With the recent success of restarting the bells, we began our long period of testing of the bells. As detailed in earlier work logs, we tested the bells for much shorter periods of time – running it for 3-4 hour stretches. The hammers were damped using rubber pads so that the general populace remained undisturbed.
We began testing the bells for a longer period and marked the rebirth of the tower’s voice. After nearly 4 days of operation, we found the beam supporting the pulleys, the cables and the weights was dislodged and damaged.
The original position of the beam tail end. It is securely held in the clamp.
Support Beam pulled out of the clamp and bent out of shape.
Preliminary analysis:
The beam was dislocated from its position due to the combined components of the bell and the clock weights acting along the axis of the beam. These forces should have pulled the beam out of its supporting bracket and the shear forces acting on the beam due to the weights must have bent it out of shape as the beam, though disengaged from one clamp was not completely dislodged.
That’s a pretty bad design flaw, but it was the friction forces that failed us are they are notoriously hard to model.
Clamp from which the beam was pulled out of. The bracket seems to be intact with no failure occurring at the bolt holes
The support beam bent out of shape. Lower-right shows the still-intact clamp supporting the beam. (Hooray for the Factor of Safety!)
Actions taken:
The weights were lowered to the ground and set down safely. The bent beam was taken to the workshop and was restored to its original shape. It is being reinforced with L-beams and will be returned to its position supporting only clock weights for now.
We have also sourced an additional support beam that we plan to use to support the bell weights and shall install this one as well so as to keep the bells running. A locking system has to be set-up in place to prevent any future axial dislodgement of either of the beams and we are currently in the process of designing it.
Additional Information:
To read further about the cable and pulley set up, refer to our earlier work log on how we prevented the interference of the bell and clock driving cables.
Comments and Discussions:
- The failure has occurred at a point where there was no evidence of a weakness or cracks or fatigue signs. The inspection that the team conducted had given the beam a green flag.
There’s no cyclic loading, the compression and expansion happened only once, so the beam should be good to go unless it was in a very bad state, to begin with.
- So far, we have only been doing small tests of ringing the gong and we found no signs of failure apart from the tension capacity of the cables which we promptly replaced by bringing the weights down
- Since this was the first time we did a long-term test(ringing lasting over a few days with repeated ringing at every hour – something which has not happened for more than a decade now), the beam failed by slipping in its clamps due to a reliance on friction in the original design, after years of static loading.
- To check if the weights swung out and/or caused damage during descent, we could begin with where the weights were wound up to, then calculate the position by using the rate of descent.
Well, if we’re checking for marks to see if there was swinging, we could be looking at marks made by the weights when this thing broke. So, even a rough estimate would give us a cylindrical area where we should be looking for marks. The whole concept of the weights swinging seem like a long shot, in the first place.
Alternate Solution:
- If we stick to the bracket support as it was initially, we would need to increase the thickness of the brackets. Instead, we could use a better material and use more support points. This would deviate considerably from the initial mounting points in order to avoid failure due to proximity with older mounting points which failed.
The team is hard at work with Bheem, and should have the clock up and running by tomorrow! Tackling the bells and the installation of the new beam shall be done once the Mid-Semester Exams are done! Keep checking our twitter handle and the website for more quick updates as we progress.
NOTE TO SELF: FRICTION IS AN UNRELIABLE ALLY!