Honda Civic Forum - 04 Civic LX pushing coolant out of the overflow tank

Hi there,

I have an 04 civic that was pushing coolant up out of the overflow tank a while back and after testing it appeared that a failing head gasket was probably pushing combustion gasses into the coolant (this car has 255,000 miles on what appeared to be OEM equipment) so I gathered the parts (gaskets, water pump, Timing belt etc.) and went to work on it and after taking the engine down to the block and having the head resurfaced I got it back together and it ran fine for about a month until today.....

Now there are a couple of items I'd like to touch on as this was my first Honda head gasket and I had one particular thing that I wanted to address in relation to the installation. When I placed the head back on the block and started the torqueing of the head bolts I had a misread on my torque wrench (I had a brand new 10-150 lb. wrench) and when I torqued 1 and 2 it seemed as though I was going past where I thought 13 lbs. should click.

I loosened them both immediately, reset the wrench and started again and this time the process was flawless from start to finish. All three sequences went smooth and there was no popping at all. The bolt holes had been blown clean and were dry, bolts were lightly oil coated and everything went as planned. If I had to guess, I would say I took them to about 25-30 lbs. before backing them off. And used Hondas torque specs and sequence for removal and installation.

I called my cousin who owns a shop a couple of towns over and told him what happened and he said he didn't think it would matter as the bolts weren't close to being max torque and they were not under pressure for more than a minute or three before I figured it out. So I went ahead and put the car back together and filled the radiator. When I was adding coolant there were a lot of bubbles and it bugged me but after filling it to where the coolant stayed in the neck of the radiator there were very few and they got smaller and almost stopped except random ones that came up every now and then.

I stood there for about 10 minutes before I felt they were not a concern and put the radiator cap on. Checked the overflow and added about a cup of coolant there to bring it to just above the lower line. I had replaced the radiator a couple of weeks before due to a hairline crack along the top. The car was then on for the next 3 hours, driving in town, quick run on the freeway and parked, idling. Ran fine, idled fine.

I live in Yakima Washington and the temp here has been between 85-100+ degrees over the past month. My daily commute is about 70 miles round trip and it is mostly highway miles, 60 mph is my meat. And the car has ran absolutely magnificently since I got it back on the road. I started out just easing on to the freeway and not pushing it at all until I had ran it a few days and a couple hundred miles then I started testing it for stability by pressing it some.

Ran it on numerous 100+ degree days with AC blasting and it has performed as expected. I had even taken this car over White Pass, one of the steepest passes in Washington State. Good test I thought to prove it would stand up to the rigors of high RPM’s and pressures. My daily commute is mostly flat across the valley bottom.

The second item I wanted to address comes in right here. When I went from Yakima to my destination 90 miles away, climbing the pass on the west side the car ran fine at 60+ mph. The temp gauge never moved an inch. Power was great, good pickup and climbed a heavy grade (about 18 miles) no problem. However, on the way down the east side, the temp gauge started going DOWN. It just dropped off until it appeared the car was stone cold. The car did not get the message though and still drove great. Because I pulled over and checked the coolant level and observed the running motor. Sounded like it always does.

Once we got on our way and the car was on back on level ground at the bottom of the pass the gauge came slowly back up to normal and when I came back about 10:30pm that night, the car ran fine and the gauge never wiggled up or down on the same pass.

I checked it out online and planned on replacing the ECR? Temperature sensor. I think that is right, maybe not. Anyways, I had not replaced it but after driving to work and back every day for the next week without an issue with the car or the gauge in 100+ temps I forgot about it and did not swap it out. I have also been checking the coolant level in the overflow every day and it never dropped or filled past what I would consider normal. And on the day last week and again today, the temp in Yakima was in the 80s when i have pushed the car up a hill.

Today I was going to Seattle from Yakima over I-90, another mountain pass over the Cascades in the center of the state. I had left town and was headed out towards Ellensburg through a series of hills on I-82 that have pretty steep grades and was doing about 70 mph. after about 30 miles I was climbing the last hill before dropping in to the valley where Ellensburg is located and the temp gauge started to rise. It never got all the way up but it was maybe 3/4 when I pulled off into a rest stop at the top of the hill and checked it.

The coolant had climbed up into the overflow, sprayed out the top past the cap, hit the battery box, splashed onto the fans and then dripped down. This is the same symptom it displayed before I did the head so I am now wondering if I might have a bigger issue. I have not replaced the temp sensor or the radiator cap yet but I will tomorrow. I also replaced the thermostat a while back when I first addressed this but again, now I am not sure if I got the right one.

So, after letting the car cool enough to put in coolant I turned around and drove back to Yakima keeping the speed at about 65 and it started to heat back up as we got to town but when I put the heater temp on hot it stayed under ¾ until I got back to my ex’s house. I ended up taking my ex-wife’s 99 Accord to Seattle to look at a Highlander she was interested in, we got back about 9pm and then I did the dance. Started the Civic, topped off the radiator, put the cap on and drove home. 10 miles on city streets and the car runs fine, didn’t see the gauge move and got back to my house.

So I am looking to see if anyone has had this particular problem and if so, could I impose on them for wisdom? Sorry this was so long but I wanted to try and paint the picture because it seems to be an odd combination of events that are related in some way. I wonder if the sensor is faulty which seems likely but would it cause the coolant to push out?

Or, is it possible there is air in the engine that I did not get out when I filled the radiator and the heavy load up the hills caused the problem? And if so, what is the best way to bleed a 2004 Honda? I’ve read that there is a bleeder valve on some but I can’t find one on mine. It’s a 1.7 motor. I saw one video where the guy loosened the temp sensor and bled it that way. Might try it, looked logical.

Then there is the question “should I bite the bullet and replace the motor?” If I had to redo the head it would be another few hundred doing it myself and I now wonder if there could be a crack somewhere, pinhole leak, the list goes on. And on that subject, does anyone have experience with purchasing engines online? I have seen 1200.00 with a 5 year warrant and it seems reasonable if I can then take my car wherever I want. I like the car but this is getting tedious.

Thanks for any help or advice.

Any questions please let me know.