Gregg Lee

I tried adding this to a similar active thread, but no reply, so I'm trying it as a new thread, reworded. .

I have a year 2000 S2000 with rough idle.. CEL codes are the same as listed in the misfire troubleshooting guide under FAQ's, DIYs, and Maintenance. Not always same cylinder.

I changed plugs (old ones looked and measured OK) and adjusted valve clearance (was too tight, adjusted to spec). Idle was initially worse but after a couple trips no load idle is nearly smooth at expected RPM. I guess the ECU had to to relearn after battery had been disconnected.

While no load idle is smooth, if I turn on the A/C, engine management holds the idle speed, but idle is rough. I can hear occasional misfires. CEL soon shows up. Does not stall. Pressing throttle pedal to raise revs stops the misfire. When A/C compressor cycles off, idle smooths out.

Adding engine load with alternator by turning on bright headlights,and running interior fan at full produces similar effect but less than A/C compressor.

Is the rough idle triggered by higher load but with no drop in revs a useful clue? is the idle rpm when AC turned on done by the Idle Air Control Valve?

Also I read that Throttle Position Sensor is sometimes involved. I used an ODB2 scan tool to look at TPS response. With engine off, key on, the percent reading moves smoothly with throttle pedal travel. No jumps or gaps. At closed throttle scan tool TPS says 5%. At full open throttle TPS says 91%. I verified that with pedal fully depressed the throttle does to the full open stop on the throttle body.

Seems like wide open ought to be 100%. Should it be 100%? Does this suggest a TPS sensor problem?

Chibo

What is your short term fuel trim when it is running poorly and what is your long term fuel trim?

VashTheStampede

Throttle position percentage is correct, and matches what I've seen. As the other poster replied, what's your fuel trim? Also, have you ever mis-shifted? Have you checked your retainers? Finally, it could also be your fuel injectors. So, to sum it up, you could still have out-of-spec valves, crumby injectors, a failing alternator or bad ground (leading to a lack of current to the coils), a vacuum leak, etc. Checking for any/all codes and telling us your LTFT% (plus or minus) will help to rule out at least one or two of those. Also, if you can, tell us what your voltage at the scan tool is reading...

Gregg Lee

Q - A

what's your fuel trim? Will check and post

Also, have you ever mis-shifted? - Do you mean something that caused an over-rev? If so, no. The only way I know of to do that is shift to a lower gear than intended. 1st instead of 3rd. 2nd instead of 4th. So that the car's momentum back drives the engine past red line. The rev limiters would prevent a power over-rev on a shift that missed by staying in neutral.

Have you checked your retainers?- No indication of retainer issue when I was adjusting the valves. All valve clearances consistent at below minimum spec by .002". Maybe one was .003." clearance response to turning adjustment was precise and consistent.

Finally, it could also be your fuel injectors.- I did read the misfire thread. So yes, but I haven't checked yet.

Out-of-spec valves - what does that mean? - I already adjusted clearance so you don't mean that I assume.

Crumby injectors - As I said no evidence of that but I will search forum for "crumbly injectors" to see what are other signs.

failing alternator - alternator output voltage is good and battery is new. Alternator passes the "alternator check" function on my battery charger.
bad ground (leading to a lack of current to the coils) - Which ground do you mean?

A vacuum leak - Already checked using the vac system map and and found nothing.

Checking for any/all codes - I did. The only codes are the misfire codes.

Telling us your LTFT% (plus or minus) will help to rule out at least one or two of those - same as trim above, right?

If you can, tell us what your voltage at the scan tool is reading. - Will check and post.
=====================
Also I re-checked all the electrical connectors that I had to disconnect to be able to remove the cam cover and for clearance to turn the torque wrench on the flywheel bolt, to be sure they are connected and still tight.

Say Chi Sin Lo

This may or may not be overlooked, but check your spark plug tube seals.

My car registered a misfire in one of the cylinder, took it apart and found oil residue on that ignition coil pack. Replaced all 4 spark plug tube seals, 4 ignition coil packs, 4 spark plugs.

Reset the ECU and no more CEL's since.

One bonus is that, these were wear items and it was relatively inexpensive maintenance items anyways.

Gregg Lee

I haven't given up but the problem got worse. Idle became worse then changed to what seems obviously a single cylinder not firing. Rhythmic engine rocking engine rocking and tail pipe "puh puh puh....." at 1/2 of engine rpm.
Car shows code P1399 temporary miss any cylinder. Oddly it occasionally goes off, particular if I raise revs in neutral, but then back on. Only after short test drive and switching A/C on off at idle could I get it to record a cylinder specific code- p0301 miss cylinder 1. So I swapped cyl 1 and 2 coils and tried again. Same difficulty getting a specific cyl code. but when I got it it was p0302 - cyl 2 miss.

That's enough for me to set it aside and wait until new coils arrive in a couple days.

It occurred to me I might try an inline spark tester that fits between plug and coil. arc is visible inside transparent plastic center. I have one for gasoline power tools but I don't know if I can get it onto the plug. I'm sure there are long neck versions around but mine is not. .

If I'm right about it being an "obvious" single cylinder miss, then ODB2 sure seems dumb. But 'm going by sound and seat of pants. Not sure which ODB2 sensors would measure those.
================================================== =====================

But coils might not fix rough idle even iif they do fix the miss. So I have added updates to some previous replies and also new requests for clarification questions below in Red.

What is your fuel trim? Will check and post With a regular miss I doubt that some sensor readings (eg O2 sensors) are relevant. However LTFT was always 0, STFTs were rapidly varying. I did erase codes repeatedly during test and retest. A Question before I check trims again: Under what conditions should I report the vales? Engine warm at idle or something else?

Out-of-spec valves - what does that mean? - I already adjusted clearance so you don't mean that I assume. And I rechecked my valve clearances after the problem became worse. All OK.

bad ground (leading to a lack of current to the coils) - Which ground do you mean? I decided you must mean the ground to the coil pack. Checked all coil pack connectors, ground wire and power wire. . Resistance was zero from battery ground terminal to coil pack connector ground wire. Voltage at the power wire was the same as across the battery terminals or 0.1V less. (all with engine off ,key on, No accessories on.)

If you can, tell us what your voltage at the scan tool is reading. - Will check and post. Which voltage do you mean? My scan tool does not report alternator/battery or ECM supply voltage. If does show a sensor voltage or two. And again measured under what conditions?
=====================================
In reply to the separate suggestion to check spark plug tubes and coli packs for contaminants, I had not noticed any when I removed the coil packs, but I rechecked. Checked both the outside and looking into bottom of coil pack and looking inside the spark plug tubes. All are free of oil, water, and dirt or anything else. Same for the electrical receptacles and connectors at the coil packs' tops.

Gregg Lee

Installed new coils and everything is fixed. No miss, idle is smooth even with A/C on.

I searched S2KI for recommendations on which coils and where to buy.

To add to the database, I bought Densos on Amazon for $38.89 each. Based on forum recommendations I bought the ones listed for 2004 model, 673-2301.


Photo sitting on air cleaner shows the slight physical differences in shaft detail between the 19 year old original (top) and new coil (bottom), but not on any critical dimensions.
Original is as it came from car. Not cleaned at all.