Remember That Jeep Cherokee That Blew Up After I Sold It To A Nice Lady? I Sent Its Oil To A Lab For Testing And Now I’m Confused

Image (58)
ADVERTISEMENT

A few weeks ago, I sold my Jeep Cherokee XJ to a nice lady who had driven five hours from Chicago. Twenty minutes after driving my gorgeous 1991 five-speed XJ from my property for what I thought would be the final time, she called me. “The Jeep is overheating and won’t move.” I towed the Jeep back to my house and took an oil sample just before discovering that the water pump had failed and apparently blown the engine’s head gasket, that I was in for a hell of a wrenching job. I sent that sample to a used oil-analysis lab; here’s what it found.

I’ve written about Fort Wayne, Indiana-based Blackstone Labs quite a few times before, because it’s a service I use to make sure my engines are in top running order. On multiple occasions, the company’s findings have helped me diagnose intake manifold leaks (which manifest themselves in elevated silicon levels in the oil — dirt and dust are silicon-based) and head gasket failures/cracked heads (which manifest themselves in elevated sodium and potassium levels – these are common additives found in engine coolant). The $30 reports also help me understand if there’s too much fuel in my oil (indicating a leaky injector) and if the motor is wearing down too quickly for whatever reason. So naturally, I was curious to see what I’d find in an engine that had just “blown up.”

At least, that’s what I thought had happened; the loud “knocking” sound coming from the block seemed like a bad rod bearing, but as I later found after a bit of diagnosis at my house, the source of the noise was just a bad water pump. Unfortunately, that water pump had caused overheating, and as I found during my compression test, this had killed the compression in multiple cylinders.

So let’s have a look at the used oil analysis datasheet to get an indication of how badly the overheating and failed head gasket/head might have damaged this beautiful Jeep Cherokee XJ’s engine:

Okay! So the second column shows the elements, in parts per million, that Blackstone found in my oil, while the third column represents how much of each element Blackstone typically finds in a 4,000 mile oil sample. Keep in mind, the oil in my Jeep’s engine only had about 150 miles on it, so it should have much, much fewer wear particles (though additives should be similar).

Unfortunately, as you can see, aluminum, iron, lead, and silicon were at levels even higher than those Blackstone normally sees in 4,000 mile samples. The Jeep 4.0-liter, and pretty much all gas engines, has aluminum pistons, though the rest of the engine is made of iron — except for the bearings, which are composed of lead, tin, copper, and other materials.

So the 14 ppm of aluminum versus the four normally seen in a 4.0-liter engine with 4,000 miles on it is a bit concerning. Hopefully my piston skirts are still intact! The 64 ppm of iron — over 2.5 times as high as what’s seen in a 4,000 mile sample of oil — ain’t great, either, so I’m hoping my camshaft lobes and piston rings are okay.

The lead, copper, and tin aren’t great, either. Lead is twice what Blackstone normally finds in a 4,000 mile sample, copper is just below typical for 4,000 miles (but again, my engine has 150 miles on it), and tin is showing a reading when it normally doesn’t. So there’s definitely some bottom-end bearing wear that resulted from this short 20-minute trot.

As far as oil contamination, part of the wear is possibly a product of dirt getting in somehow, and some might be a result of low viscosity (8.48 cSt at 100C is more like a 20-weight oil than the 30-weight oil recommended for this engine) resulting from some fuel dilution (I have no clue how fuel got into that oil), but what’s weird about this sample is that there’s something missing.

Where the hell is the sodium and potassium?

If this Jeep blew a head gasket — which my ccompression testing suggests was the case (I certainly didn’t blow holes into three pistons) — then how did no coolant get into that crankcase? I could see how dirt might have gotten in, since the pistons are now sucking air in through the block/head interface instead of through the intake manifold, and I could see how that plus too-hot oil could have caused elevated wear of rings, bearings, and pistons (heat was obviously the overwhelming contributor to wear). But how was coolant contamination not a factor, here?

I’m surprised.

I’ve had head gasket failures without obvious oil contamination (the oil still looked good (i.e. not milky), and would only slowly creep up the dipstick), but every time there was any coolant loss in one of my vehicles, Blackstone detected at least a trace of the stuff in the oil. So I called the company to talk me through this.

“I would be surprised if [a head gasket failure] was the problem based on potassium and sodium,” the representative told me initially, looking at my XJ’s report. “There probably is a problem in the cylinder area…The levels of aluminum and iron suggest quite a bit of cylinder and piston wear.”

The rep then asked me what made me think there was a head gasket leak. I told her that an overheating incident followed by loss of compression in multiple adjacent cylinders is pretty much a dead-ringer for a bad head gasket.

The rep explained that, generally, anything over 20 ppm of either sodium or potassium is a red flag for a coolant leak, and though the analysis didn’t detect any coolant, that doesn’t necessarily mean the head gasket didn’t fail. The added wear, she said confidently, is not from coolant in oil, though; it’s from elevated engine temperatures causing more friction in metal-to-metal interfaces.

What might have happened, the rep and I agreed, is that the Jeep just spewed all of its coolant out of the water pump (I tried filling the cooling system after the incident, and it all poured out of the failed pump, directly onto the ground). The strange thing is: If there was no coolant in the cooling system at the time of the overheating incident, then why is my overflow bottle filled up so high?

Hmm.

I’m going to guess the engine ran low in coolant due to the bad water pump, started overheating, blew all of its remaining coolant into the overflow bottle and onto the ground, and then the head gasket failed, possibly after the buyer parked the Jeep. So there was just no coolant to get into the cylinders/crankcase, plus there wasn’t a lot of time that the engine was running after the gasket failed.

We’ll find out when I tear into the motor. I’m hopeful that a new gasket/possibly a new head will get this vehicle back onto the road. I’m not too concerned about the elevated wear metals, since they’re roughly in line with what one would see out of three consecutive 4,000-mile oil changes. Not great, but not horrible. At least, that’s what I’ll tell myself so I can sleep the next few nights before I tear into that motor.

49 thoughts on “Remember That Jeep Cherokee That Blew Up After I Sold It To A Nice Lady? I Sent Its Oil To A Lab For Testing And Now I’m Confused

  1. I’m stunned how quickly the AMC 6 ate itself just from lack of coolant. I know from my friend’s 87 that the 4.0 can build a fair amount of heat in the engine bay but to die in half an hour seems fast. Really makes me wonder how chewed up my father’s 2.3 Lima is after I learned he drove 600+ miles with no coolant after a belt sawed its way through the upper hose. It explained why he lost heat when doing his mountain pass trips. The thing runs fine but I haven’t checked compression since… definitely have to do that and a Blackstone test next oil change.

  2. “The $30 reports also help me understand if there’s too much fuel in my oil (indicating a leaky injector)….

    Can someone explain how a leaky injector can cause fuel to mix with the oil? I’m really lost on that.

    1. Fuel is much thinner than oil, and thus moves past the piston rings more freely than more viscous oil will. A leaky (stuck open) injector will dribble fuel into the cylinder, which works its way past the rings into the crankcase.

  3. Love the call out to Fort Wayne! To my Michigan friends (go blue! BTW), us NE Indiana nerds are as exactly as terrible as you reckon them to be… and probably worse! Hey, at least our baseball stadium kicks ass. 😛

  4. Just got a reman 0331 head yesterday for my 2000 xj (no tupy available). Truck has 221k on the clock. Once this is done should be good for another 100k at least.

  5. It is also worth noting that while the length oil has been in your engine does have an effect on the trace elements in an oil sample it is not a huge one compared to actual engine life, oil change history and, when in the draining process you take the sample.

    All the foreign material in the oil falls to the bottom of the pan and is stirred up when operating; oil that has been around the block twice and drained will be closer in composition to the oil that was just drained than fresh oil out of the bottle.

    Oil samples should come from a warm engine about halfway through the draining the engine; catch the last bit of oil and you skew the results to the high end contaminate wise.

  6. A head gasket blown between two cylinders and nothing else is a very, very common type of cylinder head gasket failure and the majority of head gasket failures you see from an engine that got hot and ran dry of coolant are of this type.

    Think about it, the highest pressure generated in an engine is the combustion pressure and usually, the smallest gasket area is between cylinders.

    It’s weird you don’t know this…

  7. From past experience with 4.0’s I’ve had one with a cracked head and blown gasket that only let oil go into the coolant and not let coolant into the oil. The most recent one was the exact opposite and that one had an absolutely catastrophic failure. So, you may be wanting a coolant test now to see if you get the joy of replacing a full XJ cooling system. I did mine in the winter so I may be remembering it as even worse than it actually was.

  8. It could have a crack in the head that’s nowhere near the water passages, too. I’ve seen that before. Have to be a hell of a crack to span three cylinders, but I guess it’s possible.

  9. I would actually expect the overflow to be full. Let’s say it’s not normally quite that high. Nice Lady drives away, engine warms up to operating temp, coolant expands and some amount goes into the overflow. KABLOOEY… water pump explodes… Coolant everywhere… the cooling system is no longer closed (and it’s already depressurized) so there’s no ability for the system to suck from the overflow. Or am I totally missing something obvious?

  10. David, do you own a decent leak down tester? If so, just pressurize the cylinder and see where the air comes out. I blew a headgasket on my S6 a few years ago, and the oil was perfect. It was the coolant that got gross. Sometimes HG leaks only go one direction, aka, oil gets milky but coolant is fine, or vice versa.

  11. With the high tin and boron, I’d wonder if someone had welded that block to repair a crack or something, anf the original weakness was never really fixed.

  12. Blackstone is a terrific resource. They spotted the likely beginning of a head gasket going bad on my Evo at around 200K on the clock. It started with disappearing coolant, but not with the telltale visual cues of it in the oil. They found the elevated levels of sodium/potassium and said that I should probably look into the head gasket soon. Did just that, and while the head was off, a minor valve job was done along with stouter head studs. Pistons and cylinders looked very good according to my mechanic at the time. Car has almost 320K on it now and I’m due to send off another sample to see how it’s going (probably should do a compression and leak down test too, to get a clear picture of the innards health).

      1. I’ve had it since 2005. Mildly modified with nothing more than a downpipe, high flow cat and cat back exhaust, a better drop in air filter, a more stout recirculating BOV (to replace the OEM plastic one), some revised ECU tuning (the stock programming runs VERY rich, which is typical Mitsubishi being conservative) and I scored a set of MR Bilsteins to swap in for the harsher KYBs. I would’ve loved to do more, but keeping the modifications mild and looking after the maintenance very consistently has likely kept the car in good running shape. As stuff breaks, I try to replace it with OEM or better. The radiator went at about 150K, so in went an aluminum Koyo. The original turbo checked out at 240K so I sourced a used VIII turbo (had 40K on it) and fought the urge to go with something more potent. At that point I figured that doing so would start exposing potential weak points elsewhere that I wasn’t ready to deal with. Last things I replaced recently was a broken rear sway bar and I had an ignition coil going bad, so I found a really good full coil on plug set up to replace the stock ones with (OEM uses two coils to drive two cylinders each).

        It’s been, and continues to be, an absolutely terrific car. After all these years I still love to drive it. The bulk of the milage was from my work commutes that were 30 miles, one way. Now that I work from home, I still try to get it out on the freeway every week to “keep it fresh” and blow off some of the cobwebs of around the neighborhood driving. Car is practically a family heirloom. Both my kids came home from the hospital in it when they were babies, it’s been to track days, a bucket list drive up PCH to Laguna Seca (to watch an ALMS series race), and countless backroad therapy sessions. I hope to keep the lovable beast as long as possible.

        1. I was just thinking the other day that one thing I missed from the car forums from the the early aughts was the “ride of the month” features. Maybe something similar could be done here? Your Evo would be a fantastic candidate!

      2. I don’t know, I’ve had great luck with Mitsubishi 4 cylinder engines. The ones I abused the most were Dodge Colts, one a turbo, one not. The turbo went nearly 300,000 hard driving, pizza delivery miles and ran like new when sent to the junkyard. The other went 165,000 miles, also delivery, also running like new when the transmission gave up. Those were great little cars. Synthetic oil every time, though.

        I wouldn’t have a Mitsubishi V6 again. I don’t know how they did one type of engine so well and borked the other so badly.

    1. I’ve started using them on our Cruze Diesel to see how long I can push oil change intervals out with non-factory oil. I was so impressed that I plan to always send a sample in from any of my/our other cars, as well. Spending the few bucks each time is well worth catching a potential problem early before it turns into an absolute disaster.

      1. I agree with you about going longer between oil changes, I use Amsoil.com for the 25,000 mile warrenty, after changing from whatever anybody’s guess that the dealership puts in to sell it. (Black as tar when draining) to Amsoil it comes out only slightly darker than new.

    2. Blackstone doesn’t actually test for some things, like fuel dilution, even when they give you a number.

      I found this out a long time ago and started using OilAnalyzers and never looked back.

      Check BITOG (BobIsTheOilGuy) for details.

  13. Minor pedantry here: silicon and silicone are not the same thing. If you have detectable levels of silicone in your oil, contact your doctor as one of your headlight implants may have ruptured.

  14. Can you drop the oil pan without removing the engine? If so, just drop the pan and check one of the main bearings. Also, when you replace the head gasket, you’ll be able to assess the pistons and cylinder walls

    1. You can. Lift it up to let the front suspension sag, then loosen the bolts on the transmission mount. That will cause the engine to lean back just enough to provide room to wiggle the oil pan out.

  15. I don’t know why people keep raving about the durability of the 4.0. In my experience they are no more dependable than any other engine of the era.

    1. Because you have no experience with the 4.0 or the GM 3800 I would assume.
      I did warranty and non-warranty on both. A lot of them. Far more than David has even seen. Know how many times I had warranty or recall work on a 4.0 that was an actual mechanical problem? Once. For a loose exhaust manifold. That was it. Replaced an entire fleet’s worth of failed injectors, a couple MAPs, your usual small potatoes electrical. A few leaks here and there.
      That was true of Jeeps with 4,000 miles, 40,000 miles, and 400,000 miles.
      It was also true of the badly neglected ones with blown head gaskets. Check the head for warp or cracks, recondition or replace, and right back on the road. It was true of the ones where the water pump had failed over a thousand miles ago but they were oblivious due to a failed CTS they were too lazy to replace. (So, so many CTS jobs. Seriously. WTF people. It’s a 10 minute driveway job.)

      Actually killing a 4.0 shortblock? No. That doesn’t come easy, if at all. The 4.0 is so ridiculously bulletproof because it really is, no matter what Mopar says, based on the AMC casting. Despite increasing the bore (3.75 to to 3.875,) the cylinder walls are still absurdly thick – enough to take a 3.917+020 as I recall. (Might only be 010 at 3.917.) The webbing for the very first ones was set up for a 3.895 stroke, when it was reduced to 3.414. Which Chrysler then made even stronger later on. Just to be damn sure they had no leaks, the valve cover got 15 bolts. 15. (It’s a shitty job when it does leak, yes.)
      It literally was designed and required by AMC to go no less than 100,000 miles with no mechanical faults. The GM 3800 got there by accident; the 4.0 got there VERY intentionally. Consequently, every ‘minor’ tweak just made it run longer, cleaner, and simpler. The 4.0 is one of the only if not the only engine manufactured in 2006 that does not have an EGR – it doesn’t need one.

      But when the 4.0 does fuck up? It does not fuck around. Because it takes so much to break it. Case in point? It’s not a spun bearing.
      Based on symptoms, analysis, and coolant failure with low fuel contamination? I’d guess at cylinder bore failure. (4.0’s are wet blocks, closed top.) The water pump is directly bolted to the cylinder coolant space and likely a marginal gasket blew from overpressure. Really though, the magnesium is the giveaway – I believe the only place you’ll find any significant amounts of that in the engine are the pistons.

      1. I actually have a little experience with 4.0 engines, every one had rod bearings with visible play. Your definition of ‘not bulletproof’ and mine may be different. For most of the vehicles I was dealing with the cost to rebuild what was wrong far exceeded the value of the vehicle, that’s dead to me.

    2. The same reason people swear by the Toyota 22re which will crack its head *and* block and blow a head gasket if you give it a hint of overheating. I’ve seen them do it without ever touching the red on the gauge and with lower miles. 200k miles? Borrowed time.

      Virtually every old Toyota you see on Craigslist says “engine rebuild xx,xxx miles ago” or “new engine”, “needs rebuild” or “timing chain problems” etc BUT THEY LAST FOREVER BRO AND THEY USE THEM IN AFRICA, so…

      1. No, an LSX. If you are gonna commit blasphemy, go all the way. The LSX is pretty reliable (they run badly longer than anything else).

        This thing is nearly as bad as Kinja. It keeps telling me I have already posted the comment.

Leave a Reply