I feel like a sinner, but nearly every time I torque something, I feel like it could and should take just a little bit more.
Ive been turning wrenches long enough to have a good idea of how much stuff can take and no doubt it will bite me at some point. But so far, so good.
Am I really the fool I fear I am? Or do I at least have some good company?
Yes and/or yes.
Drivetrain and lug nuts are the only thing that gets the torque wrench in my shop. Everything else is german torque spec: gutentight.
I'm a certified A&P, so torque specs are basically gospel in my world.
If I'm nervous about torque due to a custom application it gets safety wire or loctite.
With my acquisition of my 128i last year I became considerably more attentive to torque specs. BMW decided - in its infinite wisdom - that things screwed into alloy and magnesium engine components needed to be alloy, so I am now confronted with the joys of aluminum single-use bolts that, in the smaller sizes, will very quickly let you know that you have exceeded their maximum torque.
Pet peeve alert!
I wrench on industrial equipment, usually older stuff that's been worked on a lot in the past and one thing that consistently causes problems is people over tightening stuff.
Trust the torque values. They know what works better than you
For me it does depend on the bolt/nut.
One of the few things I remembered from engineering school is that setting fasteners by torque will get you within about 25% of the desired torque value you're shooting for (under ideal conditions...oil on the threads and all). So people who get all wound up about using proper torque specs just need to realize that. It's better than nothing, but it's not perfect. It can't hurt, to be sure...but sometimes folks who read too much (vs. "doing" a little more often) can blow things out of proportion.
[Side bar: I saw a video on youtube of some guy tearing down a new crate engine and complaining about the torque specs being all wrong...but he was trying to measure it with a torque wrench. There's NO WAY that's going to be accurate/repeatable enough (done on one engine...probably the guy's first time) to use as evidence.]
I use a torque wrench on something I know is sensitive to proper (enough) torque. Rods and mains, head bolts, gears, etc. But other stuff just gets a good grunt on a combination wrench or ratchet handle and I move along.
I worked on race cars years ago and I got to check ALL the critical nuts and bolts on my car after every session. I knew which bolts might come loose and which ones never did. In a thrash, I knew which ones to check and which could get skipped...but that was rare. We didn't use torque wrenches much if at all.
Suprf1y said:Pet peeve alert!
I wrench on industrial equipment, usually older stuff that's been worked on a lot in the past and one thing that consistently causes problems is people over tightening stuff.
Trust the torque values. They know what works better than you
One of my old mechanics at work swore by "A cross-thread is better than no-thread." He was also a fan of cutting wires rather than unplugging anything. You should have seen the engineers cringe when they had to work with him...
Dusterbd13 said:Drivetrain and lug nuts are the only thing that gets the torque wrench in my shop. Everything else is german torque spec: gutentight.
ditto. lugnuts and drivetrain.
In reply to 02Pilot :
To their credit, the use of aluminum fasteners in aluminum/magnesium parts prevents 99% of the galvanic corrosion problems familiar to us plebeians with steel bolts in aluminum parts.
On smaller fasteners (like intake manifold to Aluminum heads), I find myself cringing, just hoping I hear the click before I feel the sickening sensation of the threads tearing out.
In reply to jfryjfry : How often do you have your torque wrenches calibrated? How consistent is that calibration? How do you lube the threads? When is the last time a good tap and die was run over the threads? What are the fasteners made of and who made them? What sort of documentation do you have for them?
If you can’t answer each one of those What are you really saying?
If they were kind enough to provide a torque spec, I'll be conscientious enough to follow it. I'm not the one who engineered it. Maybe the fastener can take more torque but the assembly cannot, I don't know and I don't want to turn somebody's car into a science experiment.
I had to generate torque values because some of my equipment was getting broken by others using the tighten it intil it feels right method. When the 1/2" stud broke due to over torque, I had to call everyone ont he carpet and explain the limits of the assembly. Now we use a torque wrench on all the hardware and even though the values seem low, there has not been a single bolt failure or bolt back out. This on an assembly that has everything from 1/4-20 bolts to 1/2-13 studs and a few that pull up to bearing preloads.
Toques will keep MOST people honest. Some will keep pulling the wrench even after it breeaks over a the torque setting.
Matthew Kennedy said:In reply to 02Pilot :
To their credit, the use of aluminum fasteners in aluminum/magnesium parts prevents 99% of the galvanic corrosion problems familiar to us plebeians with steel bolts in aluminum parts.
Oh, I understand their reasons, and I can see the benefits when I pull fasteners out of my nine year-old, 116k mile Northeast motor with zero corrosion on them. But I've read many, many tales of people snapping mundane things like valve cover bolts and thus adding hours to a job dealing with the consequences. They certainly have a far narrower margin than steel between tight and broken.
I use a beam type on critical engine parts and a click type on lugs. Otherwise, ehh...
I helped a guy working on an engine and watched him use the inch-pound wrench to install a valve cover. I thought that might be a bit much.
02Pilot said:...BMW decided...aluminum single-use bolts...
Shakes head in disgust.
buzzboy said:
I helped a guy working on an engine and watched him use the inch-pound wrench to install a valve cover. I thought that might be a bit much.
That guy broke the flange off a $200+ set of aluminum valve covers once.
Hey, a question relevant to my work! I'm an engineer that sometimes comes up with torque specs for engine parts.
Yes, torque specs are important! Especially on highly stressed components like engines and lugs. Someone did a lot of work to figure out exactly how much preload is needed to keep those two parts together, prevent fatigue damage, or make that gasket seal just right. Sometimes a "generic" torque spec is enough. You can google those for a given bolt size, thread, grade, and hole material. But sometimes a bolt needs more preload, less preload, or a narrower range of preload.
Preload on a bolt is just a (usually) linear function of torque and friction. More torque = higher preload. More friction for a given torque = less rotations of the screw = less preload. This friction is often called "nut factor" and it's different depending on many factors that are hard to control, like thread material, hole material, oiled vs bare vs coated, used vs freshly machined, etc etc. Because of all this variation in nut factor, a good rule of thumb is that a single-step torque value typically has a +-30% scatter in preload. Then there is also load relaxation. The bolt and flange materials will undergo "creep" after being torqued initially, plastically deforming and resulting in overall less clamp force. This can be up to 20% in certain applications that undergo a lot of heat and load cycles.
So if you know for sure that you need a minimum worst-cast preload of 10,000 lbs on a bolt, one must factor in the initial preload scatter (+-30%) and the joint relaxation (up to 20%). So you actually design the torque spec to hit roughly 17,850 lbs, -30% is 12,500 lbs, -20% is 10,000 lbs. That's just a rough conservative estimate. In this example, it's possible for the same bolt with the same torque to achieve 23,200 lbs and not relax at all. That's MORE THAN DOUBLE the minimum worst-case preload. So you can see how even if you're carefully using a torque wrench, there is already a lot of error built in. When you don't use a torque wrench, you are only compounding that error. Too little torque can cause the joint to loosen, too much torque can yield parts, break bolts, or strip out threads.
If a non-generic torque must be used, they are usually determined experimentally by measuring preload, torque, and sometimes rotation angle, and reviewing that data on a sample size of measured bolts to come up with a torque value that usually results in the target or minimum required preload. I say "usually" because of all the unpredictable factors mentioned above. There are many devices and methods to measure stretch or preload of a bolt in applications where design margins are lower, so you try to get by with the smallest, lightest, or cheapest bolt possible.
Building an engine? Torque to spec. Everything else? German specs.
In reply to maschinenbau :
i work with engineers who are responsible for bolted joint analysis. i showed them your post. they love you.
I agree that an experienced person can get close to an appropriate tightness for most applications by feel however, If you're consistently finding that you want things tighter than the spec. then either you or your torque wrench need recalibrating.
snailmont5oh said:buzzboy said:
I helped a guy working on an engine and watched him use the inch-pound wrench to install a valve cover. I thought that might be a bit much.
That guy broke the flange off a $200+ set of aluminum valve covers once.
Or maybe his engine incorporates the outer cam bearings into the valve cover? Or maybe it was a Honda D-series that has a 20 in-lb range between 'leaking valve cover,' and 'stripped out?'
I'm not going to get all holier-than-thou on breaking out the torque wrench. I use it where experience tells me it's vital. If you follow all of the service manual recommendations, you're also supposed to discard all those expensive suspension fasteners every time you disassemble the suspension. This may be because the published torque values are no longer applicable, fastener yield, or just good old liability concerns.
Torque specs are sometimes wrong in printed service manuals and shouldn't override common sense and experience. I would check for TSBs and updated info before accepting it as gospel.
And let's be honest -- there are a lot of service situations where you just can't put a torque wrench on a bolt...and if you could, the combinations of u-joints and extensions and angles would make it inaccurate.
In reply to Tyler H :
Just one thing I'd like to point out that is a commonly repeated fallacy, socket extensions do not effect torque values! (Assuming that you can hold your extensions straight. For a static case like torquing a fastener torque in must equal torque out because Mr Newton. Yes, the shaft twists but the previous statement still holds true. You can try it for yourself, go get the longest extension you can and experiment.
Matthew Kennedy said:In reply to 02Pilot :
To their credit, the use of aluminum fasteners in aluminum/magnesium parts prevents 99% of the galvanic corrosion problems familiar to us plebeians with steel bolts in aluminum parts.
Which problems?
Mind you, I'm the guy who sings the praises of lug bolts, but aluminum bolts makes little sense. It MAY be okay if BMW speced them okay, but they tend to fail in service. When your valve cover is leaking because four bolts in a row broke in service, the fasteners are just plain insufficient. And VWAG made millions of engine cases and valve covers out of magnesium and threaded steel fasteners into them and no corrosion issues even here where everything corrodes...
Torque, incidentally, is something like 4nm plus 90 degrees. They WILL break 1nm over the initial torque. That 90 degrees is all stretch.
You'll need to log in to post.
