Ok well time to figure out what to do.
We paid $1,400.00 for a 1994 Buick Regal with 129K. The previous had a maintenance log showing all the work he had done over the past 5 years. He spent quite a bit dong routine maintenance that would be expected on a car of this age. Ball joints, motor mounts, transmission service, turn signal switch, ignition switch and keys with resistor, ect.
So the car itself seemed to be in good shape. drives and handles great. The only problem is that the engine must have a cracked cylinder head, block, or bad head gasket. The PO recently had a new water pump installed. My guess is that the engine overheated while he was in traffic and it got so hot that something cracked. The PO forgot to mention anything about the engine ever overheating but the cooling system was full of stop leak when we started working on it.
Of all the stupid things, I forgot to look in the radiator before buying the car!

Any way so did a compression test as well as a leak down test and only 1 cylinder was a little weak and it was within range.
We did a chemical combustion leak test and it came back positive for exhaust gas in the radiator.
We removed the cylinder heads and sent them to a shop to be cleaned and pressure tested as well as new valve stem seals installed. Shop said they were fine.
WE put the engine back together with a new upper intake and retested for combustion gas. Same problem.

I cant believe it would be the head gaskets. We had the heads resurfaced. I cleaned the block surface with a hand held razor blade making sure not to scratch the cast iron. After that i used brake clean and a small stainless steel brush to lightly go over the surface again just before installing the head gaskets. (with the arrows pointing towards the front of the engine). My son and I practiced setting the heads on without the gaskets so we could make sure we wouldn't have any problems. Went back and installed the gaskets and set the heads in place. New bolts torqued to spec and then the additional torque angle. Pretty straight forward. (yes in the proper torque sequence).
The engine runs great except for a long start when cold. but other than that it runs great with no smoke out the back. No water in the oil. Right now the only symptoms are it fails the combustion gas test, blows air bubbles out the radiator with the cap off, blows coolant out the radiator with the cap off. We have driven the car almost 100 miles now with no problems but know there is one...

We have about 2k in the car now with the work we did.
So what do we do now? Cut out losses and run?

Cathedralclub suggested we disconnect the injectors one at a time to positively identify which cylinder is leaking. My guess is cylinder 1 which his on the front side of the engine. Now that we just did this job I could get the front cylinder head off in no time and have it retested. If not the cylinder head then it would have to be the block.
.
I found (haven't seen it yet) a 1994 Buick Regal that is supposed to only have 70k on it which I can pick up for $800. If it has a good engine in it then we could just swap it.

Ok any suggestions are welcome...

Thanks
Anna

Intake manifold could be cracked.

Is it blowing coolant/bubbles out the radiator at all times with the cap off, or only after it'* warmed up?

Same question as rjolly87 ^^^^^^^

There is an important distinction needed here. Are the air bubbles continuous over minutes or an hour at a stable operating coolant temperature?

There is an important distinction needed here. What does "blows coolant" mean in this context? Is it blasting out gallons like a lot of flow and pressure are behind it? . . . or is it overflowing like a half gallon as the engine warms up?

What is the best way to test this?

Last night we drove the car about 3 miles and the low coolant light came on. (it may be bad). We stopped in front of a closed store that had good outside lighting. I loosened the radiator cap the first notch and of course some coolant blew out but less then a cup. When we took the cap all the way off, the water level was up to the top of the filler neck of the radiator. The AC was on so that one of the fans would blow all the time and keep the coolant cooler then if it were turned off. The coolant fan will not come on by itself till around 220f. Running the AC fan keeps the temp gauge around 195F. We are still using pure water and flushing it out every couple of days.

Ok back to the radiator cap off and the engine running with the AC on. The water was pretty close to the top of the filler neck and would bounce down and back up an inch or two. I watched this for a good 3 to 4 minutes with NO bubbles coming out.
I told my son to cut the AC off. When he did the engine rpms slowed just a little and some of the water started coming out the filler neck. Not forcefully like being blown out but like happens when you rev an engine and let off. After the water stopped coming out and the water level settled down to about an inch down below the filler neck, I could see bubbles floating to the end of the radiator pretty consistent. Not a steady stream of many bubbles but just one small one every couple of seconds or so.

Thanks
Anna

One more note is that the outside temperatures are in the 90'* here and we have driven the car around 300 miles. The engine had not overheated and we have not had to add any water. The thermostat is maybe 195? what ever was speced in the manual. We have kept the radiator cap on tight. The overflow tank is not a pressurized tank. WE originally filled it to the hot line and it has been staying at that level.

Thanks
Anna

Not sure of your experience with these engines, but I will just run through a few things. Please feel free to disregard what you already know.

1. They love to trap air in the cooling system. Over the years, the forums have had the best luck with drilling a tiny hole in to the thermostat (or get one that has this concern in mind). Without doing this, it'* very hard to get enough coolant on the initial fill, and then you have to star at it until the thermostat finally opens, and the level drops, hoping that the temperature doesn't run away on you. You also need to properly bleed the air once filling is complete. This is accomplished by running it up to temperature, and while the cooling system has pressure, crack the bleeder screw open that'* on top of the thermostat housing. If air is trapped, it'* likely going to be there, and you'll see it escape before coolant start spewing out. Once you get steady coolant, close the bleed screw. If the radiator cap is functioning properly this will cause the system to pull vacuum when it cools back down, drawing coolant from the overflow. This process usually needs to be repeated 2-3 times, until you consistently get coolant when you crack the screw open when they system is pressurized. You then just keep adding coolant to the overflow, and monitor from there. I also assume you replaced the radiator cap in all of this, or at least tested it. If the system is holding pressure, and the coolant levels are not going down, there'* probably nothing to worry about.

2. To this day, you have found the same exact things I found in my Regal that had 72k miles on it when I bought it. The previous owner, even years later, is adamant they never put stop leak in it. It had all the same signs you found, but I never did a combustion gas test on it. I had an oil sludge buildup at the bottom of the overflow, the underside of the radiator cap look exactly the same, and I could see sludge/sediment buildup through multiple components. I flushed the entire system with the water hose, refilled with new coolant, and haven't had any concerns since. I believe it was just a result of improper cooling system maintenance, something that the previous owner was pretty notorious for neglecting. This sludge also plugged up the coolant level sensor, causing it to indicate low coolant most all of the time. The flush helped some, but I just replaced the sensor since it was cheap (Less than $10 on amazon )

3. Straight water boils at 212* at sea level. If you are above sea level, it'* lower. As you mentioned, the running of the A/C should drop it down below that thanks to the fans, but even then, all you are doing is saving yourself from the flash boiling. It'* still possible those bubbles could be steam.

4. Reading on the internet revealed this thread elsewhere: https://www.garagejournal.com/forum/...p/t-72706.html
Of particular interest, this post:

Honestly, I think it'* just at the point where you should just do a proper fill with coolant mix, and just run it, assuming the radiator cap, etc is still good. Just bleed the system properly. You may as well replace the level sensor while you are at it, it'* cheap. If there'* a problem, you'll see it, but I think we are at the point where we are over analyzing everything, and just seeing issues where there actually may not be any. Just finish it up right, and run it, keeping an eye on things at first.

How long did you observe this?

What rjolly87 said above ^^^^^^^

. . . and if it'* a blown head gasket there will be continuous bubbles when sitting at the same idle speed for five or ten or fifteen minutes. Running the engine up and down (even as little at the cycling compressor will do it) will push more bubbles through, but will ruin your chances of seeing if it is a steady stream or not.

If I can, when I replace or fill coolant in an engine, I'll get it full and let it warm up while watching the coolant level, wait for the thermostat to open, replenish what gets sucked down, then run the engine up to about 1,000-1,200 RPM. This sucks down a little more and I can add a little then close the cap, and only then let it come back to idle. If I 1,200-RPM fill it, leave the cap off, and let it return to idle, it will dump the same amount as I added when it was at 1,200 RPM. I'm not suggesting that you do this, just using this as an example to illustrate how varying engine speed will immediately change the coolant level . . . similarly to what you describe.

If the engine increases RPM a little then decreases a little, the increase may have lowered the coolant level a little, which would allow a little air to get trapped in a radiator tube. When the RPM decreases a little, the coolant level rises a little causing the air to slowly work its way out.

. . . and you're using mostly straight water which will be a little more volatile in this regard.

Regarding the combustion leak test: yes these tests are finicky. It could even be some vapors from whatever the [stop leak] stuff is that is in your system. They work more accurately as the problem is worse . . . at which point you probably know you've blown a gasket already.

If I were in your shoes, I'd just drive it and keep a close eye levels/performance/etc. until you build confidence. Once you get a couple thousand miles on it and the levels have stayed the same and nothing else (related) has gone wrong, you can start relaxing.

Thank everyone for the help and advice. I have a pack of drill bits sitting here on my desk and we forgot to drill the hole in the thermostat!
We will be taking the thermostat back out to back flush again. It is amazing how much of the left over gunk has circulated into the new radiator even though we flushed the coolant several times before installing the new radiator. We have a bottle of radiator flush we are going to run through the system and then go a head and add 50/50 mix. Since the boiling point i higher with antifreeze, that will help out with watching for bubbles while idling with the AC off...

When looking at the bubbles down in the radiator, we probably watched it for only 3 to 4 minutes..

This is my first experience with the GM 3.8 so thank you for rundown on how to bleed the air from the system. We used the radiator cap from my son'* 92 Regal since it was fairly new.
The coolant level sensor was full of the same gunk as the old radiator so we cleaned it fairly good and stuck it on to see if it would work. If the sensor is bad and shows low coolant, will that affect the starting of the engine? We are still having long starts when the engine is cold but I have not checked into that yet.

As to the combustion gas test... My son said that we tested his 92 Regal with a 3.8 2 years ago right after he bought it an he said the liquid turned green so perhaps it could have something to do with chemical reactions going on inside the coolant system. His 92 looks pretty dirty inside the radiator. I tested my 1988 Ford van which is all cast iron with a brass/copper radiator and very clean coolant system, and it passed just fine. I had a GMC van with a 350 engine and I remember it passed too.


Anyway,,, I tend to agree that perhaps we should just drive it and like you guys have said, keep an eye on it and perhaps in time I will become comfortable with it.


One thought right now is on the other 1994 Regal I found that is supposed to only have 70k on it. What I have learned is that the engine in the 1994 was only used 2 years, 1993 and 1994.. "The earlier “EV-6” block was revised in ’93. It had an additional pad on the driver’* side for the knock sensor and another bolt hole in the left front corner, just below the deck surface, for an accessory mount. It’* either a 24502134 or a 24502441 casting."
https://www.enginebuildermag.com/200...-buick-engine/

So that means that if we wind up needing a different engine later on, we would have to find a 93 or 94.... We have plenty of space to park the other Regal if we were to buy it. I don't see running across another one with only 70k on it. This would also give my son spare parts for anything he might need in the future that can no longer be bought.

So I am thinking, Should we pass up the opportunity or go a head and get it. It appears to be the same color too! If we do go look at it I am going to take a vacuum gauge, the combustion leak tester, and compression gauge. The 3 front plugs are fairly easy to get out..

What do you all think about having a second Regal for parts? We can always scrap one later..

Thanks
Anna

Will update later after we have some miles on it... Thanks!