Sorry this is going to be long but I wanna save you the hassle of asking about things I've already tried.

1996 Monte Carlo 3.4L DOHC V6, 103,000 miles

Everything was PERFECTLY fine except water pump was leaking. Easy enough: replace the pump. Mechanic did that, on 5/17. Picked up car, drove across town, everything fine.

Next day drove ~ 6 miles on interstate: Temp gauge has now become possessed lol. I suddenly noticed temp gauge @ almost full hot. Before water pump, normal gauge reading was ROCK steady - always at 1/3. Pulled over looked under hood; engine did not seem hot at all. By the time I got back in car gauge was close to normal again. Another half mile, idiot lite actually came on, and then got worried. Called up a buddy & filled him in, and he said "air pockets."

yep - I know; air pockets. I wouldn't have posted here if it weren't for the fact that I think I have read literally EVERY post on this already on this website, and dozens of others on other sites. Almost half of what I read, they describe almost EXACTLY the same symptoms, some even begin with the simple change of a water pump just like me. I think I have 80 hours on this so far. But when I get to the end of a thread, nobody ever says what they did to fix it. I seriously doubt it's UN-fixable.

Sitting in driveway, I cannot make it overheat to save my life; even by revving RPMS to 3000 for a full minute (temp does go up to like 1/2 but you'd expect that I think). But drive it around the block and watch out, even though RPMs don't get anywhere hear 3500 doing that.

Full blown, sauna-quality, constant heat (this is NOT a bubble preventing circulation thru the heater core).

Although I didn't remove it, I'm sure pump is correct pump. At a modest 1200-1500 RPM, I can see coolant inside radiator (thru neck) flowing like Niagara Falls. Overflow tank fills/drains as expected. So coolant IS moving around. So unless the pump is only strong enough to move coolant like that when the system is not pressurized (obviously the radiator cap is off), I think it's the right pump (he said he put the "good" one with the cast fins, not the flimsy sheet metal-like fins).

This car is supposed to have the Orange GM DEX-CON crap in it, and it's been now been filled with the "normal" green stuff. The initial leaking coolant before water pump was replaced was green, so there is not a mix between DEX-CON and "normal" coolant. Plus, I've had the car for ~25,000 miles, since 2008, without any issues. If there was DEX-CON left in there to be mixed with green (bad I guess), my radiator should have rotted out already (or something) before the pump had a chance to leak in the first place.

NO leaking coolant, NO "disappearing" coolant (burning up in engine or whatever). When I took car in for leaking water pump the coolant level was only slightly low. If my engine is burning coolant now, it would have been burning coolant before water pump replacement.

Cooling system held 120 psi for 10 minutes. ZERO visible coolant leak anywhere.

Bled/burped/purged air 35-40 times (yes really that many times, I kid you not). The guy who did the pump also did extensive bleeding attempts when I brought it back to him the next week; I was there watching him, learning about it.

I replaced thermostat yesterday just in case...I'm sure the old one was fine...

Air is NOT coming in from a combustion leak. All 6 cylinders passed a compression test, and once idling at normal operating temp, after it's clear the thermostat has opened, I can take the radiator cap off (slowly), and watch the coolant just sitting at the top of the radiator fill neck (fluctuating only slightly, as you'd expect - cuz water is pumping). ZERO bubbles. ZERO oil in coolant. ZERO coolant in oil. ZERO liquid of any kind in oil. ZERO white smoke. I did not test coolant for hydrocarbons - a test that can show that a combustion leak exists even when a compression test passes(?); I don't see how the temp gauge can jump to full hot in 30 seconds of driving time if ZERO gases are being introduced into the radiator after 30 minutes of idling. Unless hydrocarbons in the coolant under pressure CREATE a gas I can't see, I'm at a loss. Besides, if this is a head gasket or other combustion leak issue, why didn't it do this before the water pump was replaced?

No one has a clue, including the guy who did the pump (he did lots of burpin' too). I've heard they can be a pain to bleed, but this is beyond insanity....

Is there air in there that could only be removed by draining and refilling? Did the draining/filling to replace pump cause a block somewhere? (But then I would have no heat?) I didn't remove the pump to see if it's the right one, but there is good circulation in the radiator at least. Or at least it's something totally stupid that no one would ever think of, like I forgot to put my gas cap back on...NOT

I GUARANTEE there is SOMEONE out there that knows what's going on...

EDIT:

PROBLEM SOLVED

If anyone ever has this problem when everything was fine before your water pump was leaking, MAKE SURE THE PUMP IS CORRECT. The pump Bumper to Bumper found/recommended had the correct picture on their computer, but the WRONG ONE WAS IN THE BOX. I wasted 100 hours on this - don't you do the same lol. 03SSLE suggested this right away, and I also thought this. Obviously, reduced flow doesn't evacuate air into the overflow tank like it's supposed to.

 

It sounds like you have pretty much covered everything. The only thing I can think of is the possibility of a standard duty pump installed instead of a severe duty pump. Physically they look pretty much identical, but the severe duty pump has a different design impellers.

Does the car actually exhibit signs of overheating? When the gauge pegs out it is usually a sign of a short circuit. If you have a loose/dirty connection at the sending unit or the wire has some frayed insulation that would cause the gauge to intermittently peg out.

 

I'm Curious...

You said this all started... after replacing/changing the water pump...
When you say idiot light...
Are you meaning the low coolant light?

What "Process" are you doing to bleed the air out of the system?
Is the motor running when you try to bleed the air out?
Quite possibly you may have been doing it in correctly those 35 to 40 times... not saying you did..but it is possible!

When you say the temp gauge is pegged.. Is the Motor actually overheated?
If not this is a possible sign of air still in the system as well...

 

Could be an issue with the coolant temp sensor, or maybe that new water pump is faulty also... Stranger things have happened.

 

Thanks guys!!!

03SSLE: it does seem as though it may not actually be overheating. One time I pulled over with gauge spiked at 3/4 or 7/8, opened bleeder over tstat and steam & tiny bubbles come out. Maybe this would always happen since this spot is right off the engine? Since increased pressure raises the boiling point of the coolant, maybe it's not boiling till it comes out of the bleeder? I have experienced an overheated engine before; I think mine might not be actually overheated. Other than the steam from the tstat housing bleeder, it doesn't seem overheated...

STUMPMI: Sorry - I though the idiot light was any light for an "idiot" who can't read a gauge lol. My idiot light is the HOT light. That came on 8-10 times when I first got it, but after bleeding so many times the hot light has never come on again yet. But then again I haven't taken it on the interstate since this started happening so I'll bet it would. I've had this car (got it from my cousin) since 2008 & put 20-25K miles on it with no troubles (till water pump)

As for bleeding "process": trying several different things. Started out by just opening the tstat bleeder once in awhile - didn't work and I understand why now. Did get air sometimes but usually just coolant. Next tried this:

Start car from cold with rad cap OFF. Coolant level right at top of rad fill neck. Coolant just sits there, as you'd expect cuz tstat not open. As it warms up & tstat begins to open, coolant level begins to fluctuate. Up/down, up/down almost rhythmically. At this point, bubbles start coming out as it moves up & down by, say, 1 to 3 inches in the neck. The first few times I did this, the coolant level ended up going down really far, and I was able to add an extra gallon of coolant that wouldn't fit before, so I figured I was getting somewhere. I.e., air was coming out & making room for new coolant. As the temp raises more & gets to normal operating temp, it starts to want to spill out at its high point, and disappears low enough at its low point so that I can't see it anymore, looking down the rad fill neck.

So I quickly discovered a good idea to use a thick, flat piece of old conveyor belt material so I could prevent coolant loss out the rad neck when coolant level rose.conveyor belt material, which is rubber but it's still pretty stiff, almost like a rubber block. Coolant rises up enough to want to spill out, so I cover the rad neck with the rubber block. Wait a few seconds, and coolant level is back down again.

It's to the point now though that I can't make room for any more coolant. It does seem like the gauge is not quite as possessed; i.e., it's not as extreme, although it does still spike. So I had this crazy idea (don't laugh too hard lol):

Place rubber conveyor belt material over rad neck. Put block of wood over that to ensure it's nice & flat. Wedge a 28" 2x4 between wood block and hood of car to ensure it sealed. This forces any coolant into the overflow tank. This would be like cutting the spring & secondary seal off a radiator cap. I did this figuring a could rev the **** out of the engine without worrying about spilling, kinda like that no-spill funnel. I wanted to try this before buying the funnel. Sure s**t. I revved motor WAY up and it looked like literally QUARTS of air would go into the overflow tank. So again I figured I was getting somewhere. Not. Again, it does seem better than when I first got it back, but it's still not good enough.

A potentially interesting thing to note: with the setup as above, basically meaning the radiator is capped but missing it's spring so that coolant ALWAYS goes into overflow tank regardless of pressure, when I rev the engine high enough I can see the top hose (INTO the radiator), compress itself - almost flat if I rev high enough. Maybe something is blocked somewhere? I would not expect to ever see that since given the way I set it up, the system could never build pressure. Or, that just means the pump IS pumping well enough, and is pumping so fast into the engine at higher RPM that the hose on the other side of the engine just can't keep up, and that's why it collapses...

I am not 100% positive the pump is the right one. I asked the parts store where he bought it which one he got - the "good" one with cast impeller blades or the chinsy sheet metal ones. They said the sheet metal ones. I asked the guy which one he put in. He said the one with the cast blades. Hmmm...I didn't take the new pump off to see which one it is, but maybe I should.

I'm starting to wonder if by some out of this world miracle the impeller is backwards. After all, I think I've exhausted the obvious - it's time to start looking at the ridiculous lol. With enough coolant OUT so I can rev the engine with the rad cap off, I can see coolant moving from driver to passenger side at the top. This would seem to mean coolant is ENTERING on the passenger side. However, coolant should be exiting the lower hose, and the lower hose IS colder than the top, like it should be. If coolant was entering the radiator from the bottom, I would not expect that. So this is prolly not what's happening. I was not there when he changed the pump, so maybe I should just take it out and see for myself.

If it is the "chinsy" pump, maybe there is just not enough flow when driving, even though it looks like Niagara Falls in there when revving the engine? Of course then, the system is not pressurized either because the radiator cap is off...

nitehawkjcb: stranger than this? I hope not lol

 

You should get a solid stream of coolant - not steam or bubbles. That indicates there is still air in the system. Keep opening that bleeder a few times with the engine up to temp for a couple of days (being careful not to burn your self) and keep doing this until you get nothing but coolant from the bleeder valve. Then fill the overflow tank to the hot level. Take it out for a spin and see how the gauge reacts.

 

Thanks for the Clarification Mike...

Just be sure to bring the motor up to temp.... then shut it off .... Then open and close the bleeder carefully and slowly until it streams coolant...
Otherwise.... close the bleeder securely.. restart the motor til it comesback up to temp... shut the motor off once again ...and repeat the bleeding process... It may take as many as 10 times before you get all the air out of the system.... (That's why the bleeder is there) .... loosening removing the radiator cap is not the "Proper" way to bleed this type of cooling system....

Just continue this process until you get just a stream of coolant out of the bleeder valve... BESURE THE MOTOR IS UP TO TEMP AND SHUT OFF EACH TIME YOU BLEED THE SYSTEM .... WITH... TEH BLEEDER SCREW ONLY.

Once your confident its bleeding coolant only ...take it for a test drive and
and see that its is back to normal...

Sometimes...
A day or 2 after you may have to do the bleeder sequence again to remove the residual air out of the system...
After that you should be golden.

 

try hooking up a scan tool...make sure the coolant temp on the scan tool matches close to the temp on the gauge...this can isolate a gauge or coolant temp sensor issue...take off your rad cap too...stick a thermometer in the rad and see if the readings on the thermometer, gauge and scan tool pretty much match up...

 

Thanks again SUPER much for the replies... will be trying again starting tomorrow. So basically, keep a screwdriver with me, and when gauge spikes, pull over, SHUT CAR OFF, and open bleeders. Would it help to wait a few seconds or a minute to allow bubbles to collect, especially when opening the bleeder over the tstat? I think I mentioned before that there's 2 bleeders, one over tstat & one on smaller line going to heater core. Many times I get only fluid from them. Since I'm not in my driveway, I don't think I should just let it squirt fluid in the hopes that air will eventually come out...

 

I figured that. Not right away, but it makes sense now given that there's still air I can't get out. Every time I did that I prolly introduced more air lol. Thanks for that insight!!!

Quick question: would there be any benefit to totally flushing & refilling? When he refilled after changing the pump, he admitted he did NOT have the bleeders open when he filled it. I think this is why there's such a ridiculous amount of air in there...