Rep Power: 0

Hey all, I need some advice here. Car is a '98 Civic LX, 4dr, automatic, 145k miles, all stock except for a K&N drop-in.


I was at a drive-thru about a month ago when I noticed that the coolant temp. gauge starting to rise. I pulled over and popped the hood to check everything. Nothing seemed wrong and the radiator overflow tank was full. Luckily, I was just down the street from home so I made the drive with the heater on full blast (which didn't seem to make any difference at the time) and got home perfectly fine with the temp. gauge never going past the middle.

I let the car cool off and then went out on a test drive again that night and it drove perfectly fine. The next day, I dropped some people off at the airport and when I was just getting off the freeway on the drive back, the gauge started to peg and I immediately pulled over. I saw steam coming out the front and after popping the hood it was obvious a big coolant leak occurred.

I basically managed to get it home after driving a block, cooling off, driving another block, cooling off. A lot of people told me that the thermostat might have been stuck so the next day I managed to get a ride to the store to buy a new one. During the t-stat replacement, I noticed that the related hoses I had to yank were bone dry on the inside. I replaced the t-stat and opened the radiator cap and noticed the radiator wasn't full. As I was topping it off, I saw a fresh leak coming from the heater hose that leads under the dizzy. I reached under it and felt a giant rip and this was the same hose that had ripped on me before at the end which I fixed by cutting an inch off the end. This probably explains why turning the heater on earlier had no effect as my hoses didn't even have coolant flowing through them.

So I basically replaced the entire hose with an OEM one, topped off the coolant and my car was perfectly fine for an entire month until yesterday.



While in my neighborhood, the temperature gauge started to rise again. I was thinking WTF is this **** again Fortunately, after turning the heater on full blast, I got a lot of heat and the needle started to fall and I made it home fine.

1. I checked for any ripped hoses or leaks and couldn't find anything.
2. Radiator overflow tank and radiator is full of coolant.
3. Engine oil level is good.
4. T-stat I highly doubt is bad as I replaced it a month ago with a Failsafe one.
5. Radiator fan seems to be fine; with the car idling for extended periods and A/C on max, the fan periodically turns on momentarily and temp. needle never goes past middle.


I drive around today and had zero issues but I don't want to ignore this as this seems to be playing out exactly the same as it did a month ago and I wouldn't be surprised if something bad happens within the coming days during the worst time.


Any ideas on the problem?



Thanks!

 
Rep Power: 517

You should fix the oil leak that caused that hose to swell and pop.

If the heater hose popped and you only trimmed it and stuck it back on, then it popped again......It's probably been overheated at least twice, likely severely.
How bad? Bone dry is bad. That means the head was probably dry too.


Air bubble trapped in the cooling system?
Since it's been overheated, I'd probably do my standard head gasket checks first.

Then go from there.
Does the radiator cap hold pressure?

 
Rep Power: 0

Damn you're good lol. Yes, there some extremely slow oil seepage coming from somewhere in that area.


Well when When the heater hose first started to leak a few months ago, it was relatively slow and at the very end. I decided to cut off the end because I couldn't find an OEM hose or one from Autozone, and there seemed to be enough slack for it.

Now when it happened again a month ago, I replaced the hose completely as I mentioned.

What is the best way to check for head gasket damage right now? I also haven't tested the radiator cap but I have taken a look at it while the car was hot and I don't feel/see anything like coolant or hot air seeping out of it when I place my hand on it (probably not the best idea or test, but just something I did).


I drove around today, about 10 miles or so on the freeway and then through stop and go city traffic here in Los Angeles with the AC on. Car drove perfectly fine .


Thanks

 
Rep Power: 517

Check the sticky on overheating common causes somewhere around here.

For a combustion leakage check, I stick a funnel-fill in the radiator, apply shop air line pressure (170 PSI) to each cylinder and watch for changes in the coolant level. (It's not quite that simple though LOL)

 
Rep Power: 0

So I just went downstairs to check on a few things. I checked the radiator and oil and didn't see any signs of mixing.

But with the car cooled down, I started it with the radiator cap off. Didn't see any bubbles or anything. However, When I gave the motor a few gentle revs by blipping the throttle body, coolant spews out from the radiator.

Does this mean 100% blown head gasket, end of story?

 
Rep Power: 517

It would have to be horribly bad before it did that.
There's a dozen or more different ways a head gasket can fail. Mixing fluids is only one.



I can't see what you see.

Depends on if what you saw was normal or not.
Meaning, just the action of revving the engine up can cause the coolant level to change, and so can vibration from the engine running.

You need to figure out if combustion is leaking past the head gasket into the cooling system.

The big problem with finding it is most of the time the leakage is so dang slow it's tough to see it or prove it using (traditional) methods that usually work on other engines.

That's why I use air pressure applied directly into each cylinder. It works very well for me and had been an extremely reliable test for this particular failure.



Strap a rubber glove or a punching balloon over the radiator neck and run the engine? (clamp off or plug the small hose to the overflow bottle)

 
Rep Power: 0

Thanks for the response. I don't have a lot of resources or know-how on hand to do the air compressor test, but is this rubber glove test a crude testing method that I can do right now to try and determine whether or not combustion pressure is leaking into my cooling system?

So take the radiator cap off and strap a rubber glove (Is nitrile ok? that's all I have now) over the neck and start the engine and see if the glove inflates? Without revving the engine?

I didn't see bubbles surfacing when the cap was off and car was idling, which I presume is a good sign, but I can still do this test right now if I have the method down right?

 
Rep Power: 517

Yep. I was trying to think of something simple.

DISCLAIMER: I really don't know how well this idea will work, I've never done it myself.

Pretty much.

Maybe. Unless you get no results.
You may need to rev it up.

If you have an automatic trans: You may need to power brake it 3 seconds at a time (to increase cylinder pressures). Be careful though, this can destroy a trans quickly. Don't do it if you don't know what you are doing.

Sure.
Be aware that the water will expand as it heats up so you may get liquid into your balloon.


Did you find the sticky on overheating?
Check out the video scooty made where he just put the overflow hose into a pan and watched it pump bubbles.

 
Rep Power: 0

I couldn't find the sticky yet but I'll look harder.

I will do this test now, but I have a feeling that the glove will just fill up with coolant. With the radiator cap off, the coolant is pretty much to the brim (but it always has been).

As to the video you are describing.....now would it be a better test if I take the overflow cap off with the hose below it and stick it into a small plastic bottle filled with water then observe whether or not air bubbles are coming out? I'm thinking this could be a better test?

Now it doesn't matter if my engine is hot or cold or anything for this test right?


Thank you

 
Rep Power: 517

Found it

https://www.civicforums.com/forums/3...reference.html

The system is designed to keep the radiator completely full to the top.


You might need to remove a little, lower the liquid level?

This is how it was done in the video.

Problem is, the pressure cap. You would have to have a lot of leakage to get sufficient pressure to overcome the 13-15 psi radiator cap.

Maybe if you remove the cap and still seal off that opening, then use the small reservoir hose in a pan of water?
IDK.
Like I said, I haven't done this. I have a funnel that is made to fit the top of the radiator, I use this to watch for bubbles.

Depends on when it leaks. Some may leak more cold than hot, but most seem to leak worse hot. Start with the engine cold, it will warm up as it runs.

I'm not sure you should expect to find much. I just thought if it has been overheated that a blown head gasket is a good possibility. You didn't say the radiator was low and reservoir was high this last time it got hot (head gasket typically makes the radiator get low and overfills the reservoir), and I don't know what the conditions were like when you saw it get hot this last time either.

YW

 
Rep Power: 0

Now I'm not extremely automotively inclined, but I thought that the little hose that runs from the radiator neck to the overflow cap starts BELOW the radiator cap so the cap's pressure rating is not a matter here?

Anyway, I did the test I just described. At first, very little bubbles came up when I blipped the throttle body. I thought maybe there was air in the hose from when I pulled it out of the overflow tank to the water bottle. So I blipped it a few times to "purge" the hose.

Then, car eventually warmed up to the point that the bottle started to fill up with a very little bit of coolant. There pretty much weren't any bubbles coming out when the car was idling. But, when I blipped the throttle, bubbles came up.

The strange thing is I basically blipped the throttle in like "5 sets of 10 reps" with like a minute between sets, if that makes any sense. The first blip of each set produces bubbles but none thereafter.

I might be in denial here....but when I did the t-stat & hose replacement, I lost a significant amount of coolant and basically topped the radiator and overflow tank pretty much once and never had to touch it again. I periodically checked the overflow tank during the following days and didn't see the necessity to top off.

But this was like 3-4 weeks ago. Could these bubbles just be air I introduced into the system during the t-stat/hose change?

 
Rep Power: 517

Go out and study its construction. The system bleeds pressure into the reservoir after the pressure inside the radiator reaches the relief rating of the pressure cap.

It has to get past the pressure cap before it can go to the reservoir.


There are TWO seals on a radiator cap. One maintains pressure (pressure relief), the other seals from the atmosphere. The overflow tube is between those.
Plus some have a 3rd for a vacuum valve.
Wild guess: You probably introduce a small amount of air into the radiator and overflow tubing each time you remove the radiator cap.

 
Rep Power: 0

Alright, thanks. I'll study the system again, but last time with both the radiator cap and overflow cap off and lifted up, I thought I observed coolant flowing from the end of the overflow cap hose with the other end connected to the radiator neck.

I also reviewed that thread you posted and saw the video. My test pretty much yielded the same result but not really anywhere near that bad. Also, as I mentioned, successive blips made immediately after an initial one does not seem to produce any bubbles.

I guess at best case it could be air in the system and at worst, a very small head gasket leak? I'm tempted to do the test again but since I introduce air every time I remove the caps, it could yield the same results till the cows come home.

Presuming it could be a mildly bad head gasket, at this point do I have any options short of a full head gasket makeover? ie. additives, what not? I purchased this car as a cheap daily after my Mustang broke down and the previous owner didn't exactly love the car either. It's got 145k and if the car is not worth more than around $2k and it costs $1k or so to do this fix then I might be interested in cutting my losses soon and selling it as a "mechanics special".

Is the blue exhaust fluid test in order? Maybe that could be the next thing I can do? I don't really have a the means to do very complex things here.

 
Rep Power: 517

This is why I use my method whenever I can. It's a steady pressure in a stationary cylinder instead of a constantly varying pressure in a running engine. My method eliminates most of the variables and guessing.
Like being "a little bit pregnant"?
If it's a blown head gasket, no matter how big or small, it's still a blown head gasket.

If it's not a blown head gasket, then something else made it (appear to) run hot.



Again, you don't know for sure what the problem is yet.


Just like when your last car took a crap andyou sold it, you bought another used car. It was for sale for a reason. Someone else was tired of it or it broke on them, or something.... There's always a reason.
This car might have an issue, (you don't REALLY know what it is yet), now you're ready to sell it off and buy the next set of wheels with hidden unseen problems.

A used car is only a used car. If you want a perfect car, buy a brand new one with a warranty. The warranty is there because even new cars aren't always perfect.

Your car might be worth only $2k, but that's as long as it keeps running without problems.
If the head gasket craps out, it's worth scrap value minus the tow truck. If you have to buy a head gasket job for 1k, it's still only a $2k car.
If the trans craps out, it's worth scrap value minus the tow truck. If you have to buy a trans overhaul for $2k, it's still only a $2k car.

If it came down to it, you could DIY a head gasket for maybe 2-300ish and a bunch of time and energy.

Any car takes money to keep it going, no matter what the value of that car is. Some require a lot more money than others.


There's a reason I can't rely on that test. I believe it's explained in the linked thread. You can try it if you want, but it's only of value if it says there's a problem. If it can't find the problem, it's worthless.

 
Rep Power: 0

I'm going to do the blue fluid test today since it is something I can try and do for cheap (gonna rent the kit). BTW, I made a mistake when I typed, my Mustang was totaled...it didn't break down. Anyway, I've had this Civic for 2.5 years now.

I'll report back with the results on the block test.

 
Rep Power: 0

Alright, so I did the block fluid test thing and it came out good.

I think that the next thing I have decided to do is to change the engine coolant temperature sensor and the radiator cap. It appears from web browsing that my car has two...one in the T-stat housing and one near the bottom of the dizzy towards the front of the engine. Would it be best to change both? Any chance that one/both of them are "sticky" or faulty which causes the issues I mentioned that I had?

EDIT: Okay...so like an idiot, I was downstairs trying to bleed out the cooling system even more to take as much air out as I could. While idling, I decided to yank what I thought was one of the coolant temperature sensors in order to determine if it was the one for the gauge cluster. Turns out I pulled the "TW Sensor" and even after plugging it immediately back in, the CEL is on and stays on now. Is there a way to remedy this?

I also have a VERY slow ongoing oil leak somewhere in that which has since coated the plug with a bit of gunk. When I yanked the plug, I noticed that there was some oil inside the sensor plug sockets...not a good sign either I guess.

Thank you,
Brandon

 
Rep Power: 517

As I have said in the past, most of the head gasket failures I see will easily pass that test.

Even the failures with combustion leakage into the cooling system.

That's why I can't rely on it.

Replace random sensors?
Each sensor has a very specific purpose.
A real service manual that is specific to your car will be more accurate than the internet.

Your engine should have three coolant sensors:
The ECT (for the ECM) and the sensor for the temp gauge are both in the end of the head.
The fan switch is in the thermostat housing.

Use wire colors and a proper wire diagram to make certain you are correctly identifying the sensor you want.

What's TW stand for?

See above.

CEL on: Read codes, make sure the code is related to what you did, then clear (erase) the codes. (Or just disconnect the battery 5 min to clear.)

A decent scanner would let you see some live engine operating data straight from the engine computer, you could "see" some of the information the computer "sees". You could watch the engine temp as the computer sees it while driving. If the computer reading stays in a normal range when the temp gauge on the dash shows high, you might conclude that at least one of them is lying to you.


The same leak that took out the heater hose?

If you mean the plug (there are many interpretations of "plug") on the temp sensor you disconnected that is under the distributor, the distributor has probably been oozing oil for the last 50,000 miles. Oil generally doesn't directly cause electrical issues (it's not very conductive) but it can cause rubber to soften and swell, and deteriorate some plastics and some wire insulation.

 
Rep Power: 0

So I happen to have a pretty high-end OBD2 scanner on hand and managed to turn the CEL light off just fine after confirming it came on due to the associated pull of the plug yesterday.

I also went back to the store to return the block tester fluid and also bought a new radiator cap (I could probably only wish that was the issue).

During the whole day today, I decided to drive with the scanner plugged into monitor the engine coolant temperature. The car would hover at around 183-186 to as high as 206 during the dead of daytime summer with the AC on on city streets. The radiator fan appears to be kicking on in the 203-206 degree range before turning off after the temp dropps to mid 190's or so.

Also, after letting the car cool down for hours and before replacing the radiator cap, I removed the old cap and started the car with the AC on for a good 20 minutes or so to observe whether bubbles would appear from the radiator neck. Some bubbles came, especially when I squeezed the upper radiator hose. Once the car got fully hot, I was getting relatively consistant very tiny bubbles.

I'm not sure if my hose replacement job back then is taking this long to burp, otherwise these bubbles I suppose can only home from a leaking head-gasket.


Anyway, I've pretty much done all that I can and I will be calling a few shops tomorrow to see what they have to say. Hopefully I can at least go in for a check-up and diagnosis. I'll tell them everything I've already done so I expect that we can go straight to a compression test, radiator pressure test and/or the test you mentioned with a shop air line.


I'll report back with any updates.



BTW, this might be a really non-sense theory, but I was thinking about this all day today and came up with a thought. Is it possible that I have a very small head gasket leak which is causing very tiny bubbles to introduce themselves into the cooling system at a very slow rate? Thus, the problem I had least week occured after my cooling system accumulated enough tiny bubbles to develop a big air pocket? Could this explain why my car has driven perfectly fine for the last 5 days and another temperature gauge pegging is in the making?

 
Rep Power: 517

Not likely.
Ding! Ding! Ding! We have a winner!


That's exactly how this works. That's exactly why so many of these these are so difficult to figure out. That's exactly why the block testers don't work most of the time.

The leakage is so small under many driving conditions that it doesn't present a problem for long periods of time.

Many people only discover the problems after an extended highway trip.

The last one has the best chance of showing/proving the failure.

 
Rep Power: 0

So I called up a few shops and found a really good one also closeby. They explained the same exact test you recommended and said it will cost $100 and a headgasket job will run about $900-1100.

I feel like I can do this test myself. I have a buddy with a big air compressor and I can probably buy a leakdown kit (Autozone doesn't seem to rent them).

I was wondering what the best way is of finding TDC? I was browsing the net and I know some people remove the valve covers and find TDC via visualizing the cam gears. Is there a way of doing it without removing the valve covers?

Also, how warm/hot/cool should the motor be for this test? And I would be looking for leaks via the tester gauge and also checking to see if there is coolant rise/bubbles appearing through the radiator neck right?

Would this one be GTG? http://www.harborfreight.com/cylinde...ter-94190.html It only goes up to 100psi apparently.


Thank you

 
Rep Power: 517

You are thinking of a leakdown tester.
Your link to Horrid Fright is for a leakdown tester.
That's not what I use.
There is no commercial kit for my test method.

I use what I need to plumb a shop air line directly into a cylinder.
My compression tester whip will connect directly to the quick coupler fittings on the shop air hoses.

I remove the Schrader valve from the whip, screw it in the first spark plug hole.
Find TDC --valves closed-- on the first cylinder.
Gotta be exactly on TDC or the crank will spin when you apply air pressure to the cylinder.
Remove radiator cap, I install my funnel-fill to the radiator neck and add some liquid.
Connect air line to the compression tester whip.
Watch for action in the funnel.

Repeat on each cylinder.
I can be done with all 4 in about 15 minutes IF the leakage shows itself. If not, it takes more time.


I start with a hot engine. If I don't get results with a hot engine, I can walk away for a while and repeat the test after the engine has cooled a bit. No reassembling to make it run to get it hot. (That's working efficiently right there!)


Also, read the stuff in the previous link about overheating causes.