Went to look at a 92 for sale the other day and noticed that the top (plastic) reservoir of the rad was hairline cracked (front side of car).

The crack was not visible with the vehicle running, but when reving it, coolant would come out of it.

I'm concerned that perhaps a blown headgasket has resulted in excessive pressure in the cooling system causing the crack. After a test drive, I saw coolant along the bottom of the passenger side of the rad - I don't know if this was from another leak or just coolant pooling from the top leak. It was hard to check power, given the testdrive route, but did seem just a little bit gutless, but the car had been sitting for a long time and there was definitely brake drag...

I checked the oil and didn't see any cloudiness and or milkiness, and there was no white smoke. Coolant in the overflow looked ok for level and I didn't notice any gross signs of fuel or oil in the coolant. Unfortunately the seller had started the vehicle before me getting there (language barrier, I asked to make sure it was cold), so I'm going to go back and check it out again from a cold start for rough running off start, white smoke, and bubbles in the coolant. I'll also try and get my hands on a compression tester and check that, but am a long way from home and tools. I understand that not all headgasket leaks are detectable through compression or leakdown tests. I've heard about a form of leakdown test where you pressurize a cylinder with shop air and see if the coolant rises after over a period of time, but am not sure about access to air...

So am hoping to get as much info for basic, driveway diagnostics. A few questions:

1. I've never had one of the plastic reservoir rads crack on me, but I've heard stories about them leaking. I was under the impression that if they failed, it was usually where the plastic and aluminum were mated together. Is it weird that it cracked away from a joint, and is literally in the middle of the front/top of the rad reservoir?
2. I can understand joints failing due to poor workmanship, but how normal is a crack like this in the middle of the reservoir?
3. I've heard that if reving an engine causes a rise in the coolant that's a possible sign of headgasket issues; however, I seem to remember coolant rising in a funnel while reving to purge coolant in other cars and that being normal as the water pump speeds up? How much rise in coolant level is normal?
4. In terms of the above point, when reving the engine on the CB in question, it REALLY makes the coolant pour out of the crack - is this because of the normal pressure increase due to speeding up of the water pump, or the smoking gun of cylinder pressure getting into the cooling system?
5. Does one need a funnel to measure hydrocarbon concentrations in coolant? I've never tried this test before, have people had good success with it?
6. Any other thoughts/tests?

The only reason I'm considering the car with this unknown is relatively low miles and a chassis that is in really, really nice shape - I don't think that I've ever seen a CDN CB that I could afford with as little rust as this one -the body and undercarriage is in better shape now than the same model year I bought over 10 years ago!

Thanks!

 

buy a headgasket test kit. u stick it ontop of the radiator wit the engine running n suck air up into this tube thru a liquid. the liquid will change color if exhaust gas is getting into it

 

Tell me about overheating, headgasket and all hassle

Possible two things are there, it's poorly built aftermarket radiator and not an OEM one or someone might have hit/dropped something on the cheap aftermarket radiator and it started to leak

Coolant rising is really normal because of W/P in action, one thing which should be watched is, there are no bubbles/fumes like you see coming out of the soda can, if there are definitely a symptom of bad HG

Have a look on this quick video:
Yes, you can try that test, but for that you need to remove bit of coolant from the radiator so that you don't suck coolant into the tester, in my case that test wasn't helpful since the leak was very small, it's definitely good for a large leak, but for a small leak it take time for the exhaust gases to enter into the cooling system and create pressure, so you either have to warm up the car good or have to wait longer, if you know what I am trying to say?


Where does the temperature gauge stays? If you remove the engine filler cap do you see any white coffee color fumes in it?Beware there can be condensation too, which is completely different

Thanks and good luck for your purchase.

 

That's great info, thanks! I wasn't sure if it was only a valid test for bigger leaks, or whether it would work for every and all...when you say it didn't work for your leak because it was too small, do you think that if you let it warm up hot enough, or run it long enough, that it would eventually show up on the tester?

Temp gauge was fine the whole time, even on the test drive, and the motor was definitely warmed up hot. I've heard that you can have HG related overheating damage and the temp always appeared ok, due to the sensor location not being close relative to where the damaging steam pockets are in the engine.

Had a look in the filler cap and at the dipstick and didn't notice anything. Not sure if the seller is tech savvy enough to cover their tracks and did an oil change to fresh up the oil and clean up around the cap, but I hope to go back and run it some more to see if any of the tell-tale symptoms develop.

 

Seems to me it just has a cracked radiator. I have experienced that before.

 

Even with a crack away from a seam, right in the middle of the tank?

Also, what are your thoughts about the coolant pissing out of the crack on reving - is this normal due to the water pump speeding up or indicative of too much pressure increase. I mean, I'm talking pissing - it's not spraying, but it runs down the side of the rad pretty good. It's hard for me to relate it to a normal rise in a funnel when bleeding, as I don't know what the normal pressure increase on a system with the cap closed would be like?

One of the reasons I'm trying my best to pin it down is that the car is about 400 miles from home. I'm in the area till Mon./Tues for work, and have no problem throwing in a new rad with the tools I have w me, but the idea of the problem getting worse a couple hundred miles from home and a deal-breaking tow is a big sketch factor, not to mention possible damage if it really is a HG issue on a longer drive like that.

That's why I'm trying to get as much info as possible to try rule out the HG as an issue..thanks for the help!

 

Yes, if the car has been properly warmed up or running long enough, maybe 30-45 min drive, tester will be able to catch it. Because for a very small leak, exhaust gases enter after a while and creates pressure so hydrocarbons will be more prominent and tester will be able to catch immediately.

Yes, actually temp gauge can be misleading on these older accords, but that's a first start/indicator for overheating.

Good luck once again and hope you will get a good car

 

BTW, any of the bubble symptoms in the radiator for a bad HG indicator ?

 

Not sure if there's an epidemic of cooling issues going around or if I'm just getting paranoid:

Another car I looked at tonight had 0 fluid in the rad overflow reservoir - car was semi-warm, and upon starting it with the rad cap off saw bubbles that looked kind of like a kid blowing soap bubbles - the only thing was that the overall fluid level in the rad was low enough that you could actually see the coolant moving in the rad, and I'm not sure if the 'bubbles' I was seeing were in fact just due to coolant mixing with air in the rad top tank as it flowed through. The cap seemed done, and had very little spring left to it on sealing, maybe that's where some of the fluid had disappeared from....

Car test-drove really strong, temp gauge was fine, and after the test drive I checked the rad again and the overall fluid level was high enough that the 'bubbles' I'd been seeing before were no longer visible.

Is it possible that an HG leak will seal up enough with the motor heating up that bubbles from CC pressure/gas stop, or would they always get worse?

Thanks!

 

Just ask the owner to let you compression test it. The tester is $20 and you just pull the 15a ecu fuse in the engine bay fuse box and you're good to go. If there are no issues then you've ruled out blown head gasket. Even if there are issues, it doesn't mean it's a blown headgasket, but atleast you'll know you can walk away since it has an issue.

 

My understanding is that motors can still have HG issues if they test fine for compression, especially if the leak is not severe.

I'm wondering if the bubbles at start-up are because of air/coolant mix due to air in the system because of the low coolant levels, or whether they are are indicative of an HG leak that is only present at start-up and and then gets sealed as metal expands and mating surfaces get closer together compressing the leak point.

I've done more research and one poster talking about the rad-funnel/bubble test said:

So two things:

1. Trying to know if small HG leak can 'seal up' after warming due to expansion?
2. Trying to figure out/confirm if the bubbles at start-up are normal, because to the best I could see they weren't there after it was warm.

I'm trying to find a funnel to do this right, but in the meantime would love some opinion...thanks!

 

1. Yes, a leak acn be baed upon engine temp. There was a problem job I had ago on a Kia where the block had a crank in it but it would only leak halfway through it's warm up. (like 100-135F) Then after that it would stop leaking. Same thing with it's cooldown cycle but it wasn't as bad because there was no pressure due to the engine being off.

2. Bubbles during startup are not normal unless there is air in the cooling system.

 

Completely and totally 100% wrong.

Then you turn around and still theorize that you could have a leaking headgasket even though you already said you don't think you do.

It's up to you what you want to do or if you want to make more work for yourself.

 

I was just sharing what I had been taught, but don't take my word on it; it appears that there are a few members, all with much more impressive post counts than me, that must share in my delusion:

_______________
I never said that I don't think that I have a leaking headgasket. I just expressed that I didn't think that a cylinder comp test was the sole, exclusive test to rule on the presence or absence of an HG leak. Then I asked about the potential role of expansion in HG issues and for clarification on possible symptoms, such as air in coolant.

_______________
I'm not interested in more work, if anything, I'm trying to avoid more work in the future. I'm just trying to increase my understanding of the systems and processes involved, something I think we're all here for. I'm not sure if you just weren't having a great day, or if I said anything earlier that pissed you off. If the it's the former, let me know so I can make it right, if the latter, I hope tomorrow's a fresh start. Thanks as always for your help.

 

It is physically impossible for you to have coolant leaking into the combustion chamber and for a compression test not to show it. If anyone says that, then they did not do a real compression test. Period. End of story.

Just think about what is happening when coolant is able to leak into the combustion chamber(which is also visa versa), that in itself is a leak. In order for there to be a leak, you would have a pressure loss in that cylinder because pressure would be released into the cooling system instead of the combustion chamber(compression tester)

 

What could be happening is that they aren't doing a compression test when the car is at operating temperature. However, it would show up in a cold start too perhaps not as bad.

 

With regards to your original problem, as you rev the engine higher in the rpms range the water pump works harder, thus putting more pressure on the system, thus pushing coolant out of the crack in your radiator easier.

 

I totally understand what you're saying, and I've been thinking and reading more about this and thought about these possible explanations:

Perhaps it's possible that in certain cases the way that the HG is compromised does not result in an appreciable loss of compression during the compression stroke, but that instead coolant gets sucked into the cylinder during the vacuum created by the intake stroke. This post got me thinking about it:

It does seem odd to me that the massive compression created during combustion would not show up as bubbles in the coolant or affect compression, while the <15PSI in a cooling system could allow for coolant to go the other way. Maybe the intake stroke vacuum theory is the explanation, or perhaps a leakdown or other sustained cylinder pressure test may help to gradually identify leaks that don't manifest on the brief moment of combustion.

Wouldn't the problem be worse if anything during cold start, since heat expansion and sealing hasn't happened yet?

Your thought that maybe people weren't doing compression tests right (which I always considered kinda hard to mess up) made me think more about the actual testing process, and perhaps part of the problem lies in the acceptable range of compression variance. If compression of say 180-175-190-190 is considered within 10% and ok, there is nonetheless a difference of 15psi between cylinders. Perhaps this is because of differences in ring sealing, valve health, etc, OR perhaps it's enough to indicate lost compression through a minor HG leak. Weird that some get much more even compression but still appear to have HG issues, esp in the case of the one poster who was only experiencing a 5psi variance on a built 250psi motor....

Thanks for the good discussion, definitely way past the original post, but good for thinking about the issues and a great learning experience.

 

First of all, we're operating under the assumption that the car we're talking about has exhaust bubbles in the coolant. This is the assumption we're holding to be true. If we're not holding this to be true, then there is nothing to argue about.

Let's keep this in mind for the time being and throughout this discussion.

Are you saying that the head gasket itself magically seals itself during the compression stroke but comes alive in the intake stroke to suck in coolant ?

Is this what you're theorizing ?

You will never have a head gasket leak around a cylinder that only allows for 15psi less than the other. I have never seen this. I have had three engines that have had bad head gaskets and the psi on the best one was around 70psi less.

The 15psi(it's actually 10% variance not 15psi...not sure where you're getting that figure from) revolves around issues with normal wear and tear. I think this 10% variance allows you do know that the engine does NOT have a headgasket leak at that moment. Anything more than that and the manufacturer is saying you have a problem cylinder/head gasket. This is my personal opinion on this though.

If you've ever had warped rotors, you'll note that when you first brake when the car is cold and the rotors are cold the rotors will act somewhat better than they do when they've good and hot. Even so, they're still warped, even when cold and still feel warped.