• Upcoming updates
    Over the next couple of weeks we will be performing software updates on the forum. These will be completed in small steps as we upgrade individual software addons. You might occasionally see a maintenance message that will last a few minutes at most.

    If we anticipate an update will take more than a few minutes, we'll put up a notice with estimated time.

    Thank you!

Pocket Watch Regulator

The function of the regulator is not well understood by the general public or by novices. The following article explains how it works.

The Regulator

The timekeeping rate of a watch depends upon the rate of oscillation of the balance (wheel). The regulator on a watch is a device used to set that rate of oscillation of the balance. The balance, the wheel that spins back and forth rapidly, which is a watch's timekeeper, owes its oscillations to the action of the hairspring. The hairspring is located within the balance on the balance staff. The regulator, which rotates concentricly about the balance staff, functions by having two pins, extending from a short tab at the regulator's rim or from the regulator arm, that straddle the end of the hairspring. As the regulator is slightly rotated, the two pins slide along the length of the hairspring by a short amount. The pins thus change the effective length of the hairspring, changing the rate of oscillation of the balance. However, the regulator can only change the timekeeping rate of the watch by several minutes per day, at best. If a watch has three minutes error per day, or more, it is an indication that the watch requires service. Watches of lesser quality have a plain regulator, whose tip extends over a scale.

Plain Regulators

Plain regulators have just a simple arm that extends from the ring around the upper balance jewel to a scale marked on either the movement top plate; or on the balance cock, as seen on this example, a Waltham Riverside grade No. 1621. Looking carefully at the ring to which the regulator arm is attached (in the linked-to example), one can see a small tab at about the eight o'clock position. The two tiny white dots on the tab are the tops of the pins that straddle the end of the hairspring (being a Breguet overcoil hairspring, the end passes up and over the other coils of the hairspring). Earlier watches have the regulator located below the balance, as seen on this example (posted by Keith R...), instead of on top, but the principle is that same. Moving the arm with a piece of sharpened pegwood (or for amateurs, a wood toothpick) causes regulator to rotate, moving the two pins.

Patent Regulators

Finer watches require the ability to accurately move the regulator a precise, repeatable amount. The different watch companies used precision regulators whose designs were patented, hence the term, "Patent Regulator." The generic term for these precision regulators is "micrometric regulator" or “micrometer regulator.” Typically, a large change of the regulator adjusting device causes a small change in the positions of the regulator pins, resulting in a small change of balance oscillation rate and causing a small change of the watch's timekeeping rate. One example would be that one full turn of a regulator adjusting screw of a Hamilton 992B watch movement causes a change of rate of about 15 seconds per day (Hamilton Technical Data Sheet T.D. 127, bottom of page 2). Pages 1, 2, and 3 of Hamilton TDS-124 discuss regulation with Table 1 on page 3 showing that one turn of the regulator screw of their 18- or 16-size watches result in an 18-22 seconds per day change in rate. Another example is Page 51 of a 1917 South Bend Material Catalog which shows that a full turn of the regulator screw of South Bend regulator results in a rate change of 10 seconds per day.

Various Styles of Patent Regulators

Starwheel Regulator
Waltham's Star Wheel Regulator Design, also known as a Church regulator (named after D.H. Church, sometimes referred to as "… the most important technical contributor to the success of Waltham"), used on a number of medium and high grade watches in several models, is basically a rack and pinion arrangement. The regulator arm ends in a curved, toothed bar (the rack). The star wheel has an upright, toothed shaft (the pinion) which is held against the rack by a spring. To regulate the watch, a small toothpick, or other pointed wood tool, is used to push against one of the arms of the star wheel, causing it to rotate, which in turn moves the rack, causing the regulator pins to slide along the hairspring, changing the hairspring's effective length. During the 1890s, Deuber-Hampden switched to a Similar Style Regulator. Waltham started using a different style of patent regulator on its medium grade 18-size, model 1883 watches, the 1906 Ohlson Starwheel regulator. In this design, the Starwheel, and its pinion, was attached to the tip of the regulator arm while the rack was part of the regulator scale.

Richard Oliver's Regulator
The rack and pinion design was utilized in Richard Oliver's patent regulator. Instead of a starwheel, Oliver used a slotted, toothed wheel. This allowed the watch to be regulated using a screwdriver.

Earlier Columbus Regulator
The Columbus Watch Co. used another rack and pinion design, the Owen regulator (posted by luvsthetick), on earlier versions of their higher grade watches. What looks like a tiny screw at the tip of the regulator is actually a small pinion that engages the teeth on the inner edge of the regulator scale. Rotating the "screw" causes the regulator to move in a controlled, precise manner. The pinion was easily lost upon disassembly of balance cock and balance and it is missing from a significant number of surviving examples of movements having this style regulator.

Ohlson Regulator
Another Waltham patent regulator design was the Ohlson Regulator, also known as the 1908 regulator (very different from Ohlson's 1906 Starwheel design). An example is shown on a model 1908, grade No. 642 movement. It has the manufacturing advantage of the regulator and its spring being a single piece.

Whipspring Regulator
A common micrometer regulator design consists of a regulator arm nested between an adjusting screw and an anti-backlash spring (usually referred to as a whipspring). Examples may be seen on these: Illinois-Burlington Model 5 and Hamilton 992. This provides positive movement of the regulator arm (and hence the regulator pins) when the adjusting screw is turned in either direction. Tight tolerances between the adjusting screw and the threaded block within which its threaded prevent the screw from moving on its own. Earlier Rockford watches had a Johnson Regulator, a variation of the whipspring regulator.

Traveling Nut Regulator
The regulator Elgin used on higher grade watches during the last quarter of the nineteenth century and the early years of the twentieth century is a Traveling Nut Arrangement, patented by C.S. Moseley and G. Hunter (No. 157,021, Nov. 17, 1874). The spring in the right-hand picture of the example is under the balance cock, pushing the regulator arm to the right rim of the nut. The regulator arm is nested in a groove in a cylindrical nut which travels along a curved, threaded rod. As the nut is turned (by using a toothpick in the grooved rims of the nut), it moves the regulator arm, thus moving the regulator pins.

Prior to a switch to a starwheel regulator in the 1890s, Hampden used two versions of a different variation of traveling nut regulator, rotating the screw to move the nut. One example was the Tucker regulator with its diamond-shaped projections which grasped the tip of the regulator arm. The other was the very similar Teske regulator which had circular-shaped projections. Hampden's use of these is described in the NAWCC Message Board thread entitled "Tucker/Teske Regulators". The Tucker and Teske patent pdf files may be downloaded from links in the thread entitled "Teske Vs Tucker Resolved!!!" These two regulators were also sold by retail jewelers who would retrofit them their customer's watches.

Cam or Nautilus Regulator
The cam style regulator, sometimes referred to as a nautilus regulator, is seen on Swiss watches from a number of manufacturers. A spring presses the regulator arm against a cam. Rotating the cam by means of a screw driver in its center slot causes the regulator arm to move closer or further from cam center. This slides the regulator pins, on the opposite end of the regulator arm, in one direction or the other.

Regulators Used On Aurora Watches
The Aurora Watch Co. used several different patent regulators. One was the Johnson regulator (posted by Omexa). A more commonly seen regulator (on Aurora movements) is the Hurd regulator (posted by Jerry Bryant).

Extreme Regulator Position

When a watch is properly cleaned, oiled and serviced, the regulator indicator ought to be within the center 50% of its total range when keeping accurate time. Actually, it should be in a narrower portion in the center, but 50% allows for some variation of the quality of service without considering the cost of that service. If the regulator has to be positioned outside of the center 50% of its range to obtain accurate time, then it is an indication that either the watch needs cleaning and oiling, or that something is wrong with the watch. If it comes back from being serviced with the regulator outside of the center 50% of its range, it means that the person doing the service has not found, or corrected, one or more problems.

Regulating A Watch

It's not likely that many novices know how to check the regulation of their watches. Nor is it hard to set a watch's regulator to obtain the best timekeeping rate of which the watch is capable. It is just tedious if you don't have a timing machine. In order to perform well, the watch needs to have been cleaned, oiled and serviced so as to be in good running condition. If the watch is run continually, a cleaning and oiling is needed every 3-5 years. If you're only going to wear your watch occasionally, this ought to be done once at the onset and about every ten years thereafter. If you're not going to carry it (or run it), don't bother getting it serviced, but don't bother trying to regulate it, either.

It may occur to one to ask, "If my watch has been serviced recently, why does it need to be regulated?" The answer is that regardless of how well the watch was serviced and regulated by a repair person or watchmaker, the watch will run at different rates in different positions. This was true even for a brand new watch that was factory adjusted to five or six positions. The proportion of time throughout a day, day after day, that a watch resides in the different positions is different for each person. That's why even the highest grade watches have to be regulated to match each persons daily habits. Here's what Hamilton had to say on the subject in literature distributed to its dealers:

Personal Regulation

Various factors affect the performance of any watch. For this reason, regulation of a watch to its owner's routine is very necessary. As an instance, a man occupied at a desk has a routine and habits differing from those of a mechanic; or, a typist and her routine cannot be compared to a housewife. Individually, watches worn by these persons would be subjected to widely different conditions during a day's time. And a difference in watch performance would result. Some persons do not wear their watches regularly every cay, and they allow them to lie in a flat position while not being worn. Others wind their watches at irregular intervals. Even changing a pocket watch from a trouser's pocket to a vest pocket may result in a variation of several seconds in its daily rate. In addition, small watches are greatly affected by changes in climatic or temperature conditions. Accurate timekeeping is considerably influenced by these things.

Thus, varying conditions under which a watch is required to perform, and over which there is no control explains why a new watch, or even a repaired watch may need more regulation than sometimes seems reasonable.

Most important to you, therefore, it is always advisable to explain to watch purchasers or repair clients, with utmost diplomacy, that a watch may gain or lose time during the first few weeks they carry it. This will prepare and encourage customers to return their watches once a week for comparison and regulation, when necessary. Too, it will open an opportunity to make people "watch conscious"; making them realize that accurate performance in a watch is almost wholly dependent on the care it receives from them.

Hamilton Technical Data Sheet T.D. 124 (4-Sep-51)
A South Bend ad said the same thing forty years earlier, pointing out to railroaders that "The Studebaker" grade watches, watches adjusted to five positions, needed to be regulated to the owner's daily routines. Railroaders typically brought their watches to the local watch inspector (who tended to be a jeweler/watchmaker) to have their watches regulated, whereas civilians would go to their favorite watchmaker. In either case, could be the place where they bought the watch, but it need not have been, especially if the watch was purchased via mail order.

Railroaders were prohibited from regulating their watches (see Section 15 of the 1898 Big Four rules ), not because they were incapable of doing so, but because it invalidated the process of ensuring that their watches could hold the specified rate. That doesn't apply to individuals today. Regardless of whether the watch is adjusted to positions, or not, if you carry it daily, you'll want to regulate it to best match the positions in which it'll reside, and the bumps it'll receive, as you go about your daily activities.

Wind the watch fully, once a day. The Best Time to Do So is in the morning when you first begin your day's activities and start carrying the watch. This is a good time to check the rate and record the deviation, always doing so at the same time each day (give or take an hour or so - allowing for the weekends). Compare the watch to a time standard, and note the fast or slow deviation in seconds. A good standard to time your watch against is the US Naval Observatory Master Clock Time website. Only set your watch the first day by setting the minute hand to the nearest minute, then noting the deviation. After five days, or a week, of carrying the watch and noting the deviation from the time standard, determine the total number of seconds the rate has changed. Then divide by the number of days to determine the average change in seconds per day. For example:

Date - - - - Deviation From Time Standard
2-Jan-07: -27 Secs
3-Jan-07: -18 Secs
4-Jan-07: -07 Secs
5-Jan-07: +02 Secs
6-Jan-07: +13 Secs
7-Jan-07: +24 Secs
8-Jan-07: +31 Secs

Total change = 58 seconds
Number of days = 6
Average change per day = 9.7 seconds

Just guessing that a patent (micrometer) regulator changes the rate about 15 seconds/day for one full turn of the screw (or star wheel, or whatever), turn it slightly more than 1/2 of a revolution towards the 'S" (slow) marking on the scale. Swiss watches frequently have an 'R' marking, meaning retard (and 'A' meaning advance). Then, repeat the process, keeping careful note of how much you turned the regulator screw and what the result was. After a month, you'll have learned just how much timekeeping change is caused by how much you turn the screw and you'll be able to set the regulator for the best timekeeping rate that the watch can achieve. The same guess applies to plain regulators, using a move of the regulator arm of one graduation of the rate scale for an estimated change of rate of about 15 seconds/day. Remember, most regulators can be moved by the use of a toothpick (or, if you want to be more professional, by a sharpened piece of pegwood). Use of a hard, metal tool runs the risk of scratching the movement if the tool slips.


This 1917 article explains "The Regulation of Watches."

The NAWCC Message Board thread entitled "Tucker/Teske Regulators" describes Hampden's use of these regulators.

This page has been seen 29,940 times.

  • Like
Reactions: Laser
Know Your NAWCC Forums Rules!

Find member

Recent Activity

Icon Legend

  • Normal page
  • Color code

    • Content has new updates
    • Content has no updates

Share This Page

Readability Information

An automated review of this page attempts to determine its readability in English (US). The page contains roughly 144 words across 7 sentences.
Readability score
Readability score: 37.90
Grade level
For the information in this page to be accessible to the widest audience of readers, both a high readability score and a low grade level are ideal.

Forum statistics

Latest member
Encyclopedia Pages
Total wiki contributions
Last edit
Swiss Fake by Kent