Rough idle, stalls, when engine warm. Runs fine when cold.
Customer Question
1998 mitsubishi galant, manufactured fall 97. Battery remained disconnected for extended period. Car
runs fine when cold, but if left idle, once warm, idles extremely rough, and dies out when attempting
to drive after warm up. Won't pop up any codes either. Runs great with MAS unplugged, but then
the obvious codes for that show up. Any suggestions?
The majority of idle issues occurring after battery disconnection can be resolved with
a throttle cleaning. When the battery is disconnected the computer loses it's adaptation
for the idle air bypass and the idle air control valve resets to it's default position.
This is OK when cold since the computer is requesting extra air anyway, but once the
engine has warmed up and the computer reduces air bypass for normal idle, it is no longer
sufficient to keep the engine running smoothly or even at all.
Proper repair is straight forward: Pull the intake hose off and thoroughly clean the
throttle body so that there are no dark deposits at all, it should show nothing
but smooth silver. In severe cases the idle valve will need to be cleaned too, this
is not typical though.
If the problem were to maintain afterward we would want to check idle valve operation
to make sure it has not seized from age, as well as the belt timing to make sure it
has not slipped. Realistically though, based on the symptoms and how they cam about
as well as being able to unplug the MAF and regain operation, the throttle body condition
is your #1 suspect.
Ok, yeah, I'll give that a shot, and get back to you. I need to get all the systems
ready for an Illinois OBDII smog test.
On an interesting side note, when I changed the ECU, I initially let it warm up until
rough idle, then keyed off and on several times, each time waiting for the MIL to go
out, then started it again. This time it started with an idle around 1500rpm, slowly
dropped to 1k, then quickly cycled back up to 1500. It did this for a minute or two,
then dropped down to the rough idle, which seems to run between 500-800, kind of hunting
around. Is there a key sequence that is used to help it learn? The car is undriveable
when warm.
I haven't got in to the throttle body yet, but going to do that right now. Is there a way to
clean the IAC on the car? Or would the servo motor be more suspect than dirtiness?
I had a mechanic (non-mitsu) check the the MAS, he said he couldn't get a frequency reading
from the computer with his equipment. Checked line voltages, though, and they were good.
There is no set method to re-learn. It will over time on it's own, but the proper method
is to clean the throttle out fully then disconnect the battery if the adaptation needs
to be reset. Otherwise it is just the drive and wait game.
The IAC has to be removed to be cleaned. On a 98 I believe it will have three screws
holding it on. You can then slide it out and clean the pintle off. Make sure you do
not drop the o-ring if you remove it though. You will have a hard time finding a replacement
if it is lost.
If there is no improvement after cleaning both and the IAC is definitely moving and
not seized, then we can look into a MAF issue.... though they are quite rare. Typically
the only failures we see on those are when people try to clean them or close the hood
with something laying on top of it (oops!). We will need a proper scan tool however to get either a frequency or gm/s reading from
it. If we can definitely not get a reading but the reference voltages are there, then
we would have a bad MAF.
Start with the easy/free stuff first though, especially with it being more common. The
MAF is expensive and you don't want to jump right to that when other basic maintenance
issues haven't been addressed yet.
Yeah, it's been like this since I bought the car three years ago. Too late on the MAS,
I've two now.
So once I get the IAC off, do I just want to wipe it out? Everything I've read says
to plug holes on throttle body before cleaning so no solvent gets in IAC, and I'd rather
not mess it up by drenching it or trying to spray it off. By far, the IAC for this car
is the least available and most pricey part of what I'm dealing with, in chicagoland...
Remove the intake hose. Spray brake cleaner or comparable cleaning agent directly onto a rag and use it to wipe
out the throttle. If there are stubborn deposits you can use a scotch brite pad, wire
brush, etc to knock them loose and then wipe it out again with a rag soaked in cleaner.
Do not spray anything into the throttle body.
Remove the IAC and with it out of the vehicle use brake cleaner to spray directly
onto it then wipe it off. Let it dry and reinstall.
The purpose of the "block the ports" etc are assuming you are spraying into the throttle
body which you should never do. This will cause difficulty starting when the
IAC ports fill up, as well as potential flooding spraying into the intake etc. Just
do as above and you'll be OK.
Can the IAC be removed with the throttle body still attached to the manifold? Might
need a diagram of screw location if you have one handy.
Upper throttle body was gunky but it wiped right off, just a little bit of hard deposits
on the upper inside of the throttle plate, which was gunky also.
Yes you should have the three screw model IAC that you only need to remove the intake
hose to access. The IAC is then directly below the throttle opening with 3 philips screws
holding it on.
Here is the best picture I have, as you can see it is viewed from behind the throttle
body, but should still get the point across:
Yeah, I think that diagram is for a '99 and later. The IAC on mine has a straight 6
pin connector, that comes in under the throttle cable, opposite the TPS. With that being
said, I'm gonna pick up a throttle gasket and take it off, to really get it good and
clean and see what the IAC is all about. It looks like it mounts from the bottom. Replaced
the spark plugs and air filter today, as well, to see if that'd help; it'd been awhile.
After spending an hour meticulously cleaning the throttle body and plate, I put it back
together to give it a shot.
Couple of notes: Initially, it ran fine till it warmed up, at about 1200 rpm , then
back to the terribly rough idle.... feels like it's missing, too. On several cylinders.
Would cycle between 500-800 rpm, kinda hunting and gagging around, at about a 1.5 sec
interval. Let it idle for about 15 mins straight, not touching anything. Eventually,
it smoothed a little to 600-800 rpm. Finally, i keyed it off, waited and started it
again. Went up to about 1500 rpm initially, then dropped to 1000 for a bit. After a
couple mins, it dropped back down to the 600-800 range. Let it sit there for a while,
then keyed off and started again. After a couple times of this, it seemed like it smoothed
out a little, so I revved it to about 3k and held. With a steady throttle, it fluctuated
between 3 and 3.5 for a couple mins, then bogged out and stalled. Next time I started
it, it went to 2k intially then started hunting between 1k and 1500, with about a two
sec interval. After a couple mins, fluctuated around 750, but not steady. Did that a
couple times and decided to take it for a spin.
Things I noticed: Would accelerate fine, but as soon as I released the throttle,
it would bog down a bit. Additional throttle would bog, gag, backfire and start missing.
Restart, run fine till release. A couple times I managed to work through it, but still
same thing... release pedal, bog down. Restart, try again. After a good hour of this,
no codes had popped, MIL off, scan tool says pass. After a restart, while it was hunting
between 1-1.5k, I unplugged the IAC. That didn't change anything with the symptoms,
still hunted, then slowed and gagged with throttle. Restarted, same thing. Of course,
then the code P505 popped. Plugged it back in, cleared code, tried again, same thing,
fast initial rev of 2k, then hunting between 1-1.5. Waiting till it was done hunting
would leave the idle high, around 1-1.1...put it in gear and go without touching the
pedal it would idle down the road at about 900... but touch the pedal, maybe a surge,
then gagging.
I suppose the MUT-III would make quick work of this, but the nearest dealer is 60
miles away, and I'm afraid of the price.
Do I need to bother taking the throttle off to clean the IAC, or do those symptoms
lead you to believe something different? Metered the TPS, did notice there was about
30 ohms on the idle position switch when closed, but throttle position sensor seems
to follow the curve fine. Couldn't get to the IAC to meter it, that'll be when I take
it off. With the straight 6 pin configuration, will the test be the same? (between 2-1
and 2-3, 5-4 and 5-6?)
Of course, it still runs great in the open-loop once it's warmed up, you wouldn't
even know there was a problem......
Seems like one of those sensors going bad is a common problem, no junkyards in the
area have throttle bodies.
Or do I just need to be patient and keep driving it till it learns? Seems like a
ridiculous amount of time for a relearn curve, but I dunno, it' s my first mitsubishi.
It baffles me that no codes will show up with an hour of drive time and it running so
crappy....
That diagram if for the 98-03 Galant. Being a late 97 build you probably have the early
style (92-97) and then you have the butterfly style valve on the bottom of the throttle
body. In this case cleaning without removing the throttle is not possible, and I further
wouldn't recommend removal for cleaning either. If you want to do it and you have the
philips screws holding it on instead of bolts, you will want to use an impact screw
driver to remove them. The heads will strip out if you try to use a normal screw driver.
From the description it sounds like that IAC may be failing. When the problem occurs,
try tapping on the bottom of the throttle hitting the IAC to see if it wakes up.
And yes, proper diagnosis would require a MUT or a scan tool that reads IAC stepping
(very few do). At this point it is not going to get significantly better than it has
already and it is safe to assume there is an issue at play.
What are the possibilities of bad brains? I bought the car used and it was in this state
from the get-go. I called the mechanic this morning that had it on the scanner to clarify
a couple things. He said the computer wasn't receiving a frequency from the MAS, or
wasn't sending it to his scanner, so he suspects the ECU may be at fault. I've tried
5 different MAS on this car, including the original, so he kind of ruled that out. I'm
not definite on that though, as besides the original one I bought the car with, 2 were
remans and 2 were junkyard. Maybe I could verify that I have the right brains in the
car for VIN 4A3AJ56G1WE038008. The ECU I've got is MD346077. Is this right? Or would
you rule that out too?
Hi,