing belt (after I rotate a few times it would be off) I fabbed some 5mm pins for the camshafts (I hadn't been pinning the cams this whole time) and bingo. TIming set. Put everything back together and she is right on the money on the red mark (15 degrees) with a little adjustment of timing adjustment screw. I now have confidence that she will pass smog once i get a chance to take her in. She has a little more power and the idle could be more reliably set.

Seriously, I am better than this. I installed the belt with engine out of car, and after 6 attempts in the car I have no more skin on my knuckles..... That really sucked but I am confident she will pass smog.

I did not do a compression check post reinstall of timing belt. I cant fathom a tooth would have wiped a valve out, and she really does run smooth and strong now.

Wish me luck ;-)
Chris

Glad to hear you got it sorted. It's easy to miss a tooth on the TB because setting belt tension isn't as straight-forward as the service manual makes it seem. This is because the spring tensioner can "rock" on its pivot point, which can result in not enough tension between the rear exhaust cam and the tensioner when the tensioner spring is released. The "let go and let the spring tension it" requires more finesse and often repeated attempts. The pros do it several times and check belt tension at the rear exhaust cam each time until they hit it right. If it's too loose there, the car often will skip a tooth.

Good news is one tooth off isn't enough to cause valve contact. In fact there are likely many NSXs driving around with one tooth off on the belt since most Acura techs just snap the tensioner loose like the manual says and send it without checking belt tension. The key test is to put a vacuum gauge on the car. If the needle is "dancing" at idle, you're off by a tooth.

Hope it finally passes smog!

 

[nis350]

The key test is to put a vacuum gauge on the car. If the needle is "dancing" at idle, you're off by a tooth.

Hope it finally passes smog!

does it apply to both the intake and exhaust cams? I assumed the vacuum would be steady without any fluctuation for a normal engine?

which is the best port to hook up the vacuum gauge?

 

[akeyzoo]

Passed smog here in CA with flying colors! Thanks all!

 

[Honcho]

Passed smog here in CA with flying colors! Thanks all!

Glad to hear it! Congrats!

does it apply to both the intake and exhaust cams? I assumed the vacuum would be steady without any fluctuation for a normal engine?

which is the best port to hook up the vacuum gauge?

Both cams- the idea is a valve is still open when it shouldn't be. The needle should be steady. LarryB taught me this.