Pin Joint with High Friction -- Lug Analysis

[matts05]

Hi,

I am analyzing an over-center mechanism in which friction is very significant. I freebodied all of my linkages and have developed loads at the center of each pinned joint as well as a frictional moment using a friction circle method. From these moments, I can find the actual friction load at the contact point. Right now I am performing strength checks on the mechanism and am not sure what to do; specifically I don't know how to handle how friction will affect my lug strength checks.

I haven't found anything in Bruhn or my company's strength manual on this, and asked a couple people but didn't get a good answer. Given the high friction loads, I do not think that I can neglect it and perform a standard lug check with a load going through the center of the hole. I was thinking about looking at both loads--friction and bearing load--separately, generating a stress ratio based on the normal bearing load and bearing allowable load, and interacting that with another stress ratio based on friction. Maybe look at the lug section where the friction load is applied, react the friction load with a section moment and normal load then get a normal stress ratio. I could conservatively add the stress ratios to generate a margin of safety but,

a) I don't know if generating the stress ratio from friction like that is valid, and
b) If I come back with a ridiculously negative MS I would need a better way of interacting the ratios, or an entirely different method of analysis.

Any ideas? I would prefer to find a way to do this by hand rather than a model. If I need to clarify this please let me know.

Thanks,
Matt

 

[desertfox]

Hi matt05

The friction is generated by the external loads placed on the links and pins, I cannot see why they would affect the stress calculations but it would be better if you posted a free body diagram of the situation showing the forces acting.

 

[ESPcomposites]

Can you just analyze under the conservative assumption that the beneficial effects of friction are not considered (i.e. standard lug analysis as in Bruhn)? The problem with friction is that you never really know what you have. It can vary from part to part and over time. Things like oil/lubricants during the installation process, unknown clamp up forces, loss of clamp up due to vibration, environment, etc. can affect friction forces.

Brian

http://www.espcomposites.com

 

[tbuelna]

I've worked with a couple linkage designs that relied on over-center travel against a hard stop to make them self-locking. This is a common approach used with linkage systems on aircraft and spacecraft. However, it was always a problem to show the links, pins and bearings good in a stress analysis since the forces in the system can get extremely high (in theory) right as the linkage goes over center. The problem wasn't so much friction in the rotating joints, instead it was the bearing and shear stress in the pins and bushings. The problem was made worse with short stiff links and stiff structures supporting the linkage.

The only solution was to design the linkage pivots with some type of precision adjustment (usually an eccentric bushing) so that the center-to-center distance of each link could be rigged to produce a controlled amount of force at the pivots when the linkage rotated over-center. Besides the issue of over-stressing the pins and bushings of the pivots, controlling the linkage forces at over-center was important because if it required too much torque to drive the linkage over-center and back, it would cause the actuator drive motor to stall. On some of these EMAs that used an over-center linkage, the tolerance limit the linkage pivots had to be adjusted to was only +/-.001".

 

[RolMec]

Hi Matt,
for a pinned joint rotating under external load I consider the friction in the joint as an integral part of the external load and do the following:
a) for the pin
apply rotational moment MR = 4*my*F*r/PI, derive torsional stress and either overlay it with the shear stress or combine it into an equivalent stress approach
b) for the lug
from this MR additional normal stress components result in the applicable sections (of the plate o.e.) --> coupled forces.
N.B.
For a short pin of l/d < 4 a particular approach to stress calculation is required.
Regards
R.