• NAXJA is having its 18th annual March Membership Drive!!!
    Everyone who joins or renews during March will be entered into a drawing!
    More Information - Join/Renew
  • Welcome to the new NAXJA Forum! If your password does not work, please use "Forgot your password?" link on the log-in page. Please feel free to reach out to [email protected] if we can provide any assistance.

Odd types of head gasket leaks, air in radiator, water pump seal reverse leak?

Ecomike

NAXJA# 2091
NAXJA Member
Location
MilkyWay Galaxy
Most of us are familiar with some versions of head gasket leaks. I recently discovered one odd version which was a single gasket leak only between cylinders #4 and #5, with no loss of oil or coolant, and no oil or air in the coolant, no coolant in the oil and a perfectly normally exhaust, no oil or coolant in the exhaust. Even the spark plugs look normal. The only clue was 25 psi compression and on those 2 cylinders and a rough idle with no other possible cause except for low compression for the rough idle.

Now onto another engine with an odd issue and my question! In my Nissan diesel engine, I have had an odd, intermitent problem with some air getting into my coolant system, sometimes, but not always, and it is an open system by the way. I have had the problem for years, and probably 25,000 miles now, and it is getting worse lately. The most prevalent part of the problem is that frequently, like 50% of the time, or sometimes 25% of the time, the coolant in the overflow recovery bottle will fail to return to the radiator when it cools down overnight. Sometimes, if I release the air pressure from the top of the radiator through the cap to the overflow bottle at the end of the day, before the coolant cools down, the radiator will draw all the coolant back in from the overflow bottle, but not always, and not always all of it, sometimes just half of it returns.

I have checked and serviced all the other possible gremlins except for a possible intermitent very odd head gasket leak, or a coolant pump with a seal that sucks air at certain times, and then seals at other times, but does not leak out coolant under pressure, and never leaked coolant out when cold, until possibly a few weeks ago. I say the water pump seal never leaked out coolant until possibly a few weeks ago, because I had a small over night leak sitting in the driveway overnight, during an overnight cold front recently, where there was a small puddle under the harmonic balancer which could have been from the water pump seal, but I could not verify the true source.

Recently, air, foam in the radiator has gotten worse when running the engine, but it has been bad 2 out of the last 14 days, bad one day, ok for the next 6 days. I ran a pressure check on the radaitor last week, replaced the radiator cap with a presure test / pump rig, and it sat at 7 lbs of pressure for 20 minutes, and 1/4 throttle in park, with no change in the radiator coolant system pressure. In fact we had to pump the pressure up from about 4 psi to 7 psi. It has a 165 F thermostat, an runs cool until it gets loaded on the freeway, or has the A/C on during a very hot day, and this test was A/C off, at about 50 F that day. Oh, and to top off the madness of this phenomenon, it never looses any coolant!!!! :banghead:

So that is the back ground, now the question is, has anyone else ever seen a water (coolant) pump suck air periodically, engine on and or off, pump rotating, or pump not rotating and cooling down, with out ever leaking coolant externally at the water pump seal?

Has anyone ever seen a really ODD head gasket leak that did not ever over heat the cooling system, that allowed air into the cooling system during cooldown, and did not overpressure the cooling system with compression air, never lost any coolant for months and even years, yet when the coolant cools down it let air into the coolant system instead of drawing coolant back in from the overflow bottle? Or it let some air in to the coolant system, and displaced some coolant, like once a wek, then stopped for a week, and when it did let air into the coolant system it did not overpresure the system or overheat it?

I know a lot of XJ'rs with the old closed coolant system have had air in the system problems, could their problem be like mine (at least what I think mine is) namely the water pump seal sucking air in on rare occasions.

I am also wondering, beyond just a failing water pump seal and early warning signs, if the engine start up, or thermostat cycling could not be involved in a pulse effect that temporarily opens the seal allowing air to be sucked in, in combination with a weak, old spring in the seal. I have actually seen this on large self priming industrial pump seals. They would suck dry junk into the seal face on a mechanical seal in the first second of pump start up on self primmg pumps that did not have an external seal water flush. The stuff they sucked in would hold the seal face open and the pump would start leaking externally big time.

I am getting ready to replace my intake & exhaust manifold gaskets in order to service the exhaust down pipe flange, and to replace the water pump since I believe the problem is the water pump seal, but I want to make sure I am not overlooking another possible source of the mysterious air in the coolant system!

I suspect the one time, recent coolant puddle under the harmonic balance, which is under the water pump seal area on my engine, that happend on the coldest day of the year here about 2 weeks ago was my confirmation. Also keep in mind my engine is 22:1 ratio diesel diesel, not an 4.0 gasser, and the engine is not overheating, not loosing coolant, and the engine runs like it was brand new.

Looking for opinions and alternative posibilities and ideas, and offering this odd water pump seal sucking leak as an alternative to some XJ-4.0 air in the system problems reported by many others with the older closed systems.

Lastly, the idea of the water pump seal sucking air was offered to me by a nearly retired lifetime diesel engine mechanic who says he has seen it before.

I just don't want to replace the pump and find out a week later its some 1 in a billion very odd type of head gasket leak right after doing the manifold gaskets and water pump replacement. I hate :smsoap:doing the same stuff twice in the same year.

:helpme:
 
Last edited:
When I googled "water pump sucking air" I came up with a few threads in other forums ( different vehicles) that seemed to have the problem...One that I read through sounded pretty severe...they could see bubbling in coolant with radiator cap off as soon as they started the motor...
 
have you had this problem since you replaced the cap? that's the only reason I can think of besides the 1 in million headgasket or waterpump sucking air that I can think of
 
McQue said:
When I googled "water pump sucking air" I came up with a few threads in other forums ( different vehicles) that seemed to have the problem...One that I read through sounded pretty severe...they could see bubbling in coolant with radiator cap off as soon as they started the motor...

And the links are???:twak:

Just kidding, not sure why I didn't Google it already myself, guess I will do that now. I have seen small amounts of fine foam and some small bubbles here and there, with the cap off at different times and different temperatures, but it never blows coolant out like a real, normal, signifcant head gasket leak would do, and it will come and go in intensity. I can not see flow or bubbles until the thermostat opens. I guess the foam and fine bubles comes from the water pump agitating and shearing the bubbles into very tiny air bubbles, thus the foam by the time it reaches the top of the radiator and degasses.

I have tried no less 8 radiator caps, different styles and pressures, with no change, except that a 16 lb cap works better over time than a 13 lb cap does. I suspect if I had this problem on a jeep, as hard as they are to degas, and as hot as they get, they would just boil over and overheat. In my case the diesel runs very cool, until it is loaded up on a long highway trip, at which point I have the cap back on it. The air flow also seems to increase as the engine rpm increases at a full load on the highway. If I do not floor the fuel peddle the coolant stays in the radiator most of the time, But if rev it high and push WOT in all gears for a long highway trip it fills up the overflow bottle...and when I stop, and let the pressure off the radiator cap with the release lever I find air in the top of the radiator. But at 1/2 throttle and earlier shifts (it's a manual Tranny) for the same trip it stays in the radiator.

As they say, where there is one, there is another, ad infinitum, so this may be the problem with some old jeep cooling systems too that hate to give up air once it gets in the block???!!!!!
 
McQue,

Following your Google lead I found this:

"The most likely cause for the water pump impeller to fail is due to cavitation which is air bubbles forming on the suction side then collapsing on the pressure side causing erosion. Happens on boat propellers too. If the radiator cap is not holding pressure, the water boils at very low temperature
(depending on pressure) then litterally turns into water again on the pressure side. Did you test the old cap? Also, the lowest pressure point around the pump when spinning is around the shaft. If the seal around the shaft is bad, it can suck air into the system when running especially if the radiator cap is not holding pressure."

and this:

The brown coolant ordeal was a recall. If you follow the recall, you need to re route the heater hose and add a restrictor to the system along with a thurough flush and adding the anti cavitation additive for diesel engines.. I would also recomend checking the voltage of your cooling system, as electrolysis will cause repeated heater core failures, it tends to get bad on these cars.

This one caught my attention, "anti cavitation additive for diesel engines". As well as the electrolysis issue, which might be caused by a ground wire issue causing a voltage differential between the iron block and the heater core!

I did not find as many hits on Google as I expected to find.
 
Sounds to me like a cracked head. The bubbles in the coolant could possible be exhaust gas leaking through a crack. I am not real good on the Nissan but I have seen this happen.

If the head cracks between the valve seats it can let exhaust gases into the coolant system. You may never get oil/water mix or stem out the exhaust until the crack totally lets go. I have seen them never let go. It would just bubble up the coolant. Sometimes it will get worse with the engine warm and other times it does not matter how warm or cold he engine is.

If you are building pressure from the exhaust in the coolant that would explain why the radiator will not draw coolant from the recovery tank.
 
shelbyluvv said:
Sounds to me like a cracked head. The bubbles in the coolant could possible be exhaust gas leaking through a crack. I am not real good on the Nissan but I have seen this happen.

If the head cracks between the valve seats it can let exhaust gases into the coolant system. You may never get oil/water mix or stem out the exhaust until the crack totally lets go. I have seen them never let go. It would just bubble up the coolant. Sometimes it will get worse with the engine warm and other times it does not matter how warm or cold he engine is.

If you are building pressure from the exhaust in the coolant that would explain why the radiator will not draw coolant from the recovery tank.

a blown head gasket will also do this. If you have continuous bubbles in the coolant system then 95% of the time it is exhaust getting into the system.
 
rockwerks said:
a blown head gasket will also do this. If you have continuous bubbles in the coolant system then 95% of the time it is exhaust getting into the system.

But would a blown head gasket leak come and go at random intervals? We ran the engine in park at nearly 4,000 rpm for 5 minutes with a pressure tester on the radiator pumped up to 4 psi before starting the engine and the pressure held at 4 psi steady. Did the same thing again at 7 psi. But if I pull the tester and the radiator cap and run it open at just 2,500 rpm I get great coolant flow and foam at the top of the radiator, sometimes foam, but not always foam! Note that this is a top to bottom flow radiator, not a cross flow like the Jeep 4.0s.

The foam breaks up quickly at the top, and the coolant stays in the radiator!

The engine is a 1982 Nissan SD22 diesel engine. All cast iron.

It has done this off and on for 5 years now, and about 25,000 miles, so I have a hard time believing it it is the engine or head gasket. Also I had a one time, first time puddle of coolant under the water paump during a very cold overnight freeze 2 weeks ago, that at the time kinda cinched the water pump seal leak idea for me, as the air bubble rate increased the next day, but then the 2nd day it stopped leaking entirely in either direction for the next 7 days, then the 8th day it was back. The water pump has never leaked externally with the engine runing, but I seem to recall that external pressure leaks on water seals soon follow cold snap over night leaks?

shelbyluvv,

It can't be an exhaust valve head leak because it never over heats.

As I said in my original post it never overheats, hell (LOL) it barely gets over 160 F in the winter and 190 F in 100 F ambient with the A/C runing, and bubbling comes and goes, there one day, gone the next.
 
MY 01 FSM says..if coolant level is low in radiator but not in coolant reserve/overflow tank, this means the radiator is not drawing coolant from the coolant reserve/overflow tank as the engine cools...(just as you said) So, FSM says to check condition of radiator cap and cap seals, and to check condition of radiator filler neck. If the neck is bent or damaged, replace radiator. Also check the condition of the hose from the radiator to the coolant tank. It should fit tight at both ends without any kinks or tears. Replace hose if necessary. Also check coolant reserve/overflow tank and tanks hoses for blockage. Not much help, as it sounds like you did all of this already...

found this comment in a google search...

" A thermostat sticks open and closed randomly for several reasons. headgasket sucking air, water pump sucking air, frigid cold can fail even a good thermostat into bizarre actions. This after acknowledging the system is not using coolant. Also a reservoir will fill up with cold fluid if thermostat is stuck closed in the cold. An engine can do many different things and make a thermostat problem elusive."

Flakey, intermittent Thermostat on really cold days?

McQue
 
I have a 98 4.0 here right now that never overheated stayed nice and cool and had a head gasket leak that would inject air into the coolant system. never foamed. If its foaming at all you have issues. My thoughts are from working on all types of automotive engines for over 30 years. take it for what its worth
 
McQue said:
MY 01 FSM says..if coolant level is low in radiator but not in coolant reserve/overflow tank, this means the radiator is not drawing coolant from the coolant reserve/overflow tank as the engine cools...(just as you said) So, FSM says to check condition of radiator cap and cap seals, and to check condition of radiator filler neck. If the neck is bent or damaged, replace radiator. Also check the condition of the hose from the radiator to the coolant tank. It should fit tight at both ends without any kinks or tears. Replace hose if necessary. Also check coolant reserve/overflow tank and tanks hoses for blockage. Not much help, as it sounds like you did all of this already...

found this comment in a google search...

" A thermostat sticks open and closed randomly for several reasons. headgasket sucking air, water pump sucking air, frigid cold can fail even a good thermostat into bizarre actions. This after acknowledging the system is not using coolant. Also a reservoir will fill up with cold fluid if thermostat is stuck closed in the cold. An engine can do many different things and make a thermostat problem elusive."

Flakey, intermittent Thermostat on really cold days?

McQue

First half, done that tried that, the google part is interesting, I had not considered a stuck, temporarily stuck, thermostat pushing coolant out to the overflow bottle backwards through the radiator. Interesting, though it does not fit most of my data on this one.
 
rockwerks said:
I have a 98 4.0 here right now that never overheated stayed nice and cool and had a head gasket leak that would inject air into the coolant system. never foamed. If its foaming at all you have issues. My thoughts are from working on all types of automotive engines for over 30 years. take it for what its worth

It's only foamed twice that I know of, and that was only with the radiator cap off.

How do you know your air source is a head gasket leak, and not a water pump seal sucking in air? Keep in mind it is a low pressure area at the water pump seal especially at high pump flow rates!

Ah, HA! Just dawned on me, Even at low rpms I can see substantial flow across the top of my radiator, a 1/3 of WOT, rad cap off, the flow is tremendous, so at 3,500 to 4,000 rpm it might be sucking so hard at the pump inlet that it can suck air in through the shaft seal instead of water from the radiator. This is a Nissan engine and water pump attached to a Jeep 4 banger radiator. Time to pull and inspect the water pump inlet...If someone installed a high flow pump it may be pulling water too hard on the suction side at high rpms, and getting air through the shaft seal when the water flow through the radiator can't keep up.
 
trust me Id bet ya it aint a water pump sucking in air. it would have to create so much cavitation in the water flow that at higher RPM's it would not eve function.

even if it where pulling in air enough to show as bubbles it would leak when the engine is shut off.

reread the post.......so it is leaking LOL
 
Last edited:
Ecomike said:
shelbyluvv,

It can't be an exhaust valve head leak because it never over heats.

QUOTE]


When I fought the mysterious coolant loss and bubbles in the rad, my Shelby never over heated, mixed oil and antifreeze, or steamed out the tail pipe either.

Best of luck!
 
shelbyluvv said:
When I fought the mysterious coolant loss and bubbles in the rad, my Shelby never over heated, mixed oil and antifreeze, or steamed out the tail pipe either.

Best of luck!

It has not lost any coolant in 5 years and 25,000 miles, except for a table spoon once two weeks ago when it got down to 30 F overnight, and that was directly under the water pump seal. The last three days all the coolant stayed in the radiator, with less than a cup in the overflow bottle. Four days ago it pushed a quart into the overflow bottle on one 30 mile highway trip. I returned the coolant manually, and on the return trip all the coolant stayed in the radiator even though it was a little hotter outside on the return trip. The coolant never got over 160 F.

rockwrks,

I have seen industrial water pumps suck air momentarily on start up, and then the mechanical seal resealed itself once pressure built up in the pump.
 
Ecomike said:
It has not lost any coolant in 5 years and 25,000 miles, except for a table spoon once two weeks ago when it got down to 30 F overnight, and that was directly under the water pump seal. The last three days all the coolant stayed in the radiator, with less than a cup in the overflow bottle. Four days ago it pushed a quart into the overflow bottle on one 30 mile highway trip. I returned the coolant manually, and on the return trip all the coolant stayed in the radiator even though it was a little hotter outside on the return trip. The coolant never got over 160 F.

rockwrks,

I have seen industrial water pumps suck air momentarily on start up, and then the mechanical seal resealed itself once pressure built up in the pump.

lil different design and running speeds usually. So replace the water pump and see if it helps. theory is great but not very practical
 
This is a DIESEL engine, correct? RPM has little to do with load and cylinder pressure. Running 4000 rpm in neutral is way different from WOT on a highway onramp. Cylinder pressures will be much higher on heavy acceleration than idling or just steady cruising. I believe you have a head gasket leak that is only showing itself at high cylinder pressures ie:when you get on it. Drive it mild, and it won't do it, but get on it a lot, load it down and see how much water is pushed out to the overflow. In other words, the head bolts may be stretched or a little low on torque.

Go for a drive and see what happens.
 
stangrcr1 said:
This is a DIESEL engine, correct? RPM has little to do with load and cylinder pressure. Running 4000 rpm in neutral is way different from WOT on a highway onramp. Cylinder pressures will be much higher on heavy acceleration than idling or just steady cruising. I believe you have a head gasket leak that is only showing itself at high cylinder pressures ie:when you get on it. Drive it mild, and it won't do it, but get on it a lot, load it down and see how much water is pushed out to the overflow. In other words, the head bolts may be stretched or a little low on torque.

Go for a drive and see what happens.

Yes, it is a diesel. Great feedback, thanks. While rpm does not affect cylinder pressure and load compared to idling versus low rpm at heavy load, it does or can affect the water pump seal and localized pressure between the water pump inlet and outlet. I have seen foam and gas bubbles at the radiator throat (sometimes) with the radiator cap off, but no overflow to the overflow bottle right after reinstalling the radiator cap, and no pressure increase in the closed system after reinstalling the cap.

Because of the gearing and tire size, I can not get this thing over about 2,300 rpm (peak torque) in 5th gear, so I can, and have run it at WOT in the 5th gear at about 70 mph (no custom and gear speedometer cable on it yet). I have also run it near peak rpm under load. So far I have not seen a consistant load, peakload, rpm, or engine temperature that consistantly causes the problem.

Keep in mind, that for 5 years, 25,000 miles, the main problem was the occasional refusal of the coolant system to suck coolant back into the radiator from the overflow bottle during the overnight cool down. One night it would work, the next it would not. Over 2 years ago I had the radiator rebuilt, and I replaced all the hoses and clamps, which increased the frequency of its working properly (by about 75%), but did not stop the occasional failure to suck the coolant back in. IIRC it has been more erractic and unpredictable the last 4 weeks.

rockwerks,

I understand your comment regarding theory and practice, but although Edison might have agreed with you, Einstein would have disagreed! :D

Being a Chemical Engineer myself, I have suffered through way too much theory in Physics and other science classes to give up my theoretical work!:laugh3: Besides it helps me to PROCRASTINATE!!!:D
 
Ecomike said:
Being a Chemical Engineer myself, I have suffered through way too much theory in Physics and other science classes to give up my theoretical work!:laugh3: Besides it helps me to PROCRASTINATE!!!:D

Well it is entertaining to say the least, You going to take it to committee next? LOL now get away from the screen and replaced the WP and lets see if it works.

I feel like singing TINY BUBBLES..............
 
Well a few weeks ago I busted an exhaust manifold bolt, and had to pull the intake, exhaust maniflds, the AC compressor, and decided as might as well get around to the water pump too. Got a few trips on since then. In the driveway there was a noticable decrease in air bubbles in the open radiator neck even at high rpm. Took it for a spin after getting hot to get any trapped air out, and it filled the overflow bottle and then refused to draw it back in again over night like before. :mad: But, this time the radiator hose was much harder, much higher pressure than ever before, that I can recall.

So I manually transfered the coolant back to the radiator from the overflow bottle the next morning, cleaned the cap and neck area, and drove it again, 70 mile high way test trip with AC on.

This time the radiator hose was very hard, engine temp got up to about 180 F (hand held IR gauge used), but this timethe overflow coolant bottle was still almost empty (all still in the radiator), all after two 35 mile highway trips 4 hours apart. Have not driven it since then (been repairing the alternator.... since then:mad:).

My inpection of the old water pump mounting surface and mecahnical seal area and engine block mounting surface after pulling the water pump, and it's prior odd behavior in use, have me convinced the water pump was sucking in air at various times. :lecture:

Need a few more road tests to see if there is still any other source of trouble in the cooling system.
 
Last edited:
Back
Top