After leaving the mechanic, my car revs violently on cold starts. Why?
March 3, 2015 10:44 AM   Subscribe

2005 Mitsubishi Eclipse. This never happened before I took the car in. Should/Can the shop fix this?

On cold starts, my car loudly revs up to 3000 RPM, drops to about 2000 RPM, sits there for about 60 seconds, then begins to drop to 1000 RPM.
Total time from ignition to calm is about 2 minutes.

This is less dramatic when the car has been used within the last hours, non-existent if used in the last 30 minutes.

Before entering the shop, my car would just start peacefully at about 1500 RPM and stay there.

I took my car in for a 90,000 mile tune-up which included replacing the timing belt, the water pump, and a bunch of rubber tubes. Some brake work was done, too.

Shop says that my car is supposed to rev up like that to warm the engine and that I should run the car 10 minutes before driving it. So... does that mean this is good behavior, and my car had bad behavior these last 90,000 miles?

What could be causing this? Should/Can the shop fix this? What should I tell them?
posted by jander03 to Travel & Transportation (12 answers total)
Cars that have cold start control have a fast idle setting, this is correct. But only full race built engines have a fast idle that high....

1000 rpm is the proper idle speed for that car. I suspect that 1500rpm is the correct cold start fast idle. If the shop is telling you that a 3000/2000 rpm idle is normal, either you haven't explained it to them properly, or they're making stuff up. So you now have the correct idle speed when warm, but your cold start control is screwy. It should idle at 1500rpm or so for maybe a minute or two, then drop down to 1000rpm ish (usually less - maybe 800-950, but I don't know that car specifically).

Are you sure you have used the same rev numbers (and they are accurate) when explaining the issue? Or are they translating "It runs really fast and then runs slower after a while" into you having a fuzzy understanding of the fast idle process?

Shop says that my car is supposed to rev up like that to warm the engine and that I should run the car 10 minutes before driving it.

This is utter bullshit. No modern car needs to be idled for any period over 10 *seconds* before driving, never mind 10 minutes. They're spinning you a line, which makes me think they screwed something up and don't want to pay for it (if the exchange was genuine, as above). But it doesn't necessarily follow that a service screwed something up. It may be that your 1500rpm at-all0times idle from before was the car trying to overcome a timing issue from a stretched belt that has now gone away it has uncovered a hidden cold start issue that just wasn't apparent before.
posted by Brockles at 10:58 AM on March 3, 2015 [5 favorites]

Oops, what Brockles said. :)
posted by selfnoise at 10:59 AM on March 3, 2015

Hmmm, on re-read: You say a timing belt and a bunch of rubber tubes? Were they vacuum tubes? It may be that the ECU needs to relearn its idle because it has learnt how to control the engine with vacuum leaks (which affect idle speed) and a stretched belt or similar. How long have you driven the car since the service?

It may be the ECU needs to relearn, but they should maybe reset it if it is that far out of whack.
posted by Brockles at 11:00 AM on March 3, 2015 [1 favorite]

I am inclined to think something more along the lines of Brockles second post. My guess is that this is an ECU relearning issue due to big changes in either intake or vacuum. The plus side is that if they didn't swap the computer during this tune-up, the fix should be relatively straight forward (resetting the computer and letting it relearn in the manufacturer specified way, hose-related mechanicals, or an unplugged sensor or something)

The minus side is that they should know this already but instead of diagnosing it and fixing it, they are bullshitting you...
posted by milqman at 11:08 AM on March 3, 2015

I think an idle of 1000 is still to high - I think a 800ish is more correct. It should high idle on cold starts, but that should be in the1500- 2000 range. The ECU may need to relearn - as Brockles said - so drive it a bit and see what happens.

If that doesn't work, I would go back to the mechanic and tell them that the idle and high idle are too high. It can indicate other things being wrong, but it's usually a vacuum leak. While those are easy fixes, the trick is finding them.
posted by Pogo_Fuzzybutt at 11:14 AM on March 3, 2015

There may be a vacuum leak on the "tubes" that were replaced. Or the Idle Air control is malfunctioning. Or it's compensating for no longer having vacuum leaks.

Your mechanic should plug in his scan tool, and observe the requested idle vs. the actual idle. If the ECU is requesting that idle, it may still be learning.

For that matter, it might be worth it for both your sanity and your opinion of the mechanic to disconnect the battery for ten minutes, hook it back up, and see what happens. You should be able to do this at home.
posted by notsnot at 11:23 AM on March 3, 2015

On some poking around, this is something that has happened to other similar cars, but it's all people blathering about symptoms and wild guesses - nobody has come back and posted with what actually fixed the issue, so I'm still sticking with either re-learn from vacuum leaks or idle speed control issues (MAF sensor, speed controller etc, fast idle control - even coolant temp sender). Anything that affects how much fuel the car gets when cold only (as opposed to the parts that control idle in both circumstances). Either the engine doesn't know how cold it actually is (so over fuels) or the control mechanism is confused about how high to bump the idle.

There are so many modified versions of those cars its hard to get a sensible idea, but 800-900 seems to be the range those cars should idle at.
posted by Brockles at 11:31 AM on March 3, 2015

Response by poster: This is helpful! Are there negative or positive consequences to idling high?
posted by jander03 at 12:32 PM on March 3, 2015

Zero positive consequences, negative consequences because something is not working right and so may be costing you more fuel generally (not just at idle).

In addition, a high idle can make the car push on a bit when you are trying to stop (the engine is still pushing while you are trying to slow). So... not catastrophic, but not great.
posted by Brockles at 12:38 PM on March 3, 2015

I'm going to bet they caused a vacuum leak instead of fixing one when they replaced the hoses. Fixing a vacuum leak generally lowers the idle since unmetered air isn't getting into the system.

Go back and ask them to do a (free) smoke test to see which hose popped off or wasn't tightened down.
posted by hwyengr at 3:28 PM on March 3, 2015

Oh, and its not great for an engine to rev that high on startup before the oil has a chance to warm up. Not as bad in SF as in a Chicago winter, but if this is a drive-it-into-the-ground car for you, you'll want to get it fixed.
posted by hwyengr at 3:33 PM on March 3, 2015

Response by poster: Thanks everyone. Shop agreed to take another look.
posted by jander03 at 8:45 PM on March 3, 2015

« Older What's the best way to protect myself from...   |   Whiteboard anxiety Newer »
This thread is closed to new comments.