Accord surging/pulsing rhythmic idle before warmed up

[Uncle Ben]

I've been working on fixing my Honda's idle issues for a couple weeks now, but no improvements yet.

When I start the car cold, the idle is a little high right from the beginning at around 1,500rpm or sometimes a little less, but not terribly high for it being cold. After running for just a few minutes it will usually start to surge up and pulse up and down in a consistent rhythm, usually between 1,000 and 2,000 to 2,500 or so, but it has already surged up to 3,000rpm a few times. It feels just like I'm pressing on the gas and letting it back off every 2 seconds. This problem has been going on for about 3 weeks, but there have been times that I have driven it in the last week and it would not do the surging/pulsing at all, so it is not consitent, but it does it more often than not.

Last night, for instance, I started it cold and it idled at about 1300 for several minutes and then suddenly, with it still in park, it started the surging, which continued for more than 5 minutes. It finally stopped after I held on the gas around 2,500 rpm to help warm it up (that was the lowest rpm I could get out of it while holding the gas as little as possible. Any less and it would go back to surging), and I also noticed that if I put the car in reverse it would stop surging immediately and would go to regular idle speed, but as soon as I would put it in park or neutral it would start surging/pulsing again. After about 15 minutes of driving or just letting the car get up to operating temperature, the idle runs perfect at around 700rpm and there are no issues at all when FULLY warm, so it is just about always when the car it not fully up to operating temperature. There have been times where I only drove for a few minutes and the idle problems stopped, and there have been times where it won't stop until after at least 15 minutes of driving. And there are even times where it seems to run mostly fine, right from the cold start. I have also had the car run well for a short 5 minute trip (from cold) and then when I get to my nearby destination and put it in park, it immediately starts to surge up and down in the rhythmic pattern. It was running fine from cold, and only started the puling after I put it in park 5 minutes later. The engine probably was not up to full operating temperature yet, but it seems strange that it would be running fine until I put it in park after a short 5 minute trip. That happened twice this past weekend.

Also, about a week ago I had my first stall, which was while I was driving down the street around 40 mph. This was in the afternoon when it was warm out, but after only about 5 minutes of driving, so the engine was probably not up to full operating temp quite yet. The rpm went way down and the engine shook and then it stalled. It immediately restarted, but idled very rough and now, also for the first time, had the check engine light come on at that moment. Even though the car would start back up and idle very rough and low, when I pressed on the gas it would not respond and would barely more. I managed to get pulled over and turned the car off and sat there for a few minutes. When I started it again it idled normally and ran just fine and I proceeded to drive it home with no problems at all. It ran and idled perfectly for the next 20 minutes all the way home. The code I pulled was P1129 MAP Sensor Signal Higher Than Expected. The car had already been doing the poor cold idling and surging for at least 2 weeks at this point, but it had never stalled or gotten close to stalling and had never giving a trouble code up to this point. I cleared the code after doing some of the work that I'll mention below, and the check engine light has not come back on since about a week ago, but the problem still persists.

Here is what I have done already:

-tested the map sensor by backprobing and checking for the correct voltage with key on, engine off
-removed the map sensor and connected 2 of the pins to a DVOM (backprobed with electric connector still attached) and then applied vacuum to the sensor and watched the voltage go down (as it should) as vacuum was applied, and go back up as vacuum was removed

Based on these tests resulting exactly as they should, I ruled out the MAP sensor

Here are the other things I've already done:
-removed intake and thoroughly cleaned the carbon buildup off the inside of the intake chamber and egr passages
-thoroughly cleaned the bore and plate of the throttle body
-thoroughly cleaned the idle air control valve and made sure the flap was able to move freely
-bled all air out of the cooling system
-tested the egr valve according to the haynes manual: tested for continuity with the electrical connector attached (backprobed)
-tested the egr valve by removing the electrical connector, then starting the engine and applying battery voltage and ground to 2 of the terminals and the engine immediately stalled (which indicates that the egr is working)
-checked the air filter and it is pretty clean and only a few months old
-replaced the PCV valve
-checked for vacuum leaks (smoke test)

None of these tests pointed to any issues, since all the parts tested normal and no vacuum leaks were found, and I smoke tested twice and used multiple different vacuum locations each time I tested, just to be extra thorough.

All the parts that were cleaned thoroughly with carb cleaner did not seem to change the engine idle issues at all.

The only thing I can think to try next is to just throw a new MAP sensor and Idle Air Control Valve at it to see if either of those fixes it, but I hate throwing parts at it and also potentially wasting money, but I don't know what else to try.

Thanks for your input!

 

[Briansol]

I think the code you got is a byproduct of the the real symptom.

tps, iacv, and iat are the items that would be the things i would look at. seems you already hit these though.

 

[Uncle Ben]

Sorry for forgetting to give the vehicle info! It is a 2001 Honda Accord w/ a 4cyl 2.3L ULEV V-TEC engine. Everything is stock.

The problems started about 3 weeks ago, but it was not after any repairs or maintenance.

UPDATE: last night I checked for vacuum leaks (smoke test) one more time and found a very small leak where the air cleaner duct attaches to the throttle body. I cleaned the outside of the throttle bore and the inside edge of the air cleaner duct with a carb cleaner-soaked rag. I also changed the MAP sensor.

After doing this, I started the car up (cold) and it immediately seemed to be running much better than it has been. It did not idle high and it did not do any of the surging, and after only a few minutes the idle was settling down to the normal "warmed up" idle, so things were looking good...BUT, a new check engine code showed up for P1519 for "idle air control valve circuit fault". Very strange that a new code would pop up, especially not directly related to the new MAP sensor that I had just put in...and also strange because the car was running great for the first time in weeks, yet it tripped a new code.

I noted the code and then cleared it and started the car back up again. Then I tried wiggling the wiring connector to see if there was a short and to see if the code would pop up again, but it never did. I let the car cool for over an hour (50 degree night weather), then started it up again and drove it for a mile or so and came back and parked it and it never idled high and never surged even once. I also drove it this morning about 10 miles to work and again it idled just as it should from cold start and continued idling perfectly the whole trip.

I also tested the idle air control valve according to a service manual page that someone sent me, and tested the vacuum at it while idling, and then also unplugged the electrical connector on it while idling to see if the vacuum would change, and it seemed to indicate that the IACV was functioning correctly.

I know cars can be inconsistent, so I'm not claiming victory yet, but I'll continue driving it and seeing if any of the idle issues come back.

 

[Uncle Ben]

Unfortunately some of the idle issues have come back after less than a week of driving after the car seemed to be fixed. It is not doing the crazy pulsing/surging that it was doing before, but it is idling too high while cold. Yesterday it was at least 15 minutes of driving before the idle went down to the normal 700 or so rpm, and I noticed immediately when it did. During the whole drive up to that point, it would not idle at anything less than about 1,000 rpm, and it would occationally jump up or dip down 100 or 200 rpm for just a split second, and only once, but after about 15 minutes of driving I suddenly saw the rpm's fall down to 700 or so and it never idled high again for the rest of that trip, which was a total of 35 minutes or so. It was like it just suddenly reached full operating temp or something suddenly changed and then it idled perfectly from that point on.

This morning it idled a little high from cold as well, but this time it only lasted about 5 minutes and it never did the quick, single 100 or 200 rpm jumps like the previous day, and then dropped down to about 700 rpm and ran great for the next 30 minutes until I reached my destination.

No check engine light. Not sure what to think or what to check next...

 

[Briansol]

cold start higher idle is normal. The cars are programmed to do that to warm up the catalyst system for emissions gains quicker. Is it super cold where you are?
it shouldn't last after the water temp is up to just below regular operating temp

 

[Uncle Ben]

I'm in Southern California. Even in the morning around 6:30am it is only about 40 or 45 degrees at the lowest. You're right that it is normal to idle higher when cold, but if normal idle is 700 or so with this car, then a higher idle at 1200 or so seems too high. Also, there is the issue of the idle jumping up suddenly, or dropping down suddenly. At the moment, it is not back to doing the pulsed surging up and down continuously for minutes on end, but what it is currently doing after all the repairs I've done up to this point, is that it is idling at around 1200 for around 10 minutes of driving (sometimes more) and during that time occasionally the rpm will very quickly jump up 100 or 200 rpm and stay there, or it will very quickly fall down about 100 or 200 rpm and stay there, at least for several seconds. It just doesn't feel right, and I don't recall it doing this last winter in the colder weather.

Thanks again for your input and suggestions!