Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
bolting torque 7
- Thread starter woodman90
- Start date
-
2
- #2
There is no standard torque value for tensioning bolts. Download the RCSC bolt spec from
-
1
- #4
-
1
- #5
cap4000,
No!!! A325 refers to the ASTM standard that the material for the bolts must meet. A490 is the ASTM standard for the material for the A490 bolts.
taro is correct. In the RCSC spec referenced, there is no standard torque, the torque required must be determined by a trial installation of the bolts you will use (the same dia, same length, same materials, installation methods, same wrenches, same everything you will use in the actual connections). You will have to do the test in a machine like a Skidmore-Wilhelm machine, and you will have to do it at least once a day.
Suggest you use either turn of the nut method or DTI's (also known as load indicating washers) to get the correct pretension on the bolts. Correct pretension is what you need to have. Or, if it is permitted for your case, consider designing connections that use only snug tight bolts.
Regards,
chichuck
No!!! A325 refers to the ASTM standard that the material for the bolts must meet. A490 is the ASTM standard for the material for the A490 bolts.
taro is correct. In the RCSC spec referenced, there is no standard torque, the torque required must be determined by a trial installation of the bolts you will use (the same dia, same length, same materials, installation methods, same wrenches, same everything you will use in the actual connections). You will have to do the test in a machine like a Skidmore-Wilhelm machine, and you will have to do it at least once a day.
Suggest you use either turn of the nut method or DTI's (also known as load indicating washers) to get the correct pretension on the bolts. Correct pretension is what you need to have. Or, if it is permitted for your case, consider designing connections that use only snug tight bolts.
Regards,
chichuck
Can we go to the mechanics a bit.
Manufacturers and orgs like boltcouncil specify those torques and we get explanations like those torques will provide full friction capacity which can usually be calculated independent of torque. So there is something hidden here or I am missing something.
What if by mistake I overtorque or undertorque. How can I defend my design then and take a better nap?
Respects
ijr
Manufacturers and orgs like boltcouncil specify those torques and we get explanations like those torques will provide full friction capacity which can usually be calculated independent of torque. So there is something hidden here or I am missing something.
What if by mistake I overtorque or undertorque. How can I defend my design then and take a better nap?
Respects
ijr
-
1
- #7
SlideRuleEra
Structural
I'll add my 2 cents, having started as a mechanical engineer and later becoming a structural engineer. First of all - I agree with Taro & chichuck - torque measurements have no place in structural steel erection.
Other than being just a dowel in a hole, bolts work as springs following Hooke's Law - they must be stretched the proper distance to apply the correct clamping force. Torque measurements do not measure the "stretch" directly but assume certain values for friction between the bolt, nut, washers, etc. In many mechancial engineering applications torque measuements are appropriate - the parts & working conditions are relatively pristine, conditions are controlled (lubricated), and the actual clamping requirements have wide tolerances. Also many "bolts" used for mechanical applications are not bolts at all, but are cap screws or nuts on threaded studs.
For many structural applications the bolts and nuts may be staight from the supplier, OR they may have been sitting around the job site for weeks or months accumulating "light rust". In any event field conditions are not as well controlled (friction properties vary widely, making torque measurements unreliable). The fasteners are still acceptable, but to work properly they must be installed using techniques that are accurate (measure the stretch of the bolt).
Other than being just a dowel in a hole, bolts work as springs following Hooke's Law - they must be stretched the proper distance to apply the correct clamping force. Torque measurements do not measure the "stretch" directly but assume certain values for friction between the bolt, nut, washers, etc. In many mechancial engineering applications torque measuements are appropriate - the parts & working conditions are relatively pristine, conditions are controlled (lubricated), and the actual clamping requirements have wide tolerances. Also many "bolts" used for mechanical applications are not bolts at all, but are cap screws or nuts on threaded studs.
For many structural applications the bolts and nuts may be staight from the supplier, OR they may have been sitting around the job site for weeks or months accumulating "light rust". In any event field conditions are not as well controlled (friction properties vary widely, making torque measurements unreliable). The fasteners are still acceptable, but to work properly they must be installed using techniques that are accurate (measure the stretch of the bolt).
-
1
- #8
can i add my 2 penneth too?? It concernes me very much when i see people trying to find a link between torque and tension, the two methods do not go together at all, no matter what certain tool manufacturers try to say, tension is directly stretching the bolt, and therefore creates the correct loading of a bolt, or it is hoped to be that way, there is only one sure fire way to check that the tension is correct, and that is to have the bolts in question machined flat on both ends of the bolt, and use an ultrasound elongation detection machine, once the load has been put into the bolt, there is no sure fire way of testing how far its relaxed, but again the use of the ultrasonics does put you in the right ballpark so to say.
Some manufacturers (hy-torq, torq-up and raymond) will tell you that using their tools, you can check the tension in bolts by turning the nut, this is misleading and dangerous, and is totaly unprofessional, no matter how much these manufacturers of torque wrenches want you to believe that they can assist when it comes to tension, they cant, no matter how much they dress it up in long words, and scientific claptrap, we are talking about two different forces here, with no reletive point of reference, plus you have the added problem of torque being so unreliable for a true reading, there are so many influences that can alter the determined outcome, ie, is there lubrication on the bolts or not? is the threads of the nut and bolt clean, is the surface area between nuts/washers/and surface clean, all of these will efect the torque output on your machines, and there is no way of monitoring the damage done to the bolt, best thing to do is stick to the codes as layed down, remember most of these codes have been written in blood so to say, the accident has already happened, so with hindsight, we can determine what needs to be tensioned, and what needs to be torqued, and above all dont listen to anyone who says that they can use torque to determine the laoding of a bolt, they cant, and never will be able to, either.
Some manufacturers (hy-torq, torq-up and raymond) will tell you that using their tools, you can check the tension in bolts by turning the nut, this is misleading and dangerous, and is totaly unprofessional, no matter how much these manufacturers of torque wrenches want you to believe that they can assist when it comes to tension, they cant, no matter how much they dress it up in long words, and scientific claptrap, we are talking about two different forces here, with no reletive point of reference, plus you have the added problem of torque being so unreliable for a true reading, there are so many influences that can alter the determined outcome, ie, is there lubrication on the bolts or not? is the threads of the nut and bolt clean, is the surface area between nuts/washers/and surface clean, all of these will efect the torque output on your machines, and there is no way of monitoring the damage done to the bolt, best thing to do is stick to the codes as layed down, remember most of these codes have been written in blood so to say, the accident has already happened, so with hindsight, we can determine what needs to be tensioned, and what needs to be torqued, and above all dont listen to anyone who says that they can use torque to determine the laoding of a bolt, they cant, and never will be able to, either.
Although I avoid pre-tensioned/slip-critical bolting arrangements, I do like to have the bolts I specify torqued to some degree of satisfaction. The turn-of-nut procedure is reasonable, however it often sounds nicer to specify a torque value to achieve. Here are a couple of formulas that I use as a guideline.
Take with a BIG grain of salt - and do your own digging on the matter before using any torque value.
For A325 - coarse thread - zinc galv - clean - non-lubricated
Torque = 0.38 * Tension ^ 1.4725 (units of N-m)
or
Torque = 0.2 * BoltDia * Tension <- produces slightly different results
For A490 - coarse thread - zinc galv - clean - non-lubricated
Torque = 0.4 * Tension ^ 1.5025 (units of N-m)
The tension value which I use is 70% of the yield strength of the bolt.
However, as all the others have said previous, if I do require a proper slip-critical connection with properly pretensioned bolt - it is not recommended to simply use a torque value.
Anyone else have any formulas that they use for rule-of-thumb conditions?
DRW75
Take with a BIG grain of salt - and do your own digging on the matter before using any torque value.
For A325 - coarse thread - zinc galv - clean - non-lubricated
Torque = 0.38 * Tension ^ 1.4725 (units of N-m)
or
Torque = 0.2 * BoltDia * Tension <- produces slightly different results
For A490 - coarse thread - zinc galv - clean - non-lubricated
Torque = 0.4 * Tension ^ 1.5025 (units of N-m)
The tension value which I use is 70% of the yield strength of the bolt.
However, as all the others have said previous, if I do require a proper slip-critical connection with properly pretensioned bolt - it is not recommended to simply use a torque value.
Anyone else have any formulas that they use for rule-of-thumb conditions?
DRW75
Um, yeah. Meanwhile, in the real world of installing a couple hundred bolts per slip-critical connection 50 ft. up in the air, turn-of-the-nut is pretty damn close to an actual elongation measurement.
On-site inspection to ensure proper bolt lubrication is a must. Otherwise the bolts might not be able to take the proper tension no matter how you measure it.
That said, some people who are usually very conservative has recently decided to allow calibrated hydraulic torque wrenches to be used on their bridges instead of turn-of-the-nut. They included lots of caveats in the spec:
So it's still very much tension-based, but there is more attention that needs to be paid to the need to recalibrate.
Back to the original question--don't specify torque. Specify tension; it's what you design with. Specify how many turns of the nut, or direct tension indicators.
Check out the Research Council for Structural Connections. It's connections to steel, not to wood, but probably still have plenty of useful info for you:
Hg
-
1
- #11
HgTX...you beat me to the punch! There is a relationship between tension and torque, it's just that it's difficult to quantify because of all the variables. After all, turning a nut at some torque is just like sliding a box up an inclined plane. Same work principle applies.
DRW75...I've used similar in the past for "ballpark" estimates. They are surprisingly close on uncoated A325's that have not rusted but are not heavily oil coated either, when checked with a Skidmore-Wilhelm device.
Zepp....don't know where you practice, but it isn't in the construction industry.
DRW75...I've used similar in the past for "ballpark" estimates. They are surprisingly close on uncoated A325's that have not rusted but are not heavily oil coated either, when checked with a Skidmore-Wilhelm device.
Zepp....don't know where you practice, but it isn't in the construction industry.
One should also not confuse "torque" with "rotation"--I think the term "torque" gets used a little loosely, no pun intended.
Actually, that makes me wonder. Is the initial question asking about a torque measurement in ft-lb or N-m or are they really asking for how many turns of the nut?
Hg
Actually, that makes me wonder. Is the initial question asking about a torque measurement in ft-lb or N-m or are they really asking for how many turns of the nut?
Hg
HGTX
thanks for highlighting the quote, but used a little out of context, or maybe my error was in using wrong context, I was stressing on the only sure fire way of knowing the stretch in a bolt, even then it is open to some ellement of mistrust, its always difficult to determine if you have really done the job right, but so long as industry standards are adhered to, then you can sleep safely
Ron.
I am not based in the construction industry alone, infact i cover most industries, and my main concern in the work that i do is bolt tensioning and bolt torquing.
I have been involved in the construction of bridges, and buildings, using all manner of methods, load indicating washers, torque, quater turn, tension, when the method layed down by the designer is adhered to, then usualy there are no problems, the biggest problem when it comes to bolting a joint together is people like me, and im being very honest here, people like me that make a living out of selling, renting, and using bolt tensioning and torquing systems can be a nightmare on what should be a very simple job, because the loyalties are to the manufacturer of the equipment, and not to the job in question, this is why so many companies, end users are sat with pieces of equipment that look very elaborate, yet, they just sit there, and the only value they add to the company is that one day it will end up in the christmas scrap metal fund for the workforce party.
The first iron bridge ever built did not have Hy Torq, or raymond, or any other manufacturer around to tell them how best to put the nut and bolts together, they just stuck to the design, and what the manufacturer layed down about the bolts, and amazing as it may seam, its still standing.
My problem with areas like this, is when people try to tell you that there is a relation between torque and tension, if this is so, can you tell me how much load you will put in a 65mm bolt when 2500Nm torque is applied to the nut?? what?? you want to know what lubricant im using?? how far out is the pressure guage on my pump?? are the seals passing in my torque wrench?? ... you see, the questions go on and on, and still there will be no direct figure arrived at, the only relation between torque and tension is that turning the nut will stretch the bolt, to some extent, period, there is no reliable value, these are the points that i am trying to stress, for torque applications, stick to torque, for tension applications, stick to tension, dont try to over complicate the situation, because when you do, your at the mercy of the salesman.
thanks for highlighting the quote, but used a little out of context, or maybe my error was in using wrong context, I was stressing on the only sure fire way of knowing the stretch in a bolt, even then it is open to some ellement of mistrust, its always difficult to determine if you have really done the job right, but so long as industry standards are adhered to, then you can sleep safely
Ron.
I am not based in the construction industry alone, infact i cover most industries, and my main concern in the work that i do is bolt tensioning and bolt torquing.
I have been involved in the construction of bridges, and buildings, using all manner of methods, load indicating washers, torque, quater turn, tension, when the method layed down by the designer is adhered to, then usualy there are no problems, the biggest problem when it comes to bolting a joint together is people like me, and im being very honest here, people like me that make a living out of selling, renting, and using bolt tensioning and torquing systems can be a nightmare on what should be a very simple job, because the loyalties are to the manufacturer of the equipment, and not to the job in question, this is why so many companies, end users are sat with pieces of equipment that look very elaborate, yet, they just sit there, and the only value they add to the company is that one day it will end up in the christmas scrap metal fund for the workforce party.
The first iron bridge ever built did not have Hy Torq, or raymond, or any other manufacturer around to tell them how best to put the nut and bolts together, they just stuck to the design, and what the manufacturer layed down about the bolts, and amazing as it may seam, its still standing.
My problem with areas like this, is when people try to tell you that there is a relation between torque and tension, if this is so, can you tell me how much load you will put in a 65mm bolt when 2500Nm torque is applied to the nut?? what?? you want to know what lubricant im using?? how far out is the pressure guage on my pump?? are the seals passing in my torque wrench?? ... you see, the questions go on and on, and still there will be no direct figure arrived at, the only relation between torque and tension is that turning the nut will stretch the bolt, to some extent, period, there is no reliable value, these are the points that i am trying to stress, for torque applications, stick to torque, for tension applications, stick to tension, dont try to over complicate the situation, because when you do, your at the mercy of the salesman.
The first iron bridge ever built probably had rivets. But anyway...
You're absolutely right, there's no consistent relationship between torque and tension. That's one reason I don't really trust those twist-off bolts.
There IS a consistent relationship between bolt ROTATION and tension. That's why turn-of-the-nut works.
The RCSC spec does acknowledge a calibrated wrench tightening method. According to their commentary, they took it out in 1980 because people were doing the wrong thing--using standard torque values which as you say ain't no good. They put it back in in 1985 on the grounds that if people do the right thing, it does work, and it's not the method's fault that people do the wrong thing. (The RCSC method is the basis for what I quoted above.) It calls for determining a torque-tension relationship on a situation-by-situtation basis.
The RCSC spec (94 pages, 1.4 MB) is here:
Woodman90, you'll be able to find the proper nut rotation to specify in Table 8.2 of the RCSC spec, or else use direct tension indicators. DTIs are more reliable because turn-of-the-nut requires determination of "snug-tight" condition which can be an unclear concept, but they're harder to inspect than just eyeballing the turn marks, and so can be impractical for very large connections (unless you trust those squirter DTIs). You'll see that TOTN, done properly, actually allows for what would be quite a range of tensions if they were measured directly.
Hg
You're absolutely right, there's no consistent relationship between torque and tension. That's one reason I don't really trust those twist-off bolts.
There IS a consistent relationship between bolt ROTATION and tension. That's why turn-of-the-nut works.
The RCSC spec does acknowledge a calibrated wrench tightening method. According to their commentary, they took it out in 1980 because people were doing the wrong thing--using standard torque values which as you say ain't no good. They put it back in in 1985 on the grounds that if people do the right thing, it does work, and it's not the method's fault that people do the wrong thing. (The RCSC method is the basis for what I quoted above.) It calls for determining a torque-tension relationship on a situation-by-situtation basis.
The RCSC spec (94 pages, 1.4 MB) is here:
Woodman90, you'll be able to find the proper nut rotation to specify in Table 8.2 of the RCSC spec, or else use direct tension indicators. DTIs are more reliable because turn-of-the-nut requires determination of "snug-tight" condition which can be an unclear concept, but they're harder to inspect than just eyeballing the turn marks, and so can be impractical for very large connections (unless you trust those squirter DTIs). You'll see that TOTN, done properly, actually allows for what would be quite a range of tensions if they were measured directly.
Hg
Just wanted to add that the pretension in bolts, valued at 70% of the bolt material's ULTIMATE (not the yield) strength is based on the net effective tensile area and not the nominal bolt diameter.
An excellent source for understanding the mechanics involved with STRUCURAL bolts is the "Guide to Design Criteria for Bolted and Riveted Joints", second edition by G. Kulak, J.W. Fisher and J.H.A. Struik. I downloaded a free copy from the web using Google.
An excellent source for understanding the mechanics involved with STRUCURAL bolts is the "Guide to Design Criteria for Bolted and Riveted Joints", second edition by G. Kulak, J.W. Fisher and J.H.A. Struik. I downloaded a free copy from the web using Google.
HGTX
The first iron bridge has both, rivets and nut and bolt, but I agree with your last statement in as much that the bolt council have advised that turn of the nut after torquing has been done is acceptable, and as such, on site there should be a bolt load meter to test a certain number of bolts per day to ensure that they are not exceeding the yield, my chief concern was people trying to associate torque with tension, when that happens, as I said earlier, you become open to the unscrupulous salesman, and could end up with all manner of equipment that is not suitable.
The first iron bridge has both, rivets and nut and bolt, but I agree with your last statement in as much that the bolt council have advised that turn of the nut after torquing has been done is acceptable, and as such, on site there should be a bolt load meter to test a certain number of bolts per day to ensure that they are not exceeding the yield, my chief concern was people trying to associate torque with tension, when that happens, as I said earlier, you become open to the unscrupulous salesman, and could end up with all manner of equipment that is not suitable.
pmkPE,
Thanks for that, a slip on my part!
Without thinking about it too much, my gut feeling would be 70% yield since one always stays away from looking at ultimate values for design (or i do anyways)... but you're right, it's 70% ultimate...
DRW75
Thanks for that, a slip on my part!
Without thinking about it too much, my gut feeling would be 70% yield since one always stays away from looking at ultimate values for design (or i do anyways)... but you're right, it's 70% ultimate...
DRW75
I have found the turn-of-the-nut method to be impractical for most applications. I do believe that it is the most reliable, but many inspectors refuse to accept it unless they are right there watching you install each bolt. I prefer DTI's mostly because they can be check after the fact. I don't like the torque wrench test at all. On many jobs that I'm associated with, the installer will put the bolts on snug tight, and then come back days...weeks...etc later to "torque" the bolts. Miracuously, they can get the "torque values" before closing the gap in the DTI. Then they demand to know why the torque value isn't good enough. Also, inspectors are wonderful about seeing that DTI's or turn-of-the-nut method are prescribed on the drawings, then demanding that a torque test be performed. Heck, I know that torque tests have been out of favor with the RCSC for years, but once someone "knows how things are supposta be done" you have an uphill battle convincing them otherwise. Also, is it really worth the days of trying to convince someone of an alternate way when only a few bolts are comeing into question. My favorite part about a torque value is that a rusty seized bolt can pass a torque test and barely be snug tight, and a super lubricated bolt with an uncalibrated torque wrench can break before achieveing torque. My two-cents for field aplications: use DTI's or twist-off bolts. I don't believe that twist-offs are as reliable, but if an inspector sees the ends twisted off the bolt, then he usually doesn't question the bolt beyond that.
I've seen the "squirter" washers, but if your using an impact wrench, the orange markers can wander off erasing the visiual evidence that pretensioning has been achieved.
I've seen the "squirter" washers, but if your using an impact wrench, the orange markers can wander off erasing the visiual evidence that pretensioning has been achieved.
What's practical or im- depends on the application. For example, DTIs aren't as well accepted in the bridge market as their manufacturers would have hoped because it's quite impractical to go through a few hundred bolts in a deep girder splice with a feeler gauge, while eyeballing turn-of-the-nut marks is a simple affair. (Yes, it's possible to cheat by putting the marks on afterward. That's where good onsite inspection comes in.) Some agencies mandate DTIs anyway, but they're not the "obvious" choice one might think they'd be.
Twist-offs are torque-based and are subject to all the same cautions as using a calibrated torque wrench method, particular concerning bolt lubrication condition. And if for some reasoning retightening is needed, they'd need to be replaced. I Am Told this is also the case for DTIs.
"...if an inspector sees the ends twisted off the bolt, then he usually doesn't question the bolt beyond that." Is that a good thing or a bad thing?
Another problem with twistoffs is that for large connections a large number of fitup (non-twistoff) bolts must be used, which means two kinds of wrench will be needed in the field. Or so I am told.
Hg
Twist-offs are torque-based and are subject to all the same cautions as using a calibrated torque wrench method, particular concerning bolt lubrication condition. And if for some reasoning retightening is needed, they'd need to be replaced. I Am Told this is also the case for DTIs.
"...if an inspector sees the ends twisted off the bolt, then he usually doesn't question the bolt beyond that." Is that a good thing or a bad thing?
Another problem with twistoffs is that for large connections a large number of fitup (non-twistoff) bolts must be used, which means two kinds of wrench will be needed in the field. Or so I am told.
Hg
Let me recommend the Structural Bolting Handbook published by SSTC. It's a pocket-sized little book with all of the topics that have been addressed in this thread. Bob Shaw runs this firm, and is highly involved with the RCSC spec. He has been an excellent resource for our company when we have had questions regarding all types of bolting and welding.
You can find it on their website at
You can find it on their website at
- Status
Similar threads
- Locked
- Question
