Misfire but no CEL

[cccscotland]

Just picked up my 1995 NSX after being stored for 6 months, and it feels as if it has a misfire, but no check engine light shows. I'm a believer that things that go wrong with cars, and especially strong ones like the NSX, tend to be due a chain of events rather than a part failing, I'll give you my two theories.

First, the symptoms. At idle, the engine seems to hunt, up to 3k rpm it feels like a misfire, but no check engine light. Above that it seems to clear, but run with less power (because the ear can't detect the misfire at higher rpm?). At 80 mph at steady throttle it feels a bit stop/start - as if the car clears its throat momentarily and then reverts. It's quite happy at that speed though.

Also the EPS light is on. I've got the battery up to speed - a run of 300 miles - and reset the clock fuse. The light comes back on, it seems when a certain speed is reached rather than steering input or time. When the car is started with the light off, the power steering continues to work even if the light comes on. Start it with the light on, and there's no power steering.

Some background. The car's been sitting outside with no cover for 6 months. At the start of that time, the timing belt was changed. The mechanic noted the misfire, but put it down to needing a good run as it was continuously driven in and out of the garage. He also changed the plugs, as he though that may be the issue. It wasn't - I've now driven 300 miles, but it's no better.

So, theory 1. The timing is out due to incorrect fitment of the belt when changed.

Theory 2. The EPS and the engine ecu are connected somehow - I've seen some suggestion that pulling the clock fuse resets the ECU. Whether that is the steering ecu or engine I'm not able to ascertain. Could they be connected?

Other simple possibilities?
Coil packs

All help gratefully received
Cheers
Matt
1995 NSX-T - 210,000 miles...

 

[Larry Bastanza]

First thing to do is put a vacuum gauge on hose #4 from the throttle body (or the control box) and see if the vacuum is steady. If it is pulsing the timing belt is incorrect. One tooth off on the rear exhaust cam, will make every other cylinder differ, the pulse will show that.

Tell me about the weather/environment it was in. Rainy, damp or dry and hot?

Regards,
LarryB

 

[cccscotland]

Tell me about the weather/environment it was in. Rainy, damp or dry and hot?

Regards,
LarryB

Thanks for the reply. Scotland, so raining and damp!

 

[Larry Bastanza]

Do you know if the car is OBDII? This was the year in the US we went to OBDII, but not sure about your market. I ask since the 1994 model in the US(OBDI) cannot sense misfire.

 

[pgilliam1]

You need a new mechanic. He "noted a misfire" after his timing belt install. C'mon, "needs a good run". Seems too obvious to me, so I must be missing something. Be sure to post your fix.

 

[Sumeronthechain]

I would check the coil packs.

 

[cccscotland]

OK, more information. The car is not OBDII compliant - the coil packs only have 2 pins and I use a small 2 pin connector under the glove box for diagnostic. I'm yet to try a vacuum test as I don't have one to hand, but if the below fails I'll definitely hunt one down.

So, yesterday I drove from Rotterdam home to Lake Geneva, via the autobahn - about 650 miles. As you know, it's possible to go quite fast on the autobahn, so I did. And the faster I went, the better the car ran. After running at between 120mph and 140 mph for about 15 minutes, the misfire was virtually gone. After a short period of running 50-60 mph, the misfire was back.

I even ran at over 150mph (sat nav indicated) at one point. I'm guessing if I really had a permanent misfire or bad timing, that would not be feasible? So I'm now inclined to think an intermittent electric fault, rather than a permanent fault, such as mis-timing. As to the EPS, I'm ignoring it for the moment.

So, today I pulled the coil packs and plugs and had a good look. Referring the pictures below, all the plugs electrodes were a uniform colour, but on one of the plugs, the ring where the thread is was clean - all the others were smutty. It's the plug on the left in the first pic. I looked down the tubes of the coil packs, and all had a bit of shine at the end - see pic 2 - there is a small gleam of clean metal on the contact. However, there was one coil pack where the metal was dull and grey. The metal on the underside of the cap was rusty, but whether this has any relevance I don't know. See pic 3 - this was the coil pack that corresponded to the plug that had the clean base.

So, I'm thinking that my easy first port of call is to replace that coil pack and see if that fixes anything. Would you guys concur?

 

[Larry Bastanza]

Was it #1, or #2 cylinder? Seems like you do need a set of coils.

This is ASSUMING the cam timing is proper.

Regards,
LarryB

 

[cccscotland]

Was it #1, or #2 cylinder? Seems like you do need a set of coils.

This is ASSUMING the cam timing is proper.

Regards,
LarryB

Middle of the rearmost bank. I guess that's 2?

 

[acuratech27]

put the coils in and with the car running at idle start disconnecting the coils one by one. if the idle changes the coil is fine if there is no change then it is something with either that cylinder, coil, plug, injector or icm. do that then repost what the results were for further steps

 

[cccscotland]

Ok, so.

I removed the packs and plugs as detailed in my last post and reinstalled - I noticed on removal a couple of the plugs were only finger tight - and I cleaned the air filter, plus there were a couple of small deposits on the TB which I cleaned off.

I tried David's plug off test, but there was a change in idle in all cases.

I got a chance for a short test drive today, and I think it's fixed. Pulls good from just over 1k rpm in top with no stumble, and a nice steady idle at rest. Great news, and thanks for all your help.

The EPS light is still on - I pulled the codes -31 & 33. It's a 95 with the integrated power and control unit - the manual (and BrianK) suggest that the codes are related to speed sensor and battery or ig1 connection at the ignition switch.

The speedo works fine, so I'm tempted to rule out the speed sensor, and I've checked the battery and steering earth behind it, all seem fine. Section 17-40 (the flow chart for checking the ig1 connection into the EPS ECU) seem to be missing from the zipped manual on that section of prime. Does anyone have any tips as to what to look for, or am I on the hunt for a new ECU?

Thanks for all the help so far. Whilst I'm happy if my car stays strong for the foreseeable future this site is so brilliant in enabling non-mechanics to have the confidence to dive into our cars and I derive huge satisfaction from fixing stuff!

Thanks again,
M