Cooling system driving me crazy

Hello,

I have a 96 XJ with around 140k on the clock. Ive owned the car for around 3 years now. Last june, I installed a CSF 2 row all aluminum radiator, a new fan clutch, a 97+ electric fan, a Stant 195* thermostat, new upper and lower radiator hoses (with the spring in the lower radiator hose). I was in california at the time and the jeep was running in california weather on both highway and city with the AC on without ever going above 220.

Recently, I moved to virginia and in the last 2 months i started noticing that when on the highway (65-75 mph) and with the AC on, the jeeps temperature would rise to the 3/4 mark right before red zone. I would instantly turn off the AC, turn on the heat and decrease my speed. This brings my temperature down in just a minute or two max. Without the AC, the car would get close to the 3/4 mark on the gauge but would not hit it. Under acceleration (3rd gear, rpm between 3-4k and speeds above 65-70) without AC the car would hit the 3/4 mark.

I was also experience squealing noises with increasing rpm espeically when the AC was turned on and the water pump pulley had some play in it. Last weekend I replaced the water pump with a gates pump, replaced the idler pulley, and replaced my harmonic balancer. I tightened the belt to 180-200 ft.lbs using a kricket 2 and the squealing was gone (even with AC on) and everything was running great. New coolant was added and I tried making sure no air bubbles were left in the system, I squeezed the hoses and shook the car, and kept on refilling my radiator for around 2 hours (was working on another car in the mean time) whenever water level went down. Coolant reservior is full.

Today, on the highway, I wanted to test my cooling system for the first time with the new water pump, belt, etc.. so I turned on the AC and drove the car at around 70mph shifting between 3rd and 4th gear (rpm about 3000 and 2200 respectively). You guessed it, the gauge went up to the 3/4 mark and I had to turn the AC off, turn heat on, decrease speed and it cooled back down in around a minute if not less.

I am at loss. Is it not really overheating and my gauge/sensors are faulty? I am missing something in the cooling system? Do I have air bubbles? I am frustrated as all the cooling system parts are new and I cannot explain what is happening. Please let me know what you guys think.

Also, the electric fan turns on at around 220, and it never overheats in the city no matter style of driving or AC on or off.

Please help me get this figured out.

Thanks,
Laith

Fan shroud in place and intact?

You say you added coolant, what kind? 50/50 or straight ethylene glycol?

I don't think it's likely, but are you sure you got the right water pump? Buddy of mine had a CJ7 that would overheat, even with a new water pump. Turns out he got the wrong rotation water pump...

What's the pressure rating on the radiator cap? Has it been pressure tested?

Just kinda spit-balling a bit here. I don't think it's truly overheating, but I think that's only because you're shutting it down before that happens. It is definitely running hotter than it should though.

Radiator cap may not be sealing. ^^.

I only use factory stats. 83501426AC.

You may not have gotten all of the air out. I just swapped the radiator in my wife's WK and it to a good hour to get all the bubbles out (Jeep running with a funnel full of coolant in the filler neck)

His woulda purged already as many heat and cool cycles he's gone through.

Yes the fan shroud is in there and intact. I added 50/50 coolant. Id imagine if the water pump was not the correct rotation one I would also overheat driving around the city too. I will double check that though.

Im not sure what the pressure rating on the cap is but I checked the spring and it feels fine. I really hope its an easy fix like a pressure cap, I will swap in a new one today after work.

I thought I got all the air out, what suggestions do you guys have for getting the air out? any specific method to do that?

Could it be that the temperature sensor is not reading accurately? Maybe its showing me a higher reading than it actually is in there? Any way to diagnose a faulty/incorrect sensor reading other than a thermal (IR) gun?

I would really like to get this fixed as its my DD and its a total pain to drive with my eye on the temp gauge all the time, not being able to drive at the speed limit and no AC...

Cherokee cooling systems with a radiator cap are self purging. +1 on a new Stant or genuine Jeep radiator cap.

Confirming the Temp indications with an IR temp gun aimed at the thermostat housing is a good idea.

Genuine Jeep radiators always out perform any aftermarket radiators.

Re: Getting air out of coolant

There was a tip years ago to drill a small hole (1/16"D or less) in the thermostat. The hole was small enough so as not to affect the thermostat's function but enough to bleed the air out of the coolant. I did the mod on my 88XJ and it seemed to work. It seems some brands of thermostats already had a small bleed hole, while other thermostat brands did not have the small bleed hole.

Best regards,

CJR

So as long as my radiator is full after a complete heat cycle, and my reservoir is full then I should be good to go in terms of air bubbles and coolant level correct?

I will pick up a new stant/Jeep radiator cap. 16 lbs correct?

I will check and see if I can rent/borrow an IR temp gun, get it on the highway with AC on and get it super close to 3/4 mark, pull over immediately and check out the thermostat housing temperature.

Any other suggestions? Im worried above having other way more serious issues lurking like a cracked head or block, catalytic converter clogged, etc...Any thoughts?

Thank you so much guys, this forum has helped me understand everything about my XJ.

There's no need to burp/purge air on a late style system. There is no "heater core control valve" and it has fluid constantly flowing thru it and the bypass at the t-stat housing.

I find it interesting that we're reading so much about normal running XJs suddenly starting to run hot. Since you're asking about other possible problems, I have some suggestions:
1. Check the other forum threads on trying the NAPA kit to determine is exhaust gas is getting into the coolant.
2. A coolant system that pressure-tests Ok can still have a leaking head gasket. I know, because I had that happen to my 88XJ, when the head gasket would only leak on a hot day under acceleration and then would re-seal. The shop pressure-tested it and it was OK. But I noticed a white puff of smoke on a hot-day acceleration and then the white smoke stopped after acceleration.
3. If your head bolts are the new Torque-to-Yield (TTY) head bolts, it is possible that hot spots in the head, near a head bolt, are reducing the Young's Modulus of Elasticity (E) and the hotter head bolt will yield at lower bolt stresses/torques. Once that happens the ball-game's over. Why? Because plastic deformation(i.e. bolt yielding is UNPREDICTABLE. Once it happens, the TTY head bolt MUST be changed. Only the old style head bolts and the current ARP head bolts are predictable because they operate in the Elastic Region of stress. In other words, once a TTY bolt yields its length is longer and its bolt diameter is smaller after you remove the bolt stress. If you reuse the stretched TTY bolt (w/smaller diameter) it will now yield at a lower stress/torque level than it did originally. However, when you release the bolt stress in the old style/new ARP bolts the bolt length returns to it original length and original diameter. To seal any gasket, we need to provide and MAINTAIN a specific CLAMPING FORCE on the gasket for a specified operating pressure.
4. Typically, engines start to gradually run hot over a
long period of time/high mileage, i.e. the old radiator starts to build-up with scale, engine block starts to build-up with scale, water pump starts wearing out/leaking, hoses collapse, etc. When a low mileage engine start running hot, with new radiator, new water pump, new hoses, etc., we need to look elsewhere. It's natural to look at the head as that's the hottest part of the engine with the highest temperatures and generated pressures. Likewise, cracked heads/blocks tend to fail rather quickly.

Best regards,

CJR

Okay so I dont need to worry about air bubbles as long as my radiator is full after a heat cycle and my reservoir is full too.

CJR, so I guess the first step would be to try and identify if I have mixing between my coolant, oil, and exhaust gases. I checked the dipstick yesterday and it was not light tan/off-white color which is a good sign. I also cranked the engine with the cap off and coolant did not shoot out, also a good sign. I would like to do a compression test and a combustion leak test.

Would you guys think its a good idea to replace the temperature sensors? I dont think they have been replaced before. Does a specific brand matter? trying to avoid the dealer ripping me off when im not even sure if theyre the problem.

Still trying to get my hands on an IR temp gun to check the temperature on the thermostat housing...

This issue is driving me crazy

A scanner would be of great help in determining if any sensors are bad!

RCP Phx said:

A scanner would be of great help in determining if any sensors are bad!

Click to expand...

You mean an OBDII scanner? Wouldnt the check engine light come on if I had any codes?

An OBD scanner with live data, or a smart phone scanner App lets you monitor coolant temperature. All OBD sensors can be tested to see if they still function within their specifications, however all sensors have a service life and may read erratically before they fully fail.

I think that basic testing and basic cooling maintenance for ALL of the cooling system parts needs to be completed before you start worrying about potential/theroretical internal mechanical issues. You need accurate data to diagnose.

Tim_MN said:

An OBD scanner with live data, or a smart phone scanner App lets you monitor coolant temperature. All OBD sensors can be tested to see if they still function within their specifications, however all sensors have a service life and may read erratically before they fully fail.

I think that basic testing and basic cooling maintenance for ALL of the cooling system parts needs to be completed before you start worrying about potential/theroretical internal mechanical issues. You need accurate data to diagnose.

Click to expand...

So what do you recommend for basic testing and cooling maintenance? As i said, CSF 2 row all aluminum radiator, thermostat, fan clutch, and 97+ electric fan were installed a year and 2 months ago. The system was flushed before install and 50/50 coolant was added. Water pump and belt were installed last weekend. Coolant came out looking good when I drained it for water pump install. Verified electrical fan comes on with AC and at about 220.

Only overheats at highway speeds with AC on and under load (i.e. 3rd/4th gear going 2-3.5k rpm at about 65-75mph).

Will install a new radiator cap today just to make sure but if mine was bad shouldnt i see a fast increase/decrease in temperature range.

What else can I do?

SEXJ96 said:

Only overheats at highway speeds with AC on and under load (i.e. 3rd/4th gear going 2-3.5k rpm at about 65-75mph).

Click to expand...

Are you sure there ain't a problem with air movement through the radiator? or maybe a bad radiator? Years ago I had a bad 2 row radiator that failed to cool the 4.0 at highway speeds. It was returned to CSF 2 weeks after I got it. The problem there was some blocked fins diagnosed by the touch method,...some of the fins were cooler than others. You said the fan shroud is intact, how about the fan clutch itself?

Do the condenser or the radiator have bent or blocked fins? You may need to remove the grill and check and while you are there clean them with a stream of water from a garden hose.

My XJ have dual electric fans for several years. At highway speeds the fans do not run unless the AC is on. Even a 13+ years old radiator which is beginning to leak do not overheat in South Florida summer on the highway or stop and go city street.

RCP Phx said:

There's no need to burp/purge air on a late style system. There is no "heater core control valve" and it has fluid constantly flowing thru it and the bypass at the t-stat housing.

Click to expand...

Depends on what your definition of a "late model" system is. Once had a '02 Sebring with a 2.7 (pile of junk) that had a bleed valve on the ECT sensor housing, and yes, the cooling system needed to be bled if re-filled.

'96 XJ still has a heater control valve. Never cracked mine, but apparently it's a common problem on '96's. Some guys will do away with it completely.

techno1154 said:

Are you sure there ain't a problem with air movement through the radiator? or maybe a bad radiator? Years ago I had a bad 2 row radiator that failed to cool the 4.0 at highway speeds. It was returned to CSF 2 weeks after I got it. The problem there was some blocked fins diagnosed by the touch method,...some of the fins were cooler than others. You said the fan shroud is intact, how about the fan clutch itself?

Do the condenser or the radiator have bent or blocked fins? You may need to remove the grill and check and while you are there clean them with a stream of water from a garden hose.

My XJ have dual electric fans for several years. At highway speeds the fans do not run unless the AC is on. Even a 13+ years old radiator which is beginning to leak do not overheat in South Florida summer on the highway or stop and go city street.

Click to expand...

I mean the radiator was fine when I was in california even at highway speeds never went over 220 but just recently its been heating up on the highway under load and I assumed it was the water pump since I had play in the pulley but that didnt fix it. As far as the condenser and radiator, the condenser has a little bit of blockage between the fins but the radiator looks fine but I should remove the grill and clean them anyways. Would you suggest spraying them from the inside out or vice versa? I would assume inside out is better...Hopefully ill be able to get back to that point and not worry about overheating anymore, its a total pain especially being my DD.

JeepNoob said:

Depends on what your definition of a "late model" system is. Once had a '02 Sebring with a 2.7 (pile of junk) that had a bleed valve on the ECT sensor housing, and yes, the cooling system needed to be bled if re-filled.

'96 XJ still has a heater control valve. Never cracked mine, but apparently it's a common problem on '96's. Some guys will do away with it completely.

Click to expand...

I think the PO deleted the heater control valve and routed the hoses straight to the heater, or the car did not come with one (96 is finicky year). I will have to check tomorrow.

**Updates:

On my way home from work the car started overheating again as expected, a turned heater and decreased speed and it hovered around 220-230 while I was baking inside the whole way. I got home and first thing I did was pressure test the cooling system. As soon as I hit 16 psi I hear water running as if someone had a faucet open. I looked around with a flashlight and it seemed like it was from the water pump. Ofcourse the pressure dropped pretty fast and within 20 minutes was down to 6psi. Upon closer inspection I noticed that the upper radiator tube connection to the thermostat housing had water on it, so I tightened the clamp and dried everything off and pressure tested again, viola, it held pressure pretty well and no noticable leaks. It dropped from 16 to 14 psi in 20 minutes. What do you guys think of that?

I took the pressure test kit off and started the car, no bubbles or anything, waited for it to heat up (blocked the airflow infront of the grill with newspapers for faster heat up) and once it reached around 200 the thermostat opened up and the water level dropped. At least I know my thermostat isnt stuck closed/open however Im thinking that maybe, maybe, it doesnt open up all the way thus not enough demand for highway speeds? What are your thoughts?

I also used a block tested in the radiator neck to test for exhaust gases in the coolant (I know its not very reliable but I this point Im willing to try everything and anything) and the fluid did not change color. The car was running and it heated up to running temp (around 200) and still no color change. I did not drain a lot of coolant but I made sure I wasnt sucking up coolant in the block tester.

I then ran the car, waited for the thermostat to open and filled it back up with coolant while it was running (cap off ofcourse) and put on a new radiator cap becuase I couldnt find an adapter in the pressure testing kit that would fit my cap to test.

My next thoughts are either swap both temperature sensors (any brands recommended so I know I got good sensors in there), or get an IR temp gun, or some live OBDII scanner to be able to verify that I'm actually overheating and that my sensors and gauge are giving me an accurate reading.

Any suggestions based on todays tests? Maybe running a compression?

I would really like to get this issue fixed as its a total pain sitting on the highway with the heater blasting, getting baked and its my DD so I have to use it to commute every day.

Thanks guys, I appreciate your help and sorry for the long post.

Take a good look at your fan clutch, you may have replaced it but "their not all created equal".