Intermittent/cold misfire issue

jwmelvin · 13 March 2015

I'm massively frustrated and hoping there is something I am missing.

My car is a 2000 (3.2L OBD2) with CTSC that has been on it since new. It has 40k miles. It has not been smooth; I will give my history:

[TABLE="class: grid, width: 500"]
<tbody>[TR]
[TD]June 2014[/TD]
[TD]

purchased with recent TB/WP service
immediately to Ben @ Daisy for: new O2 sensors, clean injectors, new fuel filter, new seals and bypass pipe on CTSC, new spark plugs.

[/TD]
[/TR]
[TR]
[TD]Sept. 2014[/TD]
[TD]

During a hard corner with low fuel (fuel light on), flashing CEL & TCS. Car reported P1399. Filled tank and problem did not recur.
One week later, at cold startup ~60 F, same problem of rough idle ~2k rpm, same code. Smoothed out after ~20 sec.
Two weeks later, same issue at cold startup ~60 F, smoothed out.
A whine behind the firewall showed up and seemed associated with higher than normal AFRs. Replaced fuel pump with Deatschwerks and problem solved.

[/TD]
[/TR]
[TR]
[TD]Nov. 2014[/TD]
[TD]

Rough start again, smoothed out ~20 sec.
Later in drive, EPS light, codes indicated EPS 33 & ABS 61.
Later during separate drive, no power steering.
Battery tender overnight; cleared codes
No further EPS light/codes
Two days later, leaking head gasket; car to Ben @ Daisy

[/TD]
[/TR]
[TR]
[TD]Nov. 2014 - Feb. 2015[/TD]
[TD]

Head bolts had pulled the threads out of block around cylinder 2. Repaired with TimeSerts.
Engine rebuilt with all new bearings and rings, ARP head studs. Valve job.
Fueling changed to AEM F/IC and RC 550 cc/min injectors, factory FPR.
New ground cable on battery; new coils; new spark plugs; new TW sensor.

[/TD]
[/TR]
[TR]
[TD]March 2015[/TD]
[TD]

Feb. 28: set up F/IC with base map generated by AEM software, adjust fuel slightly to minimize short-term trims, drive home without incident.
Look at F/IC behavior more closely and determine that MAP sensor is insensitive to pressure but centered around atmospheric. Buy new F/IC.
New F/IC in car and it starts and runs perfectly; go for a short drive.
Next morning, cold start is rough with hunting idle. Code P0410 (secondary air injection) and P1399 (random misfire). AFR ~10 during rough idle/misfire. Smooths out and seems fine; I drive a different car to work.
That evening, I replace the battery with a new Group 35. Car immediately starts and runs perfectly. Fuel trims and AFR looks good. Drive for ~15 min including stopping for fuel and no issues whatsoever. When home and engine warm, I adjusted the idle screw per the manual (service connector shorted, idle set to ~600 rpm).
Next morning, cold start again misfire and rough running, with hunting idle. AFR ~10. Does not clear after a minute or two. Pulled codes with service connector and it flashed 71, 72, 73, 74, 75, 76. OBD2 indicated P1399.

[/TD]
[/TR]
</tbody>[/TABLE]

So, the following are new in the last ~3k miles:

Fuel pump (Deatschwerks DW200)
Fuel filter (OEM)
Fuel lines
Injectors (RC 550 cc/min) w/ AEM F/IC piggyback
Battery
Battery ground cable
Ignition coils
Spark plugs (currently Denso IK22 (heat range 7, 0.031" gap))
TW sensor
Valve clearance adjustment

Is there something I am missing?

Since the car sometimes starts and runs perfectly and other times does not, and because this issue occurred with the original CTSC fueling setup, I do not think that simply pulling fuel from the F/IC map is the solution. It further seems that the map is close given the <10% fuel trims that that OBD2 system is applying.

Since the problem (at least until this morning) cleared once things warmed up a bit, it seems like it's related to something the ECM does during cold start.

The P0410 code from a few days ago seems like a red herring, as that system only affects the flow after the exhaust ports, so shouldn't cause poor engine running.

I checked the F/IC log and the factory MAP sensor follows the F/IC MAP sensor perfectly. Both indicate ~33.5 kPa (-20 inHg) at idle.

I ordered a used ignitor (ICM) from eBay to try. My CTSC has the ignitor under the intake/blower; does that really mean I need to pull the entire blower off to try a new ignitor?

Any other ideas? Please?

I have read many threads related to hesitation, misfire, cold start, etc.

http://www.nsxprime.com/forum/showthread.php/154609-Strong-hesitation-when-cold

http://www.nsxprime.com/forum/showthread.php/135300-Problem-at-startup!

http://www.nsxprime.com/forum/showthread.php/117313-What-could-be-causing-code-P1399

http://www.nsxprime.com/forum/showthread.php/145195-Igniter-unit

http://www.nsxprime.com/forum/showthread.php/184823-Another-random-misfire-thread

http://www.nsxprime.com/forum/showthread.php/162175-HELP!-Constant-misfire-getting-worse-daily

http://www.nsxprime.com/forum/showthread.php/172286-Mystery-motor-problem-code-P1399

http://www.nsxprime.com/forum/showthread.php/180877-Ya-think-that-I-m-running-rich

jwmelvin · 14 March 2015

After all that, I'm pretty sure the issue was a loose belt. Once I tightened it up, the car started and ran perfectly. This morning it did the same! You may imagine how much of a relief this is to me. Now I just need to let the clutch break in and get it tuned.

BrianB · 14 March 2015

I had a cold idle issue but I'm obd1. I had to lower my fuel pressure and adjust fast idle valve and idle screw. It was running too rich at idle.

I'm still on the rrfpr with the fic and 550 injectors. When I go for a new tune I'm going to get rid of the rrfpr.

Good luck! Hopefully the belt fixed it, who'd you go to for the tune?

jwmelvin · 14 March 2015

Will be going to John Kerr in Carlisle, PA. http://www.j-k-tuning.com

jwmelvin · 16 March 2015

I'm curious if anyone can explain what is really going on here, spark or air?

If spark: a loose belt causes spark issues because of the alternator? I would have thought the battery fills in for a while, and my voltage gauge measures ~14 V.

If air: a loose belt causes the supercharger to spin slow, starving air? I would have thought the bypass valve allows air to go around the supercharger (in reverse of its typical flow). Or is the bypass valve effectively one way because of some aspect of its design?

I'd really like to better understand what is happening because it doesn't seem like the system should be as sensitive as it is to a loose belt.

syndicate · 16 March 2015

I think the bypass only opens a little at idle and requires more pressure to open further. Sounds like the slipping is occurring at the supercharger. Try loosening it again and see what happens

jwmelvin · 19 March 2015

progress: fuel-pressure issue

Okay, an update. Dave Dozier [MENTION=4799]DDozier[/MENTION] and Ben at Daisy [MENTION=14909]redbird92[/MENTION] have been very helpful in advising me as I get closer to the problem. See here. Notwithstanding some false correlations I was drawing, I appear to be having an issue with my fuel-pressure regulator handling the flow from my Deatschwerks DW200 pump. While that doesn't necessarily explain the intermittent nature of the issue, it needs to be addressed before moving on. Here are my latest observations:

With OEM FPR, at idle, pressure is ~48 psi.
If I use a hand vacuum pump (Mityvac) to pull the FPR down to 20 inHg vacuum, when the fuel pump primes (key turned on but engine stopped), pressure is ~38 psi. That is the expected behavior with the 1:1 FPR.
As soon as the engine is started, pressure jumps up to ~48 psi. This indicates the FPR cannot return enough fuel to the tank.
I pulled the return line off the FPR and can easily blow through it into the fuel tank, indicating that there is no restriction or kink in the return line.

So, I'm feeling pretty good when I figure this out and decide that it's time to put in the AEM regulator that I bought (part number 25-303BK). It is designed to handle higher-flow pumps, and comes with a set of three interchangeable orifices for the output so that you can accommodate a range of flows. But I quickly realize that it does not fit because it interferes with the timing-belt cover or the part of the block that bolts to. I do some searching on Prime and see this thread dealing with the exact same issue. (There was a considerable amount of frustrated yelling and cursing in the meantime.) So now I have ordered some fittings to try and get a little more room for the AEM FPR. Thank you [MENTION=18698]NSX-Files[/MENTION] for interrupting my screw-it-all-I'm-changing-everything attitude last night.

I did go through a full cycle of planning how to put another regulator in there (not so different from the original Comptech setup with the so-called rising-rate FPR). For others who may be looking at this type of thing in the future, read this web page, which explains the connectors quite well; I didn't find it until I had finally gotten up to speed on the various connections. I was about to order the required parts for a remote regulator, but remote mounting the FPR seems less desirable if I can get the AEM to work, just from a parts-count and packaging perspective. That said, it would have the benefit of minimizing the number of NPT fittings, which I have always found suboptimal. With the current plan (the approach by NSX-Files I linked to), I prefer to avoid the pressure loss from two elbows, so I'm going to first try a single 45, and if that doesn't work, a pair of 45s. Will update...

DDozier · 20 March 2015

syndicate said:
I think the bypass only opens a little at idle and requires more pressure to open further. Sounds like the slipping is occurring at the supercharger. Try loosening it again and see what happens

The bypass is controlled by Vacuum, at idle you are at peak vacuum except while decelerating with your foot off the throttle. The bypass valve is fully open while in Vacuum and closed as the engine transitions to boost.

Dave

jwmelvin · 21 March 2015

solved? new FPR mounted

I managed to get the AEM fuel-pressure regulator to fit, by using a single 45-degree, 1/4-npt street elbow. I went with this one from McMaster; it's quite good quality and has a larger passageway than the 90-degree elbows that I also bought. Here's the fitting:

I installed it so that it angled the SOS fitting over and up. Using some angle up helps the return barb line up with the factory hard line that it must connect to. Here's a picture of the install:

(full-size link)

And now the fuel pressure is 40 psi at idle, 50 psi with the vacuum line at atmospheric pressure, just like Honda intended. I will note that the AEM regulator does not hold the static pressure if the fuel pump is not running, so I'm guessing that the car will start best without a delay after the fuel pump primes. On my test, it fired right up and idled perfectly, with AFR of 15 (after the secondary air injection finished) until it shifted to closed loop.

I also cleaned the ground point where the ignitor harness bolts to the intake plenum, as suggested by Brian @Source1. Thanks.

Will update after I confirm things are back to normal, but I feel prepared for my tuning session next week. Thank you so much to Dave [MENTION=4799]DDozier[/MENTION] and Ben [MENTION=14909]redbird92[/MENTION]. I finally feel like my car will be enjoyable again, soon.

jwmelvin · 25 March 2015

solved again?

Vacuum line #4 !!!

Yes, I thought I had blown out that line; I must not have done a very good job. Given how this cause shows up in threads on Prime often enough, I feel a bit silly for chasing my tail. But without this issue, I wouldn't have looked so carefully at fuel pressure and might not have noticed the regulator being unable to keep up with the flow. In all, assuming this truly resolved the issue, I'm very happy.

Here is the OBD2 log that showed me what the issue was:

You can see the MAP sensor reading ~95 kPa upon startup, rather than ~35-40 like it should. Unsurprisingly, the situation is dire for combustion as a lot of fuel is mixed with not a lot of air. I cleaned the line with carb cleaner and compressed air, starting from within the box of vacuum components. I plan to move the pressure feed from the bottom of the CTSC plenum/manifold to the side, so that it is less likely to welcome contaminants.

Again, thanks to Dave [MENTION=4799]DDozier[/MENTION] and Ben [MENTION=14909]redbird92[/MENTION] for both suggesting the way to figure this out was to log OBD2 while the issue arose.

dlfourty2 · 13 January 2019

Was the issue with Vacuum line #4 ? Mine is having the same issue until I gas it a bit and the car warms up.

jwmelvin said:
Vacuum line [URL=http://www.nsxprime.com/forum/usertag.php?do=list&action=hash&hash=4]#4 [/URL] !!!

Yes, I thought I had blown out that line; I must not have done a very good job. Given how this cause shows up in threads on Prime often enough, I feel a bit silly for chasing my tail. But without this issue, I wouldn't have looked so carefully at fuel pressure and might not have noticed the regulator being unable to keep up with the flow. In all, assuming this truly resolved the issue, I'm very happy.

Here is the OBD2 log that showed me what the issue was:
View attachment 122069

You can see the MAP sensor reading ~95 kPa upon startup, rather than ~35-40 like it should. Unsurprisingly, the situation is dire for combustion as a lot of fuel is mixed with not a lot of air. I cleaned the line with carb cleaner and compressed air, starting from within the box of vacuum components. I plan to move the pressure feed from the bottom of the CTSC plenum/manifold to the side, so that it is less likely to welcome contaminants.

Again, thanks to Dave @DDozier and Ben @redbird92 for both suggesting the way to figure this out was to log OBD2 while the issue arose.

jwmelvin · 14 January 2019

dlfourty2 said:
Was the issue with Vacuum line #4 ? Mine is having the same issue until I gas it a bit and the car warms up.

Yes the line to the MAP sensor was fed from the bottom of the intake plenum and must have had some sort of oil in it. After figuring it out I moved the tap to the side of the plenum and haven’t had a problem since.

Intermittent/cold misfire issue

Similar threads