Torque Values (Rule of thumb?) 4

[sbozy25]

Here are the 2 bolted joint scenarios I'm looking at.

1. I have a 1"-14 bolt with a maximum service loading of 8,880lbf. The bolts are black oxide coated so I'm using a k factor of .25 - .21. My calculations show that I need torque values ranging from 714 ftlbs to 850 ftlbs to obtain the necessary clamp to resist slipping. I am recomending torque of 775 ftlbs to 825ftlbs to obtain necessary clamp. I have clamp loads calculated at 3,543 lbf to 3,771 lbf respectively.

2. I have a 7/8" bolt with a maximum service loading of 5,550lbf. Everything is same with coating. I'm showing a need of 450 ftlbs to 500 ftlbs with clamp loads calculated at 2,057 lbf and 2,286 lbf respectively.


I have been questioned if these torque values are to high for the grade 8 nuts and bolts we are using. From what I can see, I'm staying well below proof loads of both sizes.

So here is my question: Is there a rule of thumb I should consider for 1" and 7/8" bolts? A fellow engineer here as stated that I might consider dropping the 1" bolt to 500 ftlbs and the 7/8" to 400 ftls. I'm concerned that I may not get a sufficient clamp.


Any thoughts?

Cabbages, knickers, It hasn't got A BEAK!

 

[CoryPad]

A rule of thumb is to use torque values that will generate preload between 50 % and 100 % of the proof stress. It appears from simple examination of your proposed dimensions, materials and torque values that you are in the general range of acceptability. It appears that your colleague is overly cautious regarding how much preload to develop, which can lead to joints that do not resist the applied forces.

For some reference documents, try NASA RP 1228 Fastener Design Manual, RCSC's Guide to Design Criteria for Bolted and Riveted Joints, and MIL-HDBK-60 Threaded Fasteners - Tightening to Proper Tension.

 

[desertfox]

Hi sbozy25

Any bolt should be preloaded to suit the application not to what somebody thinks and the general rule of thumb where I am working is:- for a joint not intended to be taken apart at anytime it would be preloaded to 90% of proof stress and 70% proof stress for joints that need to be dismantled.

Try this site it may help;-

http://www.roymech.co.uk/Useful_Tables/Screws/index_screws.htm

desertfox

 

[sbozy25]

Cory and Desert:

Thanks to both of you. I feel better now leaving my torque values as they are. My colleague was worried that I would snap the head off the bolts, but I am relatively low in stress so I doubt that is a problem. Well, as long as the supplier truly supplies grade 8...

Cabbages, knickers, It hasn't got A BEAK!

 

[Eddycurrentguy]

Good stuff, Desertfox.

Here's another site that might be of interest:

http://www.wellheadservices.net/Technical.htm

 

[JLNJ]

You can look at bolt elongation to check your numbers since "torque" values can vary widely.

Structural people use torque values less and less these days.

 

[Duwe6]

JLNJ has 'hit it on the head'. To correlate torque input to bolt elongation & stress, you have to "ask the bolt". AISC Structural Code requires you to develop your installation torque based on measured tension. google 'Skidmore-Wilhelm'
Off the top of my head, an A325 [similar to a Grade 5] requires 28 kips of tension to be considered 'fully pretensioned'.

The problem with torque is that sliding friction a HUGE element in the equasion to mathematically correlate torque to tension/stress. Smoothness of the male & female threads, and the type of lube used make any calculated torque unreliable. For a general torque on 1"-8tpi high-strength bolts/stude, we use 370 ft-lbs.

To go significantly higher than that, like you are, I would either run a 5-bolt sample through a 'Skidmore', or use Direct Tension Load Indicating washers to develop a reasonable torque value for your exact bolts.

 

[sbozy25]

Thanks to everyone for the feedback. Part of the calculations I run is predicted bolt stretch and the correlating pre-tensions.

I have found that there are multiple ways to calculate clamp loads and multiple "k" factors that can be used. So what I have done is develop a mathcad set up that runs the different clamp load calculations over a range of "k" factors and then I look at my range to determine bolt torque.

I would love to use a skidmore and qualify my torque values, unfortunately I do not have time. I am stuck with pure calculations. I was just blown away that I was being told the bolts would break, when I can clearly see they are below proof... I'm sure I will be ok, this is a low dynamic loading vehicle and I'm using really thick steel.

Again, thanks to everyone for the great feedback!

Cabbages, knickers, It hasn't got A BEAK!

 

[Duwe6]

Again, try Direct Tension washers. Either the 'titted' ones or the 'Squirters'. They can be had to show AISC pretension for sixes up to at least 1-1/4" and both A325 [Gr.5, more or less] and A490 [Gr.8 and B7, more or less]. Your fixture can be a slice of heavy-wall pipe with 2-4 flatwashers on each end. Waaay cheaper than buying a Skidmore. You have to determain your "k" experimentally -- every bolt manufacturer has different 'smoothness' which, when considered with what lube your your shop chooses to use, gives a huge range of possible "k" factors. I've seen 200% difference in torque values when both bolt mfr and lube are changed.

google "squirter washer"

 

[sbozy25]

Again, I would if I had time.

This program has already launched and the customer will be bolting the key joints down any day now.

I have a bit of time before production starts, so I might order some bolts and go out to our skidmore.... might... lol

Cabbages, knickers, It hasn't got A BEAK!

 

[CaliSolarMechE]

The RCSC Guide to using ASTM 325 fasteners (

http://www.boltcouncil.org/files/2004RCSCSpecification.pdf)

suggests another method of controlling bolt stress--- using the "turn of the nut" method.

In theory, you get the nut tight without preloading the bolt. Then you turn the nut a predetermined number of degrees, hence pulling it to a predetermined strain and thus, tension.

The strength of this method is that there is no variation due to friction, as with the torque method.

The well-known drawback is that it's difficult to control for the starting point; one technician's "finger tight" can be way different from another's.

I hope this late reply doesn't come as overkill!

 

[Guest102023]

I just found this yesterday, it is for Grade B7 bolts with various lubricants

 

[flash3780]

Often the best torque to use is the one that worked in the last design. Absent a comparative analysis, you generally want to crank bolts down to near the yield point since the flange is what takes the vast majority of the actual loading. Separating flanges are what often lead to failures of bolted connections.
Another reason higher torques are good is it prevents loosening. Locking features are great, but start with a properly torqued fastener and it's much less likely to vibrate loose from the start.