The MIL came on (steady on, not blinking) about a month ago on a very rainy day. The car felt sluggish going from stop to start and at low speed, but then it was OK at higher speed (>40 mph). I checked the fuel cap, which was tightened appropriately. I went to my mechanic. His computer diagnosed the problem to be malfunctioning of the “throttle position sensor”. I had the throttle position sensor replaced. About two weeks after the new sensor was put in, the MIL came on again and went off in two days. Two days later it came on again (steady, not blinking) after I drove to work in a downpour. This time the car had no problem getting from stop to start and at low speed, but could not go over 45 mph. I went to my mechanic again. The computer diagnosis this time was “system too lean”. After three dry days when I drove only very short distances, the MIL went off. What is going on? are computerized diagnoses always accurate? why did the MIL seem to be related to wet weather? I bought a Honda Accord because of its reported reliability. I am very disheartened by these problems. Please help!
Computerized diagnoses are supposed to point a mechanic in the right direction, but he still needs to use logic, observation skills, and basis diagnostic skills to resolve a problem. A mechanic who automatically interprets a code to mean, “replace the ______”, is simply a parts replacer, not a true mechanic.
This is just a guess, but, from afar, I think that the problems are likely to be related to bad electrical connections that are aggravated by wet conditions. I would suggest finding a different mechanic who is willing to do a lot of wire tracing & inspection in order to find the problem.
Also…did you buy this car as a new car, or as a used car?
I ask because a used car is only as good as its care & upkeep were prior to the second (or third) owner’s purchase. If this car was purchased as a used car, it is possible that it suffered collision damage or flood damage in the past, either of which could affect the integrity of electrical connectors.
Thank you very much, VDCDriver. Electrical connection problems make great sense. I will find a different mechanic to track down the problem.
I bought the car as a used car in 2009, It had been working well for 4 1/2 years until recently. With this info, Do you still think flood damage or collision damage may be the ultimate culprit?
Any mechanic with access to either the factory service manual or alldata should be able to follow the troubleshooting logic diagrams (aka troubleshooting trees) to follow each DTC (diagnostic trouble shooting code) through all the connections.
If you are have multiple problems, especially in very wet weather, there is a chance that a single connector such as a firewall connector is involved and has a bad seal or may have been disconnected at one time and not fully seated when reconnected, so water is finding its way into the connection.
The troubleshooting trees take the mechanic from connector to connector to isolate the issue, however, you can expect to pay a diagnostic fee for this service, because unlike plugging in a code reader and reading a code, following a troubleshooting tree can involve quite a bit of time and a mechanic should be compensated for this. But in the long run, it can save you money because it finds the root cause and only that repair is needed.
But in your mechanics defense, 99% of the time, its the part.
Thank you very much, Keith. I want to get to the bottom of this and get the problem solved, rather than replacing different parts and still having the problem. I already paid $345 for the replacement of a throttle position sensor. The involvement of an electrical connector seems very likely.
You didn’t tell us how many miles are on this vehicle. It could be that two problems back to back are just coincidence too. In fact, there is an old saying that problems come in threes. Honda’s are very reliable, I have a 97 Accord with 187k miles on it, owned since new. It has had a couple of issues, but I would not call it unreliable by any means.
TPS will often burn the portion of the potentiometer that is used at idle, that is not really an uncommon repair. The lean condition can be tricky to solve as several components can cause the issue, along with any number of vacuum leaks, air leaks around the air intake ducts after the MAF (mass air flow) sensor or even an old air filter that is allowing moisture to get into the intake and contaminate the MAF.
You should also get the PCV valve checked as if it gets clogged up, oil fumes will get into the intake duct and contaminate the MAF causing the lean condition. AN O2 sensor can also cause this but usually they will pop a separate code for this eventually. An O2 sensor that is getting slow, but not slow enough to pop its own code may give a P0420 code which many mechanics will think is a bad cat.
When you get a check engine light, get the code read and your mechanics opinion of what is wrong, then post the code here (letter and 4 numbers, i.e. P0200) and we will help you decide if the mechanic is right or he should explore other possibilities first.
If your current mechanic has been giving you good service at fair prices, I think you should work with him on this before jumping ship on him. If he has a computer with internet at his shop, and if he is customer oriented as he should be, then he should allow you to log in here and view our responses.
Thank you very much, Keith. My car is approaching 83k miles. I do mixed city and highway driving. As Honda Accords go, I think this car is still considered to be in its prime. As I mentioned in the original post, the MIL is off now. Will mechanics be able to detect anything wrong at this point? There appears to be many possibilities. It looks like a complicated diagnosis process as you detailed. I think electrical wiring still a strong possibility because the MIL came on during or right after wet weather.
I consider my mechanic to be honest, but I do not know how competent he is partly because I do not have extensive knowledge on cars. He is on the pricey side. I have been going to him because he does not suggest additional things to do like a Honda dealer and another mechanic I went to. I can ask him if I need certain maintenance or replace a certain part, he would tell me not to do this or that because it is not time.
“System too lean” means what it says. And the car not wanting to go over 45 is another good indication that it’s starving for fuel. I’d have the fuel filter checked/replaced, and the fuel pressure tested. A partially clogged fuel filter might not show up as much in city driving as it will at highway speeds or during hard acceleration runs.
Thank you, oblivion. If it is “System too lean”, why did the MIL come on during or right after wet weather and go off after a few dry days?
I have no idea… It could be a coincidence, it could be your driving style is different in the rain, it could have been water intrusion into a sensor wiring harness, maybe conditions are borderline such that it takes a day with low atmospheric pressure to meet the criteria for setting the light. Or perhaps it’s cooler when it rains, and the cooler, denser air entering the intake is pushing a borderline condition over the edge.
Most likely it’s just a coincidence that it happened on a wet day. If you went on a highway run on a wet day for example, the engine starved for fuel, and the light turned on, then drove like a grandma in the city for a few days, without accelerating hard or going very fast, that would tend to make the MIL go off, as the computer would not see the problem reoccurring under those conditions.
If you rule out the fuel filter and your fuel pressure is good under the demand of going at highway speeds, then I’d start looking for connectors under the car that might be getting wet. It would also be useful to have any pending codes retrieved—the system will store codes for problems that are occurring that it’s “keeping an eye on”, that haven’t quite met the criteria for turning on the MIL yet. Someone with a scan tool can verify that your oxygen sensors are changing value as they should and are not “lazy”–preferably on a wet day–that would be one way to determine if the lean condition (or perceived lean condition) is being caused by a flaky sensor or wiring fault. (assuming the fuel delivery is good) Also the fuel trim can be looked at to see under what conditions the engine is actually running lean and the system is attempting to compensate for it.
A competent mechanic should be able to interpret what’s going on by looking at the data stream in real time.
I went to my mechanic this morning. He plugged in the computer and the P2138 code came up. It said “Throttle Pedal Senor failure”. He could not do anything now because the MIL was off. Two questions.
- Is it safe to drive?
- I just had the “throttle position sensor” (TPS) replaced. The part looked like
What are the differences between “throttle position sensor” and “throttle pedal sensor”? I looked online, but it seems that the two terms are used interchangeably.
Generally the codes will remain in the computer for 10 drive cycles after the last time the DTC is detected. Most DTCs are what are called two trip codes, that is they must be detected twice before they turn on the MIL (check engine light). If the DTC is not detected again in the next three drive cycles (engine cold, driven to warmup and shut down until cold again), then the MIL is turned off, but the code remains in the system for 10 more drive cycles. after 10 drive cycles, the code is deleted and the ready light is turned on (for smog station checks, you don’t see this).
The rain could be causing the DTC by either getting moisture into a compromised connector (loose or damaged) or getting into the MAF via an old air filter or something left off the air intake duct.
I think you have what is called a “drive by wire” throttle system. You have a TPS at the throttle body to inform the computer of the throttle position. But instead of a throttle cable that goes from the gas pedal to the throttle body and has been traditionally used since almost the invention of cars, yours uses a “Pedal Position Sensor” in the gas pedal. It sends an electrical signal tot he computer to tell the computer how far you have pressed down on the gas, then the computer sends a signal to the throttle body to tell it to open the appropriate amount.
Maybe you got some water in this sensor when you got into the car with wet shoes on. It should be sealed against this, but there might have been a seal failure. I don’t know if a false signal from this could somehow lead to a lean condition, but maybe it could. These are pretty new systems. Even the car companies are still on the learning curve for them.
There is a throttle position sensor . . . TPS
This is mounted on the throttle body
There is an accelerator pedal position sensor . . . APPS
This is mounted on the accelerator pedal
There is no “throttle pedal sensor”
Is it possible somebody got the 2 confused?
I did not make a mistake because I wrote it down directly from the computer screen. My mechanic even said that “throttle position sensor” and “throttle pedal sensor” were two different parts. I was confused when I looked up online.
Are you in the U.S.?
I live in the US.
Do you understand now. Not everyone uses the same terms, even car manufacturers.
The MIL went on again yesterday, a dry day, on the way home. This morning I went to my mechanic and the diagnosis code was P2138 again. After going on a website (appeared to be for professional mechanics), he told me that the accelerator pedal position sensor (APPS) needed to be replaced. I was told that the APPS was not on the pedal in 2007 Honda Accord. Any idea where it is? Is it possible that it was APPS all along? Unfortunately I did not have the diagnosis code the first time.
Thank you, Keith.
On my 2006 Chrysler, the electronic throttle is in the trunk, under the battery compartment. A fairly conventional throttle cable runs from the accelerator pedal to it. I don’t know where it is on a Honda, but it can be pretty much anywhere. A factory service manual can tell you where to look.