How I Finally Fixed The ‘Jeep Cherokee From Hell’ That Broke Down 20 Minutes After I Sold It To A Nice Lady

Last night I found myself in a dumb situation only I could have gotten myself into. I had sold a beautiful 1991 Jeep Cherokee to a nice Chicagoan named Tracy, who only managed 20 minutes of her 4.5-hour drive home before her new vehicle blew up. I took the Jeep back, promising Tracy I’d bring the venerable 4.0-liter engine back to life. She took my word for it. She believed in me. She ditched all of her other vehicles, and was now depending on me, giving me a deadline of June 14 or else she’d be carless. Last night, I realized that today I’d be flying to California to drive the 2023 Ford Bronco Raptor, meaning I had to fix the Jeep right now or Tracy was screwed. I wasn’t going to let that happen without a fight.

I write to you utterly fatigued from a cramped airplane on its way to Palm Springs, where I plan to off-road the crap out of a 2023 Ford Bronco Raptor. I’ve got a half-inch bloody gash in my left pinky, oil underneath my fingernails, and — though the woman sitting next to me hasn’t mentioned it — most likely a light smell of ethylene-gycol-based engine coolant emanating from my pores. I know I look like shit, my body definitely feels like shit, and yet somehow I’m on top of the world right now — not because of the Bronco Raptor, though that’s likely to be epic, but because last night I broke out every weapon in my toolbox and finally slayed the mechanical menace that is my beautiful 1991 Jeep Cherokee five-speed.

The overheating episode shortly into Tracy’s drive home had warped and cracked the Jeep’s cylinder head, killing compression in four of six cylinders. The vehicle ran on the remaining two, but it released a sound so traumatzing that I was forced to discard it from memory.

I zipped a cylinder head off the parts engine sitting in my garage, sent that into a machine-shop to be checked and shaved flat, and bolted that lid onto the exposed iron heart nestled between the shock towers of a truly gorgeous but thoroughly flawed AMC-designed box-on-wheels. Unfortunately, despite the engine now running well with its new head, the motor continued overheating and appeared to be gaining oil. I feared the engine had cooked itself during Tracy’s short drive and the block was now cracked.

There’s a reason why I feared the worst. I’d seen these exact symptoms before: A slight increase in oil volume on the dipstick accompanied by chronic overheating. And on all occasions, I’d sent an oil sample to Blackstone Labs, who told me that indeed, there was coolant in the oil. Check out these three Blackstone reports from a 1992 Jeep Cherokee, a 1995 Jeep Cherokee, and a 1996 Jeep Cherokee:

All of these Jeep Cherokees ended up requiring new cylinder heads. But since I’d just replaced the 1991 XJ’s head, it wasn’t a stretch to take the overheating+gaining oil symptoms as a sign that the block had been compromised.

I’d pressurized the cooling system with a hand pump, and found that there was a slight leak (the pressure shown on the gauge dropped slowly). But I noticed no obvious drips, so perhaps it was internal? A cracked block would explain this pressure drop, and it would explain the overheating and the increase in oil level. But why was the engine making good compression? Couldn’t a cracked block compromise this? Why wasn’t there any steam coming out of my exhaust? Why was this increase in oil not accompanied by the signature milky consistency of a water/oil mixture? Why didn’t my block testing die detect any combustion gases in the cooling system?

I was still not convinced I had a block issue.

I thought about this Jeep all day yesterday when I really should have been writing/editing/running this business. I asked myself two questions. The first was: Am I 100 percent sure that the engine is gaining oil?

After installing the new head, I’d run the engine and noticed the oil level was elevated well above what six quarts tends to read on a Jeep 4.0 dipstick. But then again, I had poured some Marvel Mystery Oil into those cylinders, and some might have gotten into the crankcase. To get rid of that factor, I then conducted an oil change, but since I feared coolant was getting into my oil, I didn’t want to waste my good stuff, so I just grabbed some spare jugs I had sitting around. I used an old gallon jug of Rotella 10W-30 to measure out six quarts of cheap lubricant, but had I been precise? Was I truly 100 percent certain that this engine was gaining oil?

On top of that, since the oil didn’t look milky, could that have been fuel in the oil? I poured some fuel injector cleaner into my gas tank, and revved the living crap out of the engine. Accelerating that little Jeep made me realize just how magnificent the machine is. There’s just so much power and torque in just a 3,350 pound package; when this engine debuted in 1987, it must have blown people’s minds.

The second question I asked myself yesterday when I was daydreaming during work hours was: Why is the Jeep only overheating under certain conditions? It seemed that high-speed highway driving and idling were the worst-case situations. The former involves lots of engine heat rejection, the latter involves little coolant or airflow.

As I’d previously had issues with coolant temperature sending units on both Jeep Cherokees and Grand Cherokees, I dropped the $8 on a new one to make sure i was at least getting the correct readings on my gauge. Deep down, I knew the sending unit wasn’t the problem, because the temperature gauge was reading right in the middle under medium-speed, low-load driving conditions — especially when the revs were above 2,000 RPM. If the needle reads appropriately under not-particularly-taxing thermal conditions, then the sensor is probably doing its job, and the issue is actually with the system’s thermal capacity.

My soccer game on Monday night had given me a lot of information. I’d driven roughly 40 mph on back roads the whole way home, and the needle stayed firmly in the middle, only elevating when I stopped at my house and let the motor idle a bit. “This cooling system’s ability to reject heat seems to be compromised,” I thought. Sure, a cracked block would prevent the system from reaching its 13 PSI pressure, and could cause the vehicle to overheat, but the pressure leak wasn’t really that bad; I was able to pressurize the system with that hand pump, and the needle remained at 13 psi for quite a while before dipping down to 12. I’d driven a Willys Jeep across the country with a cooling system pressure of fairly close to ZERO PSI (there was a little leak in the radiator), but as long as I kept the rad full, the engine never overheated. I’d had pinhole leaks in radiators before, and those engines still stayed cool as long as they had water (“water,” by the way, is what people in the auto industry call engine coolant, by the way) in them.

More importantly, as I sat there at my house at 11 PM after my soccer game on Monday, with the AMC straight-six idling and me staring at the gauges like a weirdo, I slid my heater lever to the right and immediately felt those glorious joules that were once contained in gasoline blow through my vents. The temperature needle dropped to the middle. Hmm.

I thought about all of this yesterday after work when I realized that I’d be returning from my Bronco Raptor trip the last weekend before Tracy’s deadline. There would be no time to continue diagnosing the Jeep when I got back. Tracy needed a Jeep by the 14th or she’d be walking, and I couldn’t drive to Chicago during the week due to work. I had to deliver this Jeep to Chicago as  soon as my flight landed from California, meaning I understood last night that I had to fix the Jeep RIGHT NOW. That was not the greatest realization, I will admit.

 

I ordered a radiator at about 6 PM, picked the plastic-and-aluminum heat exchanger up from the store, and began installing it. In my head, it was plausible that the oil increase on my dipstick had been either a leaky injector or just me being imprecise in my measurement, because in the past couple of days (including my drive to and from my soccer game) the Jeep appeared to have gained no oil whatsoever. As for the overheating, I reasoned that if all it took to keep the temps down was a little bit of assistance from my heater core, then maybe a more effective radiator would do the same. So I was praying that my only issue was radiator-related.

A reader named Jeremy had messaged me on Instagram to see if I wanted to meet up, as he was in town from Toronto. We’d agreed to go to a bar, but I switched plans on him. He generously came to my EPA Superfund Site and assisted with my radiator replacement, which would end up costing me a reasonable $135 in parts, all told. The job took about two hours; I removed the entire fiberglass header panel (which contains the lights and grille), took off the upper radiator support beam, undid the coolant hoses from the radiator, removed the electric fan, unfastened the mechanical fan shroud, and plucked the old radiator right out. It appeared to be in fine shape, but it also looked to be the original unit (based on the Chrysler stickers present). That’s an old heat exchanger.

I had to drill into two of the brackets on the new radiator to make it fit, but — thanks to Jeremy’s help lining the bottom mounting posts up with the holes in the lower radiator support in the Jeep’s unibody —  the installation job was largely drama-free. At around midnight, we had the cooling system back together and were pouring gallons of antifreeze into the system.

Then we hit the highway, which was empty at one in the morning. I punched the accelerator pedal until the Jeep was driving 75 mph, generating lots of heat from that 4.0. Then I waited for that needle to rise from its center position, slowly but surely. But it refused. It sat there, unwavering. I hammered the gas pedal a bit harder, but the needle remained pointed at 210. I then pulled off the highway and idled the Jeep; this is typically a taxing thermal condition because all the heat that the engine creates in order to propel the car at high speeds still needs to be rejected (it’s not an instantaneous process), except now you’re at idle with low water flow and low airflow.

The Jeep didn’t give a damn. It continued to run at its normal operating temperature.It wasn’t the relatively cold ambient temperature outside giving me false hope, either; the prior cool night, after my soccer game ended around 11 PM, I noticed overheating at idle. But now, no matter how long the Jeep sat stationary, the new radiator had no problem transferring that engine heat to the incoming air. My overheating issue was gone. It was a clogged radiator.


In retrospect, the coolant coming out of the old rad was pretty damn brown, so maybe I should have known. But I’d been burned so many times by overheating Jeep Cherokees gaining oil due to cracked engines that it was impossible for my mind not to go there.

After four years, the XJ from hell has been rescued. Now it’s time to get rid of it as quickly as I can before it tries to ruin my life further. I wouldn’t be surprised if it performed flawlessly for Tracy for years to come. I truly hope it does.