Subaru Overheating issue
2006 Legacy Outback • 2.5i • 148K miles: I was driving home after visiting my handicap sister one Sunday evening (as most other Sunday's). It was a beautiful day (55 degrees). I was 45º by now in the evening, and a little chilly so I turn on the heat. The vents blew hot air out at first, then it blew cold air (not cool) cold...That was my first indication something strange was happening! Then all of a sudden I saw the needle of the temp-gauge leave its usual "mid" position and rise up toward the red. I pulled over to check under the hood. The reservoir was overflowing and bubbling over (spilling out).
The next day the Subaru dlr said the Sub-radiator fan shorted. It needs to be replaced and you'll probably need head gaskets too.(I gave approval to replace the fan only).
The Subaru dlr called me a few days later, "Its fixed come and pick it up". I got home no problem (20 miles, the temp needle stayed in the middle as usual the whole way). That evening I went out. Same scenario occurred again. I turned on the heat and it went from blowing hot air to blowing "cold" air and again, I watched the needle leave its mid position and go up the red the same as before.
º No oil in the coolant reservoir
º Oil still honey colored (not milky)
º Leaks (very little oil on ground, No coolant though)
º No white smoke whatsoever
º No noise, no piston slap or knocking (very smooth)
I reported back to the Subaru dlr, they investigated and said, "All is working properly, you need to replace the head gaskets and also check for warped heads.
I have my doubts with my knowledge of rebuilding American V-8 engines many years ago. I asked if they did a compression check? They said, "Why, that won't tell anything". Is that so? (I thought "once a leak always a leak" no)? How is it possible that it can it leak at random?
If it is loosing coolant and it is not a leak, it is most likely a head gasket issue. When having a bad head gasket coolant sometimes goes from the head gasket leak directly to the combustion chamber and the problem will not put oil in the coolant and vice versa. With the poor history Subaru has with head gaskets, this is the most likely cause.
I understand the poor history of the Subaru "head gasket" issue but I am not completely convinced there is not something more going on here which I haven't pinpointed yet! Especially since the overheating issue only occurs when I turn the heat control knob on (then I can watch the temp gauge needle rise). This exact same scenario happened twice so far. I will pick it up from the Subaru Dlr today and see if it happens the third time. I'll go from there?
Interesting observation: The Subaru Dlr came back the second time as I noted above, "You need to replace the head gaskets sir". However, I was not completely convinced and I didn't settle for that answer right away - or even the second time for that matter! When I picked up the car tonight, I KEPT THE HEARTER KNOB TURNED ALL THE WAY "OFF" for the entire trip home (12 miles) [I swore not to touch it - at all]! The gauge reached it's normal position (middle) within three miles (35º ambient temperature), I took back roads for safety reasons incase of an emergency pull over (no Hwys). However, it never overheated, not even close! I remained in driveway at home and left the car running for about half an hour (while watching under the hood for any strange occurrences) - but nothing out of the ordinary happened. Both fans finally started running towards the end of that thirty minutes. The temp gauge still remained in the middle and didn't budge! Wondering: How can a blown head gasket/s mysteriously repair themselves? Hum Through my internet investigation, one article pointed out, "If all the air is not burped out of the cooling system (some air could remain in the heater core), it could create a vapor/air lock when the heater knob is turned on requesting hot air in the interior, whereby, flipping a switch and create the thermostat to close (when what is inside the heater core is released. Resulting in a overheat situation, which mimics blown head gaskets! Conclusion so far: Blown HG's do not seem to be the problem, but more testing tomorrow.
Good one. Listen carefully. First the smaller corrections: turning the heater control on Subies ONLY opens an air vent door allowing passage of air past the always-plumbed (valveless) heater core. IOW the engine coolant system does NOT know where you've set the control. So that makes diagnosis a bit simpler; the fact that turning the heat "up" affects the tipping point of coolant thermodynamics is simply secondary, and based upon internal temperatures of the coolant bath; the instantaneous pegging of the temp readout gauge is a function of superheated gases trapped near the heat sensor probe...possibly in sync with your "jiggling" the system with the minor change in heat-sinking via your knob gyrations...but that's immaterial. No voodoo there. The important stuff: Subaru 2000-2009 2.5i SOHC motor head gaskets MOSTLY fail because too much coolant loss from a minor breach of the HG (usually at the #4 cylinder) results in sufficient accumulative loss that the motor overheats. replenishing the coolant is usually sufficient to temporarily remedy the condition unless the motor overheats long enough to "blow" the head gaskets INTERNALLY. That's the key. Like the prevalent catastrophic (Mt Vesuvius-like) overheating scenarios very common with the first gen '96-99 DOHC 2.5i, we've learned that a minority of the newer SOHC 2.5i can also have their HGs fail between the combustion chamber and the coolant ports in the head, resulting in an initially intermittent "percolation" of the hot combustion gases at high pressure (obviously) into the coolant system, "percolating" bubbles of exhaust gases into the coolant. If temps are low (ambient and/or short duration travel) the system may be able to tolerate some of this behavior, but eventually the tipping point is reached where the coolant system (mass and radiator) can't sink all the hot bubbles, and Mt Vesuvius occurs at the radiator cap. Most owners learn to allow the system to cool, then replenish the radiator...even learning to purge ALL trapped air bubbles by topping off after each high way run...only to find another explosion occurs later on. Eventually the system becomes sufficiently "porous" that constant bubbling of gases can be seem through the (carefully opened!) radiator cap at idle...or even in the expansion tank. At that point it can be difficult to drive even a short distance before another exasperating boil-over. The ONLY solution to this high-pressure HG failure is complete replacement of both HGs, paying careful attention to machine- shop service for examining the heads for warpage and cracks. You'll end up using the newer 3ply HG that seals better from the ubiquitous EXTERNAL coolant and/or oil seepages that plague all of these motors, sometimes innocuously as owners learn to keep their coolant levels up and/or tolerate the stench of an atmospheric oil leak onto the hot cat on the passenger side. Given that you have NOT evidenced a coolant leak on the driver's side, AND your description of the overheating landscape, I'm nearly 100% convinced that you have the "old-style" high pressure internal HG breach. You can get another fairly definitive test response by having a hydrocarbon sensor "sniff" the bubbles emanating from the coolant (rad cap or expansion tank if visible) to show exhaust gases present in the coolant. That nails the diagnosis. In other less defined scenarios I've seen both blocked radiators or blocked heater hoses restrict coolant flow sufficiently to cause momentary overheating either on the highway (leaves/mud/debris blocking the radiator cooling fins surface area), or even in the city (blocked rear heater hoses). So before you bite the $1500-1800 bullet, at least spray-clean the front radiator surface of debris, and disconnect and with a garden hose BACK-FLUSH the two heater core hoses under the windshield. If you see granular or sludge- like junk in the wash then you know someone tried to stop the overheating with the wrong coolant additive that collected up at the hoses to block coolant through them. (REMEMBER: the heater core is permanently plumbed in inline with the cooling system...NOT a parallel arrangement with control valves...so a blocked heater core WILL take down the system!) If you're convinced that the coolant flow is normal through the rad and core and the overheating continues then it's the HGs...especially if you see those dastardly bubbles! (Note: to aid in testing for percolation, as well for purging air from a recently-filled coolant system, I attach a wide, tight-fitting kitchen funnel to the radiator orifice, ading water/coolant to create a wide, level surface for ease of visual testing.. Get the motor to warm idle, turning the AC fans ON to keep the system fairly isothermal so that the level in the funnel doesn't trampoline up and down as the fans switch on otherwise. You'll then have a nice mini"lake" to watch for what look like air bubbles that percolate up either steadily or in response to mild throttling. But be careful: once the system's quite compromised raising the throttle can result in so much combustion gas pushing into the system that superheated coolant/gas mix can shoot up several feet and burn you! In that case replacement of HGs is critical, as you probably shouldn't drive the car except ever so gingerly over to a wrench who's VERY experienced in this work. Note also that unlike the earlier DOHC 2.5i, the newer SOHC 2.5i "takes" to HG replacement in stride (if the heads are uncracked and straight, of course!), operating another 100k miles easily. Just make sure you didn't warp the heads, as they can take only about 0.004" planing before they'll overheat from being too small! So should you do it? I like '06-'09 OB 2.5i a lot, as the VVT and simpler emissions system introduced in '06 is a nice setup, and the aluminum hood used in '06-07 is a bonus in the salt belt. So unless you need brakes, tires, wheel bearings and CV axles all at once, replacing the HGs is a good investment. But if you need to add another $2k for all the other stuff above I'd think twice. A good analogue to your situation I've used for naive customers is to imagine boiling a lidded pot of water when cooking pasta: sometimes it boils over, sometimes not, depending how much of those air bubbles can escape or be absorbed by the water/pasta mass and how much the heat is on...and the lid is cocked. Each time you drive now you're playing that boil-over tipping point game, as you no longer have a sealed "pressure cooker", eh? Hope this helps. Ern
Note that your HG breach almost NEVER occurs into the oil passages, so you won't see contamination there. Indeed the HC bubbles are invisible, but sometimes longterm percolation can result in staining the inside of the expansion tank with exhaust resudue (just like a tailpipe). Note also that unless you've replaced the OE big thermostat with a cheaper, smaller-orificed aftermarket one, the OE t-stat is NEVER the culprit in this scenario, so don't bother to replace it if it's the big OE one. I've tested so many of these and NEVER found a bad one, so don't waste your time. Also note that Subaru water pumps are also ultra-reliable, so when you replace the HGs DO of course replace the t-belt (since it comes off!), and test its tensioners, but leave the water pump alone to save a few bucks. Sounds like heresy, but remember that Subies are unlike Hondas, Toyotas and VWs that eat water pumps for lunch.
Thank you for your excellent reply to my overheating issue. You were very poignant with several possibilities, evaluations and tests which warrant further investigation. Guru: You can get another fairly definitive test response by having a hydrocarbon sensor "sniff" the bubbles emanating from the coolant (rad cap or expansion tank if visible) to show exhaust gases present in the coolant. That nails the diagnosis. Skydiver1500: I like definitive answers! I would like to do this test. Is the "Sniff" test something that a rookie can do? Do I need to go to a garage for this test (hydrocarbon sensor)? Guru: In other less defined scenarios I've seen both blocked radiators or blocked heater hoses restrict coolant flow sufficiently to cause momentary overheating either on the highway (leaves/mud/debris blocking the radiator cooling fins surface area), or even in the city (blocked rear heater hoses). Skydiver1500: There absolutely is NO dirt or debris blocking radiator fins. The Subaru Dlr replaced one of the radiator fans (Sub-fan they told me) because it was shorted (however, it's working now). Everything was very clean when I got it back! I also watched both fans running too since it's return. Both heater hoses at the firewall are very warm when the car is running at normal temp and I wrap my hand around each one of them, which proves that coolant is flowing though the heater core (even though the heater control knob is off and interior fan is turned off as well). No Voodoo! Is it possible that there is an air pocket remaining it the cooling system? What is the solution to burp all the air out (purging all air out of the cooling sys)? I'm almost certain that the heater core is clear because the air coming out of the vents went from very hot to very cold during an overheating situation Sunday night, and again on Monday when I drove it in to the Subaru Dlr
The heater core sits a bit high in the coolant system, so increasing volumes of exhaust gases percolating into the coolant rise to settle in the core hoses, allowing the core to significantly cool quickly as blower air cools the gas-full core, whereas a liquid-filled core would cool from airflow MUCH more slowly (about 50:1 as I remember air vs water specific heat?). Therefore, to purge all gases (exhaust and/or air) in the system is a bit difficult, as Subie eliminated the radiator's opposite-end purge hole after about '99. MOST of the time a few highway runs and re-toppings of the radiator (squeeze the top hose to "burp it" too) are sufficient to purge all air. But to be sure, it's better to raise the front of the car so that the rad cap is higher than the rear core hoses, install a tight wide-mouth funnel on the rad orifice, and overfill it a bit with 50/50 mix at warm idle. Any trapped air bubbles should work their way to bubbling out the funnel "puddle". After a few minutes you'll be convinced that there's no more trapped air...IF you don't have ongoing HG-related percolation of course! (Note that I've seen a core hose filled with granular goup that MIGHT have operated as a one-way valve...in other words possibly allowing some hot flow under pressure, but causing blockage when static. Not sure that matters, but disconnecting and back-flushing these two hoses shouldn't be ignored, as it's easy. Note also that Subie heater cores tend to NOT block...perhaps helped by the fact that they're always in the loop, rather than operating via water valves only in the winter. Get back to me AFTER you run your test procedures. Ern
It might be easier to park facing significantly uphill to tip the cooling system rather than jacking the front end up. (We have hills here in New England!)
Is there a "best answer of the year" button? Cuz Ern deserves it! Jaw. Dropped.
I've decided to take on changing of the HG's! I am now waiting for the special tools to come in (should arrive Wednesday 3/17/16) in order for me to remove the harmonic balancer (crankshaft pulley), and the tool to hold the cam sprockets to remove them for machining. Once I get the heads off I will then find out if there was any HG leakage that caused the strange problem I was having and also which noone could answer (except for HG replacement). I will report back to you on my findings.
Oh yeah, I forgot to mention. As I was disassembling, I removed the thermostat. I put it in a pot of hot boiling water on the stove, and it opened fine. Then when it cooled off in open air (out of the water), it closed very smoothly...so the thermostat WAS NOT my problem in case you were wondering (because I was).
If the t-stat is OE size (big bore) you can reuse it. But if it's AFTERMARKET (small bore), although it may open and close at correct temperature, it will NOT allow full flow at highway speeds, and thus cause boilovers. So check it against a new Subaru one, or measure the orifice diameter...whatever. You do NT need special tools for remving the balancer. Just block it and crank the starter; it'll loosen quickly. When retightening use a big screwdriver or pry-bar to block the flywheel and then just torque it on TIGHTLY.
You replacing your head bolts Sky? Also a good time to replace your coolant hoses, heater core hoses, soft fuel lines, other items? I like SGB's iridium spark plug idea (it's on my list for the future)
Walth's right, but I doubt the heater core hoses are problematic (unless plugged with Prestone "oatmeal" sealer!). Seems most modern iridiums are equal quality, whether NGK, Bosch, Autolite or Denso, so you needn't get BLR6EIX OEunless you can grab under $30/set on eBay like I do.
Guru - The T-stat is the original BIG-bore T-stat (no small-bore aftermarket T-stat). Do you still recommend testing it against a new Subaru T-stat since the original seems to be working fine? Wralth - I do NOT intend to replace the Head Bolts! Guru - What brand of 'Spark Plugs' come as OEM on the Subaru - and are they any of the ones you listed above?
Nope...just use the OE t-stat you have. But DO GET NEW HEAD BOLTS (part of the full rebuild kit with gaskets and seals). ..and torque them very carefully. Lots 'o guys botch this. If you're unsure let your machine shop mount the heads (if willing), to ensure you don't blame them for failure later from a non-existent cracked or warped head, for example. Wrenches routinely do this full job in 8 hours, but as a first-timer expect to take 12-16 hours to do it right!
The cheapest set on rockauto is $65 for all 12 bolts. Only one, Beck Arnley, states differences in washers (big/small). http://www.rockauto.com/en/catalog/subaru,2006,outback,2.5l+h4, 1432741,engine
Guru - I feel confident about torqueing the heads myself. Just got my Snap-On Torque Wrench out of mothballs and I had it recalibrated for the occasion. Also I'm not in a hurry to get the job done as I have another vehicle to drive while the Subie is down! I'd rather stick w/ pre-stretched head bolts. Can you reinforce your opinion so maybe I'll change my mind? (Have you ever known any pre-stretched [used] head bolts to snap)?
skydiver, I've been experiencing the same issue since May. Was fine in the winter months, except after a really really long drive. Once the weather warmed up, it would always overheat after a few minutes of city driving. Guru described the problem I've been experiencing down to a T. Percolation and all. So I've been burping it with a no-spill funnel, probably once every 3 weeks. Eventually new air builds back up and it's stressful while driving wondering when it will blow next. Is replacing the head gaskets truly the only permanent fix? PS - I know nothing about cars but I am 100% certain what the issue is based on what Guru is describing.
That's the idea Tony. Permanent is relative but it's the goal of replacing the head gaskets. check out this thread: https://www.cargurus.com/Cars/Discussion-t16961_ds756392 where I posted a couple links to youtube. Brian shows a couple things, one is "temporary" but can prolong having to fix the HG's for however long? by drilling 2-3 extra bubbler holes in the thermostat. I thought that was pretty clever and very simple. It still allows the car to warm up and be pretty efficient verses taking out the thermostat. Another suggestion I've seen is putting a connector tube to jump the heater core hoses together and I think that was on here somewhere.
Walth, thanks a lot for the reply even though this thread was 8 months old. I stumbled across that same video from Brian months ago on YouTube but wasn't brave enough to drill the holes since I know nothing about cars and I wasn't 100% sure it would solve my problem. But at this point, it's either try this or give up and sell it (and lose thousands), cuz there ain't no way I'm dropping nearly 2 grand to replace the head gaskets after wasting $700 on a new radiator per the dealer's recommendation. Funny thing is, I told the dealer exactly what Brian and Guru were describing. He goes, "Don't believe everything you read online." Okay yeah, unless you're on CarGurus! Anyway, thanks again guys and Happy Thanksgiving.
You're correct in that "bleeding" a t-stat is a dying wish...often used by wrenches to get an HG case ready to trade in and dump at auction. I've NEVER seen it work longterm. Interestingly I serviced a non-overheating but coolant-dripping '06 OB like yours last week, prepping my friend that he should budget for HGs. After replacing oil-soaked valve solenoid pressure switches (look at yours?) I found a small puddle of coolant under the front. Lo and behold: a tiny crack near the upper hose/cap area. This is a very rare case where replacing the radiator may solve the problem...except his 11 yr olf t-belt is cracking...so we're getting in there anyway; but only after being assured the HGs aren't leaking as well, now that the cooling system is back to full pressure. Getting sick of this...ah yes...it's Thanksgiving! Let's all stay safe, grateful, and kind.
Yeah no doubt it's temporary. It's just how long before it goes. If you aren't looking to have it fixed then I would just go ahead and trade it in before it happens. Personally, I'd rather spend the 2k and fix it verses buying different and losing some large sum of money on trade in. Read up on negotiating dealers or even buy one from Guru here ;) Happy thanksgiving everyone!
Looking for a Used Outback in your area?
CarGurus has 35,956 nationwide Outback listings starting at $1,200.
Search Subaru Outback Questions
Subaru Outback Experts