Mind-bending Rough Running...

[micadevcon]

Alright, I have a 91 LS400, 205k miles on it. I am having troubles with the engine running rough.

WARNING: There is alot of information here, I have tried to format it to make it more digestible... Sorry, but I want to get you guys as much detail as you need, as I was told this was the place to find Lexus experts!

FYI: I have, as of yet, not gotten a Check Engine light to come on.

Symptoms:

Acceleration is spotty. Ease on the throttle and its fine, somewhat, but hit the throttle hard (while moving) and it feels like the engine hits some sort of resistance, it just kind of sits there wondering what to do, and then it all of a sudden (once the RPMs eventually rise or the throttle changes) gets all its power and throws me back in my seat. Now, the areas where it hesitates vary, depending on how warm the car is and what mood it is in. As mentioned before, it is pretty smooth when little throttle is applied, but if the throttle passes certain points it hesitates, overcomes, hesitates, etc.

Idle is rough. In drive, the car shakes, shudders, etc as RPMs drop randomly. In park it continues, albeit at a higher RPM, but now the transmission is disconnected and it isnt as noticable. Once RPMs are increased it smooths out.

Misfiring? You can still feel a slight 'misfire', but the engine is smoother, overall, once RPMs increase. Driving down the highway with cruise on, or just maintaining speed, it will start to progressively misfire, till it gets to the point of bucking harshly. In the past, the car would only run rough like this after it warmed up. It would be simply charming and smooth when cold. Then it wams up, and misbehaves. Hitting the throttle hard would also cause it to start having problems, even if cold. Now, it does this all the time. From the instant it starts up.

Odd Cold Starting. Maybe its the behavior of the cold-start injector, but when it is very cold, upon start up it acts like a diesel, about 300-400 rpm, very distinct CHUG CHUG CHUG, about twice the frequency of a diesel engine idling. Like its just shuddering. Smooths out after a minute or so, or if I manually rev the engine.

Things replaced: Updated June 2008

Spark Plugs (Denso Iridium IV Platinums. Old plugs were Bosch Platinums, oddly wet and smelled of gas... hmm)

Plug Wires (Duralast... I know, but the resistances were good and they fit in the clips. Car had the original wires and one was completely severed. Surely it would help replacing. Nope.)

EDIT: Have now replaced 'Duralast' wires with wires from Lexus - no change.

Distributor Caps and Rotors (Also Duralast, but look to be in good condish. Works fine. Still didnt fix it.)

Front O2 sensors (Bosch universals, spliced in connectors, also working fine. Cleared the TRAC-OFF being on all the time, but its come back (See post on dragging brakes).)

Mass Air Flow Sensor / Air Flow Meter (Used from salvage yard, no change effected. I tried this as at one time disconnecting and reconnecting caused the engine to run smooth for 2-4 minutes. Sadly, didnt fix it.)

Fuel Pressure Regulator (Old one's diaphragm was bust, leaked fuel into vacuum line and didnt regulate at all, but still no change)

Idle Air Control Valve (Resistance on existing one was low, otherwise seemed fine. Replaced with unit from SC400 - No improvement)

Diagnoses tried:

I cleaned the throttle body and intake manifold out, had the usual 1-2mm of carbon gunk buildup. No change!

I checked battery-block-vehicle ground. Seems to be okay. No change.

I have run four injector treatments of various brands, even Marvel Mystery Oil. Seafoamed the engine. No change.

I bought a DSO for diagnosing this silly thing and found a few interesting things. The ignition has an interesting pattern. The no. 1 (Passenger) coil's output is fairly similar to what I know ignition waveforms look like. Drop down, wait, spike, gradual decrease, repeat. But the no. 2 coil output is a bit odd. Usual drop, then spike, but then it plateaus for 1-2ms, spikes again, and immediately drops to zero. I can provide the waveforms if someone would like to see. What is interesting is the fact that at idle, when everything is very rough, the ignition waveform for both coils is all over the place, seemingly random, but as the throttle is increased, it smooths out and becomes a steady, mostly proper waveform. I connected the scope to the output from the ECU to the igniter, looks good and clean; then to the output from the igniter to the coil, alongside the coil output. They are the same waveform, odd cuttoff in no. 2 is on the primary and secondary. But the jittery/random waveforms at idle which smooth out arent present in the igniter output to the coil.

Coils can be eliminated since the waveform smooths out as the car increases throttle.

Igniters can be eliminated since I swapped connectors and checked the signal, the oddities move with the connectors!

ECU can be eliminated as its triggering signal is steady and looks alright. Some spikes here and there, but not severe or significant.

That leaves wiring? But I dont know of any wiring fault that would cause a perfect ignition spike to suddenly drop after 1-2ms, each and every time!

I have tested EGR system, and it seems to be alright. Suck on the vacuum line when it idles and engine dies. Checked the VSV, it works. Only VSV that doesnt work is for the fuel pressure regulator and all its just there to prevent vapor lock.

IACV replaced and tested, idles perfect for a little while when valve is manually closed, but engine operation deteriorates quickly afterwards.

Help?

[aarman4]

I hope you weren;t expecting me to solve all of your problems, but i may be able to start it off....

Pull your speedo bezel(the big one around the entire dash, go to www.lexls.com and hunt aroud, it shows how to do it there), Your CEL bulb is most likely dead...(mine was, what a differemce when you can actually read your codes). If you don;t have a spare bulb, than move one over to CEL (I used the seatbelt reminder).

That being said, I got a denso oem o2 sensor, and a bosch universal o2 sensor, did a resistance check on the like shown on lexls's website, and sure enogh the universal was outside of the values that shuld be there, while he denso was bang on. You might want to lay underneath the car and just run a quick check on the resistance of them after you replace your cel in case it throws o2 codes at you.....

Thats all I got for now, best of luck, I just hope it's not something crazy like a plug arcing across and grounding on the cylinder... never heard of it happening, but it would explain a bit, lol...

[wrpsuite]

this may seem simple but have you changed the fuel filter?

[micadevcon]

Well, I dont expect a miracle fix-all from you guys! Just some fresh insight into the causes of this problem.

I shall replace the fuel filter, if anything to remove it as a suspect. I want to do a fuel pressure check but I cannot find the fittings needed to connect a fuel pressure gauge to the odd banjo-fitting system used in this car.

When I put in the 02 sensors, I do believe they were within spec, but I can check those again.

As for the CEL, I thought it works, but with all the myriad of lights at startup that could be just an assumption. I will verify today.

Thanks for the help so far! Keep the ideas coming, should they strike.

[nc211]

One of three things is off balance, air, fuel, spark. Since you've had a significant problem with the fuel system already, I'd advise to revisit it again, as this does sound like too little fuel being delivered to the combustion chambers. The filter is a good idea, especially considering the regulator was leaking and there is no telling what might of found it's way into that leak. I assume you've thouroughly cleaned out the flooded vacuum line from the fuel? The connector on the line is tight, not leaking in air to affect it's ability?

To me, it just seems too coincidental that you've had fuel delivery problems in the past, and your symptoms now could clearly be pointed to fuel supply levels delivered to the engine... Just makes the most sense to me... Keep us posted! This is a new one one here! And good luck!

[micadevcon]

Well, that suprised me. I just checked the lights, and the CEL does not light up with the rest of the indicators! It must indeed be out, so I will pull the dash cluster as soon as I can and swap the bulb. Where can I get an oem replacement or equivalent bulb?

Its so strange, I was almost certain that light was there, what with all the other indicators that light up...

Finally, a step in the right direction! Thanks so very much, and I will keep this updated with my progress.

[micadevcon]

Alright, the CEL light was indeed burned out and I swapped it with another.

Code it puts out is 13, dealing with the "No. 2 RPM Signal Circuit", aka the cam sensor. I will look at the connector as previous Code 13 postings point to that, but I sure hope it isnt that the timing belt jumped a tooth, cause I really dont feel like re-doing the timing on this thing right now!

[wrpsuite]

i'd change the fuel filter before getting into all that...seems simple

[micadevcon]

Well, the cam position sensors checked out, resistances ok, and the wires for them seem okay, no shorts.

The code indicates to check the engine speed sensor as well, but I am not sure how to get at it, so I might just check all those signals with the car running at the ecu using the scope.

Attached is the information I could find on it, fairly cryptic but NE is Engine Speed Sensor, G1 and G2 seem to be the Cam Position Sensors.

Will do the fuel filter soon.

[aarman4]

I am sure your lexus dealership will have the bulbs. I just went on ebay and managed to find an entire speedo bezel with all bulbs included and got it with all bulbs working for $10, but I doubt you will be so lucky.

[wrpsuite]

please change the fuel filter, it will make me happy!

[micadevcon]

Yes sir! :D I too am eager to see if it makes a difference. The optimist in me hopes it does, but the realist in me feels its something else. I will give it a shot tomorrow evening as work is keeping me busy.

I have yet to test the fuel pressure... Anyone know where to get the proper fittings to check fuel pressure?

Also, today the check engine light hasnt come on, nor has the TRAC-OFF light illuminated. Yet the rough running persists...

[micadevcon]

Ignition coils ruled out. Bought a new coil to test with and it didnt make a difference.

Research by a friend indicated the IACV might be dirty/faulty? Does this happen often on LS400's? I know it does on the 6-cyl Lexus lines, but what about the 1uzfe?

Havent been able to get old fuel filter off, too much corrosion and dont want to chance damage to the line, soaking with liquid wrench for now.

I doubt it is the fuel delivery system. If the filter were indeed clogged, then logically increasing throttle (thereby increasing fuel demand) would cause the vehicle to run worse than at idle (when fuel use is minimal). In this case, higher throttle makes it run smoother.

Timing looks good, at idle and at speed, advancing properly and on the right marks for idle. Doesnt move around when its running rough.

Still no new codes.

[SKperformance]

Is the tensioner on the timing belt good and has not caused it to skip a tooth?

Fuel pump possibilities, heard of bad o2 sensors throwing it out also.

I would be tempted to check the fuel injectors.

[CUMan]

Have you checked for broken wires in the driver side trunk hinge?

The usual broken wire in this wiring harness is white with a black stripe. This wire is a ground wire. When it is broken, it creates all kinds of weird electrical problems. For example, when my 93 LS had the broken wire problem, the transmission shifted strangely and the brakes felt odd, as if they were grabbing. I have never heard of this condition causing engine problems, but it could be a possibility. If it affects the transmission shift pattern, could it not also affect the engine operation?

[micadevcon]

Have you checked for broken wires in the driver side trunk hinge?

The usual broken wire in this wiring harness is white with a black stripe. This wire is a ground wire. When it is broken, it creates all kinds of weird electrical problems. For example, when my 93 LS had the broken wire problem, the transmission shifted strangely and the brakes felt odd, as if they were grabbing. I have never heard of this condition causing engine problems, but it could be a possibility. If it affects the transmission shift pattern, could it not also affect the engine operation?

The rear brake lights sometimes dont light up on the passenger side, but the driving lights do light up. I will have to check that harness... Would I be able to see a fluctuation in the fuel pump's driving signal from the Diagnostic connector? I know there is a way to manually actuate is from the diag. connector.

SKperformance: I had thought about it skipping a tooth, but if it were bad timing the roughness would be constant. This roughness vaires widely, from a mere 'cough' misfire feeling to a near stall, and it doesnt follow any pattern. Havent checked the tensioner as I simply dont have the time to tear into the thing that deeply. Also, the code 13 was for a cam position sensor, so if it had skipped a tooth, the code would still exist as the cam shafts would be out of sync.

Fuel injectors are consistently coming to mind but they are a major chore to swap out/clean, since I think you have to remove the intake manifold to get at them? Hopefully my last resort. I will check the o2 sensor readings with the scope, its possible the new ones I put in are faulty, but improbable, as no code was thrown (Does not necessarily rule them out).

[SKperformance]

the manifold is easy to remove to get at the injectors.

[wrpsuite]

I'm no expert but think about all the time and expense toy are going through without changing a simple cheap part. Soak it in PB blaster and get er done!

[ryanmaher]

any updates I am having a problem with my 94 LS400 acting starved

[Harky]

...[Igniters can be eliminated since I swapped connectors and checked the signal, the oddities move with the connectors!]...

This is interesting...I assume that you switched both ends of the connectors...between the igniters and the coils...so that you did not produce a misfire. I don't know if a difference in the output of the coils is normal or not.

If the wierd signals moved with the connectors, that would point to the connectors...right? Have you checked continuity? connector condition? broken or partially grounded wire?

Also, since your problems seem to relate to throttle position, have you checked or adjusted your TPS???

Did you have it off the throttle body when you cleaned it?? How are the connections?? Check for continuity on the idle contact and a smooth linear change in resistance reading throughout the full stroke on the throttle position pot.

If it checks OK, it wouldn't hurt to adjust it a little. I read in another thread that an out-of-synch TPS caused rough running.

There is an adjustment procedure on the lexls.com site.

[micadevcon]

Just an update, I am continuing to have the same issues.

Replaced injectors, no change.

I ordered and installed OEM Ignition Wires, just to be sure (Had Autozone "Duralast" wires in it) and that didnt resolve the issue. I didnt think it would anyway.

I am back to wondering if this is all an EGR/Pipe issue. When the car has sit for a long time (8+ hours) and is started, idle is smooth. While the car warms up, the idle begins to drop, ever so slowly.

Around 500 rpm there starts a strong vibration in the cabin, and soon the misfiring starts. First a single misfire, a few seconds pass, and two more, and gradually the misfiring becomes constant. When revved, it goes away. Most of the time I have to put it in neutral at stop lights because the whole car jerks and shakes if its kept in drive.

This seems consistent with a lean misfire, most probably caused by the egr pipe from the exhaust leaking, adding too much air to the intake. It would seem that either the pipe is wholly cracked, or just a little, as the warmer the car gets (longer it runs) the misfire gets worse, affecting highway driving and cruising. Thus the crack must be expanding...

Then, when the car cools down, its like everything resets.

My concerns: If it were an EGR pipe leak, shouldnt I be hearing the sound of an exhaust leak? Popping, dieseling, etc? It doesnt seem to be doing that... or perhaps I am just used to it. There seems to be quite a bit of warm/hot air coming from the passenger side of the engine bay.

Any thoughts? I would like to see if I can confirm this diagnosis before I buy the pipe and spend a day putting it in.

Thanks!

[python]

where did u get ur injectors? you can check injectors by doing an ohm test between the 2 pins

[xoneinax]

OP, you ever figure this out ?

[micadevcon]

OP, you ever figure this out ?

Nope, still trying to figure it out. EGR pipe doesnt seem to be the issue. I checked the EGR, cleaned carbon off, soaked in Seafoam, no change. I did notice what could be a slow leak, VERY slow leak, when I had the fluid in the egr valve, but nothing drastic enough to cause these issues.

My next item to check is the IACV, which lends credence as the problem starts out slowly, as the car warms up, and stumbling starts in 'ticks' - progressing to stumbling and rough idle.

[micadevcon]

Seems the IACV can be ruled out...

I pulled the IACV off, cleaned it with some carb cleaner and really inspected it. Nothing looked too bad. Only issue I could find was that the coil resistances were low, about 22.5-25 ohms. Spec is 30-50 ohms.

So I manually closed the valve all the way, reinstalled and tried starting.

It Died.

And a second time. So... looks like the ECU doesnt have knowlege of the position of the valve at all times, its a 'reference' thing.

After starting it again with my foot on the throttle, I let off and it fell to... 600rpm! Smooth and steady! It stayed that way for quite a few minutes! It has never idled that smoothly, I had to double-take as I couldnt hear it or feel it, so THATS how a Lexus should be! :D

Only problem: With more time or as the car is driven, the valve seems to progressively open (or close?) more and more, causing it to run lean (or starve?) and sputter at idle. I thought that perhaps the low ohms were causing the valve to move in ways the ECU couldnt compensate. So I ordered a replacement.

Got an IACV from a SC400, same engine, only difference is in the intake hose diameter and configuration of the valve tip- but otherwise the same. Coil resistances were 45 ohms, so dead on!

After installing it I started it (did not manually close the valve this time, though) and she ran rough as usual.

I am really at a loss here, when I manually closed the valve, it ran perfectly for a little while, and gradually got worse from there. It tries really hard to stay at 600, but dies down and surges up, bouncing around 600 rpm. Hitting the throttle smoothes it all out... till it starts idling again...

I will next check out the TPS as Harky pointed out - perhaps it isnt getting the right signals for idle. Also doesnt seem to be any wiring issues on the trunk harness...