What I'd suggest is disconnecting the battery for a while to reset the engine computer. Then recheck to see if it's still showing open when it's up to normal operating temperature. Also check to see how quickly it gets up to temperature. The engine computer expects the temps to be high enough for closed loop control within a certain time frame. If it's taking too long to reach the correct temp perhaps that's what's causing it to stay in open loop...I've never seen this situation before so just guessing here.

I suspect the lack of power may be an separate issue and it might log a code to give you a clue about that once you get it into closed loop.

Regards.
Scottie.

 

Thanks for your ideas Scott-
I have taken your suggestion and disconnected the battery. I'll let it rest for ~ 2-hours or so. If this doesn't change things I'll let it dissipate overnight and see what happens. I also favor your idea that the open loop situation and the lack of power are separate issues and I'll chase down the fuel pump tests to see what I learn from that.

You suggested the "computer expects the temps to be high enough for closed loop control within a certain time frame", that "time frame" would be useful information to know, is there a specification on that printed anywhere? I don't see mention of it in the service manual.

Best regards,
Tom C

 

Hi Tom.

You suggested the "computer expects the temps to be high enough for closed loop control within a certain time frame", that "time frame" would be useful information to know, is there a specification on that printed anywhere? I don't see mention of it in the service manual.

There isn't a set time as such. The time would depend on what the coolant temp was at engine start. Obviously the colder the engine is when started, the longer it would take the coolant temperature to reach the target value for closed loop control.

I was asked to check a car (non Hyundai) a while back that had the engine check light on. I can't remember the code that was logged but the description was something like "excessive time lapsed for closed loop control". Anyway, it turned out the problem was a stuck open thermostat and it was taking a long time for the coolant heat up enough for the fuel system to enter closed loop. I can't remember now if that car ever did enter closed loop even after the coolant eventually did heat up. I just thought your fault sounded kinda similar to that one and that's why I suggested you check how quickly the coolant heats up.

Let us know how you get on with the ECU reset.

Scottie.

 

The link below isn't directly related to your engine but I thought it might be of interest :

Hyundai DTC P0125 - Insufficient Coolant Temperature for Closed Loop Fuel Control

 

I have encountered this problem twice on friends cars. Both times a new thermostat fixed it. They were stuck open. The codes led directly to this solution.

 

Compression readings are very low.
An engine that is only a few years old would have a compression reading between 180 to 210
Compression readings in the high 120's or low 130's means poor fuel economy and no power.
It can be due to carbon deposits on the valves or worn out rings on the pistons or even ....
Over heats like those that can crack exhaust manifolds can also ruin the heads and/or gaskets which also can cause low compression readings.
This is only some information. Don't go by me.

It seems that you want to keep this vehicle and already spent some time and money trying to fix this.
Let us know how it goes ... update us, please
Wishing you good luck.

 

test your thermostat

 

Hey Folks-
Thanks for all of your interest in my problems!!

avisitor, I originally thought that compression was a bit low but I don't think even in tip-top condition this engine could ever produce 180-210psi, the service manual doesn't provide specification for this but they do tell me the engine has a 10:1 compression ratio, so I am guessing that ~145psi is the best it could ever do. However I will retry the compression test soon and report what I find.


I have tried Autospark's advice and upon a cold start, the system started up in closed loop but within 1-minute it dropped off to open loop control again. The engine temp reach 200 deg F operating temperature in about 13 minutes.
Hopefully my screen shots tell the story in sequence. It seemed to reach final temp at 200 deg F during idle in a reasonable amount of time, so it does not indicate to me that the thermostat is at fault.

No trouble codes showed up.

I am thinking now that my first next step is the test or replace the ECT sensor which the original P0115 code may have indicated. Then move on to a fuel system diagnosis. What do you all think about that idea?

 

Thermostate and new gasket is a cheap fix. I would try this then move to the wire testing between the cts and computer then the sensor. No one tests a thermostat these days. If you open you must put a new gasket on anyway so put in an OEM thermostat. Do not use aftermarket on this part!

Google testing thermostat if you must...lol

 

This is bazaar. It goes into closed loop very early (probably programmed to do so) then goes out as the temp rises. I don't think it has anything to do with the temp sensor because it looks like it's pretty spot on with the 200 deg reading. It's not going into limp mode because you can see things like timing changes with revs etc the CEL and a code would register. My questions are 1) what scantool are you using and 2) does that car go into closed loop at idle. I know from past experience some cars don't to improve idle quality. Increase the revs and see if it goes into closed loop. At 2001it may well be the case.
Getting back to your timing belt. Did you replace it as it was or go through the timing procedure to ensure it was correct?

 

The scans you've uploaded make me suspicious of your upstream O2 sensor. It seems to be stuck lean despite of the short term trim being maxed out. Unfortunately your scan tool isn't able to read the O2 voltages once the system drops out of closed loop making diagnosis more difficult.

Perhaps we're all barking up the wrong tree chasing a problem with the coolant temperature when the fault is an O2 sensor temperature problem...just another guess but the data seems to point that direction. I'd suggest you check the O2 sensor heater circuits, voltage supply & ground (ground is a PWM control from the PCM). If access to both is easy it might be an idea to swap both sensors over to see if that helps.

O2 sensor connector pinout below. Terminal 3 is the heater 12V supply, 4 is the heater ground. Terminal 1 is the sensor signal & 2 is the sensor ground.

Scottie.

 

Bank 1 sensor 1 is a lame duck for some reason but my concern is there is nothing from either sensor when the temp has increased. I recognise the software and I have it so when I can get my hands on a OBD2 I will see what mine does.

 

Looks like bank1 sensor1 is a lame duck for some reason. My concern is neither sensor has output when the temp has come up yet sensor2 had output in the first frameshot. I looked at a Holden program and the timer for closed loop is a minimum time delay after starting. It's in seconds only to prevent the system from going into closed loop too quickly for some reason possible for idle stabisiation. I have the same scan software so when I can get my hands on a OBD2 car I will check it out.

 

OK I just checked a car with your software and it displayed the O2 sensors output voltages all the time in closed and open loop modes. I would check the wiring between the O2 sensors and the ECU and after that it may well be an internal ECU fault. Did Scotties suggested the reset do anything?

 

Hi Charles-
Thanks for your investigation and suggestions. The screen shot sequence that I posted was after the "Scottie reset". I also tried this a second time after an overnight reset with near identical results.

Looks like I'm tied up til Friday now so that is when I'll start probing and testing with my VOM, then if I can't figure anything definite I'll start changing parts. I have an O2 sensor that was originally in the junk yard exhaust manifold that I mentioned in my original post so maybe I'll just throw that in first and see what I see. THEN maybe I'll get to the fuel system side troubleshooting.

As to your question on the timing belt change, I lined up all the timing marks and checked/double checked the best I could. I am quite sure it is all good.

Thanks again.\ and best regards
Tom C

P.S, what are your opinions of the TouchScan program (I'm still in free trial mode)

 

I would measure the output voltage at the sensor first. If all is working ok it should be fluctuating all over the place (I am assuming it's not using a wideband). If it's putting out a steady voltage then the ECU is not responding for one reason or another. This will only be after it's warmed up but that shouldn't take much time. I had a customers car a Toyota Corolla many years ago where the woodruf key had sheared on the crank allowing the crank pulley to move altering the valve timing and it lost so much power as I drove around the block I was glad it was a downhill run back. I have no idea what readings it would have produced because it was back in the dark ages where we had no data output but it was food for thought. The customer complained it was loosing power and getting worse.
My software is called OBDwiz which is the keyboard driven version and I find it very good software. It came with a genuine Elmscan I bought some years ago now but it still can do a lot. I am playing with Torque at the moment and last night I downloaded the plugin for Hyundai and it stopped the program working so I had to remove it to get it back.
Good luck with the testing.

 

The best time to check the sensor output is while the system is operating in closed loop. You know what the sensor voltage should be during that period if the PCM is adding as much fuel to the mixture as it was in the earlier data list. I'd expect the voltage to be fairly constant and probably close to 1V (fully rich).

I'm not sure the voltage would fluctuate much in open loop either since I'd expect fuel delivery to be fairly constant when the engine is at idle. I can't say I've ever bothered to measure O2 voltage in open loop though, so maybe it does.

 

That's basically what I meant if its working it should be fluctuating all over the place but forgot to mention it's when in closed loop. Open loop they still put out a voltage but it moves slowly. The No1 sensor only has I think a max of 0.2 of a volt which is saying extremely lean. I would give the throttle a few jabs when in closed loop and see what happens.

 

After a 3-day ECU reset, started the beast up and saw that it quickly went into closed loop and both O2 sensors seemed to be printing out reasonable values for a good 5-6 minutes. Just as I began to think this problem had resolved itself (of course I know better than that!!) it went into open loop again. But this time I got a little bit lucky as it threw out the original P0115 code. I installed a new cheap Autozone ECT sensor and was back in business for about another 5-minutes (in closed loop!!) and then the upstream O2 sensor flatlined. Within a couple of minutes the other O2 sensor flatlined so out of desperation I replaced both with a couple of junkyard sensors that I took when I replaced the cracked manifold. Seems to be working fine, so now on to investigate the other half of my problem - no power at acceleration.

First thought was to take a fuel pump pressure test and see what's what, but this seemed to be impossible because there is no convenient place to connect a pressure tap. The service manual shows a special flanged adapter (p/n 09353-38000) that bolts in between the existing flange on the fuel rail and the flange on the fuel line. But no one on this earth (at least in NJ) knows where to get this adapter. I didn't want to cut into the hard molded fuel line to install a tee with hose clamps so I gave up this idea and went straight for the in-tank filter replacement.

Filter swap went pretty smoothly except the new filter seemed to be about 3/16" shorter than the original and I had to modify few features to make it work. In the end, it all fit together reasonably well.

BUT now there is a new problem - the pump isn't working. I am not getting any voltage at the fuel pump plug. The fuel relay clicks when the key is turned on and I swapped a couple of relays around to rule out the possibility of a bad relay. The 15 amp fuse is OK and no new trouble codes are being reported.

Any thoughts on what or where to look next ?
Thanks.
Tom C

 

BUT now there is a new problem - the pump isn't working....Any thoughts on what or where to look next ?
Thanks.

It's a long time since I've worked on a Santa Fe so I'm not sure if it has one but could you have perhaps bumped the inertia switch?

On some models there's a switch, usually on one of the strut towers in the engine bay, that opens the power supply to the fuel pump in the event of an accident. Make sure the button on the switch is pressed in or the pump wont work.

 

This is what I use...... for fuel rail.

http://www.toolaid.com/content/11mm-16mm-fuel-rail-pressure-test-adapter-kit

 

I am wondering if the fuel pump stops working after the 5 minutes. I have had cars running with no power to the fuel pump. They must have the pump to start but then after disconnecting power to the pump they continue running due to manifold vacuum sucking the fuel in. It happened to my own car when I was driving home. Mind you it had almost no power caused a major traffic jamb and took ages to get home.

 

Its a curios question Charles, I have been wondering what ever turns the pump off ? I don't see any signs of a fuel pressure sensor or a timer that turns the pump off. Until now I was assuming the pump runs constantly and whatever fuel isn't used just flows over the pressure regulator valve and back to tank.

Autospark - I checked that inertia (?) switch as you suggested and it seemed to be engaged. I pulled the connector off that switch and found there was no voltage reaching it. I pulled the fuel pump relay and jumpered the #1 socket (hot) to #2 socket and then was able to measure full battery voltage at the inertia switch connector (this means the relay isn't pulling in as I originally thought). Then I reconnected that connector back onto the inertia switch and checked voltage at the fuel pump connector - it was seeing full battery voltage with the relay jumper still in place. Reconnected the fuel pump connector and heard it running. I let it prime itself for about 1-minute and started the car up and she ran pretty good for a few minutes.
After that, each time I started it, it no longer stayed running for more than a few seconds.

Going back to the relay - when ignition key is turned on, battery voltage appears on relay socket #3 which is the input side of the relay coil but the circuit isn't being completed for some reason. I am at a loss to figure out why the relay isn't pulling in. Is there something else that controls the ground side of the relay coil?

Getting late, I guess I'll disconnect the battery and hope for another magical reset !!

 

The pump runs all the time the engine is running. It uses the signal from the crankshaft sensor to turn the pump on and off via a relay. Some cars prime the pump for a set time when you first turn the key on. Don't know if it's possible with your scantool setup to monitor injector duration. What I would look for is a big increase in duration when it goes out of closed to open loop. This would indicate a very lean mixture has occurred for some reason and the ECU is trying to compensate for it by holding the injectors open for a longer period of time. This duration would from past experience go from around 2 when all is well to over 10 milliseconds at idle if the engine is starving for fuel. The first engine I saw running without power to the pump was when I was teaching. Another teacher had disconnected the pump on a Ford to reduce the line pressure before disconnecting the hoses and it just kept running and perfectly normal. You could rev it up and it didn't even splutter that's when we got our brains together and worked out that the manifold vacuum was enough to suck enough fuel through to keep the engine running. Most cars won't do this they cut out when the fuel pressure drops. Your one is a bit of a brain teaser.
With your relay question it should turn on when you start cranking the engine. This is a safety feature to cut the fuel in the event of an accident and the engine stops running. As I said some cars prime the pump for around 20 seconds just after you turn the key on and unless the ECU gets the signal that the engine is rotating the signal to the relay to turn the pump on stops.
Time to give up on the magic reset. it's not going to occur now.

 

Is there something else that controls the ground side of the relay coil?

No and it's unlikely to be a problem with the crank sensor since you have spark and injector control. You'll need to check the continuity of the wire between the ECU & relay.

Don't know if it's possible with your scantool setup to monitor injector duration. What I would look for is a big increase in duration when it goes out of closed to open loop. This would indicate a very lean mixture has occurred for some reason and the ECU is trying to compensate for it by holding the injectors open for a longer period of time.

I assume that's another one of those posts where you type one thing but meant something different :laugh:
When the ECU is operating in open loop how would it know the air:fuel ratio had gone very lean? The ECU adjusts the injection duration (via the fuel trims) in response to the signals from the upstream O2 sensors but that only happens when the ECU is operating in closed loop control. The O2 sensors are ignored in open loop which is why the fuel trim values show zero.

And apart from that it's a pointless test, IMHO. Of course the mixture will be lean if the pump isn't running.

 

The FI duty cycle is always being reported. It will either be in response to the O2 when in closed loop or when in open loop the fixed tables in the ECU fuel maps will provide the number..

 

I assume when you say duty cycle you actually mean injector duration? And if so, who said injector duration wasn't always being reported?

 

OK monitor the injector duration in milliseconds and see if there is any change occurring and if so how much when it goes from closed to open loop without touching anything like throttle. I would like to know what happens. Boy this is getting complicated.