Talk:Bolted joint

From Wikipedia, the free encyclopedia

This article is part of WikiProject Metalworking, an attempt to cleanup, categorize and expand Wikipedia articles related to metalworking. If you would like to participate, you can edit the article attached to this page or visit the project page, where a list of open tasks is maintained.

,

This article is part of WikiProject Woodworking, a project to improve Wikipedia's articles related to Woodworking. For guidelines see the project page and Wikipedia:Contributing FAQ.

[edit] Dispute given spring analogy

I believe the illustrated explanation is oversimplified. It is true that a bolted joint is surprisingly counterintuitive, but I think this is just wrong. As I learned it, the tensioned bolt and the compressed material act as parallel springs. Given that, the third picture is incorrect; applying a compressive force to the joint will necessarily compress it. The joint will, however, resist the compression based on the sum of the effective spring constant of the bolt and the plates. Once the bolt has no tension in it, the joint would continue to compress, then resisting with an effective spring constant just from the plates.

The more interesting case is when the joint is in tension. As I understand it, the same idea holds (since we are assuming linear elasticity. That is, the joint will have an effective spring constant of the sum of the spring constant of the bolt and of the plates. The unintuitive "magic" of a bolted joint, then, is that if the joint is designed such that the spring constant of the plates is much greater than that of the bolt, any tensile load is carried by the joint more than by the bolt by the ratio of their spring constants. Obviously all of the tensile load in the joint is carried by the bolt, but compared to the joint at rest, the bolt gets very little additional load as the joint is loaded in tension.

Two things come out of this: First, because the bolt holds the joint together, giving you friction against shear, you get tensile strength "for free". Second, I think (although I need to check my Stress Analysis notes) that the nature of the loading on the bolt means that a correctly-designed joint will have particularly good fatigue behavior, because the average stress on the bolt is high, but the amplitude of variation is low. —BenFrantzDale 16:50, September 12, 2005 (UTC)

I mostly agree with the above comments, the illustration is over simplified (see below). But, also, there are two concepts being mixed in the comments above;

Concept 1) As long as the preload isn't exceeded by the external load, the bolt doesn't see any of the external load. This is a typical concept used in bolted joint design. It's a simplifying assumption based on k_member being much, much greater than k_bolt. The illustration is intended to show this concept.

Concept 2) The external load seen by the bolt is a function of k_m and k_b. This concept is rarely used because it k_m is usually unknown. The illustration in not intended to show this concept.

So, I think it's wrong to apply concept 2 criteria to an illustration of concept one.

With regards to the illustration, I agree it is over simplified (there should be two compressed springs representing the grey members, with Q trying to separate them, instead of a single spring. This would show them in parallel with the bolt). Or, I could re-work the illustration along the lines of the spring scale / block analogy in Shigley's Mechanical Engineering Design, or at http://www.boltscience.com/pages/basics3.htm. Let me know your preference.

Also, the section could be expanded to include; concept 2), greater pre-loads reducing fatigue loading and compressed gaskets where neither concept holds. --Duk 18:22, 15 September 2005 (UTC)

With regard to Concept 1, the way it was explained to me is this: the prload greatly exceeds the external load, but the bolt relaxes somewhat (probably due to creep) slightly, so at the instant you tighten the bolt, that's the highest load the bolt will ever see because, as a rule of thumb, that relaxation is roughly the same as the maximum external load. Thus, even though the external load is small compared with the preload, the bolt doesn't stop acting elastically. The counterintuitive thing is that once it's preloaded, the plates act as though they are holding the joint together with a spring constant higher than that of the bolt. The class notes I have have a diagram that I don't remember from Shigley, but are in the page you linked. —BenFrantzDale 20:27, 15 September 2005 (UTC)

Image has been updated--Duk 19:50, 18 September 2005 (UTC)

Looks good. The page could do with a joint diagram when you or someone else has a chance [1]. At least for me, what makes the zeroth-order model confusing is its approximation. For me, the first-order linear model shown by a joint diagram isn't too counterintuitive. —BenFrantzDale 03:22, 27 September 2005 (UTC)

[edit] Circular (self) Reference

"Bolt Banging" re-directs back to this same page. So what is it?