Tightening bolts to 90% yield stress. 2

[wannadie]

When tightening a bolt, I have read that I should tighten to provide preload of the order to 90% of the yield stress in the fastener.

I understand the importance of preload, but find the 90% extremely high with little safety factor due to the unpredictable nature of coefficient of friction and tightening controls.

What do other people do? Are there any other standards besides VDI 2230 which is really far to complex for my needs.

 

[rb1957]

the preload should depend on the tension load the bolt sees in service. if the load is loaded primarily in shear, then the preload should be much lower and this is where i'd use the typical design guidelines for bolt torques, here's a typical link ...

http://www.boltdepot.com/fastener-information/US-Recommended-Torque.aspx

this is something much discussed here see javascriptpenindex 450,450,'

http://www.tipmaster.com/includes/refinfo.cfm?w=450&h=450')

there are some (me included) who see value in these typical tables (where used with a bit of knowledge), there are others who would always calculate.

if the joint is loaded in tension, and particularly if it is loaded repeatedly in tension (fatigue critical) then the preload becomes a key design parameter. strictly the preload should be designed around the service load. if the preload is less than the service load, then the ultimate condition of the bolt is gapped and the bolt strength is designed by the srevice load. if the preload is higher than the service load then the ultimate load in the bolt can be higher than the service load (as the joint isn't gapped). that being said i don't mind using design guidelines (with a bit of knowledge). 90% yield seems a little high, i'd typically use something between 50% to 2/3rds. torque preload can be off (high or low) by 1/3rd, so 90% could yield the bolt. this is not such a bad thing, but it probably isn't a good thing !

i'd start my bolt design with the loads ... shear or tension, static or fatigue. this should size the bolt. then determine the preload based on the application.

 

[wannadie]

A few comments:

If the bolted joint was loaded in shear with shear service loads I see no reason why the preload would be less significant. The preload is providing the clamping force, this is creating a 'friction joint' resisting shear forces being transmitted to the fastner.

It is alright saying ok I use between 50% to blah, but this is certainly something I can't quote, my job requires me to do calculations and reference my source, I need strong references.

I am required to design many bolted joints and am really after a reference or guide I can work to.

I am working in Europe so 99% of the joints are in metric.

Regards,

Matt.

 

[imagitec]

Any machine design book should cover bolted joint design adequately and will shed some light on the reason for the high preloads.

A bolt clamping two members together acts as an extension spring, and the clamped members act as compression springs. If an external load that tends to separate the members is applied, some of the compression on the plates will be relieved. In turn, some of the tension on the bolt will be relieved. Thus the force on the bolt does not increase by the same amount as the applied load. What actually happens depends on the effective spring stiffnesses of the bolt and the members.

Rob Campbell, PE
Finite Monkeys -

http://www.livejournal.com/users/robcampbell

 

[imagitec]

I understand the importance of preload, but find the 90% extremely high with little safety factor due to the unpredictable nature of coefficient of friction and tightening controls.

I've created a spread sheet and a MathCAD workbook that go through these calculations. It considers tightening tolerances and the resulting range of preloads on the performance of the joint. I'll dig them up, if you're interested. Bossard's catalog of metric fasteners (

http://www.bossard.com)

has a great technical section on this.

Torque is an indirect way of measuring preload. There are alternatives: the turn of the screw method; the stretching of the screw; fasteners that shear at a specified torque.

Rob

Rob Campbell, PE
Finite Monkeys -

http://www.livejournal.com/users/robcampbell

 

[JStephen]

I would think that in a lot of cases, this is covered under various industry-specific loads. Structural bolts, flange bolts are two varieties that come to mind.

 

[Sparweb]

"...I need strong references."

Mil-Handbook-60

Use the ASSIST website:

[

http://assist.daps.dla.mil/quicksearch/fsc_quicksearch.cfm

]


Steven Fahey, CET

 

[CoryPad]

While the entire VDI 2230 method may be more detailed than you need, you can use the one page version that is used for initial fastener sizing.

Also, you can refer to Handbook of Bolts and Bolted Joints edited by Bickford and Nassar for more succinct tools to select preload.

Or NASA RP 1228 Fastener Design Guide.

If you use the turn of the nut technique, you shouldn't have to worry about exceeding the plastic strain capacity of a fastener even if you use 90% of yield stress as a target. As long as you can apply angles on the order of 30 to 60 degrees, even stiff joints shouldn't result in a fastener that has had its plastic strain capacity exceeded.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Ron]

In building construction, we generally use a 70% preload. I would agree that 90% is a bit high.

 

[desertfox]

Hi wannadie

Try this site for information:-

http://www.boltscience.com/pages/info.htm?

regards desertfox

 

[cbrn]

Hi,
if you work in Europe, then VDI 2230 is certainly the most detailed Norm you can refer to, though it is known that for some joints (particularly when the bolt is "short" and Da/d > 3) it can be "dangerous" because it under-estimates the influence of alternate loads (complete explanation is a bit long to put it here...).
Alternatively, I see no reason why not to use the ISO 10011 (now Eurocode 10).

Regards

 

[CoryPad]

cbrn,

Could you share more details on why you refer to ISO 10011? It is a withdrawn series of standards relating to auditing quality systems. Does it have information regarding mechanically fastened joints? Also, can you provide any information regarding Eurocode 10. CEN does not show this to be an active document.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[rb1957]

cory,

could you post a link to NASA RP 1228 ... i've tried finding it but no joy.

thx

 

[CoryPad]

http://ntrs.nasa.gov/index.cgi?method=advanced

In the Title box, type Fastener Design Manual.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[rb1957]

cheers ...

i'd used "fastener design guide" as the title ... and got zero hits !

 

[Philrock]

One rule of thumb is that tension developed by tightening torque is predictable to an accuracy of about +/-30%.

Another rule of thumb is to torque to a calculated 70% of yield.

If you combine these two rules of thumb, the highest tension you should develop is 91% of yield (1.3 x 0.7).

 

[TVP]

rb1957

Fastener Design MANUAL not Guide...

 

[rb1957]

thx TVP ... i'd picked up the difference with cory's later post and got the doc

 

[cbrn]

Hi,
sorry for the mistake regarding "ISO 10011" in my previous post. The error is due to the fact that where I work we refer to VDI 2230 modified by internal research data.

So, it's not ISO 10011 but CNR-UNI 10011, it's an Italian norm which was proposed to be included in an Eurocode, but was never included "as such". The correct, newest Eurocode is 3 - Design of Steel Structures - part 1-8: Design of Joints (prEN 1993-1-8:2003). As you can see, it's been issued as "work-in-progress" and is still not released...

Sorry again, but I think anyway that this norm is worth a look.

Regards

 

[plasgears]

In machinery subjected to high temp and transient temps, preload helps retain the joint and prevent loosening. It's interesting to calculate preload reqd in the presence of a bolt with higher thermal expansion than the base metal.