Explain Torque+Angle

[T45]

Can Ya'll explain how this works?

Testing determines the torque at which the transition phase ends and the linear portion of the trace begins. Allowing for part-to-part variation, this “snug” torque is determined. The specified angle of rotation, which must be measured with sufficient accuracy, starts from this point. As bolt length-to-diameter ratios get smaller, more tension is generated per degree of rotation, so greater accuracy is required.

Does anybody use this in place of regular torque settings, or is it simply a technique applied when the manual/spec calls for it? Are there tables which calculate what's referred to as the linear trace somewhere (the effective torque of 1/4 turn etc) ?

Thanks !

 

[Heavymetalmechanic]

Is this in reference to a torque-then-turn application?

 

[Hybridss]

This would be in reference to something like a "tq to yield" bolt like many modern headbolts. You would tq the bolt to a specific tq to begin. This would get you to a place where the bolt is very tight...but has not yet began to stretch...but any additional movement and it will begin the stretch. You then use degrees of rotation to further track the applied tq. Many time is something like the following:
1) tq all head bolts to 25 ft/lbs using the proper bolt sequence
2) using the same sequence turn each bolt 90 degrees
3) final pass using same pattern tq an additional 60 degrees

There are accurate tools to do this. but general degrees like 90 or 60 can be eyeballed. Usually I would mark a line with a sharpie.

Other methods are to measure the actual bolt stretch using dial indicator...such as rod bolts. Stretch is probably the most accurate. This compensates for variances in "run on tq" and friction. Run on tq is the tq required to turn the bolt or nut, especially if nuts or bolt have "self locking" features.

I have had procedures where is was required to measure run on tq...then apply the actual tq over and above measured run on tq also.

 

[joe_padavano]

The problem is that tension (preload) in the bolt is what you really need to measure. Torque is a convenient but somewhat inaccurate indication of preload, as torque vs. bolt preload can vary considerably depending on friction. In extremely critical aerospace applications, we use either instrumented bolts (there's a strain gauge mounted inside the bolt) or load sensing washers in place of a torque measurement. The torque plus turn concept simply uses the torque value to get the bolt snug (friction force is lower as bolt preload is lower, so at this level torque is a little more accurate indicator of preload) then uses the threads of the bolt as a micrometer to add in a known stretch to the bolt. Since the tread pitch, bolt diameter, and axial stiffness are all known for a given fastener, turning the nut (or bolt) a certain number of degrees after snug introduces a known stretch and thus preload in a given bolt.

 

[Crown Imperial]

There are accurate tools to do this. but general degrees like 90 or 60 can be eyeballed. Usually I would mark a line with a sharpie.

I would not recommend you do this. Use a degree wheel / dial gauge. Not very likely you will get a 90 degree TTY with a +/- 2.5 deg. tolerance accurately with a sharpie.

 

[T45]

Since the tread pitch, bolt diameter, and axial stiffness are all known for a given fastener, turning the nut (or bolt) a certain number of degrees after snug introduces a known stretch and thus preload in a given bolt.

This is awesome explanation, thanks ! I've seen tables where standard fasteners have max torqu values (like m6 x 1.0 = 10nm max, etc). Is there an engineering table for this kind of information? based on torqu+angle?

Thanks again to everyone who for helping to explain all this

 

[ttpete]

This is awesome explanation, thanks ! I've seen tables where standard fasteners have max torqu values (like m6 x 1.0 = 10nm max, etc). Is there an engineering table for this kind of information? based on torqu+angle?

Thanks again to everyone who for helping to explain all this

It depends on the application. Always think of a bolt or stud as a spring. What you're doing when you tighten it is actually tensioning it. Torquing it is one way to induce that tension. It's not the best way because of many variables, things like application, lubrication, thread finish, tensile strength and shank diameter. So, there's no really valid general rule for a given thread size. It's only a rough guide.

If the application is something like holding a cover plate on, the torque and material required would be very different from the same size fastener used as a connecting rod bolt. When I worked at Ford Engineering, we had a fastener lab that tested fasteners and specified fastener types and torques for all applications on the vehicle.

 

[Wakefield]

Torque plus angle might be better at getting to maximum tension on bolts without taking as much chance of breaking them. Not as much variation due to friction in the threads?
Tried to tighten a grade 8 coarse thread 3/8" "bolt" (hex head capscrew) it stopped getting harder to turn before clicking the wrench which was set at 52 ft.lb. -it turned a little bit more without felling any tighter and then it began to have less resistance and then-BUSTED! Got another one and nut and clicked at 48 ft. lb. Clamping a bracket to a mower deck.
Perhaps when a bolt or nut is tensioned to where it is giving its maximum possible performance it needs to be discarded and replaced with a new one if disassembly and reassembly is done.

 

[steves_auto]

I have been working on vehicles for 37 years and I don't see anything wrong with simply torqueing the bolts. worked for 50 years.

 

[dnschmidt]

Torque to yield is here to stay. The reason that is true is it allows smaller and in some cases fewer fasteners to get the same job done. With the emphasis on fuel economy weight matters and this lowers weight. It also saves space which on modern vehicles is in short supply.

What you're primarily doing with this is maximizing the compressive force of a bolt. The bolt has the most compression when it is stretched to it's optimum point. That's all this method is trying to do. Torque to yield is a pain in the *** to implement but when you think about it the concept makes sense.

 

[gtermini]

I have been working on vehicles for 37 years and I don't see anything wrong with simply torqueing the bolts. worked for 50 years.

It was fine before, but now everything is being reduced to absolute minimum size to keep cars light and meet fuel efficiency standards. What was once held on by a3/8 bolt because that looked right, is now held with a 6mm that needs to be tightened to it limit to hold. These are non-reusable fasteners since they were taken to yield and are permanently deformed.

EDIT: dnschmidt beat me to it.

Greyson

 

[Daves69]

The type of lubricant make a difference also.

www.youtube.com/watch?v=FjBaWo0QMYU

 

[gearhead1]

Actually, torque to yield method provides for more accurate and thus more even clamp load. Yes, torque worked for 50 years and is good but torque to yield is better for critical fasteners in an application.

joe_padavo hit it right on the head. I'll say it a slightly different way which hopefully someone understands it if they didn't already.

The overall goal of the fastener is to provide clamp load to the design. The more you torque the fastener (a screw is an inclined plane of the simple machines just wrapped around) the more it stretches (microscopically). The stretch is what provides the clamp load. You can't stretch it past the yield point of the material or it will permanently deform. If you stretch it up to the elastic limit of the material, the fastener will go back to its original length, think of it like a rubber band.

The bottom line is, you're really after the most accurate way to consistently get the same bolt stretch in an application. When the design is being set up, fasteners are ground parallel on each end for a bolt gauge to work. Those fasteners are used in a study that determines the torque and angle. The bolt gauge sends a wave in the bolt which goes to the other end and back up. Based on how long it takes is the length of the bolt and has a resolution of +/– 0.00001 inch. (http://www.dakotaultrasonics.com/product/bolting-products-2/bolting-products/)

Consider a cylinder head, long, with multiple bolts in it. The goal is not just a tight clamp load, it's also even clamp load along the length of it so the gasket doesn't fail. So it's very critical to have even clamp load on the entire length. Notice I say clamp load, not torque. Torque relates to clamp load but that's the rest of the story......

If you take an application, torque the fastener you will get a certain amount of clamp load, let's say 100lbs of force.
Now, if you were to take a fastener and polish the threads smooth and put oil on it and then torque it with the same torque, the clamp load will be 150 pounds of force! Why? Because you changed the friction in the application. This is not optimal for a cylinder head for example. In the old days the clamp load across the cylinder head was very uneven, we just didn't know it.

Gasket technology has improved, which drove the need to increase the clamp load of the cylinder head. When the clamp load was increased, the application is more susceptible to the variation now. Friction is much less of a factor when the torque is low. But when you increase the torque, the effects of friction are a bigger factor right along with it. This is why the process always starts with a small torque value. Then it goes to angle. Why? Because angle can be accurately controlled. When you torque to an angle it has absolutely nothing to do with friction! But when you tighten by torque alone, the bolts that had more oil on them and the holes with smoother threads will have more clamp load.

How it works in the factory I worked at on engine cylinder heads is the machine would do the torque-angle-angle process, then the machine would measure the 'resultant' torque that it took to get it there. The resultant torque was in a wide window due to the factors of lubrication and thread quality. One time when the resultant torque was out of the window, the shop floor raised the flag. Engineering told them to ship the product. But manufacturing was flabbergasted! How can this be!? The clamp load was the same, the difference was the new bolt supplier had a slipperier coating on the bolts.

It's actually quite simple, I can explain if you have questions.....

 

[Wakefield]

Think of the hexhead capscrew "bolt" and nut on the end being tightened to clamp what is between them,as the "bolt" stretches its threads distort by getting longer between gyres but the threads in the nut don't or even get compressed together,so the threads might start to interfere causing much more friction resistance,sort of like a pipe thread

 

[jmm]

Don't know how prevalent it is outside of my field, but the current line-up of Volvo/Mack diesel engines take it a step further. Take rocker arm hold-down bolts. The service literature specifies that you clean the bolt holes and apply engine oil to the bolts, then torque to 48 ft/lbs, tighten another 180 degrees, back off 110 degrees, check to see that the bolt is at 7-11 ft/lbs, and tighten another 90 degrees. You end up at about 145 ft/lbs with all the stretch removed from the bolt...in theory at least. A lot of the critical engine hardware is torqued this way, and as best I can tell it works. As to why and the reason for, that's above my pay grade.

 

[gearhead1]

Adding lubrication to the bolts minimizes the friction effects. If you do it every time, it's consistent. If you don't do it, then some will have lube, some won't, which will equate into some have more clamp load than others.

The place I used to work lubricated head bolts for that very reason - to minimize friction effects. It doesn't completely remove the variable, but it significantly reduces it.

 

[dnschmidt]

Actually ARP gives you lube when you buy their bolts. In fact TTY should to a large degree eliminate most of the negative effects of different lubrication. Here's why. You start off by torqueing to a substantial but low value. In short, most of the slop is gone one you've achieved whatever the first value is. Once you get to the degree part the amount of stretch is completely determined by the thread pitch. If you've got a 3/8" by 16 bolt for example and move it another 90 degrees you have stretched the bolt by 1/16" divided by 4 = 1/64" regardless of lubrication.

This more precise stretching is why TTY works better regardless of bolt condition. Once you've made up the slop the stretching is simple math. Using only torque a lot more gets into the equation.

 

[chillrich]

Wouldn't the degree torque be just as inaccurate if it's starting point is a torque value?

 

[joe_padavano]

Wouldn't the degree torque be just as inaccurate if it's starting point is a torque value?

The whole point is that the starting point torque is low enough that variations due to friction are small and thus within the design tolerances. The higher the torque, the higher the friction force and thus the greater the variability in each fastener's preload. Again, the only way to get preload set perfectly is to either use an instrumented bolt or measure bolt stretch. That's not practical in most cases, thus this is the next best method.

As for using torque measurement only for 50+ years, the big difference is that in the last decade or so, automakers have been forced to dramatically reduce structural margins in components to reduce weight and increase mileage. When a cast iron block was so overdesigned that a 50% difference in installed fastener preload didn't matter, torque measurement only was sufficient. Today, those margins are shaved to the bare minimum and more accurate fastener torque measurements are necessary.

 

[joe_padavano]

Always think of a bolt or stud as a spring. What you're doing when you tighten it is actually tensioning it.

It's not just the fastener that's a spring. The material being clamped also has some compliance and is thus a spring. Obviously, the axial stiffness of a given fastener depends on the fastener material properties and diameter, but preload also depends on the material properties of the item being clamped. Since (for example) cast iron is three times stiffer than aluminum, a given fastener clamping a cast iron part requires less of a turn angle to achieve a given preload than it does when clamping an aluminum part, since the clamping force in the bolt must be reacted against the part being clamped. The aluminum part is "springier" than the iron part and thus compresses further for a given load.

 

[Crown Imperial]

Don't know how prevalent it is outside of my field, but the current line-up of Volvo/Mack diesel engines take it a step further. Take rocker arm hold-down bolts. The service literature specifies that you clean the bolt holes and apply engine oil to the bolts, then torque to 48 ft/lbs, tighten another 180 degrees, back off 110 degrees, check to see that the bolt is at 7-11 ft/lbs, and tighten another 90 degrees. You end up at about 145 ft/lbs with all the stretch removed from the bolt...in theory at least. A lot of the critical engine hardware is torqued this way, and as best I can tell it works. As to why and the reason for, that's above my pay grade.

This doesen't compute to me.
What they are basically doing is either overstretching the bolt or not clamping it tight enough at the end. I just don't have any other way of explaining it.

 

[zmotorsports]

joe_padavano has given the best articulated description I have heard to date.

The point is to get the initial torque within the range prior to friction becoming too much of a contributing factor, then turning the additional degrees of rotation to stretch the fastener because like was already mentioned these are easily determined by the specifications of the fastener. It is a much more accurate means of setting the preload of the fastener and applying the clamping forces necessary.

Mike.

 

[Hybridss]

I would not recommend you do this. Use a degree wheel / dial gauge. Not very likely you will get a 90 degree TTY with a +/- 2.5 deg. tolerance accurately with a sharpie.

The sharpie is not to line up the 90 degrees. Its to keep track of which bolts have been tq'ed when doing multiples passes. That way if you get a phone call or something you know where you left off. The long breaker bar or ratchet is typically enough to get within that 5 degree range fairly easy if you are a competent mechanic and use some basic techniques that aid in accuracy.

I have Tq'ed head bolts on Chevy Gen III small blocks making substantial hp using this method many many many times and have never had a failure due to tq issues. When making big power you'll typically use something like ARP head studs anyways. Then you are back to regular Tq values.

But it would depend on what I was assembling and the critical nature of it.

 

[veks]

Thank you guys for the lesson on torque and angle.

The sharpie is not to line up the 90 degrees. Its to keep track of which bolts have been tq'ed when doing multiples passes. That way if you get a phone call or something you know where you left off.

nice save

 

[chillrich]

The whole point is that the starting point torque is low enough that variations due to friction are small and thus within the design tolerances. The higher the torque, the higher the friction force and thus the greater the variability in each fastener's preload. Again, the only way to get preload set perfectly is to either use an instrumented bolt or measure bolt stretch. That's not practical in most cases, thus this is the next best method.

As for using torque measurement only for 50+ years, the big difference is that in the last decade or so, automakers have been forced to dramatically reduce structural margins in components to reduce weight and increase mileage. When a cast iron block was so overdesigned that a 50% difference in installed fastener preload didn't matter, torque measurement only was sufficient. Today, those margins are shaved to the bare minimum and more accurate fastener torque measurements are necessary.

Excellent. Thanks

 

[67King]

Maybe this will help give a visual. If you look at the graph, you'll see the straight line that doesn't curve, which represents the material's modulus. No matter how far you go, that slope will always be followed, as it is how much the material stretches up to the yield point. In a TTY fastener, you intentionally yield the fastener, and the clamp load is going to be equal to the modulus, no matter how much you yield it. In other words, it doesn't really have to be exactly 90 degrees (if it did, you couldn't re-use them).

The clamp load is determined by the smallest area of the bolt. A TTY bolt almost always has a necked down area that is precisely controlled in manufacturing, as that is the area that will yield and the return from the stretched condition (the thicker portion will not yield). As previously stated, this gives tighter control and more uniform clampload that is robust to variability. It is kind of like using the stretch method to determine clampload where you can't get a strain gauge on the fastener.