Torque and angle? Have seen a few examples where one does first a torque of say 150Nm(on say a M12 thread) and afterwards add a angle of say 45 degrees, this procedure will add a total of xNm(perhaps 200Nm in this case), I did see a example of this procedure where the initial torque was 300Nm and with the addition of a angle resulted in 450Nm. Is there any source/litterature for these procedures? I have searched but can't find any source on this
Cheers!
k.

 

Not sure exactly what detail your looking for. Here I think is a pretty good explanation of the process.

https://www.norbar.com/News-Events/B...ngle-Explained

To paraphrase. It's really not about getting a final torque number. Torque is really just sort of a convenient analogue for bolt tension or "stretch" because you can measure the torque easily but you can't measure the tension. However, some unknown percentage of the total torque applied goes into overcoming friction which can be affected by lots of factors like how clean the threads are, if there is any lubricant or even temperature and so its not always an accurate indicator of final tension. Torque angle procedures are often specified for things like cylinder head bolts because they are quite long and so for getting that tension just right, torque measurement alone is not accurate enough. Hope that helps.

 

Thanks!, what i'm looking for is some form of relationship between what a given torque and subseqeunt angle and what total torque is produced. Fully aware of different tensile strenghts, ex 8.8, 12.9 equals permanent stretch at about 900 & 1300N/mm2. Must be some litterature about this torque angle procedure somewhere?
Cheers!
k.

 

There was a TV commercial not long ago, don't remember what it was advertising, but it showed a couple of old people using cell phones and one says to the other "Remember when you used to have to go to the library to look things up?" That sounds like the kind of thing that's probably in an engineering textbook somewhere because I would think rearranging the equation would be how they come up with stepped torque specifications for other applications, starting with the final tension they want and making a bunch of assumptions about the friction. Hopefully somebody will come along with a link or reference source for you. Wish I could help further but I took calculus over 30 years ago and can't remember anything about it now except that it was in a huge auditorium with about 175 students.

 

slugger, what you are after is not a direct answer. What I mean by this is that you take a 1/4" bolt and a 1" bolt, you torque both of them to say 40 ft-lbs, then you turn them an additional 90 degrees, they are going to apply 2 different torques (the 1/4" bolt will probably drop to around 20 ft-bls due to exceeding its yield strength and the 1" bolt is going to go up to around 100ish ft-lbs. You have to think about bolts kinda like cold taffy. They are one length when you don't pull on them. As you pull on them, they are going to start to stretch some. The harder you pull, the more they are going to stretch. If you pull a certain amount (length wise), the thinner taffy is going to neck somewhere and become very thin (loosing any strength) and the thicker taffy is going to narrow slightly, but keep its strength.

Really, torquing by angular degrees can be more easily thought of as torquing by length change. With this, you have to know the diameter of the bolt, the thread pitch, and in an ideal world, the temperature of the bolt (granted, as long as you are not changing the temp, then this can be disregarded in a lot of cases). The thread pitch is going to control how much length you are going to add once the bolt is under load (the distance between the head of the bolt and where the threads engage is the same, but there are going to be more threads into the bolted surface) and you need to have the bolt under an initial amount of load (this puts an in a state where the bolt has not started stretching, but is at the point that it is going to start). So, once you start turning the bolt, the torque starts going up dramatically because the bolt is now being forced to stretch as you apply more degrees of movement.

Contrary to what you may think, you want the bolt to flex some. A very stiff bolt is very brittle. When the bolt is brittle, it will support a lot of load, but when it fails, it fails catastrophically. You want a bolt that is a bit elastic for most applications. But, for say a head on an engine, if the bolt stretches, then you have the head lifting off the engine, resulting in head gasket failure. This is where you try and take that stretch out. The trick is figuring out how much stretch you can put into a bolt. This is where the type of material, how much it has been hardened, etc is needed to be known. As you stretch a bolt, its strength actually goes up for a period of time and you will see an increase in what is referred to as the clamping strength (ie, how much it is pushing the two surfaces together). But, there is a breakover point (called the yield point) where the bolt will suddenly loose a significant amount of its strength and loose a lot of its clamping strength. You can see this in the bolt as you will see a portion of the bolt neck down (become narrower). The bolt has not physically broken, but a lot of the molecular connections inside the bolt have been broken. Some old timers will say that you can tighten a bolt and you can feel it get weak as you exceed the yield point as the wrench will become slightly easier to turn just before the head of the bolt goes flying across the room as the bolt actually breaks. If you keep the stretch of a bolt such that you never reach the yield point, the bolt will spring back to its initial length once the torque is removed. The goal of most torque by angle bolts is to put them right at the yield point. This is where the bolt has stretched as far as it can while maintaining its strength, but once you remove the torque, it is not going to spring back to its original length. It is going to remain stretched. This is why you never reuse bolts that are torqued by angle. You stretch them again, you are going to push them beyond their yield point and now you have a bolt that is just waiting to fail.

I hope I am not going over your head as angular torquing of a bolt is a very complex way of torquing a bolt and has a lot of variables in it. Granted, I deal with bolts that are also heated as part of this process (adding just one more variable into this madness), but then, where I work, we need to torque bolts to millions of ft-lbs of torque and you don't simply do that with a wrench. You have to heat up a bolt to make it more elastic (want to stretch more), torque it to some high torque that you can reach with the tools available and then by knowing how hot the bolt is, you can calculate if this is going to result in the necessary torque. In my application, this is how we torque the bolts for the vessel cap of a nuclear reactor (not something you want coming apart that is under extreme stresses from the internal pressures. If you have additional questions, let me know. I will attempt to answer them. I have received a fair amount of training in how you torque bolts as one of my jobs was doing certification of nuclear and SUBSAFE systems on submarines. There torque is key to almost everything. Too little or too much torque can cost lives.

 

Excellent answer Sir. Just enough tech, with just enough down to earth language. Completely understood. Thank You, and I'm not even the guy who asked the question. LOL!

Jack

 

Wow! Thanks!
Ok, so when engineers are calculating the torque settings for a specific bolt(for us common people that use a torque wrench), say a wheel bolt(or nut) their aim is to have a clamping force of xxx Newton per bolt/nut. the wheel nut/bolt are reusable and not near the stretch point for most cars I think. So say "they" want a 1000N flat clamping force per 1/2"(or M12) bolt or nut, they then start thinking what amount of stretch would that be on a 1/2" bolt with a surface area of about 60mm2? that will produce about 20N/mm2, but how do you go from that to the amount of stretch that is required to acheive the tension?, here is where my thought process stops...as linked in previously the torque when used on a wrench is doing not only the stretch of the bolts but also doing mostly friction, only 10-15% of torque goes into the stretching of the bolt/nut,
complicated stuff this is....
Cheers!
k.

 

slugger, do not worry about stretch. As long as you are not putting the bolt at the yield point (ie, where you have taken all the stretch out), you are just down to the metal properties. So, lets take the example that you gave and put it to reality. There is an equation that does what you are after: Torque (T) = a material constant for the type of bolt (K, which in this case, assuming a standard bolt with no lube, K=0.2) x axial bolt force (F, in newtons) x bolt diameter (d in meters) x (1- i/100) (where i is a lubrication factor, closer to 100 is less friction, most lubes are rated at about 80, since we are assuming a dry bolt, it will be 0 since no lube is being used). So, taking out the extra wording: T = KFd (1-i/100). So, putting in numbers, T = 0.2 x 1000 newtons x 0.012 meters x (1-0/100). So, working out the math, this comes out to be 2.4 Nm of torque. Considering that if you are talking about a standard bolt with a tensile strength of say 77000 psi (or 54,250 newtons), this bolt would only be strained to about 2% of its maximum design strength. Now, if you wanted to put the bolt at the design strength to maximize what you are getting out of the bolt, this would occur at 130 N-m. Hence if you torque your wheel nuts to above 130 N-m, you will most likely break them. Now, this assumes that you don't have any resistance from friction turning the bolt. This is where you can get the torque wrench to read out above 130 N-m and not break the bolt. But, you lubricate it to allow it to move freely, you could possibly break it at the 130 N-m depending on how well you greased things.

You want to have a quick calculator for you, then check out: https://www.engineeringtoolbox.com/b...or-d_2065.html. Really, what you want to know is on this page: https://www.engineeringtoolbox.com/m...ue-d_2054.html. This page tells you the maximum torque that you can apply to various size bolts based on their grade ratings. You stay below these numbers, you should be good. This is where torquing the bolt being square on the bolt is critical. You start getting into having the torque wrench at a slight angle and now you have opened up a totally new can of worms.

 

Thanks Thermo! owe you a beer! very very useful information here, had a read up on the links you provided, seems that torque wrenches is not so accurate as one might think(+-25%!), better off using a bathroom scale and fixed lenght to get to the "right" torque(in the improvised torque wrench related document), also in the related documents there is tensile stress and hooks law, I guess that is where one could calculate backwards towards a angular torque with thread pitch(will have to think for a while on this and do some math excersises), again very useful information thanks for sharing,
Cheers!
k.
,

 

Slugger, if you are really wanting an accurate torque wrench, I have 2 solutions for you. The first will cost you around $100. But, you can send the torque wrench that you have out to a calibration shop and they will put it on one of their calibrators and it will tell you the error in the wrench. You can then adjust the torque as needed to compensate for the error (ie, your 100 ft-lbf torque wrench reads 10% low, you need to torque to 50 ft-lbs, you would take the torque wrench to read 55 ft-lbf - the initial 50 + 10% of the 50).

The second way is to get a large spring scale. You can then make a little rig with a foot long bar (center to center on the holes) where one end is going to have the spring scale attached and the other end is going to have the square hole for the torque wrench. As you apply a torque, the torque wrench and the scale should read the same. You plot the scale vs torque wrench, you can see the error in your wrench. Some of us are fortunate enough that we have access to one of the calibrators and can simply slap their torque wrench on the calibrator and do this for free. One of the bennies I can take advantage of.

As you mentioned, torque wrenches may not be all that accurate. This is where the style of torque wrench can become critical. Most people like the "Clicker" style torque wrenches. Great for the home mechanic because they are easy to use, don't even have to look at them. You simply set the torque you want and pull till it goes 'click'. But, these are some of the most inaccurate torque wrenches out there. They should have their handle dialed back after each use to prevent introducing errors in the torque and if you go beyond the click, what did you torque the bolt to. You have no idea. The better torque wrench is the dial torque wrench. Now, you do need to look at the dial as you are torquing as the dial reflects what force is being applied. Some even have a little light in them that you can see when you hit your desired torque. But, if you go over, as long as you are looking at the dial, you know what you went to. These also tend to be built with slightly better parts (but the price reflects that too). The final kind of torque wrench you have is the bending beam style. These are easy to spot as you have 2 bars of metal stacked one on top of the other. The one bar is very thin and has a pointer on the end that points to a scale. The other bar is made of spring steel and will bend slightly as you apply a torque. Very accurate (atleast at first) and are not that expensive, assuming you can find them (not a common design as they are seen as "old tech". The issue with these is they wear out. Like the springs in your car. The more you use it, the spring steel will slowly loose its springiness and not be accurate. But, you need to use the snot out of them for this to be significant. They can also suffer from the small rod being bent from being dropped. So, they need a bit more TLC.

With all this being said, you can help minimize any errors in torquing. Do the following:
- apply the torque in a nice, steady way. No jerking.
- use both hands, one hand should be pulling as the other is pushing. So, for example, if you are needing to pull the handle towards you to tighten a bolt, that hand will obviously be pulling, your other hand should be pushing just enough to keep the head of the torque wrench stationary such that there is no twisting of the torque wrench side to side. All twisting should be straight down on to the bolt. This is especially true if you are using any sort of extension and the torque wrench is not directly in line with the bolt head.
-use of crow's feet do not have a significant impact on the final torque as long as you are using a short crow's foot. Do not create some foot long adaption to make things work. There are ways to make it work, but that is a different discussion all together.

 

Thanks again Thermo! oh yes, I have the clicker type wrench, are slacking the lock mechanism and turn down to zero load after use. I will borrow a "proffesional" wrench when the time comes to do rear hub nut at about 310Nm. I was thinking of comparing that pro wrench with my "hobbyist" wrench. I know the old style type(torsion bar?) you mention, have newer used them, but have a store that stock them nearby at about 10 EUR, have alway thought of them as inferior to the clicker type...probably becuase they are cheaper, will get one of those shortly. As a test of Newton physics I will test them against a bathroom scale, my weight(as a zero mark in terms of Force(N)) x lenght braker bar(or shaft of wrench), so if wrench klicks(if the klickar type) at 300Nm and is 500mm in lenght that should give a reading on the bathroom scale(centered under end of wrench) my own weight plus 60kg(600N), and at 90 degress/square to centre of axis, this should get a reasonable accuracy or control of how accurate the wrench is, if the numbers dont add up I would then rather trust the bathroom scale and math, and adjust accordingly, will post how things go with this "experiment"
Cheers!
k.