Hyundai Excel Ambiguous Codes Read - continued

This is a continuation of a post with the same name. Briefly, the 91 Hyundai Excel I own idles roughly and smokes sometimes even though it seems mechanically OK (cylinder pressure within spec).

I am hoping COMBOVERFISH responds to this post as he was absolutly right in suggesting a multimeter would show a trouble code when a scope did not. After the last post, I replaced the oxygen sensor and the symptoms are similar.

Diagnostic Codes

1 2 5 2 5 2 5 2 5 ... which I take to mean Barometric pressure sensor. So far I have no idea where this is or how to test it.

Oxygen Sensor I replaced the sensor although a good ground may have prevented me from detecting that the old one was working. In any case the behavior is unchanged. On warm start, I measure between 150-300 Mv initially. After

5 minutes the signal has increased to the range 890-1000 Mv. The engine idles roughly and when accelerated, smokes. Unhooking the oxygen sensor seems to produce no change in the smoking behavior or the oxygen sensor reading.

Am I reading the codes correctly and what is the barometric pressure sensor? Is it possible that the oxygen sensor is not connected to the ECM? Any suggestions happily accepted. Thanks

Reply to
redryderridesagain
Loading thread data ...

Further tests;

After disconnecting the negative battery terminal for 30 secs, the codes seem to be reset. The diagnostic codes are 11111111111111111...

The behavior at the exhaust pipe/oxygen sensor is the same - an initial reading of 0-100Mv and no exhaust pipe smoke rising to 850-950Mv with smoking.

Disconnecting the oxygen sensor does not change these symptoms and does not change the diagnostic code which is still 11111111... No overall engine warning light comes on.

Why does disconnection of the oxygen sensor not change the diagnostic?

Thanks for any ideas.

Reply to
redryderridesagain

============= =============

Put a new PCM in the car.

~:~ MarshMonster ~:~

Reply to
Marsh Monster

Do you mean PCM as in the following;

"2002 Hyundai XG350 Even though a cable is connected to the accelerator pedal, this vehicle is drive-by-wire. The cable leads to a pedal position sensor under the hood, and the powertrain control module (PCM) operates the throttle."

I can't find any reference to a PCM in the Haynes manual, however the following components are described; Fuel rail and injectors Fuel pressure regulator Throttle position sensor ECM Idle speed control actuator Air flow sensor Engine coolant Temperature sensor Intake air temperature sensor Crankshaft position sensorMFI control relay Vehicle speed sensor Oxygen sensor

This is a 91 Hyundai Excel - would it have one? Any suggestions on what its function is or where to find it? Any suggestions regarding those other components?

Tkx

Reply to
redryderridesagain

Consider the PCM to be the same as the ECM.

Powertrain Control Module versus Engine Control Module...

Reply to
<HLS

Is there some conclusive test for an ECM failure? Is there something significant in the car's behavior that I described or are these units prone to failure?

The following is originally from a vendor forum, I think.

"...But, I have to ask, if you drove your car to the dealer with this problem what part of the computer failed? My experience with bad computers is really bad, as in you can hardly get anywhere with the vehicle. Or you just can't get anywhere. Car computers are real tough pieces of silicone. I've seen wiring harnesses melted together and after the repair the computer shook itself and went back to work...."

Are there some other steps that I can go through to test the sub-components and rule out a simpler problem? What are the likely sources, given the symptoms or are there no other possible explanations?

Thanks

Reply to
redryderridesagain

I just happened on your new posts by chance. RE: the ECM, you basically have to verify that all inputs, outputs, powers and grounds are functioning properly before you condemn it. If your Hyundai ECM is the same build type and manufacture as similar year Mitsubishis, then I would highly suspect it as the cause of nonsensical garbage codes. You still need to check powers and grounds to the ECM, and ATLEAST make sure that the input signals are sensible.

It is difficult to test an O2 sensor on a vehicle with a concurrent funky ECM problem. You can force it to read rich by pinching off the fuel pressure return hose, and you can force it to read lean by introducing a large vacuum leak. Those two methods will tell you if it is capable of reacting to known conditions by producing from 0 - 1 volt accordingly. Chances are that your new sensor is OK, so you can move on to other suspect areas for now.

The possible code 25 refers to a failure in the barometric circuit inside the airflow meter. This could cause a large shift in fuel trim if it was way off calibration, or shorted/open. Similar story: I had a Land Cruiser in the shop that was setting a baro code. This particular vehicle, unlike other Toyotas, inferred calculated Baro readings from other sensors. IOW, it set a bizarre baro code even though it didn't have a real baro sensor. It turns out that the ECM was so far off in it's baro calculation that it ran extremely lean to the point of near misfire and very high NOx (failed an emissions inspection). A new ECM fixed the Land Cruiser.

If you are still suffering black smoke out the exaust, perhaps you could find a junkyard airflow meter to try out. Your car uses a Karmen Vortex design that outputs a frequency to the ECM based on airflow, plus discrete air temp and baro inputs. There's plenty to go wrong with this design that could send a false richening command to the ECM. If the AFM is hard to find or too expensive to purchase as a quick cheap parts swap, then it's back to more diagnosis. I am having too difficult a time with deciphering these codes over usenet to be of much use. I don't feel confident that you are getting consistent results from the code retrieval process - and I blame Hyundai for that, not you. Their service info seems pretty useless.

Toyota MDT in MO

Reply to
Comboverfish

========= ========= what he said..... ditto

~:~ marsh

Reply to
Marsh Monster

Here are a post or two from a tech service I subscribe to; The common thing I noticed is bad grounds to and sor the computer and bad computers.

Good luck

Reply to
Stephen H

Thank-you all for you help so far. Until I read Stephen's post, I would have assumed the MAF sensor for the reasons below - now I feel I should, test or replace the ground splice.

MAF Tests I performed the MAF tests in Haynes 4-24 (Air Flow Sensor 1994 and earlier) Output voltage of the sensor when disconnected between pins 3,

6 should be 2.7-3.2 V (was .15-.03V). Voltage between pins 5,6 should decrease when air intake is obstructed - it was 0V. The codes shown after this test with MAF disconnected are 2,5 - Baro pressure sensor 1,2 - Airflow sensor 1,5 - Motor position sensor

Is this a good enough indication that the MAF has failed?

ECU Resetting the ECU and disconnecting the coolant temperature sensor for a general test of the diagnostic fcn of the ECU produces

1,5 - Motor position sensor

I guess the ECU is not meant to be interpreted by mugles :>

Ground Is there any good test of the bad ground theory - such as connecting the test ground on the ECU to the negative battery or the block with a good thick conductor and seeing if the problem disappears or looking for a voltage W.R.T. ground on the ECU test ground?

Thanks again for all the help - I'm proceeding with MAF replacement.

Reply to
redryderridesagain

Sorry that should be Air Flow Sensor and not Mass Air Flow sensor (1995 and later)

Reply to
redryderridesagain

Sorry that should be Air Flow Sensor and not Mass Air Flow sensor (1995 and later)

Reply to
redryderridesagain

Replace the AFS or NOT.

AFS I believe that the Haynes instructions are misleading and wrong. Many sensors including the AFS, cannot be disconnected and tested but must be backprobed. The voltage between 6-5 is power and 3-6 is output, not vice versa. The scope shows a great looking square wave which varies in frequency as airflow. The Karmen Vortex sensor is a complicated piece of engineering but it seems to still work as designed.

CKP Crankshaft position sensor (in the distributor) seems to be working as speced and gives a regular looking square wave. The code being registered appears consistently to be 1,5 = Motor position sensor. How is this different from "Crankshaft position sensor" and "Camshaft (TDC) position sensor"? Is it possible that the code is an indication that the cam and crank are not synced? Would this engine run if the two were not synced?

Oxygen sensor and street behavior The oxygen sensor swings between .2-.8V when idling for short periods of time, however, moves to the range .9-1.0V when the gas pedal is depressed.

Next steps I will replace the ECU when I get one (used) and will continue to test sensors.

Reply to
redryderridesagain

======== ======== I do some work for a used car lot.

They've brought me 2 used ECUs in the last 8 months to install because they didn't want to spend the money on a reman unit.

NEITHER of the 2 vehicals were repaired by installing the used ECUs.

The first was a Honda Accord. When I told them the ECU was bad, gave them a replacement price, they requested for me to install the used ECU. No prob. They paid my labor on the diagnosis. Supplied the used ECU. I installed it. Then I called them up and asked if they wanted me to charge another $289 to diagnose the symptoms that the vehical had now that the used ECU was

in it, or would they like to opt for a reman? Their call, not mine. I ended up letting them provide a reman unit, just to prove a point.

The second vehical was a Nissan Pathfinder. When I called them up and told them the ECU was bad, they did the same thing. So did I. I got paid for my diagnosis, paid to install their used ECU, made my phone call and asked them if they wanted me to charge them for diagnosing the new symptoms on the truck, or if they would like ME to install the ECU. They took the vehical, put a reman ECU in it. THEN, brought it back to me because they couldn't figure out why the damn thing was'nt spitting out codes, tranny was now shifting, but the idle was rough as a cog. I got paid $129 for driving the truck for 2 hours to a seminar. (Relearn the computers memory) Something that would have been done if they had allowed me to install the ECU in the first place.

But, if they're customers don't mind, neither do I. lol

any whoooo....... this is just a warning to you. Make sure that the ECU get has a warranty.

Pull codes on the vehical BEFORE you change the ECU, and make special note of the symptoms the vehical has.

Then, pull codes on the vehical AFTER you install the used ECU if you have further drivability concerns. That's the quickest way to detect a bad ECU when using used ones on interchanges. I've seen bad used units quite a bit. Not all the time, but enough to where I will NOT use them. Unless the customer supplies it, and then I get payed for my initial diagnosis and the jobs done. When the used ECU is installed, you HAVE TO...look at the vehical as a new job if it poses ANY symptoms what so ever. It's the foundation of diagnosis. Diagnose the problem in a systematic way, each time, every time. And, never "guess" that the part you just installed is good when it could be bad.

again, just a warning and some advice.

hope all works out well for you.

Reply to
Marsh Monster

As previously stated. You still aren't addressing the possible Baro code, and the fact that there is a Baro sensor built into your AFM, which could shift fuel strategy if out-of-range. All you have to do is test from the Baro wire to ground as determined by your wiring diagram. I think they are terminals 6 and 4 (from memory)

If this doesn't fix it, you may well have spent more than the car is worth. Keep things in perspective before you do something pound foolish.

Toyota MDT in MO

Reply to
Comboverfish

This is about a normal operastion for an o2 sensor; it could be lazy or have other problems but for interner diagnostics it appears to be ok

But if you hold 2000 rom does it start switching again or does it stay at those numbers. Near 1 volt the o2 sensor is seeing a rich condition; OK if you just romped on it; but it should stabalise and start going from .2ish to.9 ish. (do not be concerned about where it it in this range.0-.9 as long as it goes from low to high. IF it stayes at one range or another then IS the O2 sensor sticking or is it actually reading the resulth of the post combustion burn and is showing you a clue to the real problem. No one said auto diagnostics was easy.

Reply to
Stephen H

I feel a little bit stupid for not being able to tell black from blue smoke however, It now appears that there is a mechanical problem. My cousin (a BMW tech) has suggested oil control rings may be a problem after looking at the car and the results of my efforts.

Symtoms: Idles roughly Starts with blue smoke Accelerates with blue smoke Some smoke when idling.

Sensors tested: Air flow sensor - OK Crank position - OK Temperature sensor - swings between 10K and infinity after 5 mins - not OK but will leave it for now Fuel pump, injectors, pressure valve- tested as suggested by Haynes and my cousin -OK Plugs - clean after 20 mile of freeway driving black after city driving idling Leak down - 1-10%; 2-6;3-10;4-20 (leaking through crankcase and cylinder head)

The initial cylinder pressure readings plus my lack of observational skills put me off initially, however, I will do a re-ring and re-valve stem seal with the engine in the vehicle.

I would like to know how the sellers mechanic made it pass Aircare (they tried twice, the second time it passed with the NOX readings raised from 84 ppm to 517 and the idle HC lowered from 2000 ppm to 16) and whether a recent timing belt replacement had anything to do with it.

Thank-you for your help. I'll keep the group posted.

Reply to
redryderridesagain

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.