Thousands of verified Experts are ready to answer your specific questions 24/7.
Satisfaction is guaranteed and you pay your Expert only if you are satisfied.
Just type your detailed question and click "Get an Answer."
In minutes you'll get a response from an Expert. You can always ask follow-up questions.
Happy with your answer? Just click "Accept" to pay your Expert.
Apparent timing problem on 1998 Dodge Avenger 2.5 V6 coupe. that has defeated all my DIY efforts. It suddenly died in a Kroger parking lot about six months ago. I could smell gas but see no spark. Replaced the distributor (built in coil), rotor and cap and it fired up again. Hasn't run right since then, and I probably misdiagnosed the problem, anyway. There are two main symptoms. (1) Below about 600 rpm the engine misses badly. A little bit of load with A/C on and transmission in drive makes it worse, enought to shake the car and you expect it to die at any moment. Problem is slightly less when engine is cold. Over several months the bad RPM range has increased and missing now occurs as high as 900. (2) From low RPM, the engine will immediately lose power if you attempt to give it much gas, a classic timing symptom. The computer never turned on the check engine light. The car runs fine going down the road, but after about 20 seconds at a stop light it
1998 Dodge Avenger coupe 2.5L V6 Already Tried: Text was cut short. (1) Below about 600 rpm the engine misses badly. A little bit of load with A/C on and transmission in drive makes it worse, enought to shake the car and you expect it to die at any moment. Problem is slightly less when engine is cold. Over several months the bad RPM range has increased and missing now occurs as high as 900. (2) From low RPM, the engine will immediately lose power if you attempt to give it much gas, a classic timing symptom. The computer never turned on the check engine light. The car runs fine going down the road, but after about 20 seconds at a stop light it begins bucking and you have to put it in neutral and baby it till the light turns green. People who've looked at it have been evenly divided on whether it is a spark problem or bad fuel delivery. I've replaced the distributor again, along with new plugs and wires this time, with no effect. Put on a new fuel filter, and then a new fuel pump, with no effect. Pinching the fuel pressure regulator's return line had no effect. Disconnected the MAP sensor out of curiosity, and got a check engine light but the engine was a little stronger, so I put in a new MAP sensor. Didn't solve the problem but the computer is now quicker at trying to keep the RPM's up at idle, and the lugging problem is very slightly better. A borrowed scan tool showed absolutely no codes from the computer, but below 1200 rpm it refused to advance the timing when you put your foot on the gas. Apparently it thinks this is the right thing to do. Since you can't actually adjust anything on one of these cars, I've run out of solutions. Seems likely it's a sensor sending a wrong-but-legal value for something, but I have no idea what it might be.
Customer welcome to Just Answer!.
Whether I can help or not, Texan... I've got to congratulate you on the superb post you've sent. Fantastic information, sent by someone who knows his way around a car.
I don't think you misdiagnosed the orignal problem necessarily. It quit... wouldn't run... the distrubutor got it going again. It sounds like a classic case of shorted ignition coil, the bane of these units. You can't argue with success and you DID have success.
The continued misfire problem since then seems a bit hard to nail down since it appears to have started at the same time. I'd start by checking spark in a coupla ways...
The front bank is the only handy place to work, so let's hope it shows up there. Start the engine and pull each of the plug wires out, one at a time. Hold them near a conductive surface and watch the spark delivery on each circuit. I'd expect to see 3/4 to 1 full inch of spark from each of the boot ends of the plug wires on a healthy ignition system. If you see nothing near this, you got a bad distributor.
If it's good on some plug wires, not on others, I'd replace the distributor cap if the plug wires check OK.
The cap is designed for asthetics, giving the front bank and rear bank clean connections with no criss-crossing. This means conductors need to be cast within the plastic body of the cap and spark may have blown through the plastic insulation, allowing short-between-cylinders.
Good output on some plugs mean you have good insulation. Low output from the same coil (and rotor) mean spark is going someplace else or you have a burned out plug wire.
High resistance from a burned out plug wire will cause intermittent misfire, particularly on a warm engine. In addition, the resistance will cause the coil to build to a much higher output voltage and may be responsible for failure of the distributor cap (if it's occurred).
Measure each plug wire with a digital ohm meter set on the 20K scale. Anything that shows up as not open (infinity or "1") checks OK. The actual resistance of the circuit isn't that important, not so much as just seeing actual continuity between ends.
Of course, testing plug wires is a pain on this engine... particularly the back bank. Do what you can, but if you do decide to remove the plenum to test the rear bank, you might just be better off replacing the set anyway. I wouldn't want to do this more often than absolutely necessary lol!
The plenum-to-manifold gasket is metal covered rubber and doesn't necessarily require replacement... your call.
Write back if you have any questions, Texan. I'd be happy to hear from you!
Thanks,
Ed
Hi Ed,I did in fact replace the plugs and wires after the second distributor didn't help. Not surprised you missed this in such a long description. Spark appears hot and long on a borrowed spark tester. The only thing I now know for certain is that computer is reading something that tells it sometimes not to advance the spark. You could see this on the scan gadget . I think if we could find out what conditions could cause this, and solve it, then the idle problem might go away as well.Marty
Ooops. Sorry Marty... information of the sort you sent was like Thanksgiving and I did miss it.
Low spark advance might not be the issue though. It won't cause misfire like what you're seeing and may be completely normal. I'd look at the two strongest inputs to advance (after rpm), TPS and MAP. TPS is fairly easy, just requiring checking the voltage output to be linear as the throttle is opened, but the MAP is a bit tougher. With a new MAP in place, I think you can pretty much skip this step.
Something that hadn't occurred to me until now is that you might be having EGR problems, flowing more just off-idle than what it's rated for. Try disconnecting the vacuum hose at the (metal) EGR valve and see what happens. It's a bit of a stretch... don't know why it would pick this moment to start acting up, but it's an easy check. The same thing is true of my original distributor cap pick, but don't discount it just yet because it isn't pure speculation... it does happen.
I've gotta crash for now, Marty. Talk tomorrow!
Good luck,
Me, too. I don't think it's the TPS since throttle seems to work fine once you're going down the road, however I'll see if I can check that, and the EGR, and will get back to you. Probably in the afternoon. - Marty
Sounds like a plan. I'll be at work but I can check my JA mail from time to time (depending how bad they're beating me lol).
Talk later,
Hi Ed, I've had an interesting afternoon. You've already earned your pay by having me look at the problem in a different way. Your suggestion to check the EGR valve made me think of the anit-polution system as a possible cause, and led me to check the PCV valve first off. It was totally gunked up. I plugged the hose and did a short test drive. Problem didn't go away but you can tell it is better.Couldn't find the EGR. My Haynes car manual shows it in the wrong place. Just located it a few minutes ago on the back of the upper intake manifold, thanks to Google. Some other fellow coudn't find it either.I also had put in a strong dose of fuel injector cleaner two days ago. This started the check engine light flashing at idle, and the light stayed on today. I'm going to replace the PCV valve this evening so I'll know that's right before I go on. Then I want to see if I can get someone to read the computer code, befor I procede to the EGR. There's a chance it might be useful.I'm very pleased with the way this is going. Although the problem is not solved, I'm willing to pay off this ticket and then start another one after checking the EGR tomorrow. Tell me your preference.Off to AutoZone.Marty
AutoZone rented me a code-reader. Didn't know they did that. Hope to have good info from it tomorrow. - M
Oooo! I didn't either! Hopefully it also has data capability...
Sorry about the EGR thing. I figured you knew all about the placement from when you replaced the plug wires... not a handy thing.
One thing I'd caution you about with the EGR vacuum solenoid/ pressure transducer (yes, it has those names) is that the vacuum source needs to enter the FAT end of the transducer. It's not uncommon to see the vacuum/ output ends of the transducer swapped, so the first time the solenoid is opened it locks vacuum at the (metal) EGR valve nipple. Runs for crap, I'll tell you that. It's not so much misfire, as just plain sickness.
A flashing MIL is normally a sign of misfire, something the PCM figures to be catalyst-threatening. I somehow doubt the cleaner you put into the fuel system had anything to do with it... info only.
Avoid doing a battery disconnect for a while to retain system calibrations if at all possible. Fuel system adaptives and the adaptive numerator (used for the Misfire Monitor) are learned over several miles of driving and may not be restored in time for being helpful. Seeing the flashing MIL tells me that you've got your PCM fully updated, so let it do its job... storing info for us.
Last, hold off on the accept, Marty. I prefer to see the problem to its conclusion (one way or the other) before seeing a customer accept. Rarely, things can't be solved over the internet and I feel crappy about it when I've been paid for something that didn't take care of the problem. I'm confident with your abilities we won't have that problem but I'd rather that you not click the button until we're done... one way or the other. K?
But thanks. Talk later,
Ed, I'm uncertain about which hose to disconect from the EGR valve to test it.The solenoid is on top of the valve. A hose goes from the end of the solenoid to the intake manifold. From the little top hat on the valve itself, a hose goes downward to what looks like a connection on the exhaust manifold.Which hose am I disconnecting, and also should it be plugged or left open for the test?Marty
Hi Marty. To just disable to EGR system for testing, disconnect either the hose from the intake or the one at the metal EGR valve itself. The tube that goes to the exhaust manifold is the backpressure signal source, used for load sensing.
Got some daylight left. Will give it a shot.
Disconnected the small hose that leads from the EGR solenoid to the upper air intake manifold. Car wouldn't start and got a small puff of smoke from the open main air input opening. Reconnected the hose, then started car, then pulled the hose again. Thought the missing had gone away, but that was just because the engine was cold. As soon as it began to warm up the missing was back.I've been reading about the throttle position sensor and realize I misunderstood it. Probably wouldn't hurt to test it as you suggested. The accelleration (timing) problem is less after I put that new MAP sensor in, although it is still there. The rough idle hasn't been helped by anything. MAP and PCV have simply made the engine a bit stronger.Can I read the codes now, if I promise not to erase them?Marty
Describe the missing for me Marty. Misfire comes in a number of flavors.
By definition, misfire is complete loss of combustion in a particular cylinder. That's no longer the case. We might have misfires that are 100% loss of performance on a cylinder, 70%, 20%... it goes on. A "hard miss" is something I'd categorize as 100%, but you might have something far less than that.
Worse, it can be erratic, coming and going with multiple cylinders involved.
Misfire caused by fuel or EGR system problems can often fall into the less-than-100% category, weakening cylinder performance without completely cancelling it out. Ignition misfire is typically all or nothing... it hits or it doesn't. Neighboring cylinders are often unaffected by the bad luck one hole is having but EGR and fuel-related misfire can affect a number of cylinders more or less equally at the same time.
Adding throttle often sorts these things out as well. If it's a hard miss, putting the trans in gear (auto) and giving it light throttle might push a weak part over the edge so that it misfires more often if it's not a constant miss already.
EGR problems tend to improve with added throttle because the incoming charge dilutes the EGR flow. Fuel system misfires may become worse with more load but it depends whether the injector is restricted or it's having problems firing properly.
Lots O'flavors...
By all means, check codes. Nothing bad can come of a code check and new info can come along at any time. Sorry to keep you waiting...(company).
Hi Ed,Don't be concerned about delays - I'm a very slow mechanic (took me two days to change those spark plugs) and often only have an hour or two during any particular day to work on a car problem.Missing:You asked me to tell you more about the missing. As far as I can tell the miss is 100% and random. There's just enough noise around the engine to confuse things, so I go back and listen at the tail pipe. At idle speed you can clearly hear the missing explosion. It's not rhythmic, but it is happening a LOT. Then a slightly longer gap tells you that two cyl's in a row didn't fire. Then about once every 15 seconds or so three cyls in row will not fire. That's when the RPM needle drops to 200 or below and the computer gooses it to keep it running.When I changed the plugs they were all the same, and not in very bad shape.Codes:This morning I got out there with the rented code reader. The code that has been showing up is P0300 "Random/Multiple Cylinder Misfire Detected". I wrote down the "snapshot" info. Only thing that looked odd was coolant temp, which was too low to be right. Cleared the codes, ran the engine about three minutes. Got a P0300 pending, and the snapshot said coolant was at 40 degrees. Well - not in Texas anytime soon.The rented code reader does not show what's actually happening at the moment, like the one that was borrowed overnight from a Nissan dealership, so I couldn't see the timing like I did that time.Timing:Put it on the road last night to blow it out a bit. Very surprised to find that at 2000 rpm and 40 mph the engine fell on it's face when I asked for quick acceleration. I thought this only occurred below 6-700 rpm.An Idea:After listening to the miss last night, I woke up this morning with a strong picture of a distributor cap in my head. This one had the center wire that used to come from the coil. The Avenger's cap has a metal rod that goes down into the distributor instead. I originally replaced the distributor, rotor and cap. That's when the missing and timing issues appeared. When they got progressively worse I took the distributor back to Autozone and got a free replacement. It didn't solve the problem. Now I'm remembering that because the rotor and cap still looked new, I did NOT replace either one of them. I may have made a huge mistake there. Do you know of any instances where that cap-to-coil connection has intermittent failure?
Good work with going back to the tailpipe, Marty. That's where I go if I'm looking for misfire and can't tell what's going on up front. Holding your hand over the pipe will give you a good idea if the miss is total or a lesser one and the cadence between misses will also help decide how many at a time are involved.
The distributor caps DO definitely have their problems! Because they shuttle spark back and forth within the cap molding, spark can sneak between cylinder circuits and cause problems like what you're seeing. Typically it shows up at the spark plug ends with weak spark from one or two plug wires but it's the most plausible reason I can see for your intermittent misfires.
You won't necessarily see a miss on every power stroke on an affected cylinder, but adding load like you did at 2000 rpm may have forced the issue lol! If you're seeing 3/4" of spark or better at any of the front plug wires, you won't have a problem with the rotor but I can't let the cap off the hook just yet. They aren't cheap (about $75 thru Mopar) and they're not handy to test either.
If you have a second vehicle with plug wires, you can check the cap for between cylinder leakage...
Pull one plug wire off the second vehicle and attach it to any of the six plug wire outlets of the cap. Start the engine and then "sniff" around the cap with a grounded test light to see if there are any places where spark will come out. By definition, the only place you should see spark exit is at the rotor contact for just that cylinder within the cap. If it leaks out of any of the other contact points... you've got your boy. Try all six wire outlets in the same way and see what happens.
Sometimes the only way to test something like this is with the voltage it's used to seeing. An ohmmeter would be worthless, not knowing the difference between a millimeter and three cm, although both are technically open circuits. Not to 60,000 volts they're not...
No worries about not writing back often enough, Marty. "Often enough" is different for everybody and I can see you're busy, so don't sweat it. I'll be missing for a few days next week (camping), but I think we'll get it sorted out long before that.
Forty degrees?
I'll see what more I can learn on the dist. cap issue. Replacement I put in was definitely not Mopar. Back to you when I learn something. - Marty
Just discovered that I never threw away the original distributor cap. Replaced it because the contacts were pretty worn, but if I slap it on and there's a significant change in behavior the answer may be right there. We sure didn't have a missing problem when it was in use.
I like it!
Well, the word for today is "disappointment". No change in behaviour at all with old distributor cap. I would have bet money on that one!While the engine was warming up I went back and listened at the tailpipe again, and I did hear a sort of rythm before the missing got worse.baaaaaaaaaaaaa but but baaaaaaaaaaaaaaaa but but baaaaaaaaaaaaaaHadn't really noticed this before, but then it doesn't mean anything to me either.Didn't get any codes during this brief run.Not sure where this puts us on where to look next. Take your time. Wife says I have at least 3-4 weeks before it goes to the scrap yard. A woman who feels betrayed is a fearful thing!Marty
Ooof! I feel your pain... I had high hopes too!
I can't help feeling this is still a secondary ignition problem, Marty. Hard misfire separated by intervals of normal running isn't uncommon to the spark world. The fuel system will be cycling its feedback control of mixture, possibly accounting for the occasional miss/ no-miss performance. The slight change in mixture may be all it takes to push it over the edge.
Quick question: When you replaced plugs and wires, did you do both at the same time? Flashover misfire isn't super-common on the 2.5 engine but is something that I always worry about with recessed-type spark plug engines. The nature of sealing off atmosphere in a small area like this can create conditions that make it more attractive for spark to follow the outside of the spark plug to ground rather than take the intended path. Continued flashover will etch a track into both the spark plug upper insulator and the inner plug wire boot, giving spark a shorter path to ground each time it uses it.
If you replace one part... but not the other, the tracking will quickly ruin the new part. If the plugs and wires were both completely new to each other I think we can disregard this possibility.
We're getting a little short on possibilities now so I'll have to ask The Question... Did you set spark plug gap before installation? A spark plug gap that's almost-closed will behave this way, sometimes firing... sometimes not. There's no way to crack a spark plug on one of these recessed heads, so I don't think that's an issue (unless it was bad out of the box). I'd consider a gap of .015-up to be OK for misfire prevention; the real problem comes in the .002" areas...
Sorry to hear your FJ is on Death Row, Marty. No need to tell me about the betrayal thing... We'll figure it out, sometimes things just take time.
Shocking as it may be, we seem to have had a spontaneous healing here.I went out about an hour ago and put the like-new replacement distributor cap back on.Started her up just to make sure I had everything back together right.No sign of the missing. Still some very minor stumbling if you suddenly depress the accelerator about half way when idling under load - but even then it only occurred about one time in three.Will road test it shortly, then have more details.Marty
Well.. No Joy. After about two wonderful miles I got stopped by my second red light where the symptoms returned pretty strongly. Turned around. Actually experienced a few moments of loss of power at 40 mph that had me wondering if I was going to make it back, but this did not happen again.Back in the driveway I let it idle in Park with the A/C on. Idle got increasingly erratic, check engine light started flashing, and then I turned it off. Got the reader and found the usual multiple misfire code. Looked at the snapshot. All numbers related to fuel read -100.6. RPM was zero, Load was zero, Map number was also zero. Seems to me they've been that way before when the code was only pending. However, the coolant temp was again 40 degrees.I don't know why it took so long this time for the symptoms to appear, but my wife had remarked that every day was different when she was still driving it to work.I'm wondering about that coolant temperature. The snapshot from a P0300 that was firm gave me rpm 704, Load 5.4 pct, Map 58 KPA, Coolant Temp 109 F.You'll have to tell me: if the computer thinks the engine is cold, would it be supplying a mixture that is much too rich - and could that in turn lead to erratic idle and a situation where the computer wouldn't want to advance the timing when called on?I'm done with it for the day - but tomorrow I might check the price on a temperature sensor...
The temperature guage on the instrument panel behaves normally. If there's only one sender then my idea is probably just wishful thinking. - Marty
Hi Marty. Sorry about the blackout... I drew Satur-duty at the shop so I've been tied up. I followed some of the saga from work but didn't have time for a reply.
Your Avenger is quickly heading into the realm of "gut feeling" rather than hard, linear testing. We've pretty much been through the logical steps and I've got some opinions.
Gawd I hate opinions! I much prefer to base advice on scientific testing or the Diagnostic Dart Board. The Board's in the shop, so we'll have to go with intuition...
You've done enough work on the secondary ignition system that I think it can finally have a get-out-of-jail-free card. The distributor has been replaced (twice), you've put new plugs and wires, cap and rotor into the system, yet there's intermittent hard misfire. I have a hard time believing you've got flashover at this point (too quick) and the chances of getting two bad distributors in a row are almost lottery-type.
I have to go with a crank sensor at this point. It's not a comfortable decision, one that may be wrong, but it seems to be the logical decision.
Your distributors came with new cam sensors... nothing changed. You replaced plugs and wires... nothing changed. You replaced the distributor cap and rotor (ouch!)... nothing changed.
Besides the PCM, the final ingredient in the recipe is the crank signal, which is known to be a failing point on the 2.5 engine. It doesn't normally assume this role, the intermittent misfire that sets codes... typically it kills the engine and leaves you walking.
The nature of the misfire as you listened was rhythmic, but not steady. Large spans of good running are interrupted by intermittent double-misfiries for no apparent reason. It sounds like it gets considerably worse while on the road, so engine or transmission heat may play a part in this.
The act of driving the car will generate heat in the torque converter area, which is also the environment the crank sensor lives in. The extra heat may be enough to push it over the edge, producing erratic signals that the PCM interprets as real, then setting misfire codes as a result. It's a stretch, but it's my best advice right now.
The crank sensor is located almost directly below the distributor on the transmission bellhousing. Take the battery out (yet again) and you should have fairly decent access to the sensor.
The new one should come with a paper "dot" stuck to the sensor end of the unit, serving as the clearancing adjuster for the operation. Push the sensor into the hole until it bottoms, tighten the bolt... done. You may hear a ticking noise for a short while as the flexplate actually contacts the paper dot, but it will fall off eventually... don't freak out.
They're not cheap, running around $110. This is well over my confort zone for flippant advice, but it would be my first choice if it was sitting in front of me right now.
My feeling on the coolant sensor is that it's a code reader error, Marty. I've seen this in the past, where the data stream isn't completely synced with the scan tool and the default is something like what you're seeing. The lack of coolant temperature rise while driving, plus the HUGE difference between ambient and intake air temperature sensors should have set a code if it was real. I think it's not.
If you want to know for sure, check key-on voltage for the intake air temp sensor and the coolant sensor with a DVOM after they've sat for 8 hours or better to normalize readings. The sensors should read pretty much the same thing (voltage-wise) once they're all room (or outside) temperature. The common wire to all three would be the black/ light blue circuit, which is sensor ground. The other wire will be your sensor signal circuit in all three cases. Seeing a major difference in voltage (more than .5v) would be a failure of the sensor. I think you'll see something much more alike.
<whew> I can talk, can't I! Your turn, get back with me when you have news.
Company coming and I got dragooned into the preparations. Crank sensor sounds very promising. I had read about it but got a different image in my head of how it might work and how it might fail. Think it's definitely worth a shot after reading your explanation.Can't tell you how much I appreciate your input!Marty
Company coming might have been a good thing! I was walking the dogs (more like swimming them..) and it occurred to me that I'd completely glossed over the fuel system. Is there a chance that this started within 10 miles of a fuel stop? Have you run more than one tank out since it started acting up?
Water in moderate quantities can act this way, causing intermittent misfire that can even move between cylinders. The FJ uses a returnless type fuel system (after the external regulator) so you'd have to pull a line off at the engine and crank it to get a fuel sample. The fuel supply line is held onto the cast aluminum rail by two small bolts, then pulls off (o-ring sealed).
A large protrusion on the end of the supply tube would make it difficult to attach a drain hose but you might be able to direct it straight into a plastic pop bottle or something. You won't need a great deal of fuel for the test, probably no more than a pint at the most and you can pour anything that looks OK back into the tank. Maybe you could enlist some help for this...
Have a fire extinguisher on hand just in case.
Look for cloudiness in the fuel at first (dead giveaway) that will separate out into water and fuel if left sitting for a while. With the amount of trouble your FJ is giving you it should be pretty obvious whether the fuel is contaminated right away, so don't worry about pumping a whole lot out.
Water tends to settle in the fuel system on systems like yours because the flow rate is so slow. With no return stream to keep water in suspension it tends to separate out and fall into low-lying areas, like the first few injectors it comes to. If the car is tilted, water will flow toward the low areas in that way, too.
The nature of your misfire events seems to suggest random selection of cylinders or at least varying participants. Once again... water.
Hold off on that crank sensor until you can get a sample, Marty. Sorry it hadn't occurred to me earlier!
Ed, this problem as been present for at least six months - only recently getting as bad as it is now. Many tankfulls have been run through it. It was about a month ago that it got bad enough I wouldn't let my wife drive it anymore. When I did the fuel pump (a fine example of throwing parts at a problem!) I looked down in the tank briefly, and the fuel seemed clear. Didn't take any out, of course.I don't mind throwing another sensor at it - just couldn't throw too many more after that. Still think I should check for water in the gas first?Marty
No, unless you're getting a steady diet of water at each fill it won't be the problem. It was a longshot but does happen from hurricane to hurricane lol! Your fuel system should be OK.
Just another thought (oh crap..). I wonder if your problem might be tied to electrical noise... an output from the alternator that isn't smooth. Alternator brushes are good for about 150,000 miles, then can begin to skip and lose contact with the rotor slip rings. Your output can get to be a bit noisy, with harsh spikes that some PCMs don't take kindly to. If you've noticed a charging light (battery icon) coming on at times (or not) this might be the case. I'm not aware of this type of reaction on your electrical system, but I've seen it happen on others.
One thing you might try if you get the engine in one of its moods is to look at battery voltage as it idles to see if there's significant voltage variance (more than .2v) as it sits there with no electrical loads on it. Regardless, tap lightly on the alternator case with a hammer and see if it changes anything. The tapping will normally settle the brushes and restore perfect operation for a short time... a very effective test.
Party go OK?
Dinner went fine, but left us worn out,Voltage irregularities seems worth looking into - haven't had any lights, though. I'll check on this.Think I better go get that crank sensor so you'll stop worrying about it. I agree 100% with your original feeling that it's the logical next step. And replacing parts because you can't test 'em is just the way it is for most DIY mechanics. Perfectly normal. Rules of the Game. Still save money in the long run, etc.Marty
Ran into transportation issues yesterday (surprise!) but managed to pick up a crank sensor late in the evening. Looking for my DVM to run that alternator test first - can take the sensor back if I don't install it. Used the DVM last week - can't figure out what I did with it!If you are off on your camping trip I hope you're having a great time!
My problem is with needle nose clamping pliers. I use 'em for evaporative system leak testing... leave 'em on the vehicle. I've fought the problem by trying to put a yellow flag on the vehicle when a clamp is installed but it's only as effective as how often I use the thing.
Good luck with the testing (once you find the meter..)! I'm just about to head out, looks like the weather is gonna cooperate for a little while at least.
Later,
We are expecting Hurricane Ike to pass right over us, and it's pretty sure I'll be off the air for a while. Could be without electricity for quite a while. Put in the crank position censor because I got bored - it didn't solve the problem. May take another look at the throttle position sensor as you originally suggested. Will give you a holler when we are functioning again. Hope you enjoyed the camping trip!Marty
Man, sorry to hear that Marty. Everyone likes a little rain on the roof but this is ridiculous. Not being able to drive the Avenger to higher ground might complicate matters beyond the present ones...
I'm crushed by the results of my last advice. Hang in there and live to see this to the next level!
Hi Ed, greetings from Houston.Back in the saddle again after a very long month. Had no Hurricane damage to deal with at home, but went 8 days without electricity. Boy can that mess you up! Spent two weeks helping friends and relatives, then another two weeks getting all sorts of personal business back in order.Never got to put the meter on the Avenger's throttle position sensor. Hope to do that tomorrow. Things are still busy, but it's time to solve the Avenger problem and get the poor thing back on the road.Marty
You had me worried Bud! That must've been a mess, but it sure could've been a whole lot worse. Glad you got through it OK.
Ed, I'm having trouble trying to figure out how to measure output voltage on the TPS. There won't be any voltage if it is disconnected, and the plug (carrying three wires) completely shields the connections from the fat pins on my meter.Do I need something with sharp pins on it that will punch throught the insulation, or is there a better approach?Marty
Forget the question about how to measure. I decided there was no reason to be picky and simply stripped off a small bit of insulation from each wire. Electrical tape hides the damage just fine.One combination of wires shows 4.98 volts all the time.Combination B goes from .77 to 3.57 volt as throttle increases.Combination C goes from 4.21 down to 1.26 as throttle increasesChanges seemed smooth.Measured with engine cold. Thinking later I should have tried to see if vibration had any effect, but I didn't have enought fingers.Marty
That sounds about right... you should see about 5.0 volts if you measure the supply voltage against the sensor ground and .77 to 3.57 is just fine for TPS range. You're just looking for a consistent voltage change as you change throttle position and that voltage doesn't change at steady throttle. Tapping on the TPS body as you mentioned will sometimes cause a weak sensor to act up, showing a quick voltage change (usually greater). I'd be unhappy with any TPS that shows more than .04v change when rattled.
I'd like to know who's idea this whole two hands-ten fingers thing was!
I'll try tapping on it tomorrow. As it was, I was holding two probes like chopsticks in my left hand, and rotating the throttle thingy with the other.However, we have to come up with another step. It's been so long, you might want to review the whole exercise so far as it is detailed above. I did start the car after my testing, first time it's run in a month, and there was no change. Started quickly. Would not idle smoothly (even when cold!) and died once when I suddenly depressed the pedal about 1/3.Put your thinking cap on...Marty
If the TPS voltage followed in a linear manner when you pushed the throttle (killing the engine) it's not the TPS. Its one and only job is to follow throttle motion.
Funny you'd mention a review! I went back through it a couple hours ago... still don't have anything concrete. One thing I'd like you to try came to mind.
Over near the left hood hinge is a row of four relays, neatly hung on a metal frame. I've had intermittent problems with the relay that supplies power to the coil, the autoshutdown relay (ASD for short). It hasn't acted like what you're seeing, with loss of power and intermittent misfire... you just won't have any spark, like what you saw with the bad distributor (which happens MUCH more often). I'm wondering if the points contacts have developed resistance, decreasing output to the coil and reducing spark at times.
The quickest test is to give the relay bank a rap when the engine's acting up. Sometimes the mechanic's best tool is a small wrench used as a hammer. I'll check some wiring out tomorrow to better identify which one's which so you can take a direct voltage measurement if needed.
It's a bit of a stretch, but every other quasi-American car we build can recognize the absence of ASD output except the Avenger. I'd expect ASD power to be OK simply because no ASD codes had been set on any other car, can't do that with the FJ.
While we're on the subject and the cap's still on, I was also wondering if you'd changed the timing belt in the last 3-4 years. If the timing belt skipped a tooth, it might still be close enough to run, but be right on the edge of cam/ crank signals being out of sync with one another. It's yet another stretch, but something we hadn't talked about yet.
The distributor offers no adjustability, so it couldn't have been put in wrong no matter what.
I'm off this Saturday and it looks like I'll be stuck in the house with crappy weather. Perfect Avenger weather, though!
Hi Ed,Got bogged down with other stuff but I located the relays. There's 3 and I cannot determine which is the right one. I also thought I'd remove them and just look at the contacts, clean them, etc. BUT just using my fingers to release the clip on the lower front of each one - and tugging this way and that - doesn't seem to release them. I may not be understanding what I am seeing. Rather than break something, I thought I'd better check with you before becoming more violent.Sure don't like the suggestion about the main belt (because of the work involved with a mechanical timing issue, so I'm hoping the relay is it.
Releases on these things usually don't work the way they're 'poda once they're out in the world for a bit. Sand and stuff gets inside the voids and makes it tough to push the tang far enough for easy removal. I don't think you can hurt anything if you don't break out the torch, so have at it.
As for cam timing... it's a stretch. If during the times when it runs OK... it ran like it did from before when it started acting up... it's probably not a cam timing issue. You'd have a rougher idle, less performace off the line and a generally poopy running engine all the time, even when it's not cutting out. There might even be a code to back this up, but it varies year-to-year and model-to-model. Mixtures would swing toward the rich side, also possibly causing a code to set.
My base instincts still tell me this is an ignition problem, but it's not going to be easy to sort out. If you can catch it in a bad mood, try pulling the front three plug wires out one at a time and watch the spark length. I can't help feeling that one of the three might have a substantially reduced spark output, if not all three. But you have to catch it when it's cranky for it to matter.
For the time being, I'd just disconnect any vacuum source to the EGR just to eliminate it as a suspect. You might eventually set a code... but that's OK. If the EPA slaps you with a fine, tell 'em I said it was OK. Use get-out-of-jail-free code 109djql;ouf09883.
1. Any way I can tell which of the three relays is the ASD relay2. Any way to jumper across a couple of pins and send good voltage to coil and fuel pump?Marty
Oh sure. But I'd have to actually get off my duff and pull something up. Hope you're happy....
The wiring diagram and colors go as such...
Wire colors are pretty regular except for the fuel pump relay output (black/ light blue). The rest are close to their lettering.... black for fuel pump power IN to the relay... black for power IN to the ASD relay (spliced upstream of there, sharing the same feed)... black/ red for ASD power OUT.
These look like ISO 40 relays, should have three terminals facing the same way with one that... isn't. The odd terminal and the one directly opposite it are power in and out. Jumper these two terminals/ wires if you're looking to eliminate the relay.
As an aid... the wiring diagram shows power (+) flowing from top to bottom (ground). Note that the ASD relay shares the same power source for both control and output, where the fuel pump uses an ignition-on source for control power to the relay. Funny how they do that sometimes...
That's terrific. No way I could have found that info on my own. Should have some feedback for you tomorrow.Marty
Hi Ed,Well I had two days last week when I could play with those relays and it rained both days - steady enough to get soaked - so I had to put it off. Today, after dismounting the bracket that holds the relays I could finally understand what I was looking at. Disconnected all three and inspected the terminals, and found no obvious problems.After spending quite a bit of time trying to decide how best to jump or short circuit the output for the ASD relay, it finally dawned on me that all three relays are the same. They all have the same number on the top 83159, and on the front all the numbers are the same except for a couple of digits on one of them.Light bulb went off. If they are the same, then I can quickly and easily test for ASD malfunction simply by switching the relays. In fact, it looks like I can just switch the connectors.Sound reasonable to you?Marty
Absolutely! I wasn't aware that they were the same thing X3. It sounds like they're just ISO 40 relays with hangers, something that I've got in abundance (without hangers). If you swap relays and problems move with them, it's a very good way of sorting things out without a parts bill. Somehow, I figured they'd have keyed connectors but never really checked it like you have.
It's a neat system. The wiring plugs are generic. The builder just has to make sure the wires are installed in the right slots. Will try this in the morning, and if no joy I'll be spending time with manual to review the steps needed to find TDC on everything!Let you know what happens.
Not sure if I have good news or bad news. Tried the relay switch yesterday, but got shanghied before I could get back to the computer.I switched the connections between the ASD relay and EATX relay on the right. They appear to be identical relays.The avenger would not start at all. On the first grind the engine continued for a couple revolutions and sent some blue smoke out the main air intake. This didn't repeat.Thought I'd confirmed a bad relay, but when I switched the connectors back there was no return to previous behaviour. Grinding away produced no indication of anything firing. Needless to say, this has me seriously spooked.Charged the battery. Wondering how to proceed.Marty
Good question lol!
I think it might be time to do some circuit testing at the PCM. The engine controller is sometimes the last place I look on some vehicles, but there are indicators I've dismissed that might be relevant. Like the oddball temperature readings you get from the ECT. If they're true, you might have a bad sensor or circuit problems which we should take a look at. PCM power sources and grounds, along with other circuits, can have some serious affects on engine operation and even cause intermittent problems depending on temperature and such.
I'll put together a few things for you to check tomorrow; let's see where it takes us. Get the DVOM charged up too!
Well, we wanted to find out if there was problem with the ASD relay, and I'd say that there is now. Acts like no spark + no gas. However I haven't had anyone to turn the key for me today so I can't really confirm. Granted, we wanted the ASD to maybe work better if I banged on it, but failure might be just as good. I don't know what the EATX does.I figure I've got to find a Dodge dealer and get one of those relays now. However, I'm going to dismount them again and see if maybe I stressed those connectors enough to pull a wire loose, or something.Marty
Given the choice, I'd take a dead/ broken vehicle every time. Intermittent stuff is WAY harder to chase.
That said, let's check a few things. To start, leave the key off and disconnect both connectors at the PCM (unless you have a scan tool... get codes first if possible).
This might seem a bit disjointed, but I'll try to group voltages and resistance measurements into groups to avoid too much switching of the meter. I think it goes easier that way.
The terminal pin numbers are conveniently cast into the open ends of the PCM connectors, going 1-40 on one and 41-80 on the other. It doesn't get much better than that!
Set the meter to 20v DC, ground the black lead and check voltage from Pin 46 first. This is B+, the constant battery feed that should be hot in this state. If at all possible, use a small load device for testing the circuit besides the voltmeter. I like to push some amps through power and ground supply circuits to see if they're up to the task. A test light will suck about .5 amp (not bad), but a headlamp or fog lamp bulb will draw close to 4 amps, a better measurement of circuit integrity. If you're not blinded, something is wrong.
Turn the key on and check Pin 20 for voltage now. This is key-on voltage, the wakeup signal. It should be very close to battery voltage (as was 46) and the load test applies.
Grounds are next. Set the meter to 200 ohms and leave the black lead grounded.
Pins 10, 47 and 50 are all power grounds, needing to have 5 ohms or less of circuit resistance to battery negative. Do the load thing-y here also.
Back to the top of the list...
Pin 26 is your engine coolant temp sensor signal circuit. Set the ohmmeter to 2K and check to see if you have continuity to ground, which there should be. It appears (news to me!) that sensor ground is external to the PCM in this model. Write down your measurement.
Pin 32 is your crank sensor signal wire, which should have no continuity to ground with the PCM disconnected. You might set your meter to something higher, like 10K for this. It needs to be an open circuit.
Pin 33 is cam sensor, same-same.
Pin 35 is TPS. In light of the external sensor ground (still reeling), I'd expect to see some sort of continuity to ground. Disconnecting the TPS at the sensor should put an end to that.
36 is MAP sensor, same-same.
37 is intake air temp signal, should read very similar to 26 (ECT). Disconnect the sensor to see if it goes open-circuit.
44 is an important pin. This is the 8v power supply to the cam and crank sensors and can make the PCM really cranky if it's shorted. In cases where it's completely shorted to ground, the PCM becomes a no-response dead chunk o'metal. In cases where it's shorted a lesser amount, it can kill the engine and leave no codes because it takes down BOTH the cam and crank sensors.
Check first to see if it's got continuity to ground in any context... should be none. We'll come back to this circuit later.
48, 49, 57 and 58 are idle air control circuits. These are the PCM power feeds to the idle motor and can cause some havoc of their own if shorted to ground. It will kill the engine and leave few clues. See if any of these have continuity to ground at any level... should be none.
It's a dual circuit, with 47 and 48 completing a loop through the IAC motor and 57/ 58 doing the other one. You will typically have something like 45-60 ohms of resistance through the intended circuits (thru the IAC and back) but there should be zero continuity between the two driver circuits. Any continuity is bad, disconnect the IAC first to see if things change. If not, you've probably got wires melted together WITHIN the harness... not handy. Look in the hot areas first...
61 is also an important circuit. This is the 5v signal to the major sensors, TPS/ MAP/ A-C pressure transducer. It appears you'll have continuity to ground, just be alarmed if you see less than 100 ohms here. I'd disconnect all three to see if the circuit becomes open (as it should) and reconnect each individually to see what their connectivity to ground really is. The A/C pressure transducer has shown some problems in the past with engine killing...
That completes the open connector tests of the PCM. We'd need to power the system now to see anything else that might be goofy.
Everything reconnected, key ON.
Check your voltage at the 5v source for your choice of MAP, TPS or A/C pressure transducer. The wire color will be dark green/ yellow. I'd backprobe it first using a ground-down paper clip or something through the weatherpack at the rear of the sensor to see what operating voltage really is. If you see 4.95 to 5.05v, it checks OK. Tell me if something is outside that range.
The 8v circuit (I promised!) is actually OK between the range of 8v to 9v... The circuit feeds only two sensors, cam and crank. Look for the yellow wire at the back side of the PCM connector and backprobe it there... it's just about impossible to test at the source. The covers come off the PCM connectors, but not without a fight. I'm sure you'll win.
While you're there, check the cam and crank signal circuits. You should see either 0v or 5v... not much in between. This is important because the transition between one state to the other is what determines the actual signal. 2v to 4v won't work. Zero to five. Slowly bumping the engine will occasionally stop the tone wheels on an open hole, which will change voltage state.
Let's check the sensor ground circuit again now that the electrical system is awake again. Black/ green appears to be sensor ground in this system and should show something very, very close to 0v with the key on. If it's draining sensor volts off as it should, you should see almost zero residual voltage on this circuit from any sensor you choose to test.
This is sometimes a handier test than simple ohmmeter testing because you're looking at circuit performance. Think of the ground circuit as a plumbing drain... if it works, no pressure or backup is left.
I'm gonna send this now, Marty. I'm scared sh*tless the site will <woosh> the whole damn thing, as it's done to me on countless other occasions. Take your time and see what you can find. I'm actually encouraged that it's not running... unlike your wife!
Hi Ed,I've printed out your instructions on checking the PCM. I see you truly are an expert, and many thanks for the time you are putting in on my problem.I'd like to stop and review, though, in light of the current situation. I was finally able to do a check on spark and gas. To my surprise I have both. This really stunned me, because I've never had an engine that wouldn't fire to some degree if both were present. This one does absolutely nothing. I even put ether in the intake. Got some very half-hearted and wierd sounding dieseling from the ether after cranking stopped, and that's all.I'm really kicking myself that I didn't start the car before I tried playing with the ASD relay. This would have saved a lot of confusion. But it had always started right up all along, and it just didn't occur to me. Anyway, since there is spark and I could smell the gas (with two plugs removed) the ASD is working at least as well as it was before. The engine should at least fire!I'm thinking that your earlier suggestion about a timing belt slip may have been right on - and that it slipped another notch when I tried starting it up the other day after the car had been sitting for a week. Mis-timed valves is the only thing I can think of that would prevent at least token firing if spark and gas are present. Of course I don't have the experience of working on hundreds or thousands of cars, but I remember a slipping timing chain on a car I owned in the 70s and how performance declined in very obvious steps...You'll have to tell me if you think the PCM could still be responsible, but I'm wondering what the easiest way might be to confirm a mechanical timing fault. Do you think a compression tester would tell the story?Marty
Hey Marty!
You might be right about the timing belt thing. If it's finally decided combustion won't happen again, yet you have spark and fuel... it seems pretty plausible.
And as you asked, a compression test will be a very valid test to decide whether cam timing is on or not.
The problem (always one..) is that just about every compression tester seems to work differently. I'd expect to see something close to 170 psi on MY tester, but the guy across the aisle from me will see something different.
Let's say anything 140 or less is suspect. Do your testing at wide-open-throttle (ASD removed) and give it 10 hits to give the compression tester the best chance possible to give true compression for the cylinder. My feeling is that if you've lost THAT much cam timing, it'll wind up well below the 140 cutpoint.
The engine does appear to be an interference-type, so be careful of cranking or starting the engine at this time... bent valves may follow.
The cam reference marks are pretty easy to access by removing each of the cam timing covers, but the crank mark will remain a mystery unless the bottom end is disassembled. That said, a fairly accurate TDC positioning can be obtained by using a long screwdriver down the #4 spark plug hole (companion to #1, under the intake). By slowly rotating the engine backward and forward, you will be able to come within a few degrees of perfect TDC, enough to tell if the engine is grossly out of time or not.
I'd expect any timing belt tooth loss to come at the crank end, affecting both cams equally. The tooth area on the cams are roughly twice the crank, so it's more likely to be the place a skip takes place.
Last, you're giving me WAY too much credit bud! If I was that good, you'd be burnin gas by now. Have a great Tuesday <sigh> , talk later.
Thanks, Ed.I'll call around first to see if I can borrow a compression tester. Hoperfully, there's one that will get down through that tube in valve cover.
Clicked the button too soon!I think I was one belt notch off, and now too notches off. Definitely concerned about valve crash but praying I'm still safe.The timing belt was new about 25,000 miles ago. But about 18000 miles back the water pump froze up solid. I knew the belt had been dragging across the pump's drive thingy but when I and a mechanic friend looked at it, it seemed OK. Now I think the decision not to replace that belt may have come back to haunt me.Will let you know as soon as soon as I've got information.Marty
It shouldn't be a problem. Most testers use an extended hose with a coupler or just the hose. Either way, it should reach pretty easily. If there's a question, your spark plug holes are 14x1.5mm. The only other automotive choice that I know of are the old Ford spark plugs at 18x1.75 (I think). If it looks right, it probably is.
Tough call on the belt even with the water pump problem. If it was in good shape then... shouldn't be a problem now.
Got the compression tester. Readings on the front three cyl. are almost too good, ranging between 150 and 160. Obviously no problem there. However it's the other head that hooks up to that distributor. I'm thinking I'll take off the input manifold again, and go for the rear cyl.s as well.Marty
The actual specs for compression (believe it or not) are around 200 psi for this engine! I can't believe it, but that's what it says. I've never tested anything with that kind of compression with my meter, which tells a little bit about tester variance.
Hi Ed,Thought you'd like to know I haven't given up. The wife got a raise and a bonus and has bought a new car. Had to get the avenger out of the driveway, and my only choice was to clean out and rearrange one side of the garage. It's in there now, and at least I can work on it whenever I have some time without dealing with issues like lighting and weather.It's made me mad, so I'm determined to make it run, but I've been told that I only have a month to save it from a cruel fate.Marty
You crack me up Marty! I read this at work earlier, couldn't stop laughing long enough to respond! Maybe camo netting or something. Push it around back and do some creative landscaping.... Congrats to the wife for her promotion and bonus! Way good! But you're right, this thing needs to be fixed and we'll get to the bottom of it eventually. I'd explore the cam timing thing next. www.savethebrokeavengers.com No... it won't work. EdDodgerench39764.1824350694