My first two instalments about forming a hairspring, in summary, covered pinning the spring to collet and vibrating it to find the timing point. The next step in the process is to form the terminal curve, this curve forms the basis for adjusting the timekeeping of the watch via regulation. The terminal curve is a key part of the hairspring and must be near perfect in order for it to function correctly. After hours of practice its something that can become quick and easy but to begin with, as you would expect, its the complete opposite!
Before we can make a start on forming the terminal curve, the hairspring has to be removed from the balance wheel it was fitted to while finding its timing point. Once the hairspring is free of the balance wheel again, it can be placed back onto the protractor like scale ready for the next step. The timing point should be found about half way along the length of the terminal curve and if we formed the terminal curve at this point that outcome wouldn’t be possible. If we go back to my last instalment for a moment, finding the timing point of the hairspring, you will remember that I cut the outer coil of the hairspring off at the timing point in order to find it easily later on. As a result of this, the outer coil doesn’t extend past the timing point and we need it to so some more adjustment to the spring is required at this point.
In order to extend this outer coil past the timing point we need to make a new outer coil by removing some more of the hairspring. We move one coil in from the outer coil, in line with the timing point (current end of the spring) and using the protractor scale we move 45 degrees in an anti-clockwise direction. The spring is cut off at this point to form a new outer coil, the significance of the 45 degrees will become clearer in a moment.
The next step is to form the two bends that create the terminal curve, both of these bends should have a angle of around 45 degrees and must be perfectly vertical in order to keep the terminal curve flat in relation to the spring. Using the protractor scale again, the jaws of a pair of tweezers are placed on the spring, 95 degrees in a clockwise direction from the end of the outer coil, a second pair of tweezers is used to push the spring outwards to create the first bend. This can be tricky and its easy to bend the terminal curve upwards due to the bend not being perfectly vertical. The reasoning behind forming this first bend at 95 degrees is to ensure the timing point lies in the middle of the terminal curve. Allowing 5 degrees for pinning to the stud, the 45 degrees we extended the spring past the timing point earlier is exactly half of the remaining 90 degrees of spring, ensuring the timing point is where we want it.
Now we come to making the crucial second bend, the point at which this bend is made determines the width of the terminal curve and its important to get the width right at this stage because its tricky to adjust it later on. It can be difficult to get the location of this bend right, especially at the beginning, so we were told to make a tool that would make things slightly easier. The tool in question is a flat piece of brass with a steel post that accommodates the collet and circle scribed around the post which shows the width that the terminal should be. Its a simple tool but it has made this part of the process a lot easier over the last month or so.
Forming the bend is a similar process to the first one but this time its done with the aid of the brass tool. The spring is placed onto the brass tool and then held with the jaws of a pair of tweezers at the point at which it crosses the scribed circle, a second pair of tweezers is then used to push the spring towards the centre until the terminal curve looks evenly spaced along its whole length. Its easy to tell at this point if the two bends are in the right places because if they are not the width of the terminal curve is larger at one end than the other. Likewise if either bend is slightly off from being perfectly vertical, the terminal curve will be bent upwards or downwards. Its important to correct any of these problems before moving on to the next stage because they will cause even more problems later on.
Correcting the bends when the terminal curve is bent upwards/downwards is no easy task, the reason for the curve being out of flat is often the first or second band being slightly off of being perfectly vertical. This problem stems from the jaws of the tweezers not being perfectly vertical when the bend was formed, the easiest way to correct this is to pinch the offending bend with a pair of tweezers to bring the spring back into flat. It takes hours of practice to be able to make these corrections well enough to move on to the next stage and a spring which was perfect up to this point can be completely ruined. Too many corrections can cause the spring to become weak at the bend and eventually it just breaks off rendering the spring useless!
For this reason over the first few weeks we had a number of springs prepared with the idea being to take each spring to the next stage, casualties would be discarded along the way and hopefully at the end we would have at least one completed hairspring. Working this way certainly helped, if one spring went wrong you could pick up another and avoid having to start all over again, and the result was the chance to experience each stage of the process to this point in a short space of time.
I will continue with the next stage in Part 4 which will include pinning the hairspring to the stud and adjusting the terminal curve to suit the curb pins. I’m sure you will agree by now that forming a hairspring is a tricky process with many stages at which things can go wrong and its certainly not something to be taken lightly. I having to confess its not something I particularly enjoy doing but its an important part of watchmaking and something I need to learn, for that reason it has my full attention at all times. Anyway thats it for now, keep checking back for the next instalment.