lilrob1213

...unless it is, and I'm just dumb. Alright guys, I'm finally coming to the forum to post specifically about my car because I've been reading the forums and trying to figure this one out for 3 weeks now. I've gone through just about every thread of every civic forum about coolant issues with no sign of anyone else having this problem and it not being fixed by the suggestions that I've already tried. Let me get more detailed, but I warn you, this will be VERY long.

The car is a 2003 Civic EX coupe 1.7 liter vtec engine with 164,000 miles (50,000 miles put on since I've owned the car a year and a half mostly highway). I went on a mission trip to Nicaragua for 6 days, came back and was about to drive from Memphis to Nashville, checked the car's oil, brake fluid, and looked at the overflow tank quickly just to see if there was fluid and there was. It was a little higher than normal, but I didn't think anything of it, and I didn't even bother looking at the coolant level in the actual radiator cuz I figured the overflow had plenty. So we took off and got about 10 miles from the house on the highway and I noticed the temp gauge rising and by the time I noticed it was about 3/4 of the way to hot. I pulled over, checked that turning the A/C on turned the fans on and it did, I looked at the overflow tank/reservoir for the coolant and it was about to overflow! I let it cool down a bit, then drove it home with temp dialed to hot, blowers full blast and the temp gauge rose and fell throughout the drive home all on the highway. The next day when we got back, I checked the radiator and I had to add about a half a quart of fluid before the level was normal, but the overflow tank still looked really high. I saw no external leaks. So, I looked around the forums and thought maybe the fluid's not flowing back into the radiator because of a bad cap, maybe that's why the radiator was low and caused an overheat. So, replaced the cap. Drove to work, which is about an hour long drive (42 miles) mostly highway. When I got to work I noticed the overflow was still high, and after work I figured it would have been sucked back into the radiator, but it wasn't. By the way, every day I had to add fluid to keep the radiator level where it was supposed to be so I didn't overheat, but I wasn't adding coolant, I added just regular tap water, and now after lots more searching I found that tap water is bad, I'm supposed to add distilled water or coolant. So, after the new cap didn't work, I looked around more on forums and found that air in the system could cause similar symptoms, so I purged the air out, running the car on jacks in front with cap off adding water til the fans kicked on twice and the adding water in the morning after it cooled again to top it off. The bubbles never really stopped though. Seemed to have worked for about a day. Then back to same problem, overflow fills up and spills out water, dropping the radiator fluid level and not sucking back into the radiator after cooling down. So I thought well maybe it's the thermostat, replaced that, purged air out of the system, no joy. So I thought maybe it's a headgasket. I went and got a block tester, did the test for about 10 minutes from cold engine to somewhat warm and the fluid was kind of clear from starting out blue, but didn't turn yellow, so poured that out and did the test again 2 minutes like the directions say and it stayed blue, so I don't believe I have a bad headgasket. So my next idea was maybe there's a clog in the radiator that is causing pressure to build up and it holds that pressure in the radiator not allowing the coolant to flow back into it from the overflow after it cools back down because I did notice that sometimes when I open the cap, I hear psssh like air is either coming out of the radiator or being sucked in, but I don't know which one, and sometimes since the overflow tank is so full, when I open the cap to the radiator water will spew out of the overflow tank?! I don't know why this happens! I have video of that if anyone wants to see, it's kinda neat actually. So, anyway, I got some prestone radiator flush cleaner, drained my coolant (which looked good, no debris or oil), filled with distilled water and put the cleaner/flush fluid in. It drove fine for two days, I drained it today, put distilled water in, ran it to operating temp and turned the heater on, etc. Let the engine cool down, drained it and the oil because I was doing an oil change. The oil looked normal, not milky or anything. The coolant was clear, with some bubbles from the cleaner but no debris. Filled the radiator back up with distilled water and I was going to wait to drain and refill tomorrow another time with distilled water to make sure I get the cleaner all out. I purged the air out and made sure the system was full. Then, on my way home from work today, the car overheated! Now it hasn't overheated AT ALL since the very first time when I started looking into this problem because I was keeping the fluid levels normal adding fluid every day, but today it decided to overheat?! Even less driving than I have done sometimes without topping up the system with water (I've gone two whole days before and was fine, no overheats). So, my thought was maybe the coolant system REALLY doesn't like running on straight water (but then why was it fine with more driving than todays over the last two days with straight water??). I'm pretty much down to maybe it's the water pump failed or something, but I really want some input to see if maybe I missed something in diagnosing it or maybe it just is that I need coolant not water, which I am going to flush the system again, make sure it's not still bubbly like this last time was and then add honda type 2 coolant to see what happens. But I want to know if there's something I need to replace or fix like radiator or water pump before I put that coolant in there and have to spend another 20 or 40 on coolant Also, I still have about 26,000 miles before my timing belt, water pump, etc. is due on the car.

I forgot to mention that before my mission trip to Nicaragua, actually about 2 or 3 months ago, I had the coolant system flushed by a Honda dealership, and it has run fine since then even driving 6 to 10 hours at a time, up until this overheat happened after letting the car sit for a week. And I'm pretty sure I had green coolant before and now it's the blue stuff... I like the green better since I associate it with never having overheat problems when this blue stuff has lol. Oh, and I tried to do a pressure test on the system, but the tester I got didn't have the right adapter for my radiator (it's the stock radiator as far as I know, it says denso on it?) and I couldn't find that adapter anywhere, so if you know where I could get a loaner tool cooling system pressure tester that fits my car, that'd be great!

ezone

By symptom descriptions: Head gasket.
Just gotta prove it now.

lilrob1213

Not to be rude, but I thought I'd already disproved that by the block test? What other test do you suggest? If I could know where to get ahold of a leak down gauge I would try that one .

Zed

When all is said and done, it's going to be the headgasket.

pjb3

The test for combustion products in the coolant and the leak down test are not always accurate. Do a search for Ezone's procedure to test the head gasket using compressed air in the cylinders and watching for bubbles or rising level in the radiator.

lilrob1213

Alright, will do thanks

arokz89

I work in a automoitve repair shop, we work on all vehicle's makes & models,but honda is known to blow head gaskets. if ur car is overheating with no leaks more then likely it will be the cylinder head gasket, block test aren't always accurate especially on hondas.

ezone

From this thread:
https://www.civicforums.com/forums/3...reference.html

lilrob1213

Alright here's an update, did a compression test (cold engine) got 89, 80, 85, 87 so not too far off from each other and I found a differential cylinder pressure tester we use for radials and used that to pressurize my cylinders at tdc and saw no bubbles come through my radiator at all! Woohoo! So I don't think it's a bad head gasket.

ezone

How much pressure were you putting into the cylinders? It needs to be as high as possible in order to push past any head gasket breach.

These typically leak extremely slowly in the beginning, sometimes it can take several hundred miles before the leak displaces enough liquid to cause any noticeable issues.

lilrob1213

The regulator I had available only went up to 100psi so I put that into the cylinders.

Today, I drained the radiator and the reservoir, refilled the radiator and ran the car til it warmed up and fans kicked on twice refilling as it went and then let it cool for 4 hours with the cap off and topped it off coming back several times to top off more since the fluid kept going down (but not leaking out, just finding it's place within the radiator) I did that until fluid stayed topped off and I filled the overflow to the minimum line. This is all with distilled water BTW, and then I drove home and when I got home, the overflow tank was full to about an inch below the cap already! It hasn't cooled enough for me to check the radiator to see if it was just extra that was pushed out (which seems like a lot) but idk what's going on anymore.

My other thought was maybe it's a water pump failure that one of the blades broke off and is partially blocking the flow?

ezone

That's quite low.
I'd find a way to connect the shop air line directly into the cylinders.


That alone screams head gasket to me. There should not be that much change in level from cold to hot.


When the engine cools back down check the reservoir level first. It probably will stay way too high.


If you don't want to believe it's a head gasket blown, maybe check out if there is a problem in the vacuum side, so when the radiator is under vacuum (when it cools down) it can't pull liquid out of the reservoir.



Forgot to mention, this is probably a bad gauge or something. Maybe a tire Schrader got stuck in the hose instead of the correct Schrader valve for a compression tester.
I fully expect 180 PSI or higher.

lilrob1213

What pressure would you recommend? I will see what I can do, but unfortunately its gonna be hard to find time to test this the rest of this week

I wouldn't doubt it was a bad gauge cuz the last one I borrowed showed 0 psi all four cylinders! And before that the one showed 42 psi... Just no good luck with loaner tools for this stuff.

ezone

Everything you can get. Read post #8.

lilrob1213

So, you do this on a hot engine? How long do you let it warm up for? Isn't that dangerous for stripping out the spark plugs out of the head? Granted I have had them out a few times recently and put anti-seize when reinstalling them, but still I've heard that's not good, also why I did my compression test on a cold engine. If I'm wrong and those are myths please let me know. That might make a difference I suppose, if the headgasket is only leaking at operating temperature. So 150-170psi... I'm not sure what our shop air puts out, but I don't think it's that high. If it is... is there a safe way to turn up the pressure slowly in case I'm not exactly at TDC compression stroke when I add the full blown pressure to the cylinder? Would that cause harm?

Oh and this morning when I checked my fluids, the reservoir was about 2 inches lower than it was yesterday, but still really high, and the radiator needed quite a bit of water to top off before leaving for work.

ezone

Absolutely.
Operating temp.
Customer just drove in off the street.
Customer just drove in from Akron at 95 MPH.
Whatever.
I have no fear of yanking plugs on a HOT engine. (I also do not deal with GMs on any regular basis LOL, I'm talking about strictly Hondas.)
I WANT the engine to be hot at the start of the test.

I can have the first round of checks done in under 15 minutes.
I also do all standard compression testing on a hot engine.....

If there are problems with the way the engine starts and runs when it's stone cold, then I might do a compression test cold.

If I were to notice the symptom (bubbles in the cooling system) when the engine were cold, I would do my testing under those conditions.

However, I never see them do this, they are always running and hot while they leak. Therefore I do my testing as close as I can to those conditions.

Leakage may or may not be more prevalent under one condition or the other, but I KNOW they (usually) leak when the engine is hot so I do my tests that way.
I start with the symptoms I see, then try to force it to happen under a somewhat controlled condition.


OTOH
If an engine were to ..... fill a cylinder with antifreeze while it sits overnight, I would want to conduct a set of tests that would force the issue, under the same conditions the problem occurs.
The owner would get home from work, drive it into their garage and park, slam the door and go in the house. Next morning it's got a problem. I would try to conduct my tests using the same routine.
Make sense?

(This would be completely different from how the D17 head gaskets routinely fail.)


Get as high as you can with whatever you have to work with. Plumb it direct.

My compression tester hoses have the same quick connect fittings as our shop air hoses, so this is easy for me.
Remove the check valve and plug it right onto the air hose.
If it's not exactly at TDC, it won't matter if you apply pressure quickly or slowly.
It WILL push the piston down---- and you start over.


I use the crank pulley TDC timing marks for cyls 1 and 4, that's always safe.
I drop a screwdriver in the plug hole to watch the piston come to the top and rock the crank to find TDC on cyls 2 and 3.




This is the symptoms of the blown head gasket, breached between combustion chamber and water jacket.. Slow combustion leakage slowly displaces liquid as you drive, and it can't pull liquid out of the reservoir as it cools because of the air space in the radiator.
(Combustion leakage is usually so slow, the block check testers can't find it.)

When it all works as intended, the radiator stays completely full at all times and the reservoir always returns to the same level when cold.


HTH

lilrob1213

Do you have your own shop or something? I really appreciate all your help! I will try to do this soon, but if I don't reply over the next few days it's because I'm expecting to be pretty busy, but I'll get back with the results as soon as I have them.
I used the screwdriver method for all of them last time but it was only 100 psi. Didn't think about looking for timing marks lol

ezone

Or something. LOL

ezone

Bad grammar. Sorry.


Let me add to this partial thought:

A head gasket can fail in many different ways, there are probably a dozen or more ways one can fail.

The head gasket in this particular engine consistently fails in only one of those ways, and it's not the way most people would expect or have experience with.
Every clue most people look for will never be found in this engine until the problem has become quite extreme.

The test I described has not failed me yet.

Of course, by the time I get to the point of doing the test I already know what to expect because I already saw the usual symptoms -- that led me to the testing.
My tests are usually done to prove (or disprove) the problem and pinpoint the exact cylinder(s) that are leaking, so I know where to look once the head is off the engine.


HTH

lilrob1213

Oh kool! Well I hope that it disproves it but then idk what to do then lol but thanks so much for all the help!

ezone

You should hope it DOES prove it, at least that gives you a solid explanation for what's been going on.

Otherwise you may spend a lot more time and money chasing your tail.

lilrob1213

That's true, I am getting worn out trying to chase my tail lol
So if it does turn out to be the head gasket, which it is looking like it will be, then where is a good place to buy cheap OEM parts? And what should I look for in a machine shop to resurface the head? What to search for?

ezone

There are many Honda dealers selling parts online at wholesale prices.

This is one of the first I found long ago, and I constantly link their catalog pictures in posts here. Majestic Honda in Rhode Island:
http://www.hondaautomotiveparts.com/



Machinist needs to be able measure for flatness first.
If it is ok then don't touch it.
If it needs machined, he needs to be able to leave a nearly glass smooth finish on the aluminum head, same as the original (you will see when you get it apart). 15-30Ra if I remember correctly.
You should not be able to feel grooves in the machined finish.




There aren't any machine shops left in my town that can do a decent job on this, I was getting heads back that looked like they were hit with a belt sander.

I have been sending heads to a reputable machinist in a town 50 miles away JUST so I can get a GOOD job done, so I can produce a repair I can have confidence in.

lilrob1213

Wow, that's a a great thing you do to ensure that people have a reliable car! Glad to see there are good mechanics out there haha

Well I did the test from your other post, 150 psi is all I could get and did it on a hot engine, got no air bubbles... Didn't hear leaking at all... Idk what to do at this point?

pjb3

I would be looking for a rising coolant level in the radiator rather than air bubbles. As the air enters the cooling system due to the head gasket it will displace fluid and cause the level in the radiator to rise but the bubbles will be at the highest point which may be in the cylinder head and not the radiator.

lilrob1213

hmm... darn... I had made something similar to the lisle spill-free funnel out of an old radiator cap and a funnel, and the cap had a hole drilled through it so I could see air bubbles, and also use this to bleed the coolant system, but I didn't see the level change any... could have been restricted by the home-made funnel though? hmm...

ezone

^This^


It could take a significant volume of leakage (and a corresponding amount of liquid displacement) before any air actually makes its way to the radiator.
IF your apparatus leaves the overflow nipple and hose uncovered and exposed to liquid, any liquid level change might be unseen if escapes through there.

Just thinking out loud here.

lilrob1213

Well... I did actually notice a lot more dripping underneath the car than normal now that I think about it... could the pressure have been sent through the overflow and come out that way if the pressure of the water in the funnel was greater? I wasn't sure exactly where that fluid was coming from but it very well could have been the overflow.

The problem is, I can no longer replicate this test because I no longer have access to that compressor or any compressed air...

ezone

Has anyone suggested a head gasket yet?

lilrob1213

I wish there was a like button... lol ok, going to have to take a look at the finances and get this fixed, asap! What do you think as far as doing the timing belt at the same time? Still got about 30k miles left til it's due, but figure why not?