Drive Accord Honda Forums banner

2003 I4 idling problem, code P0507

66K views 23 replies 12 participants last post by  sickopsycho 
#1 ·
I haven't started this car in 6 weeks, I went to start it a week ago and the battery was dead. Called a friend over this weekend and got a boost, she started right up. As we were putting the boost cables away, the engine started idling weirdly. Got in it and saw the needle bounce from about 1100 RPM to 2200, the drop, then as soon as it got to 1100 bounce back up again, sounded like someone was revving it. We decided to drive it anyway to charge the battery, got on the highway and drove about 30-35 miles, at RPMs over 1100 it was just its normal self, no problems on the highway in any gears, hit 85 at one point, no prob. No check engine light at this point. We thought maybe the problem would clear up, when we got off the highway and took it out of gear it started doing it again. We parked it and went to a restaurant, 2 hours later we came back. She started by herself (yay, the battery got enough charge!), bu the problem still existed. Drove back the same distance on the highway, again no probs on the highway, I tried to pull codes while my friend was driving her and there were no codes.

JUST as we got to my house, the last light before pulling into my complex, she started doing it differently, very quickly between (I think) 1400 and 1700 RPM, whereas before it was between 1100 and 2200 and shouded like a vroooooooom, vroooooooom, now it was a vroom, vroom, much faster, and the tach needle looked weird bouncing up and down in such a narrow range. CEL did come on at this point. I parked it w/in 3 minutes of CEL coming on and then pulled the codes again, P0507 came up as the only code. In the book and on the net, P0507 is listed as

P0507 - Idle Control System RPM Higher Than Expected

Duh! I could've told you something is wrong with the idle. What do you guys think this is? Some sort of sensor that went "stale" in 6 weeks of sitting there? The car has been extremely reliable, has 60K now, I got her at 18K 2 years ago, no issues whatsoever, had a bad O2 sensor once and that was it over this period. It almost seems like the computer thinks the engine's gonna die when it drops toward 1100 RPM and it accelerates to keep it alive. When coming out of gear it initially went to 3100-3500 once or twice, then settled in a fixed 1100-2200 pattern, until the very end when it switched to the fast frequency of 1400-1700 RPM.

Note that I used to drive the car ~2000 miles/month prior to not even starting it once in 6 weeks.
 
See less See more
#2 ·
I forgot to say that driving in gear at low RPM, say under 2000-2500, the car feels very weird, when I lift my foot off the gas I experience very strong engine braking, like it's actively pulling the RPMs down.

Any ideas anyone? What sensors-gone-bad could lead to this?
 
#3 ·
At the Honda shop I used to work at, we did the 'Idle Learn' procedure on newer Accords (98+) that would come in with these problems. The tech's hated the process! - So I don't think it will be much fun.

http://www.driveaccord.net/forums/showpost.php?p=177150&postcount=4

I'd guess your ECM was reset while the battery was dead. :yes:
 
  • Like
Reactions: frogvtec
#4 ·
Thanks, I want to try this. Q:

  • I don't have an HDS, how do I re-reset the ECM? I have a ScanGauge 2, which is what I used to pull the codes, is it possible to reset ECM with it? Or should I just disconnect the positive battery terminal for X minutes?
  • Is it normal during idle learn for it to do what I've experienced (the RPM bouncing), for those who've done this before?
  • So if the battery dies on an Accord, people have to go thru this? Nice! NOT.
 
#5 ·
I would just try it as is... chances are the ECM is already reset, which would mean the HDS is unnecessary. - That's theoretically speaking, hopefully princess will chime in, or someone else who has done it will. - I'll call the tech I used to work with tomorrow if things still aren't resolved.
 
#6 ·
Much appreciated!

I tried the "idle learn" procedure. I disconnected the battery for about 15 min (it's a DX so no worries about radio code) and hooked her back up. It didn't work. Here are the specifics, maybe I did something wrong:

1. When I turned her into the ON position (pos II), the stupid aftermarket radio wouldn't turn off, it has a demo mode which is activated by default, and since the aftermarket radio got reset too it just wouldn't turn off; I tried taking the face off but couldn't. This led to being in the ON pos for longer than 2 seconds, maybe 10-15s. I started her with the aftermarket radio on in demo mode, I eventually managed to rip the face off about a minute into my 3000 RPM escapade.

2. The fans wouldn't come on (I had someone outside look at them). I have a ScanGaugeII which can read the coolant temp, I plugged it in (while holding the pedal @ 3000 RPM!) and waited till the temp hit 198F.

3. Left her to "learn" for 6-7 min. The fans did come on for about 20s, but total time was more than 5 min after subtracting the fan on time.

She sounded just like yesterday, sick, vrooooom, vrooooom, this time bouncing between about 1050 and 1650 RPM. Nothing changed for the duration of the 5 min, it didn't seem like she was "learning" squat, she was doing the same thing over and over until I couldn't bear hearing her like this and turned her off.

Thoughts?
 
#7 ·
Take her in before it turns into a more serious issue, I'm sure that problem can lead to others, And im sure you don't want that. Have honda take a look see.
 
#8 ·
Check Your Coolant

A car's idle will bounce up and down if you've insufficient coolant in the radiator, thus leading to air bubbles/pockets. Each time an air pocket passes by the coolant temperature sensor, the sensor suddenly thinks the engine is cold, which prompts the ECU to rev up the engine (it's normal for a car to fast idle when cold). Then, when hot coolant passes over the sensor again, the ECU brings the idle down to normal. The ultimate result is a bouncing/hunting idle.

This of course might not be your problem, but I'd check your coolant level, and check the radiator for leaks.
 
#9 ·
I just had a somewhat similiar problem 2 weeks ago except that in park it would idle hunt but the difference was it wouldnt come out of park. I had to use the shift lock every time. Also it would hard shift during driving. Well all in all after a couple days at the dealer the problem ended up being a bad throttle body. I never asked for an explanation of why the throttle body would cause this because I only had 245 miles left on my extended warranty and since Honda Care covered it I figured I would leave it alone.

But like Geo said you may also want to check your coolant as well.
 
#10 ·
Thanks guys, I will check coolant level this weekend, could be a little low. I found some interesting info, Haynes lists P0507 as:

P0507 Idle air control system, RPM higher than expected

Could it be the Idle Air Control (IAC) valve gone bad? I already reset the code using my tool and my "idle learn" attempt on Sun did not throw another CEL, so even though it was bouncing like crazy I've got no CEL. Haynes shows the IAC valve under the throttle body. Anybody know much about these valves? Could it be clogged or do they just die? I'm wondering if it's possible to attempt to clean it or something. Although they say you need to use a fresh gasket if you unmount it, so any cleaning would have to be in-place, otherwise I'd rather just replace it.

The reason I don't think it's the coolant is because this was a very precise RPM bounce - it was between exactly 1100 and 1650 RPM every single time, for 5+ minutes. It didn't sound like it was sputtering or anything, it sounded like the computer was telling it to do precisely that, to bounce up and down like that. But I will definitely top off the coolant with some distilled water this wknd -- worth a shot.
 
#11 ·
Clarification

...it sounded like the computer was telling it to do precisely that, to bounce up and down like that.
Right. Like I said. The temperature sensor conveys data to the ECU ("Electronic Control Unit"; aka "Computer"), which in turn directly controls engine RPM, based, in part, on what temperature data the ECU is receiving from the coolant temperature sensor. So this is a case where "the computer is telling the engine to bounce up and down", because the temperature sensor is sending alternating hot/cold data.

Again, this doesn't mean my theory is correct, but air in your cooling system is a common cause of RPM bouncing/hunting. It's also a condition that won't always throw a Check Engine light, because aside from quickly fluctuating temperature data, everything's normal-- So the ECU doesn't think anything's wrong (and hence doesn't throw a CEL).

Anyway, let us know what the final verdict is, eh? I'm curious about this one!

Here's hoping it's just a minor issue...
 
#12 ·
I bet it is a stuck IACV or it's not responding properly. I'm speaking with knowledge of 6th gen accords but your symptoms sound identical. Its basically a vacuum leak. when rpm rise high enough with a closed throttle, the pcm shuts the injectors off and the cycle repeats. The cure is to replace the valve or some have had luck taking it off and cleaning it with Simple Green. I think that engine braking feeling during driving and between shifts is also related to it.
 
#13 ·
The reason I don't think it's the coolant is because, again, this is a very precise bounce. If there were bubbles of air, they would be random and the bounce would not be perfectly timed every time - there is no randomness to the bounce pattern, it's like a metronome, 1100-1650 RPM everytime, with the same amount of time elapsed in between. Of course it could affect the system in other ways that would cause this symptom -- that's why I will definitely top off with some distilled water. I checked and the level is a little below the MIN line. Thing is, it probably was like this before this problem appeared, 6 weeks ago when I last started the car.

At this point the IACV seems the most likely culprit, I will take a look at it this wknd. I can't seem to find the part online, I'd rather order one and have it shipped than go to the dealer's (since I have no car! duh..). The only ones that show up online anywhere are for 6th gen -- nobody sells them for 7th gen or above, not on eBay, nowhere! What gives?

P.S. Is princess on vacation? :banana:
 
#14 ·
Update: coolant was low, at least 2 inches below MIN line. Topped off to MAX (engine completely cold) with distilled water. Started her up, it did the same thing, but instead of bouncing all the time, it would only bounce once in maybe 30s, between the bounces it would stay revved at a high 1600 RPM (this is not normal idle, it's much higher, more than 2X). This was with engine fully warmed up.

I got the IACV and will be replacing it this weekend; had to go to dealer, $100+ part for 7th gen (!!!, it's only like $30 for older Accords on eBay, but you can't find it anywhere online for 7th gens).

Questions:
* If it isn't the IACV, what else could this be?
* Could the low coolant have contributed to killing my existing IACV, if it does turn out to be it?
 
#15 ·
Wouldn't a dealers $100 (at most) diagnostic which would've been credited toward the repair been a lot cheaper than randomly replacing parts?
 
#16 ·
Cheaper than a tow, and I prefer to fix it myself if possible. No biggie, if it's not it, I'll sell it and recoup part of the cost. I didn't think it would be so much either, since the 6th gen one is so much cheaper.
 
#17 ·
Problem SOLVED!

So I changed out the IACV today, and guess what? I fixed the car in the process! w00t! :banana: :nuts: :naughty: :biggrin: :salue: :jumping: :dude: :banana:

I was able to remove and replace the IACV w/o taking off the throttle body, or really w/o doing anything else other than taking IACV off and putting new one on. I spent quite a bit of time finding the right tools to do the job w/o removing the whole TB. The space is very tight but with the right tools and technique it can be done. If anyone needs tips let me know.

Did the "idle learn", but the car was pretty much idling perfect from the getgo -- the very first time I started the engine on new IACV, battery was disconnected for days, hooked her up, ignition in II for 3 seconds, started and kept her at 3000 RPM till coolant reached 200°F, then I took my foot of the gas and she dropped down to her usual ~650-725 RPM she's always done since day one. I knew right away I had fixed her. I let her "learn" for 6 minutes, although it didn't seem like she was doing anything but I figured it couldn't hurt. Note that I had someone outside look for coolant leaks since I was in the car keeping the engine @ 3000 RPM and this was first time on new IACV.

I'm double happy because I'm pretty much the one who diagnosed the problem as being the IACV. I did an additional test yesterday, I started the engine and disconnected the IACV while she was running, and she started doing something I had observed previously (see 1st post in this thread, 2nd paragraph, "...she started doing it differently, very quickly between (I think) 1400 and 1700 RPM..."). This made it very likely to be a IACV gone bad.

Whoopeee! :banana: :banana: :banana: :banana: :banana: :banana: :banana:
 
#22 ·
So I changed out the IACV today, and guess what? I fixed the car in the process! w00t! :banana: :nuts: :naughty: :biggrin: :salue: :jumping: :dude: :banana:

I was able to remove and replace the IACV w/o taking off the throttle body, or really w/o doing anything else other than taking IACV off and putting new one on. I spent quite a bit of time finding the right tools to do the job w/o removing the whole TB. The space is very tight but with the right tools and technique it can be done. If anyone needs tips let me know.

Did the "idle learn", but the car was pretty much idling perfect from the getgo -- the very first time I started the engine on new IACV, battery was disconnected for days, hooked her up, ignition in II for 3 seconds, started and kept her at 3000 RPM till coolant reached 200°F, then I took my foot of the gas and she dropped down to her usual ~650-725 RPM she's always done since day one. I knew right away I had fixed her. I let her "learn" for 6 minutes, although it didn't seem like she was doing anything but I figured it couldn't hurt. Note that I had someone outside look for coolant leaks since I was in the car keeping the engine @ 3000 RPM and this was first time on new IACV.

I'm double happy because I'm pretty much the one who diagnosed the problem as being the IACV. I did an additional test yesterday, I started the engine and disconnected the IACV while she was running, and she started doing something I had observed previously (see 1st post in this thread, 2nd paragraph, "...she started doing it differently, very quickly between (I think) 1400 and 1700 RPM..."). This made it very likely to be a IACV gone bad.

Whoopeee! :banana: :banana: :banana: :banana: :banana: :banana: :banana:
I'm having similar issues, after swaping the stock intake manifold on my 2003 I4 M/T with the 04 TSX IM. The idle is jumping up an down. I checked my coolant, it's topped off and I did the "idle learn" but still erratic idle. What tools did you need to change the IACV without removing the TB?
 
#19 ·
I'm not sure, I heard people mention a screen that could be clogged however I didn't see one. I didn't want to take any chances either, and don't have the time nor desire to further investigate the old IACV.

I was quoted $105 by one dealer but I later managed to find it at another for $90 + tax, for a total of $96 in FLA. I think online, if you can find it, it should be $15-20 cheaper (pre-tax). This sux because for the 6th gen it's like $25 and the part doesn't look more complex for the 7th gen, it looks like they just want to keep the price high (I guess they consider the 7th gen to still be somewhat semi-current?). Oh well, can't win them all.
 
#20 ·
can you tell me the part # for the 7 gen IACV

thanks
 
#21 ·
On a V6 I would say that bouncing RPMs are because the map sensor is disconnected. That sensor is located on the throttle body. Not sure about I4 though, it may be the same. On the I4 it is located on the side of the throttle body. I have a hard time believing the IACV itself was defected.
 
#24 ·
I know this post is super old, but I just wanted to bump it in case somebody finds it like I just did. I replaced my starter, and while I was at it, I cleaned my throttle body (de-coked) since it has excessive build up. Got it nice and shiny clean, put everything back together... and got the weird surging idle. Not only that, but in my manual trans, I would let off the gas and 2 seconds later get a hard engine braking effect. Idle would surge up and down when I had the clutch pedal pressed, and never drop below 1k rpms, when it normally sat at about 800. Tried the idle re-learn, no dice. Got the P0507 code (high idle) after maybe 40 miles or so. I read about the TPS reset procedure, and rather than pay Honda $200 to perform this (and have them try to sell me everything else under the sun), I found a scantool that could accomplish this task for a similar price ($160, actually) on Amazon. Next day delivery, too! Best part? If I wanted to, I'm sure I could receive the tool, perform the reset... and return the tool. Would I do that? Never- I am not that devious (snicker snicker)... in fact, assuming this works, I'll hang onto the tool because you can pay $60 and add another manufacture onto it to receive specific functions for that car too. The tool is made by foxwell- the model number is a NT510. Just make sure that you pick the one specific to your car (Honda in this case) and another thing to consider is that the TPS reset is not supported in years 2008 and below (I think this is correct). Hope this helps someone save some time... tool me a while to put these pieces together. =D I'll report back tomorrow if I'm successful.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top