Engine light - does it turn off by itself?

Hi Folks,

98 Sienna, 185K miles. I had what felt like a low RPM engine miss or a clutch slippage upon start from stop when I put my foot down pretty hard. It was 'skippy'. Autozone read my codes which were all different versions of cylinder miss. I changed the plugs and wires (and those 3 plugs in the rear were an adventure, but I got them - just a 6 inch extension and reaching around the manifold and patience did it). Runs better now, but after maybe 15-20 starts the engine light remains on.

So my question is this: Does the light reset itself or does one need to do something to reset it and then see if I get any more codes? I know my Jeep TJ resets itself after a while and I wonder if the 98 Sienna does something similar.

Also - does that Autozone reader have a reset function or what need I do?

Thanks folks, Tomes

Reply to
Tomes
Loading thread data ...

It depends on what the condition is. It's safest to assume "no".

If you are just getting Autozone's staff to read the codes for you, they are not permitted to clear them. Yes, the reader has a clear function. You can [on all the cars I've tried it on, anyway] also clear codes by disconnecting the battery for say an hour. The car will perform below optimal for a while after this - it needs to relearn the engine parameters.

Reply to
larwe

It must be reset by your dealer or Autozone

Reply to
Hauli

There are 3 ways to turn off the light You can disconnect the battery for

60 seconds to clear the codes, you can go back to AutoZone, or you can wait for the light to turn itself off. The problem is, if the trouble remains, then the light will not go out.

After you turn off the light, if it comes back on, you should go back and have it re-scanned. If it is the same trouble codes, check the coil for the identified cylinders. There are 3 coils on the front side of the engine, and each coil fires 2 spark plugs at once. If the diagnostic trouble code points to cylinders that share a coil, then check coil primary and secondary resistance. I think that some Siennas have had bad coils, which is why I'm suggesting that you check them out.

Reply to
Ray O

About a minute or two is all it takes to clear codes. You can also pull the EFI fuse to clear the codes. The car does not have to "relearn" engine parameters because they are programmed into read only memory. If you are going for an emissions test, the ECU has to go through 2 trip cycles without detecting an error to register as "ready," but there is no effect engine performance.

Reply to
Ray O

The guys at Autozone can reset the codes using the same tool they used to extract them.

Reply to
Jeff Strickland

Read the manual of any modern car; it needs to recalibrate something when you replace the battery (they don't specify exactly what). Observably when I pull the battery from one of my cars, it behaves oddly ("feels" wrong) for at least the first workday (which is two

25-mile highway trips for me).
Reply to
larwe

The advice I provide is not correct 100% of the time, but I like to think that I am correct more often than not. The regulars in this ng who read the posts about Toyotas can probably tell you that I will make corrections to incorrect information and say that I was in error when that is the case.

Although I did not read the owner's manual for each of 100 or so new Toyotas I've driven, I am pretty familiar with owner's manuals since I often had to describe their contents to customers or write product reports on errors in the manuals during my time as a Toyota district service manager. I have never noticed a reference to relearning engine parameters in a Toyota owner's manual. Perhaps you can point out what model and model year Toyota owner's manual has the information about learning engine parameters when the battery is replaced?

I am also pretty familiar with the format of the factory service manuals since I sometimes had to refer to them to assist a dealer that had a problem fixing a customer's car. I sat through every technical class the automaker offered and passed all 8 National Institute of Automotive Service Excellence (ASE) tests to qualify as an ASE Master Technician and Toyota Master Technician so that I would have a little more credibility with the dealership technicians. Along the way, I think I managed to pick up a little understanding of automotive operating principles.

The OP's vehicle has the SAE compliant on-board diagnostic generation II (OBD II) electronic control module (ECM). OBD II ECM's store operating parameters in non-volatile read-only-memory because without those parameters, the engine wouldn't even operate. If the ECM had to "learn" engine parameters, how would it know what proper O2 sensor voltage and signal frequency is? Even if it knew that the optimal air:fuel ratio for unleaded fuel is 14.67:1? How would it know what injector pulse duration to set during open loop operation? How would the ECM even know whether it should be in open loop or closed loop operation? Believe me, all that information is stored in ROM and does not need to be "learned" or "calibrated."

The only thing that is stored in the engine ECM's RAM is whether the vehicle has been operated for at least 2 drive cycles so that it can report that it is in a ready state; what, if any DTC's have been stored; and adjusted ignition timing advance if the piezo electric knock sensor has signaled the ECM to reduce spark advance. The third parameter will be stored in the ECU's RAM as soon as the ECM detects a continuous signal from the knock sensor. If the driver uses gas with an octane rating that is lower than what the manufacturer recommends, the ignition timing adjustment is stored, but if the ECM doesn't detect any pinging, then there is nothing to store.

Reply to
Ray O

Thanks Jeff.

Reply to
Tomes

Thanks RayO, much appreciated. I am off to disconnect for a while and I will post what happens next. Could very well be a coil or two or 3. Do we know a ballpark number as to what they cost? If they are cheap enough I would not be adverse to replacing them all after 185K. Dealer item or Parts store? Thanks again (and thanks for the detail in the reply to larwe as well as I learned stuff there too), Tomes

Reply to
Tomes

You're welcome.

You can buy the same tool on eBay for about $60. There are 3 or 4 tools from the same maker that run in a range of about $50 to about $120, with the more expensive models offering greater detail in the code readouts and other such features. If I was going to buy one for a post-1996 (OBD II) car or truck, I'd buy one of the mid-scale models. I recall there being 4 models, I'd not get the cheapest but the next one above. I recall it as providing all of the functionality that I would ever want at home, and it represents a good value to me. Maybe I'd get teh second above the cheapest, but I know I would avoid the cheapest and the most expensive and select one of the middle models.

Autozone sells them too. I have an Actron scanner for my '95 Ford (OBD I), but it only works on Fords, and it only flashes a light that I have to count to arrive at the code displayed. The units for OBD II cars and trucks will give a text display of the code.

Reply to
Jeff Strickland

Cool. This sounds like a Christmas present. They never know what to get for me.... Since I have this 98 Sienna, the 2002 TJ and we are going to get a Prius, this OBD II reader is the way to go. Tomes

Reply to
Tomes

I'm with Ray on this one (for what that's worth). OBD I vehicles have to relearn stuff, but I'm pretty sure that OBD II systems store the parameters in non-volitile memory so that when the battery is changed out -- or dies -- the parameters do not have to be relearned.

Having said that, IF the parameters must be relearned, then the firmware would store the less-than-optimal program so the vehicle would run until the optimal settings could be relearned. I think this is how OBD I works though, and is moot when discussing OBD II.

Reply to
Jeff Strickland

Absolutely. My fleet is mostly pre'96, so I don't need an OBD II reader yet. My daughter has an '00 BMW 3 Series, and my bro-in-law has an '02 Ford and is shopping for another vehicle. I can almost justify buying a new tool ...

Reply to
Jeff Strickland

New tools are ALWAYS justified....LOL

Reply to
Scott in Florida

Indeed.

Reply to
Tomes

My information was from the instruction manuals for 1998, 2002, 2004 and 2006 vehicles that I own or operate, although they are not Toyotas. They all say the same thing, in more or less the same words; both in the context of driving a brand-new vehicle off the lot and retraining it if the battery needs to be replaced. All these vehicles are OBD2 also. So if what you're saying is "no OBD2 vehicle needs to self-adjust" then I am confident you're wrong. If what you're saying is "no Toyota OBD2 vehicle needs to self-adjust" then I am skeptical.

The factory defaults are in ROM, yes, but the user literature claims that operational parameters are adjusted dynamically with the factory defaults as a known-good but suboptimal starting point. I don't know if Visteon electronics are fundamentally different from Toyota, but it seems most unlikely. If Ford powertrains need "training", I'd bet money that Toyota powertrains do, also. If Ford could hardcode everything, they would.

I can't make more intelligent rebuttal to what exactly is being "trained", "learned", whatever because I'm not sure what it is; I'm an electronic engineer, but not in the automotive industry. I would guess that it is self-calibration of sensor offset error but that is just a guess. I've read that this phase is calibrating the airflow vs. fuel delivery but that was not in a formal context.

The same "training" code is what copes with sensor aging - for example, a MAF sensor getting slowly dirtier. Haven't you observed your vehicles behaving differently - shifting at different points, generally "feeling" different - after cleaning a MAF, even if you took care to dry it off before reassembly? I sure have.

Reply to
larwe

I have never replace one of those coils so I can only guess what they cost. (my guess is $100 ~ $120 each).

I recommend OEM parts in general, but especially for ignition and brake parts. I have seen some drivability problems caused by aftermarket ignition parts, which is why I stick to OEM.

Glad to be able to help!

Reply to
Ray O

Perhaps the others in this ng can look in their owner's manuals and share if this info is in their Toyota or other manuals for 1996 and later vehicles. It wasn't in any of the manuals I looked at.

All these vehicles are OBD2

All 1996 and later vehicles sold in the U.S. are OBD II compliant.

So if what you're saying is "no OBD2 vehicle needs to

I did not say that no OBD II vehicle needs to self-adjust. I said that no OBD II vehicles need to relearn engine parameters. If you understand how an electronically controlled system works, that is a BIG difference.

Engine parameters do not change so they are permanently stored in read-only memory. Engine parameters are things like firing order; fuel injector pulse duration for a given amount of air; the conditions under which the system operates in open or closed loop; correct signal voltages and resistances from sensors, etc.

Once an engine is operating, the various sensors provide variable feedback to the ECM, and the ECM will make adjustments to the things it controls. For example, by monitoring the voltage coming from the throttle position sensor, air flow sensor, and O2 sensor, the ECM will adjust injector pulse duration to provide the correct amount of fuel, based on the engine parameters that were stored in its ROM. This "self adjustment" process is continuous and in real time so the ECM is not learning those variables, it is controlling something based on those variables. The ECM does not store those variables because there is no need to know what the variables were in the past; it only needs to know what those variable are now.

Perhaps you are confusing "parameters" with "ready state." The only information the ECM stores are diagnostic trouble codes or lack of DTC's from each sensor. Most DTC's have 2-trip detection logic - a fault has to appear for 2 consecutive trips before the check engine light comes on, and if the fault is absent for 2 trips, then it turns the light off. When the ECM's memory is cleared, it is no longer in a ready state until the vehicle has been driven for 2 trips and the ECM has not noted any faults from any of the sensors. Whether or not the ECM is in a ready state does not have any effect on drivability or engine performance. A vehicle has to be in a ready state for emissions tests so that the tester can determine that ECM is communicating with all sensors.

I think we have a difference in what we consider to be "operational parameters." I consider an operational parameter to be an if-then statement. For example, "if coolant temp is below XX degrees, then enrichen the fuel/air mixture." The engine ECM is nothing more than a fancy controller, and while it adjusts output based on input, it does not self-addjust or vary output for a given input.

This is not self calibration of sensor offset error. Automotive ECM's have

2 modes of operation - open and closed loop. A vehicle will be in open loop before it is fully warmed up or when throttle opening is above a certain amount so it delivers fuel based on MAF sensor and throttle position sensor. It ignores the signal from the oxygen sensor during open loop because the sensor output is not usable before it is warmed up and when the fuel mixture is enrichended under heavy acceleration. After the engine is warmed up, it enters closed loop mode, where the ECM "fine tunes" the air-fuel mixture to reduce emissions based on input from the O2 sensor. There is no calibration taking place, just a decision on whether or not O2 sensor feedback is valid or not.

The improvement in performance after cleaning the MAF sensor is not due to training or re-calibration. A MAF sensor is basically a potentiometer. The MAF sensor is basically a heated wire sitting in the air stream. As more air flows past the heated wire, the wire cools down, and as the wire cools down, resistance goes down, and as resistance goes down, voltage coming out of the MAF sensor goes up. The ECM calculates how much fuel to deliver based on voltage from the MAF sensor. As the wire in the MAF sensor gets dirty, the dirt acts as a thermal insulator so air flowing past the dirty wire doesn't cool down the dirty wire, so less voltage comes out of the MAF sensor. With less voltage, less fuel is delivered and so performance goes down. When you clean the dirt off the wire, air will once again cool the wire and so voltage will be higher coming out and performance improves. The ECM doesn't know whether the MAF sensor is clean or dirty, it only monitors voltage coming from it.

Another way to explain this effect is with a photocell. The more light shining on the cell, the more voltage coming out. If the cell gets dirty, even with the same amount of light shining on it, less voltage comes out. If you clean the cell, then voltage output is returned to original levels. No learning was involved when you clean the MAF sensor or clean the photocell.

You are giving automotive ECMs credit for having more processing power, or "brains" than they actually have. Some automatic transmission ECM's learn a driver's driving style, but that is more like setting a preference than actual learning.

Reply to
Ray O

a nice explanation of the subject. many people don't recognise the difference between setpoint and feedback. sensor errors creep in and muddle the real needs of the engine. sammm, an old field engineer

Reply to
SAMMM

MotorsForum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.