Retorquing headbolts?

[axxiom]

I've got a 2000 Legacy with 36000 miles. In light of the apparent prevalence of blown headgaskets/serious gasket leaks on late model Subies, often well before the 100k mile mark, I'm guessing that there's a good chance of the problem showing up, sooner if not later, on mine.

Would having the car's headbolts retorqued every so often lessen the chances of headgasket problems?

[cookie]

frankly on that car it is not far from a retorque to replace the head gaskets. If it was not such a pain I am sure folks would do it. They do say not to retorque a lot of one use head bolts, but I am not sure what yours uses.

[frag]

Pardon my intrusion, but I think I'm staying on the subject.

Is just removing the valve covers enough to get at the head bolts and retroque them ?

Even if I have a 2.2, my engine goes thru a lot of cold and heat cycles during the winter (often - 25C to 100 C) and I'm thinking of retorking the head bolts as cheap insurance against HG failure.

Any opinions on this?

[cookie]

it ought to be a lot easier on your car. We used to retorque heads as a matter of practice years ago.

Let us know how it works with these bizzarre new headbolts. Wait till after winter though, because we don't want you to freeze if it does not work.

[NOMAD327]

On the DOHC engines where failures seem to be more common, it's a moot point, as the head bolts are all hidden under the camshafts. (major disassembly!) On the SOHC engines which this thread is really dealing with, the bolts look like they may just barely be accessible. The issue of what to torque to is still there, as the bolts are tightened initially in a weird mixture of torques, back offs, and retighten to a stated degree of rotation.

[avk]

Frag, on 1996 and earlier 2.2s the head bolts are in plain view. They use a special socket type (called Torx Plus, if memory serves), same as on later engines. But since the bolts are torqued-to-yield, I'm not sure they can be made any tighter.

[cookie]

on my BMW and the last Mercedes I rebuilt this was a no no. Bolts stretch to get thier holding power and a lot of the new ones are only supposed to stretch once. This is sort of like when you stretch an elastic band. The head bolt is always pulling the head down at a predetermined pressure. Bolts used to come loose more because gaskets would compress, dirt in the holes would come loose, or the lesser quality of the bolt steel would stretch.

Heads often were off several times in the life of a car becuse you had to grind the valves regulary and decarbon. For these reasons checking the torque during a tune up was common.

With a modern car the head often stays on for the life of the car.

[svxpert]

here is my take on the question. if your going to re-tourque the headbolts, you should buy new headgaskets as well. one reason is once you take the pressure off the headgasket (done by re-tourquing) the headgasket loses the sealing power. you never want to use HG's twice. the other part of my answer would be that there is no need to retourque them unless you had the engine rebuilt and don't trust the person who rebuilt it. engines out of the factory are made very well, just the design of some certain engines might be questionable. on the part of using the headbolts again is as long as you clean them up good, i have always re-used the headbolts when rebuilding my engines.

[Setright]

WHAT?

 

The head bolts must not be used again! Once they have been stretched, that's it. Makes me wonder if this point could explain why some people have had their EJ25 HG's fail repeatedly?

[NOMAD327]

The headbolts on modern Subaru's at least, are not torqued past the yield strength of the bolt and are reusable. Some engines do torque the bolts past the yield point, and require new bolts every time, I don't know why the odd sequence of backing off and retightening the bolts is used, maybe it is intended to relax the joint to allow the open deck block to shift at the gasket contact point after initial torquing for better cylinder to gasket fit.

[avk]

The factory procedure involves an angle gauge. That tells me the bolts are torqued to yield.

[frag]

The factory procedure involves an angle gauge. That tells me the bolts are torqued to yield.

Could you explain what this expression means exactly?

Thanks in advance.

[NOMAD327]

Metal is elastic to an extent and when placed under tension it stretches. A given metal will stretch a certain amount when pulled and this is called it's elastic zone. As more tension is applied to the metal it will stretch farther and farther, and when released it will return to it's original length. This is what makes joints stay tight when bolted is the stretched bolts trying to return to their original length. This is only a thousandth or so on a long bolt. There is always a point with metal where it has stretched to it's limit, and with application of more tension it will start to elongate, that is known as the metal's elastic limit or yield point. After passing the elastic limit by applying even more force, the metal will draw like taffy and will not return to it's original length, it will neck down at one area of weakness and will eventually break in two at some point determined by a natural flaw in the grain pattern. This is the fracture point of the metal. Bolts that torqued to yield by design are stretched to the point they have permanently elongated a tiny bit and must be replaced after each use. They are normally tightened to this extent by initial torque followed by additional tightening in measured degrees of rotation. The way the Subaru bolts are tightened, then loosened, then re-tightened by degrees apparently does not result in tightening the bolts beyond yield. The service manual is specific about replacing some components, but does not require replacing the bolts after each use. The benefit of torque to yield bolting is that it allows a more precise torque that will allow movement between parts yet maintain even clamping loads and sealing. This is supposed to be very beneficial when clamping aluminum to steel where expansion rates are different.

 

 

[frag]

Thanks for a very clear explanation. My brain did'nt even come close to its «yield point».

[svxpert]

<<WHAT?

 

The head bolts must not be used again! Once they have been stretched, that's it. Makes me wonder if this point could explain why some people have had their EJ25 HG's fail repeatedly?>>

 

can you show us where in a factory manual it says to use new ones?