Help diagnose my misfire

[elijahcole92]

hello all,

i'm trying to find my misfire problem on my 91 nsx. i bought the car in january and drove it from fl to nc, when i bought it it had a cel 42 for bank 2 o2 sensor. So far i've now replaced the rear o2 sensor, new plugs, checked the coilpack ohms (all read the same) and was going to check compression but can't find my compression tester.

At idle i've tried unplugging coilpacks one by one, the front 3 all create a significant change, however the rear three don't make much of a change at all if any. I tossed a front coilpack into one of the rears and it didn't change anything. What else should i be checking?

Thanks

 

[Jinks]

hello all,

i'm trying to find my misfire problem on my 91 nsx. i bought the car in january and drove it from fl to nc, when i bought it it had a cel 42 for bank 2 o2 sensor. So far i've now replaced the rear o2 sensor, new plugs, checked the coilpack ohms (all read the same) and was going to check compression but can't find my compression tester.

At idle i've tried unplugging coilpacks one by one, the front 3 all create a significant change, however the rear three don't make much of a change at all if any. I tossed a front coilpack into one of the rears and it didn't change anything. What else should i be checking?

Thanks

Mine threw a misfire code due to messed up connector at the ignition control module.

 

[Old Guy]

Two questions.

First, since the 1991 MYr is not equipped with the misfire detection module, how do you know that you have a misfire? What are the specific symptoms?

Second, you say that disconnecting the rear three coil packs (presumably one at a time) does not make a difference. Is that implying that with a coil pack disconnected on the rear cylinder bank, there is no change to engine RPM or the roughness of the idle? If you can disconnect any of the three rear coils and there is no noticeable change you may have a bigger issue than a misfire. You may have an underlying mechanical problem.

Measuring the internal resistance of the coil pack windings is effective at determining if one of the windings is open circuited (in which case it will be completely inoperative as opposed to an intermittent misfire). Insulation failure of the coil pack cannot be detected with the resistance measurement. Check for tracking or contamination on the 'nose' of the coil pack which could be a sign of external insulation failure. Internal insulation failure is not easily measured, best confirmed by replacement with a known good coil.

 

[Jinks]

While the 1991 nsx does not have the advanced misfire detection system of a obd2 car. It can still detect a misfire if the ICM is at fault. Ask me how I know. From what I gather though, on a 91 is if the ICM is working as normal, you will never receive a misfire on the CEL past the module itself. However I could be wrong,just going off of past issues. Doesn't hurt to check. Both obd1 and obd2 cars have a ICM which controls.monitors ignition. I think the obd2 car just has additional circuitry to measure faults pasts the ICM module.

- - - Updated - - -

As for the coil packs, I had two of them bad on my 1991. They both passed the ohms check on the primary side. The car would shudder and misfire, but as noted previously the car would never throw a CEL code being a 1991. It was very frustrating for me. So being if the car is a 1991 and it throws a misfire, I'd check into the ICM.

 

[Old Guy]

The ICM or ignitor has no self monitoring function built into it. It is just a box with six power transistors that switches the six coils primary current based upon the control signals it gets from the ECU. One of the most common problems with the ICM is poor grounding. Except for its miserable location, confirming the grounding is easy to do. The coils on the OBDII cars have a three terminal plug (pre OBDII cars have a two terminal plug). The third connection in the plug goes to a separate misfire detection module which generates an error signal for the ECU indicating misfire. Just for good measure, the misfire detection module appearance is almost identical to the ICM; but, in a different location.

I am not particularly in the know about the pre OBDII cars. If the ICM module has failed, the ECU may be able to generate an error. However, a failed ICM would be a total ignition failure on one or more cylinders which would be different than what most people interpret as misfire. Bad / intermittent grounding on the ICM could cause an intermittent ignition problem which could result in a misfire like behaviour.

 

[elijahcole92]

Two questions.

First, since the 1991 MYr is not equipped with the misfire detection module, how do you know that you have a misfire? What are the specific symptoms?

Second, you say that disconnecting the rear three coil packs (presumably one at a time) does not make a difference. Is that implying that with a coil pack disconnected on the rear cylinder bank, there is no change to engine RPM or the roughness of the idle? If you can disconnect any of the three rear coils and there is no noticeable change you may have a bigger issue than a misfire. You may have an underlying mechanical problem.

Measuring the internal resistance of the coil pack windings is effective at determining if one of the windings is open circuited (in which case it will be completely inoperative as opposed to an intermittent misfire). Insulation failure of the coil pack cannot be detected with the resistance measurement. Check for tracking or contamination on the 'nose' of the coil pack which could be a sign of external insulation failure. Internal insulation failure is not easily measured, best confirmed by replacement with a known good coil.

My car is 91 so pre-OBDII. Well i guess i can't say 100% sure, but it most definitely sounds and feels like i have a mis. It does not sound smooth at idle or cruise, but it's not terrible bucking or anything. Sadly, i can't check engine rpm as my tach doesn't work and i have yet to find a reasonably priced oem gauge cluster. I am planning on doing a compression test and diy clean the injectors today to see if i can find anything else. i will take a video.

I will also switch the other two rears to the front and see if it follows the coil or if it's the cylinder itself. Where is the ICM and ground located?

- - - Updated - - -

Well compression test was good, cold numbers are all between 165-175.

Rear plugs look browner than the fronts here are two pics to show the difference. I also uploaded two videos to show the sound and also the change when unplugging coils on either side.

rear

front

https://www.youtube.com/watch?v=Bj-kcGjqJfs
https://www.youtube.com/watch?v=KQnIl2ecYrQ

note: the videos were shot after i switched the front coils to the rear and it's doing the same thing so i don't believe it's the coils.

 

[Old Guy]

I am guessing that those are your new plugs and that you have not done any significant driving (mainly idling) since they were installed. If so, hard to make any determination about mixture strength issues.

Notwithstanding the somewhat odd idle sound, does the car accelerate smoothly? If so, then I doubt that you have a misfire. My 2000 has a rather manic idle with a real thump to it which completely disappears when warm (a friend of mine thought I had a con rod bearing going!). Its had it since day 1 and has not been an issue. You also have what appears to be a non stock exhaust so its hard to make relative comparisons about the engine noise with my car which has the stock muffler.

The ICM is sort of stuck under the throttle body area - hence the pain in the ass part of checking the grounding!

 

[elijahcole92]

Yep, nothing more than idling on them. I have not drove the car since i bought it and brought it to NC. Yes, i have a comptech header with RM exhaust. Ahh, okay that's what i thought. How do i check the grounding if i can get to it?

Did the change in idle/sound not sound odd to you when unplugging coils between the front and rear?

 

[Old Guy]

Somebody on Prime (from Ireland I recall) had a problem with their comptech header which was creating all kinds of problems for them. I recall that it turns out there was a crack in the header which was screwing up the O2 sensor operation by admitting atmospheric air. Give yours a check - which will be worse than trying to get at the ICM to check its grounding!

Hard to check the ICM ground through measurement, mainly its a matter of checking to make sure the mounting screws are tight and that there is no corrosion at the grounding point. Its a matter of remove, clean and reassemble. If things improve then you had a grounding issue. If nothing changes time to move on to something else.

The fact that the rear plugs have some brown on them indicates that combustion is occurring in the rear cylinders. If after a drive on the highway you have a significant color differential between the front and back plugs with all of the back plugs being consistent, I would start looking at mixture issues and might be a little suspect about your header. Is the 42 code a mixture out of range code or a heater failure code? If its a mixture code, the problem could be caused by the aforementioned manifold leak as opposed to being a bum sensor. Also, is the ECU and engine tune stock or has somebody modified it?

 

[elijahcole92]

Somebody on Prime (from Ireland I recall) had a problem with their comptech header which was creating all kinds of problems for them. I recall that it turns out there was a crack in the header which was screwing up the O2 sensor operation by admitting atmospheric air. Give yours a check - which will be worse than trying to get at the ICM to check its grounding!

Hard to check the ICM ground through measurement, mainly its a matter of checking to make sure the mounting screws are tight and that there is no corrosion at the grounding point. Its a matter of remove, clean and reassemble. If things improve then you had a grounding issue. If nothing changes time to move on to something else.

The fact that the rear plugs have some brown on them indicates that combustion is occurring in the rear cylinders. If after a drive on the highway you have a significant color differential between the front and back plugs with all of the back plugs being consistent, I would start looking at mixture issues and might be a little suspect about your header. Is the 42 code a mixture out of range code or a heater failure code? If its a mixture code, the problem could be caused by the aforementioned manifold leak as opposed to being a bum sensor. Also, is the ECU and engine tune stock or has somebody modified it?

I just installed the header on the car, but i did buy it used, though i checked over and didn't see any cracks. The car has the stock ecu tune as far as i know. The 42 code just said "rear primary heater circuit malfunction" so i put a new o2 in when i did the header.

 

[Jinks]

The rear bank coil packs are notorious for getting wet then failing. Only way to check the secondary though is to buy new ones and test it. The coil pack covers on our cars have gaskets on the top. If they are missing or not correctly installed, every time you wash your car the coils get soaked. Just something else to check.

 

[Old Guy]

The heater failure (42) would just make the O2 sensor slow to respond during the initial warm up. Once up to temperature it should work so the failure is probably not indicative of where the problem lies.

Until you can get a few highway miles on the car you won't be able to use the plugs to give you any indication of mixture issues. At idle, most cars run AFRs right around stoich or perhaps a little higher once the car is warm and off of start up and warm up enrichment. This should leave the sparkplugs pretty white; but, not sugary looking - that's death! The photos of your plugs look like new plugs coming out of a car that has been idled for about 10 minutes. Just to confirm, the plugs are not wet with gasoline? That would be a sure sign of protracted failure to achieve ignition.

Maybe its not the problem; but, I would do a double check to make sure that you don't have any header leaks. You installed it with a new gasket - correct?

 

[elijahcole92]

So we unplugged all three rear cylinders and the car did not change significantly, however unplugging just one of the front made a significant change. This is with either the front or rear coils installed on the rear bank so i don't think its the coils, noid light shows power for both coil and injector. I mean the plugs definitely smelled gasey, but did look slightly wet. I'll double check header leaks tomorrow and take the car off the lift and drive it some.

 

[Old Guy]

So we unplugged all three rear cylinders and the car did not change significantly, .

As in you unplugged all three cylinders at the same time and there was no significant change?

 

[redbird92]

Misfire Stuff to Do

Here's my take on this: take into account each situation is different , but if I OR Larry B,I totally respect that guy can help we try
How many miles and Time/age are on the coils? If you plan on keeping the car , replace them, its a wearable IMO, 20 plus years old , how many heat cycles etc ...
The body of the rear bank coils rust up over time as most people dont cover the engine when they wash it, the small drain outs built into the valve covers is higher water just sits there and combine with old dried up and or missing coil cover gaskets furthers the problem, causing them to fail , I have only replaced one or two igniter packs in my career fixing these cars .

This is what I check next , First have the Fuel injectors ever been cleaned ? see attached sheet , Dripping is BAD , misfire/rough idle hello!, Also The TW sensor P/N 37870-PJ7-003 , located in the front bank cylinder head can be problematic , When was the last time , If ever , the Valves adjusted?

The early cars are senstive and problems can hide behind other issues/ lack of maintenace, bad prior repairs , I have seen it all over the last 20 years fixing NSX's

I recently had a NSX customer call me , the car was at the dealer , misfiring , THEY wanted to pull the motor, told him it had a " burnt" valve, Really???

Long story short , Compression and Leakdown were stellar, I sent the injectors out, the top Orings were dead also, new fuel filter , 3 rear bank coils , Valve adjustment, TW Sensor and Thermostat , gasket and Coolant , hadnt been done in years , although someone did all the hoses , When the valve covers were off I did check the Camshaft timing as I have seen a Few cars lately with the rear bank exhaust cam off 1 or more teeth off

Car ran Awesome

Car is still running great

Ben Daisy auto Pm or call me if needed 410-489-9334

 

[jwmelvin]

...it had a cel 42 for bank 2 o2 sensor. So far i've now replaced the rear o2 sensor...

Bank 2 is the front O2 sensor, isn't it?

 

[Old Guy]

Redbird provides an excellent list of some of the deferred maintenance issues that could cause poor operation. If you are focusing on deferred maintenance, just to add to the list, a fouled intake air sensor and dirty throttle body can contribute to mixture problems and poor operation.

It would be interesting to know whether this poor operation issue has existed since the original poster purchased the car (in which case the problem could be any of those maintenance issues) or whether the problem emerged only after the original poster started their modifications, in which case you might have some more specific things to look at.

 

[elijahcole92]

I haven't had the injectors cleaned, although i did pull the rear 3 and manually clean them with brake cleaner and a small battery. I don't really want to pull them and pay to get them cleaned since i'm planning on installing my RC550s with CTSC once i figure out this misfire problem. I don't believe the valves have been adjusted and am going to go ahead and do my tb/wp and adjust the valves while i'm at it before trying to change anything else.

Would the IAT sensor not throw a CEL? Can i just be cleaned or does it have to be replaced? I did clean the main area of the TB with brake cleaner and a rag (didn't spray it in any vacuum lines).

Regarding the coilpacks, The rear coils did have some rust which i sanded/polished off, but that didn't affect it at all. Since it seems to be a rear misfire and i tried unplugging coils up front with significant change, but then moved those coils to the rear and it did the same thing, would that not rule out coils? I bought a new legend coil since they're only $45, should i try it on one of the rears and see if it improves?

Also, the code 42 is for rear H02S heater, it doesn't mention bank, sorry.

 

[Old Guy]

A coating on the outside of the IAT can cause it to read the air temperature incorrectly (usually lower than actual) without causing an error since it is still within range. The IAT can be cleaned with throttle body cleaner or something similar.

If you do a search, you can find a historical thread discussing the cleaning of the IAT.

 

[elijahcole92]

Well i cleaned the iat, it was dirty but not terrible. I also tried the fuel pump resistor bypass, didn't help. I then unplugged both o2 sensors and there was no change at all? What does that mean

 

[jwmelvin]

Since the rear bank seems not to be firing but the coils are not the issue, it is probably worth it to try a new ignitor (ICM). They are available used on eBay for $30, as a test. Do you have access to an oscilloscope (wishful thinking perhaps) to check the signals coming out the connector to the coils?

 

[elijahcole92]

Since the rear bank seems not to be firing but the coils are not the issue, it is probably worth it to try a new ignitor (ICM). They are available used on eBay for $30, as a test. Do you have access to an oscilloscope (wishful thinking perhaps) to check the signals coming out the connector to the coils?

Well i when i pulled the connectors on the coils i checked them with a noid light (lights up when it receives power and ground), and they all lit up as well as the injectors. I cleaned the ICM ground while i was at it. I do not have an oscilloscope :/

 

[jwmelvin]

Noid light. Interesting, and seemingly perfect for the task (I learned something). Can you describe more about the nature of the misfire and conditions under which it exists or does not? Has it always been there or appeared after the drive? All rpm? Cold/hot?

 

[elijahcole92]

Haha, yeah i've only used it on fuel injectors before but figured it would work well for that. http://www.autozone.com/test-scan-a...l-tester/oem-6-pcs-noid-light-set/948078_0_0/

It has been there since i bought the car. At idle it isn't smooth, but stumbles, more noticable when it's cold. Warm idle still stumbling, driving it's doesn't jerk or anything just has hesitation. Did you look at the two videos i posted earlier?

 

[jwmelvin]

Sorry, I had missed the videos. They are interesting but I have no ideas right now, sorry.