Classic Escapement - Graham (BHI)

by GrabCAD
This projects is connected with my former projects Classic Escapement - Graham (Ideal) and (Modified), so you need to learn from there the ground information.

BHI stands for „British Horological Institute”. After posting the Classic Escapement projects I found a lot of new friends. One of them is Ken Kuo from Taiwan. He is a prodigious author of a lot of Escapement simulations. Convince for yourself by consulting his productions on Youtube, and an example is this list: By reading the first one data (Circular Pallets Deadbeat Escapement) we understand now terms like Escaping angle, Lock Impulse and so on, isn't it? More than this, by reading the description of the same project, you can learn a lot about the dead-beat escapements, with fine references. For instance, George Graham has lived between 1673 - 1751, and his dead-beat invention was released in 1715.

Ken Huo recommends us a book from the British Horological Institute as a good reference. The title is: Drawing Clock and Watch Escapements, Distance Learning Course, Author David Poole, FBHI 2011- You can read more about this book here:

After a careful study of the BHI book, I decided to make another one version of my „Graham Escapement (Modified)”. This version uses the information from the BHI book. I dared to add a greek motif on the face of plate - for fun, indeed...

When you decide to create a precise mechanical escapement you have to choose Graham dead beat escapement.

The rest of details are similar with my mentioned projects (both Ideal and Modified). However, since all we are doing is just for fun, a notice can be useful. All clips (Video) of my escapement projects present the movement from the starting point, instead of the current stable motion of the mobile parts. The starting moment is very important because we want to see how the mechanism overcome the inertia and come to a stable movement. If you work with Inventor you can change settings of Dynamic Simulation or even dimensions of the parts to see how sensitive is the resulting movement. For instance, a torque of 1 N mm seems to be the proper value when the rod is 250 mm long. Changing one or other value (or both!) you can see how the system stops or the pendulum amplitude become too great. Besides, the starting slope of the pendulum is also very important.

In other words, after coiling, a launching at a moderate slope of the pendulum must produce a stable movement of the mechanism in a short time.

Having no expertise in horology, I am sure that I did not choose proper values for all parameters, as you surely will notice if you study the picture „Output Grapher (20s).jpg” (remember points A and B from my previous project). As you can see, the blue graph (corresponding to palette, that is pendulum) presents a non-uniform maximum velocities. For an expert eye this is not good! Sorry, I have no time to make more searches for repairing these irregularities. By fortune, you have all the data to refine them for yourself - or for telling us about...