Originally Posted By: onion
Can you give any more detail about your research into fastener clamp loads?
I'm of a similar school of thought- I know what you're getting at when you talk about how much clamp load can be affected by corrosion, worn parts, lube or lack thereof, etc. In many cases (though definitely not all), I trust my own feel more than I trust a torque wrench- but I wouldn't recommend that for everybody (I've been tightening all manner of bolts on a daily basis for 15 years, so I'm pretty confident that I can get them 'close enough').
Lots of people on this site are very "by the book" in their mentality. And that's fine- 'by the book' will usually get you by just fine. But there are factors and effects that are not easily distilled into a book... nor are the authors of said 'book' necessarily concerned with how well a given procedure will work on a 10-year-old vehicle. So long as their advice is effective on a relatively new vehicle using OEM-approved parts... their job is done.
Well, I don't want to bore anyone to death, or attempt to boil down a 1,000 page book into an internet post--but the bottom line is that torque is a rough proxy for bolt tension--very rough...
If you were to take 100 new spark plugs, from the same batch, identical model, same manufacturer into newly-tapped threads, and tighten them with a torque wrench, the actual joint tension would vary by an average of roughly 20-40% based on the literature (and much of the literature isn't even consistent). Of course, that's the average--there would be scatter points with much larger errors. Different manufacturers use different platings on the threads, which confounds things even more.
The big variability is in the coefficient of friction, or nut factor. There are several variables in the spark plug example: thread condition, plating material, underhead smoothness... To give you an idea of the complexities, if you took THE SAME spark plug and threaded it into a new cylinder head 4 times in a row, each time at the same torque reading, do you think the joint would have the same, less or more tension on it by the 4th installation? If you don't know the answeer to that question, how can you say that you're being "precise" by using a torque wrench?
A spark plug installation joint isn't that complex. It needs to be tight enough to withstand cylinder head pressure and engine vibration (so it doesn't pop out), and the forces need to be low enough that you don't hit the yield point of the cylinder head on install or the spark plug on removal (so you don't strip threads or break the spark plug). I think it makes sense to look at each specific joint and determine just what sort of precision is necessary. Since most spark plugs also give angle tightening specs (X turn after seating), I think that's your safest bet.
Lastly, a previous poster mentioned that K varies with lubrication. That's true. However, in general, (and in the field in actual practice), most of the literature on the subject suggests that lubrication also lowers the variability--so as long as you account for the lubrication, your chance of accuracy actually increases with lubrication.
This is a great read on the subject, if you're an insomniac: http://www.amazon.com/Introduction-Design-Behavior-Mechanical-Engineering/dp/0824792971
The more you read, the more you'll realize how little you actually "know".
Can you give any more detail about your research into fastener clamp loads?
I'm of a similar school of thought- I know what you're getting at when you talk about how much clamp load can be affected by corrosion, worn parts, lube or lack thereof, etc. In many cases (though definitely not all), I trust my own feel more than I trust a torque wrench- but I wouldn't recommend that for everybody (I've been tightening all manner of bolts on a daily basis for 15 years, so I'm pretty confident that I can get them 'close enough').
Lots of people on this site are very "by the book" in their mentality. And that's fine- 'by the book' will usually get you by just fine. But there are factors and effects that are not easily distilled into a book... nor are the authors of said 'book' necessarily concerned with how well a given procedure will work on a 10-year-old vehicle. So long as their advice is effective on a relatively new vehicle using OEM-approved parts... their job is done.
Well, I don't want to bore anyone to death, or attempt to boil down a 1,000 page book into an internet post--but the bottom line is that torque is a rough proxy for bolt tension--very rough...
If you were to take 100 new spark plugs, from the same batch, identical model, same manufacturer into newly-tapped threads, and tighten them with a torque wrench, the actual joint tension would vary by an average of roughly 20-40% based on the literature (and much of the literature isn't even consistent). Of course, that's the average--there would be scatter points with much larger errors. Different manufacturers use different platings on the threads, which confounds things even more.
The big variability is in the coefficient of friction, or nut factor. There are several variables in the spark plug example: thread condition, plating material, underhead smoothness... To give you an idea of the complexities, if you took THE SAME spark plug and threaded it into a new cylinder head 4 times in a row, each time at the same torque reading, do you think the joint would have the same, less or more tension on it by the 4th installation? If you don't know the answeer to that question, how can you say that you're being "precise" by using a torque wrench?
A spark plug installation joint isn't that complex. It needs to be tight enough to withstand cylinder head pressure and engine vibration (so it doesn't pop out), and the forces need to be low enough that you don't hit the yield point of the cylinder head on install or the spark plug on removal (so you don't strip threads or break the spark plug). I think it makes sense to look at each specific joint and determine just what sort of precision is necessary. Since most spark plugs also give angle tightening specs (X turn after seating), I think that's your safest bet.
Lastly, a previous poster mentioned that K varies with lubrication. That's true. However, in general, (and in the field in actual practice), most of the literature on the subject suggests that lubrication also lowers the variability--so as long as you account for the lubrication, your chance of accuracy actually increases with lubrication.
This is a great read on the subject, if you're an insomniac: http://www.amazon.com/Introduction-Design-Behavior-Mechanical-Engineering/dp/0824792971
The more you read, the more you'll realize how little you actually "know".
Last edited: