Torque vs. axial force on UNC threads

[datahead]

I am trying to locate a formula for determining the axial (linear) force generated by the torque used on a 1/4 -20 UNC screw.

I can derive that force for Acme and ball lead screws, but the pressure angle is different than the UNC thread. I can also derive the clamping force or tension/preload in a screw that is seated, but this existing application (misapplication) is used for inserting large PWA's into a motherboard, similar to a lead screw.

 

[CoryPad]

Typical equations for threads like UNC show three components: 1) pitch torque; 2) thread friction torque; 3) head friction torque. If you have no head, then you have only 1) and 2). The equation should be similar to this (for metric threads):

M_A = F_M (0.16 · P + 0.58 · d₂ · ?_G)

where

M_A is the assembly torque in N m
F_M is the preload in kN
P is the pitch in mm
d₂ is the pitch diameter in mm
?G is the thread friction coefficient


Regards,

Cory

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

 

[Tunalover]

A simplified version of Corypad/s formula is:
T=kPD where T=applied torque, k=the empirical "nut factor", P=the axial force developed, and D=the nominal diameter (e.g. .164 for No.8-32 UNC).

A very broad rule of thumb for unlibricated, plated steel or CRES fastners is that k=.15.

If you have a routine need to do this kind of calculation you may want to get TORKSENSE for about $120 per seat. This program contains many metal and non-metal materials of many different alloys and grades (provides yield points). You enter the percentage of the .2% offset yield you wish to develope. It also includes numerous lubricants and finishes. It's a good deal for the money. A Google search will turn up the webpage address.


Tunalover

 

[Kenneth]

Wish I had a dollar for every time I saw someone get themselves in trouble with an assumed, or even "published," nut factor (.15 or otherwise). In a typical "fastener" as described by tunalover, most of the torque (approximately half) is consumed in overcoming the "head" friction torque. Those kinds of "nut factor" numbers (which hopefully are experimentally verified) include the (very significant) effect of that friction. As correctly pointed out by Cory (as usual), that isn't the situation here. If you simply must use an equation, use Cory's, or other similar "long" equation, not a simplified "nut fator" value.

See thread31-14960 for additional "nut factor" background...