I can be obsessive/compulsive - exhaust manifold bolt torque

[Ed S.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

I work wit very high water pressure and we always tourqe above the head pressure. I would say the the exhaust gas pressure and the fact that the 2 center cap screws have to do the same job as the 4 outer cap screws calls for a higher tourqe to keep gas sealed between the head and the manifold. My 2 cents

And with respect to the exhaust manifold, it might be because the outer bolts are closer together than the center bolts.

[Ed H.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

I work wit very high water pressure and we always tourqe above the head pressure. I would say the the exhaust gas pressure and the fact that the 2 center cap screws have to do the same job as the 4 outer cap screws calls for a higher tourqe to keep gas sealed between the head and the manifold. My 2 cents

Now this makes perfect sense; the shop manual isn't clear. It should say outer pair and inner pair.

Thank you! I'll be able to sleep tonight!

[Larry M.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

Then what is the criteria or rationale for torque specs for applications that do not have a) gaskets or b) pressure concerns..... such as suspension components or internal engine components, bearing caps? There has to be a sound science that dictates torque specs in general.

Ed:

Ideally it is not torque but BOLT STRETCH that determine the forces exerted by the bolt/fastener. Torque values approximate this, are much easier for us to do, and generally get the job done. But for critical applications, fasteners are selected, their tensile/compressive/shear properties tested and measured to make sure they are known, and then the bolt/fastener is stretched to a certain point to provide the calculated force required.

For most of our work, the bolts are stretched in the ELASTIC zone, so they can be reused again. But in some cases, these bolts are stretched into the PLASTIC zone and are forever deformed.......although they do provide the force required. Late model Chevrolet engines use this latter approach and machines stretch them in groups all at once (i.e. head bolts).

Torque is good, but actual bolt stretch is the preferred way. Torque has too many variables associated with it such as how the threads are cut, are the threads lubricated and by what compound, accuracy of the torque wrench, etc. Bolt stretch measured by micrometer (or its equivalent) gives the precise result required.

Larry

[Robert G.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

The end bolts use French locks. Maybe that has something to do with it.

[Ed H.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

The end bolts use French locks. Maybe that has something to do with it.

At only 20'/lbs maybe insufficient stretch to stay tight on their own. Another thoughtful observation, and I don't have them... (the French locks, to be clear )

[George W.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

more torque on end bolts will/could break the ends off in the bolt holes.

[Michael G.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

I would guess that the difference is that the shorter outside bolts see less friction from the bolt contacting the sides of the manifold holes during tightening. Friction eats up clamp-load, meaning that the same torque would deliver lower clamp-load on the inside bolts. As such, more torque would have to be applied to the center bolts to achieve the same clamp load.

Torque specs like this were developed back then a bit differently than you might do it today, due to the differences in sophistication of the equipment available. These were probably developed to ensure that the range of torque allowable could be maximized without yielding the bolts during tightening. As such, tests were probably run where a series of these bolts were tightened to failure while actually tightening the manifold to the head, while monitoring the torque during tightening. The torque at which each bolt in the test yielded would be noted and this data would be used to calculate the yield range of the series. The upper limit of the actual tightening torque range would be set such that it was comfortably below this lower yield limit, and the lower limit would be set relative to what range was achievable in production. There were also other ways to do it, depending on the criticality of the joint and many other factors. The specification thus derived would then be validated by tightening actual attachments (on cars, or in test fixtures) to the minimum spec for the test.

There were lots of other ways to do this, this is just one possible, very common, scenario GM used back then.

[Bill M.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

Then what is the criteria or rationale for torque specs for applications that do not have a) gaskets or b) pressure concerns..... such as suspension components or internal engine components, bearing caps? There has to be a sound science that dictates torque specs in general.

In general, each suspension fastener goes through a test in a fastener lab. to determine the torque spec. The test involves sophisticated lab equipment and a statistical analysis of the (at least 10 sample) test data. The criteria is max. clamp load without yielding of the fastener.

[Michael G.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

In general, each suspension fastener goes through a test in a fastener lab. to determine the torque spec. The test involves sophisticated lab equipment and a statistical analysis of the (at least 10 sample) test data. The criteria is max. clamp load without yielding of the fastener.

[Bill M.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

The way I remember the lab test on pre-production fasteners being run was clamp load and torque were plotted. The clamp load tapered off when the fastener began to yield. We made sure that none of the fasteners would yield (3 sigma based on 10 samples) at the maximum torque spec. I did not run the test. I do remember that I had to provide full-production-level parts to get an accurate test. The time period would have been in the 90s.

[Michael G.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

The way I remember the lab test on pre-production fasteners being run was clamp load and torque were plotted. The clamp load tapered off when the fastener began to yield. We made sure that none of the fasteners would yield (3 sigma based on 10 samples) at the maximum torque spec. I did not run the test. I do remember that I had to provide full-production-level parts to get an accurate test. The time period would have been in the 90s.

Bill, you are remembering one way of doing it, yes. But not the only way, and certainly not the most efficient.

Unless you’re going to use the load data from the test for some specific purpose, its much more efficient to plot torque vs time of revolution (or torque vs angle of revolution) to find the same yield inflection point. This is because when the fastener yields it changes all those graphs in a similar manner. Using either of these allows you to determine a plant-useable tightening spec relative to yield on the production-representative joint.

The point is, if you need the load information because you know what load you’re looking for, great, do the load test, but it was not usually done. If your goal is to ultimately create a production torque specification, torque vs time (or torque vs angle also) is the piece of information you need to base your yield test on, as you can get that information on unaltered samples of the production-representative joint. You cannot do that with torque vs load, because you need to instrument the joint for load, and that changes the tightening curve.

Of course, whichever test you choose to utilize, you still have to validate the attachment at the tightening spec you developed, so it really matters not which you method you select.

[Bill M.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

OK Mike, thanks for the explanation. I don't think we needed clamp load specifically, but I can't remember the application.

[Ed H.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

This has been, all in all, a quite fascinating bit of learning for me. But, of course, that leads to another question!

When I get French locks for each end I'll need to remove the bolts (I think, or, at the very least, loosen them.) Will that require replacing the gaskets?

Again, thank you all!

[Larry M.]

Re: I can be obsessive/compulsive - exhaust manifold bolt torque

This has been, all in all, a quite fascinating bit of learning for me. But, of course, that leads to another question!

When I get French locks for each end I'll need to remove the bolts (I think, or, at the very least, loosen them.) Will that require replacing the gaskets?

Again, thank you all!

Originally these exhaust manifolds had NO gaskets. But if you did install gaskets, then you would simply remove the bolts (two at a time), install the french lock, re-install the bolts, and then tighten to torque value specified. No new gaskets would be needed.

Larry