siliconvalley

Hi, today when I got off highway, on city streets, the temperature needle raised close to the red. When the car was moving, it started coming down. When I stopped the car and checked the fan, it wasn't running.

I jumped the 2 wires out of the temperature sensor on the thermostat housing, the fan ran.

Does it mean I have a bad coolant temperature sensor? Or the head gasket is blown that caused the high temperature?

Thanks.

----------------------
2004 EX Coupe auto transmission

Poopies

iTrader: (4)

How's your coolant level looking? You should try hooking up an OBDII sensor to your vehicle; it can detect the coolant temperature if the sensor works. You probably would have thrown a if it's broken though.

ezone

Need more info. Was the (fan) temp sensor itself actually hot enough to run the fans?
Low coolant level can hinder fan operation and cause the engine to run hot.
And the fan temp sensor is not located in the engine dump stream (top hose), it's located in the radiator lower hose stream. That means the liquid exiting the radiator would have to be very high in order to run the fans.

See if the radiator and cooling system is low on coolant.
See if the reservoir has barfed liquid out and made a mess.
Refill and burp the air out, see if that makes the overheat problem goes away for the moment.


I'll run with the standard 'blown head gasket' guess.

Sir, the radiator fan circuit has its own temp switch. This is not monitored by the PCM as far as setting any codes.
IMO The only data he could really make use of is the engine coolant temp (ECT), which should confirm the gauge on the dash probably isn't lying.

siliconvalley

Thanks Ezone and Poopies.

^^^^^^^^^^^^^^^^^^^^^^^

I touched both hoses and it seemed they were both hot to the touch. I thought thermostat was working.

^^^^^^^^^^^^^^^^^^^^^^^

I added some coolant before the trip so I think coolant level was OK.

^^^^^^^^^^^^^^^^^^^^^^^

Yes, it did barf a small amount of liquid.

^^^^^^^^^^^^^^^^^^^^^^^

I will do that and test it.

^^^^^^^^^^^^^^^^^^^^^^^

I heard of a "balloon" test for head gasket. I will read the forum and figure out what the "standard" test is. I thought of buying the combustion test kit.

Today, I plan to go to Autozone and get a new thermostat. Yeah I know it's bad to throw parts on it before figuring out what's really going on. But the wife is a typical blonde and she drives it when I am at work. I am sure when it happens she has no clue and the only way to stop her is a full blown engine
I planned to do tax today, hope it doesn't take long to put the new sensor on.

ezone

It likely is working, and burning your fingers isn't really an accurate test of temperature.

Several years ago, I could pull an ear of corn out of the pan of nearly boiling water without pain. What was the real actual temp of the water? (The one time I stuck a digital thermometer in the water it read 206*F.)
Checking BEFORE is not good enough.
Gotta KNOW for certain where it's at NOW that there is a problem.
Coolant gets pushed out slowly during the course of the trip. If the trip is long enough, quite a bit of coolant can be pushed out.


BZZZZZT! Game over! Thanks for playing...


(Unless the reservoir was overfilled to begin with, then all bets are off.)

It's low again, right?

If refilling keeps the temperature gauge in the right place, the thermostat is not the likely issue.

Note: There can be 'air' pockets trapped in the high spots of the system which can make the radiator APPEAR completely full as you loosen the cap, when in reality the system is low.

That's one method, but you have to block off the overflow tube and get the balloon to seal good, AND force the leakage to occur during the test. Sometimes these leaks don't show themselves until the engine is working hard.


I apply shop air line pressure to each cylinder and watch what the cooling system does. This test method has not failed me yet, but is not necessarily easy for the home user to do.





Check out this thread: Overheating common causes - reference

Most of these HG failures will easily pass that test. These are usually too small or slow to trigger a fail result.
Too unreliable for me to use in the shop, I jump directly to the test that proves it for me every time.

NOTE: This is a pattern failure on this engine. There might be any of a dozen ways a typical head gasket can fail, but there is one common way this particular engine does it. I test for that failure first because it is most likely to give me the results I am looking for.



I wouldn't bother putting a cheapo stat in at all. You may just cause more headaches when it fails far too soon.

Use a factory thermostat........ and it might be easier to change it while the head is off.

siliconvalley

ooops my bad. I meant to put a new temperature sensor. Actually, I got the idea of testing the sensor from you in another thread about jumping the wires. Thanks.

It looks like the sensor fails to turn the fan on. But I am not sure if that's the real cause of the high temperature. 3 likely causes:

• temperature sensor failure, fan not on

• air pocket in cooling system, need refill/flush it out

• head gasket leak

The 4th, I hope it didn't happen, is the wife baked it so hot the engine block warped beyond repair and I have to get a new engine.

ezone

Looks like.
LOOKS like.

But does it?

The liquid exiting the bottom of the radiator would have to be 203*F or higher before the fan is switched on. ANY breeze blowing through the radiator could keep the liquid cooled enough to not need the fan operating.


Jumping the wires at the temp sensor only verified everything EXCEPT the sensor can work (fans, relays, etc.). It did not test the sensor itself at all.


Remove it, connect an ohm meter to it, and drop it in boiling water?





The head would normally warp first, long before there is block damage.

siliconvalley

Hey Ezone, I tried to refill/flush the coolant first and it seems to fix the problem. I put the car on 2 jack stands and refilled it over 30 minutes with engine running and full heat. I saw some small bubbles at the radiator opening. When I increased the RPM and watched the bubbles, some bigger(~1/2" in diameter). Everytime when I increased RPM, big bubbles came out. After I repeated it about 10 times, it's still there so I gave up. I refilled coolant whenever I saw the level was below the opening. Maybe I have a leak head gasket since the bubbles never stopped.

After the refill, I drove it to highway and back to town. No overheating occurred.

Honda user manual says don't fill coolant at the radiator opening. I think it's a mistake? That's what I did, only fill it in the reservoir. I think the air in the radiator causes air pockets in the engine. From now on, I will fill coolant in radiator and reservoir.

Thanks for your help, Ezone.

ezone

I think you may have fixed the symptom, but not the cause.

Your overheat complaint was the result of low coolant level.....which is usually caused by a blown head gasket.

Sounds blown to me.

Give it time enough to displace a sufficient volume of liquid out of the radiator, it will run hot again.

WAT?!?

For a new car buyer in an ideal world, with a meticulously maintained car, that might work. It stops working as soon as the radiator is low.

The radiator MUST be filled first.

ezone

Make sure you check out this thread: Overheating common causes - reference

siliconvalley

Thanks for the link, Ezone.

I checked my user manual. Actually, it doesn't say we can't add coolant into the radiator. It's the piece of white paper under the hood labeled "Emission Control System" that says:

"CHECK OR ADD THE COOLANT AT THE RESERVE TANK, NOT THE RADIATOR"


It's likely my HG is blown because when the engine is in idle, I didn't see bubbles. But when I gave more gas, I always saw ~1/2" in diameter bubbles.

ezone

If you know how the system works, you can then decide when to ignore that label.