SOBaru question

[rapidoxidationman]

I got to make an 800 mile road trip last week to rescue my daughter from her broken down car. It overheated and was spitting coolant from the exhaust pipe.
BTW, 1997 Subaru Legacy Outback, 2.5 liter 4 banger automatic, 205,000 miles.

At the Safeway parking lot in Anaconda Montana, I drained the coolant, replaced the thermostat, and refilled it with fresh clean 50/50 glycol. No Joy. So I dragged the car onto the trailer I brought with me, transported my girl back to college, and towed the car home.
"Sounds like head gaskets" you say? Not so fast, grasshopper. They were done about 20,000 miles ago.
No intermixing of the oil/coolant, no pressurizing the coolant.
The car runs OK cold, but as soon as the thermostat opens, the coolant starts burping from the tailpipe.
There's a bit of a vacuum (not pressure, like you'd expect to see from a head gasket failure) forming in the radiator when the car runs, seemingly sucking the coolant into the only place a vacuum forms on an engine: the intake.
No obvious leaks of coolant other than to the exhaust.

The local Subaru shop (Sue's Roos, experts in all things Suby and good friends of mine) are pointing at a possible gasket failure at the throttle body, just past the air cleaner. There's coolant ports in there that keep the atomized fuel from freezing up the body.

Anybody seen anything like this before? Got ideas on where to look in addition to the above suggestions?

[darkstarmoto]

I've been messing with Subaru's for 12 years and this is the 1st time I've heard this one. I replaced several dozen head gaskets on the 2.5's and that would be the #1 thought but your description is interesting. It only starts blowing coolant as the thermostat opens you say?? Hmmm...well the throttle body gasket is the cheapest and easiest effort, I'd start there and see if you get a fix. It is possible that you have a direct leak to the cylinder that only pulls coolant on the exhaust stroke but you'd certainly think it would allow gases to move into the coolant under compression. If it just started leaking it may not have allowed enough gases to show signs in the coolant overflow...although that usually doesn't take long to leave that tell tale oily residue.

Quick test: you could bypass the throttle body by connecting coolant lines to eachother and see if it starts spewing coolant from the exhaust when the thermostat opens...if no, it's the throttle body gasket...if yes...pull the plugs and find out which cylinder is getting wet. Atleast then you only have to replace one side HG.

[rapidoxidationman]

Originally Posted by darkstarmoto

I've been messing with Subaru's for 12 years and this is the 1st time I've heard this one. I replaced several dozen head gaskets on the 2.5's and that would be the #1 thought but your description is interesting. It only starts blowing coolant as the thermostat opens you say?? Hmmm...well the throttle body gasket is the cheapest and easiest effort, I'd start there and see if you get a fix. It is possible that you have a direct leak to the cylinder that only pulls coolant on the exhaust stroke but you'd certainly think it would allow gases to move into the coolant under compression. If it just started leaking it may not have allowed enough gases to show signs in the coolant overflow...although that usually doesn't take long to leave that tell tale oily residue.

Quick test: you could bypass the throttle body by connecting coolant lines to eachother and see if it starts spewing coolant from the exhaust when the thermostat opens...if no, it's the throttle body gasket...if yes...pull the plugs and find out which cylinder is getting wet. Atleast then you only have to replace one side HG.

Excellent idea.

[Bayner]

I've seen failed intake manifold gaskets or cracked intakes allow coolant into the exhaust.
Never seen it on a Subaru though.

[GreaseMonkey]

Have you pulled the spark plugs? If all 4 are sparkly clean or are just 2 of them?

Use a radiator pressure tester and pressurise the system, then start looking for where the coolant is coming from. You can peek into the intakes, and then later remove plugs to see if any are wet.

[Dave]

Originally Posted by darkstarmoto

Quick test: you could bypass the throttle body by connecting coolant lines to eachother and see if it starts spewing coolant from the exhaust when the thermostat opens...if no, it's the throttle body gasket...if yes...pull the plugs and find out which cylinder is getting wet. Atleast then you only have to replace one side HG.

Mrs. Dave has an '03 Forester (132K mi.) with a failing head gasket. We were told by the dealer and two independent mechanics that if you do one head gasket, you MUST do the other at the same time.

We were given the Subaru branded 'gasket treatment' stuff in the blue bottle. Hopefully, that will buy us some time before we sell it.

[darkstarmoto]

It is standard practice to do both for the simple fact the engine is out of the car, but if a budget won't allow it...it just won't allow it and you can do just one side. Here's the thing though if one side has gone bad the other is not far behind, especially when we are talking about the factory installed gaskets. Honestly I always advise against doing one side just because I'd hate to have to charge all the labor again when the other side does go.

I prefer using Fel-Pro Perma Torque MLS head gaskets which we have never had an issue with once installed. Gaskets are relatively cheap figure $30-ish a side, it's the labor that makes this job expensive because in order to a quality job you have to pull the motor. I've heard of guys doing it in the engine bay and I cannot see how the heck you can do a good job that way.

[kpt4321]

Originally Posted by rapidoxidationman

The local Subaru shop (Sue's Roos, experts in all things Suby and good friends of mine) are pointing at a possible gasket failure at the throttle body, just past the air cleaner. There's coolant ports in there that keep the atomized fuel from freezing up the body.

This isn't a bad suggestion, but I'm fairly confident that is not the purpose of those coolant passages. Doesn't a 1997 Subaru have regular port injection? Assuming so, there is no fuel in the intake air at the throttle body.

It's somewhat common to have a high idle circuit that is coolant-controlled in the throttle body. That may be what the lines are for. Those can definitely fail, either at the gasket or in the system itself.

I would pull the throttle body and look for coolant inside it and inside the intake manifold. Should be pretty easy to tell if it's present, and if so, replace the TB and/or gasket (depending on where it looks like it may be coming from).

Bypassing the lines by looping them is also a great idea, as a test.

[thechickencow]

So did you try the bypass yet?

[rapidoxidationman]

Still in learn mode. This is one of those projects to fill my spare time, of which there is little lately. Hopefully this coming week or Saturday. I will post back here on what I find out though.

[rapidoxidationman]

Finally swapped out the intake manifold parts that have coolant flowing through them. The throttle body with the butterfly valve (has a throttle position sensor on it) and a smaller device - I forget what Lucio said it is called - both had coolant paths. That has stopped the vacuum in the cooling system, now I just need to evaporate the ponds of glycol in the exhaust system. Maybe lifting up the front of the car with a backhoe and pouring it out would be the better way

The problem now is the car does not like to get up and move from idle. Any sudden load (pushing the accelerator, dropping it into gear, engaging the A/C) will cause the engine to bog and sometimes die. It'll idle all day smoothly and with no load can be revved up by gently throttling up, but any sudden load change will bog it and kill it. Granted, the exhaust is not exactly free flowing with all the glycol in there, but would an O2 sensor that's been overwhelmed with water cause this? I swapped the MAF sensor with a known good one, no change. Cleaning the spark plugs is on the slate, but it sure seems more like a fuel related problem...

[trumpet]

clogged air filter? fuel filter maybe

[GreaseMonkey]

Originally Posted by rapidoxidationman

Finally swapped out the intake manifold parts that have coolant flowing through them. The throttle body with the butterfly valve (has a throttle position sensor on it) and a smaller device - I forget what Lucio said it is called - both had coolant paths. That has stopped the vacuum in the cooling system, now I just need to evaporate the ponds of glycol in the exhaust system. Maybe lifting up the front of the car with a backhoe and pouring it out would be the better way

The problem now is the car does not like to get up and move from idle. Any sudden load (pushing the accelerator, dropping it into gear, engaging the A/C) will cause the engine to bog and sometimes die. It'll idle all day smoothly and with no load can be revved up by gently throttling up, but any sudden load change will bog it and kill it. Granted, the exhaust is not exactly free flowing with all the glycol in there, but would an O2 sensor that's been overwhelmed with water cause this? I swapped the MAF sensor with a known good one, no change. Cleaning the spark plugs is on the slate, but it sure seems more like a fuel related problem...

Man, I'd like to help you but as I posted last month you really need to pull the plugs and look at them. Coolant in the cylinders, no matter how it gets there, will do a really good job of scrubbing the spark plug electrode clean. if all 4 of your plugs are clean then maybe you are on the correct path but if only 2 of them are then I am afraid you are most likely headed down a dead end street.

Anyway, at this point if it is still running poorly (the O2 sensor issue won't kick in until the engine is warmed up) it would probably be best to get a cooling system pressure tester and see if it will hold pressure over several days. If it does, then great and move forward but if it does not then you will need to spend a little more time and money.

[schnutzy]

Originally Posted by rapidoxidationman

The local Subaru shop (Sue's Roos, experts in all things Suby and good friends of mine) are pointing at a possible gasket failure at the throttle body, just past the air cleaner. There's coolant ports in there that keep the atomized fuel from freezing up the body.

Anybody seen anything like this before? Got ideas on where to look in addition to the above suggestions?

yes actually i have. on my old car, a 93 pontiac land yacht
had ports in the throttle body for that same thing. and when the gasket failed it began to do just what your daughters subaru has done, i was lucky enough to catch it quicker. replacing the gasket solved my problem. gave the car to my mom and she has put another 10,000 miles on it with out a hitch and drives it a couple hundred miles every week.

[darkstarmoto]

You should be able to split the exhaust BEFORE the catalytic convertor and see how the car runs. If that much coolant is in the pipes the cat is probably clogged (with the white chalky residue from evaporating coolant) which will choke out the car.