97 Civic starts then dies when engine is warm and is parked under the sun

[squall14]

Hi, I drive a 1997 Civic AT. The problem is that after parking the car under the sun and go get something in a store, when I try to start the car, the engine starts then dies. After this the car just cranks and does not start anymore. Car starts fine when engine is not warm (points at C or parked overnight). In order to prevent this, I do the following:
1. I press the gas pedal when cranking up the car. ( It starts somehow)
2. When starting the car, when it feels that it is about to die, I press the gas pedal that prevents it from dying.

Just cleaned the throttle body and replaced the condenser fan but problem is still there. Air Filter is also clean

 

[CO671]

Any CEL? When you put the key into the ignition, do you wait for the CEL to click before cranking (to let fuel pump prime)

 

[squall14]

Check engine light lights up then turns off before I try to start the engine. Problem is very intermittent. Only happened twice this month and when the sun is up

 

[CO671]

Alrighty, because the first thing that comes to my mind is the main relay. Makes sense because the heat can affect an already bad solder point. It's a very common problem actually.

Read more into this, it sounds like your problem
http://honda-tech.com/showthread.php?t=1320931

 

[squall14]

Is it possible that the main relay is damaged even if I can hear the clink when the CEL disappears?

I don't know why this happens more often now after performing a Periodic Maintenance Service (change oil, throttle body cleaning etc). Before that, after starting in the heat, the engine feels that it is about to die but then starts.

 

[RonJ]

The main relay isn't a good fit because you can keep the engine running by pressing the throttle pedal.

Try cleaning the IACV and adjusting the idle speed as per service manual instructions.

 

[squall14]

Idle speed is already adjusted to 1k rpm. I just noticed today that problem also occurrs even if engine is pointing at C if parked under the sun. Will try to clean the IACV as instructed.

 

[squall14]

Brought it to a honda service center. Main relay is the cause of the problem. Unfortunately, they dont have the part.

 

[RonJ]

Main relay is the cause of the problem.

I'm doubtful. But good luck. Main relays can be resoldered.

 

[Killa_CiViC]

Check the idle air control valve on the throttle body... when they get older and carbon builds up on them and they get stuck in one position and can make the car hard to start. Sometimes cold, sometimes hot.

If there is carbon build up, you can try cleaning it with a good carbon removing spray or throttle plate cleaner, but usually by the time they're at the point of getting stuck, the motor is already weak and probably not moving well anyways and should just be replaced.

A bad main relay would cause the fuel pump not to come on and the car wouldn't start at all, no matter how much you feather the throttle, and although 4th gen main relays seem to last about 40k miles, it's not a common problem on 6th gen civics.

 

[squall14]

Iacv and throttle body are already cleaned. Engine still starts then dies after a 30 min - 1 hr drive when you try to start it again after parking. I really dont have problems starting my car in the morning or night when engine is not warm

 

[squall14]

Also sometimes when putting the key in the ignition and turning it once without even starting, i hear something running like a fan. Does anybody know what it is? However this rarely happens.

 

[Killa_CiViC]

Also sometimes when putting the key in the ignition and turning it once without even starting, i hear something running like a fan. Does anybody know what it is? However this rarely happens.

When you turn the key to run.... the position just before start... the fuel pump turns on for 2 seconds to pressure up the fuel lines. If the sound is coming from the rear of your car, that's probably what you're hearing.

I would still try a different/new IAC valve if I were you.

 

[squall14]

Replaced my main relay. 1 month of testing and it seems that problem does not occur anymore