Hyundai electrical gremlin...I can't figure it out

[Jim G]

Greetings folks. This is my first post on this forum and I hope someone here can help me figure this out. My issue is with a 95' Hyundai Accent. This is my commuter car. It has 215k miles but still runs great. It's pretty gutless but I get 37 mpg and that works for me. Here is what is happening. The car was starting to run hot and I tracked that down to a bad cooling fan. Replaced the cooling fan with a good unit from the local pic-n-pull. So now I have a fan that works...but not always. In the a.m. when I leave for work and the ambient temps are cooler, everything works perfectly and the car does not overheat. My commute is about 35 min so the engine has plenty of time to get up to operating temp. When I get off work at the end of the day is when my problems begin. I live in NV and it has been getting into the mid 90's. Here is what I can't figure out. When I use the left turn signal the temp guage bounces in time with the blinker. The guage bounces from whatever the engine temp is to all the way hot and the indicator arrow does not light up. It also does this when turning on the emergency flashers. Doesn't matter if the car is running or not with the flashers. Also the cooling fan won't come on unless I turn on the A/C and then both fans run constantly. What is really odd is that the A/C won't blow cold when I am pushing on the gas pedal. When I'm sitting at a light with my foot on the brake or coasting down hill it will blow cold. Almost seems like the TPS has some sort of involvement with the A/C issue. As stated earlier everything works perfectly in the morning when I drive the car to work. A/C included. At the end of the day after the car has sat in the sun all day is when it all acts up. I have pulled the insturment cluster and can not see any obvious damage in the circuit. This car has two coolant temp sensors. One for the temp guage and one for the control module. I would think that if one of these were bad it would act up constantly. The cooling fan is controlled by the ECU and that coolant sensor tests out as good. The cooling fan works as it should in the a.m. Sorry for the long post. I'm totally frusterated and could use some suggestions.

[carsandcycles]

I believe your A/C issue is a vacuum line or reservoir problem (leaky lines or broken reservoir); at idle your vacuum is at max but when accelerating, vacuum goes down (and the reservoir is there to hold vacuum till your engine creates more).

[thcardoc]

Originally Posted by Jim G

Here is what is happening. The car was starting to run hot and I tracked that down to a bad cooling fan. Replaced the cooling fan with a good unit from the local pic-n-pull. So now I have a fan that works...but not always.

Some things like you are describing need to be observed first hand to be positive of exactly what the circumstances are. You have done a good job explaining the symptom that as a technician I would be looking to confirm. Cooling fans don't run 100% of the time, even with the AC on. They actually turn off based on vehicle speed since there is more air coming through the front of the car from it moving than the fans could ever pull.

In the a.m. when I leave for work and the ambient temps are cooler, everything works perfectly and the car does not overheat. My commute is about 35 min so the engine has plenty of time to get up to operating temp. When I get off work at the end of the day is when my problems begin.

Armed with this information I would be thinking an airflow issue. Things like the airdam under the radiator support, side seals along the ends of the condensor and radiator. How about damaged cooling fins, or a restriction of the air through the condensor by dirt, or leaves. Not knowing the area that you live very well I don't know if you would have any plants like the cottonwood poplar. The seeds from those trees float in the air with cottony fibers (hence the name) and do a great job of plugging up condensors and radiators.

I live in NV and it has been getting into the mid 90's. Here is what I can't figure out. When I use the left turn signal the temp guage bounces in time with the blinker. The guage bounces from whatever the engine temp is to all the way hot and the indicator arrow does not light up. It also does this when turning on the emergency flashers. Doesn't matter if the car is running or not with the flashers.

OK this part is easy and may have nothing at all to do with the possible overheating. You have a poor ground connection inside the instrument cluster that the temperature gage and the turn signal indicator share. At this momemt I have to question if the car is really overheating, or if the gage is simply innacurate because of the poor ground connection? We teach voltage drop testing techniques to pin point the location that needs cleaned/repaired.

Also the cooling fan won't come on unless I turn on the A/C and then both fans run constantly. What is really odd is that the A/C won't blow cold when I am pushing on the gas pedal. When I'm sitting at a light with my foot on the brake or coasting down hill it will blow cold. Almost seems like the TPS has some sort of involvement with the A/C issue.

What's bad about these is they don't have the scan data that allows you to see inputs to the PCM such as the AC request, and outputs from the PCM such as the AC and cooling fans command that most cars did back then. That forces the skills to step your way through this to have to be much more polished.

Poor airflow "COULD" (doesn't mean it is this time) be causing the high side pressure to be too high and the moment you increase the engine speed above a given point the clutch might be getting commanded off to protect the system.

As stated earlier everything works perfectly in the morning when I drive the car to work. A/C included. At the end of the day after the car has sat in the sun all day is when it all acts up.

Obviously the temperature differential is having an impact on how hard the system has to work in order to provide comfort. That's where observing this first hand would make a difference. Measuring what the low and high side AC pressures are, and by measuring the surface temperature of the AC components I would be able to tell exactly how the system is running and why it's not comfortable. The condensor inlet and outlet temperatures, combined with the high and low side pressure readings reveal a lot of information. The inlet temperature should be between 20f to 50f higher than the outlet.

I have pulled the insturment cluster and can not see any obvious damage in the circuit.

That's why you need to measure the voltages in the circuit during the problem. You might need to provide your own power to the indicator bulb, so that the groud circuit voltage drop is constant. Measure the ground circuit with one meter lead connected to the battery negative post and then starting at the indicator when it doesn't work at various points until you pinpoint the failure.

This car has two coolant temp sensors. One for the temp guage and one for the control module. I would think that if one of these were bad it would act up constantly. The cooling fan is controlled by the ECU and that coolant sensor tests out as good. The cooling fan works as it should in the a.m. Sorry for the long post. I'm totally frusterated and could use some suggestions.

Welcome to my world. Imagine seeing things like this day in and day out. With a good plan your's is easy, you should see some of the things that we have to figure out these days that really challenge us.

Until proven otherwise, treat the turnsignal/coolant gage issue as a totally seperate problem from the AC poor performance under higher temperatures. Instead of trying to solve both at once, pick one and see it all the way through, then with that answer and solution in hand see if there could be any correlation to the other. Right now we don't know if the engine is actually overheating or not. All we have is a reported symptom and it needs verified. Our routines would include scan data if it was available, and an infrared temperature probe to directly measure the engines temperature, plus other visual and sensible cues.

I must stress caution when it comes to doing anymore diagnostics underhood when suspecting airflow issues for a possible overheating condition. This older car would be subject to certain components failing under stress, and if this is genuinely overheating they are definately experiencing more stress than they might be able to control. A sudden failure of an AC hose, or radiator or hose or other cooling system component can cause serious injuries, so proper measures must be taken. (saftey glasses, face shield, apron, gloves, etc)

[Draken]

Hi, this is my first post here, the reason is I was looking for information about my own problem before taking the car to a garage.

My everyday car is a faithful 1995 Hyundai Accent, and I could write down exactly the same issues posted by Jim G about the turn signal lamp / crazy gauge needle. Also with the left turn signal. It started five weeks ago, it means starting summer in the southern hemisphere, with about 87 degrees.
As mentioned, everything seems to be fine in the morning going to my work, but at the end of the day the problem starts. Just read the original post and you have the exact picture -identical.

Except for one thing: I don't have AC unit in this car.

Another thing to be noted is the left indicator lamp in the instruments panel is slightly but permanently illuminated but only when the temp gauge is showing the false "overheating". This is hard to see in daylight, but at night you can note the light.

Maybe an electrical shortcut that dissapears at lower temperatures? Two pieces making contact bacause of dilatation?

Any ideas?

Many thanks in advance.