The long awaited details from my Co-Axial training at Omega, USA.
First things first, if you haven’t taken the time to look at a Omega’s (Geo. Daniel’s) Co-axial escapement you are in for a treat. The principles of servicing the swiss lever escapement do not apply to the co-axial escapement, don’t try and apply them.
To understand the co-axial escapement we should check out a model: For the best watch related animated models on the internet always check out clockwatch.de. For this model we want a model of the George Daniels’ Extra Flat Co-Axial Escapement. The description of the escapement on the linked page is also very good. You might also check out the description at Alliance Horlogere or this one at Omega.
First, some theory. Then, the practical stuff.
If you take a look at the co-axial escapement you’ll notice its very different. First, it has 3 jewels on the pallet fork and one on the balance wheel. In the swiss lever escapement each of the jewels on the pallet fork serve two roles. First, they receive the impulse from the escape wheel, second they lock the gear train until the balance returns to unlock it so it can receive the next impulse. — In the co-axial escapement these roles are separated. There are two jewels whose only function is to receive impulses and two jewels just to perform the locking action. This isn’t the big deal though! This seems to be a side effect of achieving what George Daniels set out to do.
In the Swiss Lever Escapement the impulse is delivered to the balance via the pallet fork by way of a sliding action where the escape wheel slides across the impulse face of the jewel. The escape wheel tooth moves the direction of its force being applied tangential to it’s circumference, but the pallet fork moves away almost at a right angle to this force, making it about the least efficient transmission of force possible.
In the coaxial escapement the impulse is delivered from the coaxial wheel to the pallet fork in a radial fashion. As the impulse is delivered to the impulse jewel, it moves in the same direction, receiving almost all of the energy and with very little sliding action. In the opposite direction the impulse is received directly by the impulse jewel attached to the balance.
Because there is no sliding action in the coaxial escapement there is no need for lubrication. Because there is no lubrication to break down the timekeeping will be more consistent over the service interval – which Omega claims is longer due to the use of synthetic oils, and their absence in the escapement.
Now, the practical information: Please take note, what I am about to tell you is not sufficient information to get you started servicing coaxial movements. This information is intended to help you see that there are in fact some real reasons why hands-on training would be a wise investment before tackling these watches.
First, The “oil-free” coaxial escapement has a little oil on it. Yes, I know, what about what I said above. The impulse jewels are left dry, there is no need for the oil there. The locking jewels have a very, very, very small amount of oil on them to help cushion the locking action. (To gauge the proper quantity of oil, it is applied under 50 power magnification.) This oil acts, not as a lubricant, but as a cushion to absorb part of the shock from the locking action. Leaving this oil off does not affect the timekeeping of the watch, but it will likely result in damage to the upper coaxial wheel teeth over time. . . . But Geo. Daniels didn’t put any oil on his movements? The coaxial wheels in his watches are made from a Gold alloy which is softer (and springier?) than the steel used by Omega. This material choice serves to absorb the extra shock without the need of a “hydraulic cushion.”
Second, didn’t Omega have issues with the first coaxial movements? Yes, they did. The 2500A had some critical problems, which have been resolved. The tolerances are so tight on these watches that the slightest shift in the escapement will result in the watch coming to a stop. The 2500A allowed the pallet fork to shift slightly when it received a shock. Because of this many of the 2500A movements were replaced with “B” movements making the “A” series all that more collectible. The “B” movement has a sturdier pallet bridge (among other changes). In their most recent caliber Omega has implemented a new shock system which limits side to side movement of the balance wheel when it receives shocks. Because one of the impulse jewels is on the balance a shift at the wrong moment will allow the coaxial wheel to pass by the impulse jewel without delivering the impulse and causing the watch to stop. My opinion is that Omega has finally figured out how to make reliable coaxial movements, but they will continue to make improvements for many years to come.
Third, you may be able to time your coaxial watch on a conventional timing machine but the amplitude readings will probably be wrong. Modern timing machines from Witschi have special programs for coaxial escapements. Calibers which operate at 28,800 bph will correctly display their timing, even on an old Vibrograph B200, but amplitude numbers will be incorrect, even if you set the appropriate lift angle (30 or 38 degrees depending on the caliber.) Newer, 25,200 bph calibers are more difficult to time because older timing machines do not have a setting for this frequency. My Witschi Professional will allow me to adjust the rate to any frequency, but it takes about 15 minutes to cycle up to 25,200 from 19,800 or down from 28,800 (the nearest frequencies.) My New Tech Handy II (Quartz Tester) automatically recognizes the frequency, but it doesn’t have an adjustable microphone so it is difficult to check different positions.
Fourth, Drop and Lock may be way different on the entry stone, than they are on the exit stone. This is fine. Drop may be very shallow, this is fine. The tolerances are so tight that rather than adjust stones that don’t lock correctly, or which have been knocked loose, you should replace the fork.
Fifth, IMPORTANT! do not reverse the travel of the balance wheel until all the actions of the escapement have taken place. Unlike with the swiss lever escapement if you reverse the direction of travel of the balance in the middle of the escapement action you WILL DAMAGE the pallet stones.
Sixth, if you have a modern timing machine which accommodates the coaxial escapement the amplitude can be very low on newer 25,200 bph models. Check the Omega technical information for acceptable ranges.
So, where do I find the coaxial escapement? Omega is putting it in a lot of watches these days. The have developed their calibers 2500A, B, and C -all based on the ETA 2892. It is in caliber 2627 and 2628 and their chronograph caliber 3313. They have installed in a Valjoux 7750 in an Olympic edition model introduced at Beijing (Omega caliber 3888). It is in the new double barrel calibers 8500 for men’s 8520 for ladies and 8601 with annual calendar. I believe it was also placed in a tourbillon (but I might be wrong about this one).
My overall impressions. They’ve finally figured out how to mass-produce the coaxial escapement and it offers some minor improvements over the swiss lever escapement, but nothing drastic. As a watchmaker, it is very interesting to work on, and it boggles my mind how somebody could develop this from an idea in their head. Don’t be looking for a J.Peter escapement any time soon.