 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Car Forum / Honda Cars / November 20051999 Integra: erratic idle problem
My 1999 Integra has an annoying problem which has yet to be diagnosed by Acura service. I'm about to make my 3rd trip there. Here are the symptoms: after the car has been driven for about 5 minutes and is put in park, the RPM's fluctuate between 1500 and 2000, back and forth, about once every two seconds. The "check engine" light is usually triggered (if it isn't on already), and the engine runs roughly after that. However, the problem isn't consistent and doesn't always appear during my 8-minute commute to work. Sometimes, after several days, the "check engine" light will turn off...only to return again with the symptoms described. Acura service has scanned the code twice now. The first time, they claimed that it was "noise" and cleared the code. The car ran fine until I got it home, at which time the fluctuating RPM's appeared when I put the car in park. The second time, they claimed that I needed a new air filter and spark plugs. The bill came to $248 ($54 of which was to scan the code.) The "check engine" light was gone when I drove it away, but returned when I placed the car in park in my driveway. Does this problem sound familiar to anybody? I'm worried that I'm going to be spending a lot more time and money for Acura to scan codes but not fix the problem. Thanks in advance! Dan > My 1999 Integra has an annoying problem which has yet to > be diagnosed by Acura service. I'm about to make my 3rd trip [quoted text clipped - 20 lines] > be spending a lot more time and money for Acura to scan codes but not fix > the problem. It could be that you have an air bubble in your coolant system. Not sure if your integra has an air bleeder valve (check your repair manual for the procedure) - if so, it can be opened to let the air out. Another way is to burp the system by opening the radiator, start the car, let it run and watch the coolant level drop. As it drops, add more coolant. Goose the engine from time to time. Eventually you'll hear the system burp, removing the air from the system. Be sure to use honda approved coolant fluid. Also check for vacuum leaks, as that is also a common cause. Remco >>My 1999 Integra has an annoying problem which has yet to >>be diagnosed by Acura service. I'm about to make my 3rd trip [quoted text clipped - 33 lines] > > Remco i second all that. and suggest the op finds a different dealer. this should be a simple problem to fix. to charge all that money and go through the effort of 3 visits /not/ to get it fixed is just ridiculous. > > Here are the symptoms: after the car has been driven for about 5 minutes > > and is put in park, the RPM's fluctuate between 1500 and 2000, back and [quoted text clipped - 12 lines] > system burp, removing the air from the system. > Be sure to use honda approved coolant fluid. Might that be my problem too? Here's my story: Last week I ended up running my 90 Civic for ~20 minutes without coolant due to a radiator leak; temp gauge was pegged :^( (I was trying to get to a concert, FWIW...) I had the radiator replaced, but now it runs rough when accelerating :^( Idling is generally OK, and the problem sometimes takes a few minutes to show up after starting (but before the temp gauge needle starts to move). Reservoir was full after the rad. change, but I haven't checked it since this started... Some brief googling in this group indicates it could be any of the following: Air bubble Blown head gasket Cracked head Cracked block Bad O2 sensor/PCV Bad main relay (I resoldered it a couple months ago for not starting). Fuel filter Fuel pump Spark plugs (originals are still on it) I haven't noticed the dreaded "white smoke" in the exhaust, nor the "milky oil" indicating a gasket/head/block problem, but maybe they don't always cause those symptoms? It's also due for an oil change, but I doubt that will help - it's been dripping a few drops of oil after I park it, and was right at the add mark last night - I added a quart but no help... I also threw in a bottle of dry gas thinking there might be water in the tank, but again, no help. Fuel tank was replaced about six months ago. If it's not the filter/plugs (can I change either myself, or let the local garage do them?), I'll probably have to take it to the dealer, but that's such a pain - they're on the other side of town, and I'm not sure how well the buses run out there :^( (It's probably too much to expect to get a loaner...) And I'm not thrilled with the typical troubleshooting method of "change this, if that doesn't fix it, change something else, repeat" :^( If it _is_ "just" the head gasket (or maybe the head), should I go ahead and have that done, and maybe also change the timing belt and flush the transmission? Car has 117k miles on it, original belt, and transmission was only flushed once (around 25k miles, I think). Or is it time to replace it? I hate to do that after sinking $1k into it this year between the fuel tank, body work to get through inspection, and now the radiator :^( Plus I'm still bummed that there're no hatchback Civics w/ automatic transmissions, but maybe I could choke down one of the Civic coupes, maybe even a hybrid... If it's the engine block, that will probably make up my mind for me :^( Ron > > > Here are the symptoms: after the car has been driven for about 5 minutes > > > and is put in park, the RPM's fluctuate between 1500 and 2000, back and [quoted text clipped - 81 lines] > > Ron If the thermosensor is not really in contact with coolant, it doesn't see the real temperature and tells the computer about it. The computer tries to make an adjustment based on that bad measurement and the car can start to run like a heap. Not sure on your car, but usually there are two sensors: one for the dashboard, one for the computer - so not seeing your dash temp move may not mean very much. Yes, I'd try and burp the system on your car, especially if you've had it running hot (you are one lucky man if you did not do damage). As a side point, you mentioned you didn't have the belt replaced. You are talking about the timing belt, I take it? If so, at 117K you are past due - civics/integras should them replaced at 90K, I think. They do break without warning and can/will do a lot of damage to your car. While I doubt it has anything to do with the rough running at this time, do yourself a favor and replace it ASAP so you won't have to go through a really expensive repair. Remco > If the thermosensor is not really in contact with coolant, it doesn't > see the real temperature and tells the computer about it. The computer > tries to make an adjustment based on that bad measurement and the car > can start to run like a heap. Not sure on your car, but usually there > are two sensors: one for the dashboard, one for the computer - so not > seeing your dash temp move may not mean very much. Could the overheating have damaged the computer's sensor? Could it have gotten unhooked when the radiator was changed? > Yes, I'd try and burp the system on your car, especially if you've > had it running hot (you are one lucky man if you did not do damage). It was a cold night and I didn't run it above 30 mph. But burping didn't help - I popped the rad cap and ran it for a couple minutes, but nothing happened :^( Level was still at the top of the radiator. Could it be the fuel filter or plugs or injectors? Seems like it should run poorly soon after starting for any of them, not take several minutes... I doubt it's the air filter as I popped the cover over the fuel injectors (or whatever the hose from the air filter is feeding) and that didn't help :^( > As a side point, you mentioned you didn't have the belt replaced. > You are talking about the timing belt, I take it? If so, at 117K > you are past due - civics/integras should them replaced at 90K, I think. > They do break without warning and can/will do a lot of damage to your car. I hear you, but I wonder what safety factor is built into that 90k recommendation? I.e., what's the range where they break, or don't people keep them long enough for useful data? Ron > > If the thermosensor is not really in contact with coolant, it doesn't > > see the real temperature and tells the computer about it. The computer [quoted text clipped - 32 lines] > I.e., what's the range where they break, or don't people keep them long > enough for useful data? I wonder if maybe you need a head gasget... Overheating a car like that is a really bad idea. Do you ever see white smoke when you start the car or while driving? Any different color crud in your coolant overflow tank? (brownish color) Well, I don't know what the Main Time Between Failure of a belt is but general consensus is to replace it at 90K -- one would hope that's probably on the lower portion of the bell curve where you see failure. I'd get it done soon, seriously. Remco > I wonder if maybe you need a head gasget... > Overheating a car like that is a really bad idea. > > Do you ever see white smoke when you start the car or while driving? > Any different color crud in your coolant overflow tank? (brownish color) Turned out to be just the plugs; the local garage hooked some electronic widget up that told them there was no spark - guess at 117k they were done :^) Don't know if $78 to find and change them (and tighten up the valve cover - there was oil in the plug tubes) is reasonable, but I'm not complaining... But the question remains whether I've stressed the gasket (or something else) and it may let go on me some day :^( > Well, I don't know what the Main Time Between Failure of a belt is but > general consensus is to replace it at 90K -- one would hope that's > probably on the lower portion of the bell curve where you see failure. > I'd get it done soon, seriously. Here's what the dealer says: "The service department estimates about $82 to diagnose the problem. Also, about the timing belt, if you just want to replace the belt, its $252. We recommend replacing the belt, water pump, cam and crank seals for $420." I'm OK with the $252 for the belt and maybe some more for the pump, but do I really need to do the cam and crank seals? Thanks for the help, Ron > > I wonder if maybe you need a head gasget... > > Overheating a car like that is a really bad idea. [quoted text clipped - 9 lines] > was oil in the > plug tubes) is reasonable, but I'm not complaining... That's good news. If I were you, get a repair manual on your car and also a car repair for dummies book. Not suggesting you are a dummy, but those books do seem to make some of the simple things easier to understand. Something like detecting whether you have spark would have been shown how to do. > But the question remains whether I've stressed the gasket > (or something else) and it may let go on me some day :^( [quoted text clipped - 14 lines] > I'm OK with the $252 for the belt and maybe some more for the pump, > but do I really need to do the cam and crank seals? It is one of those things that is just easy to do - "while we're there". Basically, you are saving on the labor. The MTBF for the pump is similar to that of the belt so will eventually fail and then it will cost you 2-3 hours labor to replace. If you intend to hold on to the car, you may want to consider it. Remco > > Here's what the dealer says: > > [quoted text clipped - 11 lines] > > If you intend to hold on to the car, you may want to consider it. I already agreed to the water pump - I was wondering about the cam/crank seals :^) Are they something I should have changed, or if I could save $100 by only doing the belt & pump, skip them? Or are they _also_ "easy to do while we're there"? :^) I'd like to get the car to 150k miles, or about 20 years old (it's 15 years now with 117k). Ron, tempted to start another thread on how long people kept their Hondas... >>>Here's what the dealer says: >>> [quoted text clipped - 22 lines] > Ron, tempted to start another thread on how long people kept their > Hondas... ron, on the subject of seals, bear in mind 2 things: 1. the "correct" way of extracting the seal is to disassemble the its housing and drive it out from behind. that's never done in this kind of replcement scenario and a hook tool is used to drag it out fron the front. that can lead to scratching of the sealing surface, which while not a problem inthe short term with a new seal, is not a good plan for long life in the grand scheme of things. 2. seals are sensitive to oil quality in terms of their ability to continue sealing. good oils have special additives that condition the seals, and a lot of cheaper oils [and some no so cheap ones] skimp on those additives leaving the car to leak like a demon. my old civic is a case in point. with cheap [but well reviewed] motorcraft oil, the oil leaked from the main seal so bad, clouds of blue smoke were constantly rising where it dripped onto the exhaust. i bought a new set of seals in anticipation of changing them all but put the job off and did a quickie oil change with some castrol gtx. suddenly, the car was not leaking any more!!! and it hasn't leaked since. bottom line, seals are routinely done by a lot of shops much like brake disks are routinely cut when they don't need to be. if this were my car, i'd /not/ change the seals, or at least, not using the hook tool method, and make sure i used a decent oil. i used to have a crx with 320k miles on the original motor. no seal replacement. no leaks. yours could be there too. but /do/ change the water pump - siezed pump bearings are no joke. > Remco wrote: > > > [quoted text clipped - 17 lines] > cam/crank seals :^) Are they something I should have changed, > or if I could save $100 by only doing the belt & pump, skip them? > Or are they _also_ "easy to do while we're there"? :^) Yes. The parts (the two seals) should be under $10. The labor should be a few minutes each. I would be surprised if the dealer said you could save more than $20 by not doing the seals. Are you seeing any oil on your garage floor? If not, maybe you could get away with skipping them. Then again, maybe next month these seals will start failing. That $420 is a very fair price your dealer wants for the timing belt etc. job. The $78 for the diagnosis work and new plugs sounds very fair, as well. About oil in the spark plug tubes: I'd keep an eye on this, checking the tubes once a week for awhile. Tightening up the valve cover shouldn't have an effect on this. Most or all Hondas have a special gasket for each tube that happens to be located in the valve cover. They can't be tightened. In addition there may be an O-ring further down the spark plug tube, but IIRC that's not likely for a car as young as yours. > I'd like to get the car to 150k miles, or about 20 years old > (it's 15 years now with 117k). > > Ron, tempted to start another thread on how long people kept their > Hondas... Elle Original owner, 1991 Civic, 172k miles, manual tranny. Thinking I'll easily get to 250k miles and probably five more years. > I.e., what's the range where they break, or don't people keep them long > enough for useful data? it *could* break at any time. it could last as long as you own the car. i wouldnt be surprised if half the hondas out there never get a timing belt replacement. might be 200k or longer before the neglected belt snaps, and by then the car is usually pretty ratty looking. the manual for my 98 CX says something like 108k miles *or* 7 years, whichever comes first. right now ive got 51k on it, and this past may was the 7 year mark. im not sweating it. the car is garaged most of the time, driven only a couple times a week and is in a climate thats not too harsh on rubber. basically, its my gamble. i doubt ill wait another 7 years for it to hit 108k miles. but im not going to replace it at 51k, either. | Could the overheating have damaged the computer's sensor? unless u use synthetic oil, mineral oil would have turned into black gum all over tappets & valves, [i] better open rocker cover & see inside, [ii] oil flow would be blocked, timing belt will soon break > My 1999 Integra has an annoying problem which has yet to > be diagnosed by Acura service. I'm about to make my 3rd trip [quoted text clipped - 20 lines] > be spending a lot more time and money for Acura to scan codes but not fix > the problem. what were the exact codes? for $54 they should have at least told you the code/s that were triggered, even if they were "noise", whatever that means. > My 1999 Integra has an annoying problem which has yet to > be diagnosed by Acura service. I'm about to make my 3rd trip [quoted text clipped - 24 lines] > > Dan ------------------------------- If you 'burp' the cooling system, be sure to fill the reservoir to MAX, since the engine will want to take in coolant to replenish the air it burps out. Check the reservoir next morning for a couple of days and top up again. Use a flashlight and look from the side of the reservoir. 'Curly' | RPM's fluctuate between 1500 and 2000, back and | forth, about once every two seconds. clean / replace fast idle valve : may be same as mine, #8 in www.hondaautomotiveparts.com/auto/jsp/mws/catimgs/13sm40_d03.gif www.hondaautomotiveparts.com/auto/jsp/mws/prddisplay.jsp?inputstate=5&catcgry1=A ccord&catcgry2=1990&catcgry3=4DR+DX&catcgry4=KA4AT&c atcgry5=THROTTLE+BODY or throttle angle sensor ( leftest part of #6 ). | The "check engine" light is usually triggered Open rocker cover & chk for clogging, or add a can of flush solvent, idle for 10 minutes, then change oil. | engine runs roughly after that. use a voltmeter to read O2 sensor's output http://en.wikipedia.org/wiki/Oxygen_sensor http://home.flash.net/~lorint/lorin/fuel/lambda.htm : if too high / low then try cleaning injectors via removal / solvent. | Does this problem sound familiar to anybody? yes on my F20A, I presume the air intake of yours is similar >| RPM's fluctuate between 1500 and 2000, back and > | forth, about once every two seconds. [quoted text clipped - 16 lines] > | Does this problem sound familiar to anybody? > yes on my F20A, I presume the air intake of yours is similar Thanks for the many detailed suggestions! Here's an update to the problem, which may perhaps narrow things down a bit. I work within 3 miles of my house and rarely fill my gas tank. But I filled it tonight, and when I put it in park in my driveway, the erratic idle problem disappeared! I have no doubt that the problem will return, as it predictably does, but could this somehow be related to the amount of gas in the tank? Dan > >| RPM's fluctuate between 1500 and 2000, back and > > | forth, about once every two seconds. [quoted text clipped - 26 lines] > I have no doubt that the problem will return, as it predictably does, but could this > somehow be related to the amount of gas in the tank? Fuel does cool the fuel pump.. Maybe your pump is having an issue, but that's pure conjecture on my part. | could this somehow be related to the amount of gas in the tank? Fuel pump may have dirt* clogging it : when u fill gas, * gets pushed away fr pump, but slowly sinks back into pump ( i heard is @ bottom of tank ). My throttle angle sensor was bad, www.d-a-p.com/askus.htm says MAP sensor too can be the cause. |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2008 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.