How time flies: maintaining our 159 year-old Time Ball tradition

Adam Schaefer is an astronomy guide at Sydney Observatory and a PhD candidate at the Sydney Institute for Astronomy, University of Sydney. In this post he discusses our historic Time Ball and the appeal to help restore it to its former glory.

When building observatories astronomers go to great lengths to escape the limiting effects of city lights and humid coastal weather. So atop a small hill overlooking the busiest harbour in one of Australia’s fastest growing cities seems like an odd place to build an observatory. Paradoxically, it was this proximity to the heart of commerce in a young city that allowed Sydney Observatory to fulfil its primary function.

Sydney Observatory was built in 1858 to provide accurate time to the people of Sydney and, more importantly, to the navigators of ships who used the local time to determine their position on the high seas. Astronomers would communicate the time to people in the city and harbour by dropping the time ball at exactly 1pm every day.

Sydney Observatory's Time Ball has been telling the time since 1858.

To celebrate the important part that Sydney Observatory played in the city’s history we keep the tradition alive and continue to drop the time ball every afternoon. This event can be witnessed at 1 p.m. from anywhere around Sydney Harbour where the Observatory is visible.

Keeping the tradition of the time ball alive is something we take very seriously. Yet a century and a half of use has taken its toll and the mechanism is in need of major repairs and conservation. Any donations to this cause are greatly appreciated.

Donations over $2 are tax deductible.



6 responses to “How time flies: maintaining our 159 year-old Time Ball tradition

  • Correction: I said “However, apparently the ball was not dropped on Sundays, which end on 1st October 1880 (8156 days, 1165 Sundays), making the maximum now 56,877 drops.”

    It actually ended on 1st October 1860, though this was formally reported in December 1860 lasting 851 days or 121 Sundays. I also left out Saturdays during the Depression, so the number is 260 days not 130 days

    1) 3rd June 1858 to 1st May 2017: 58,877 days
    2) 3rd June 1858 to 1st Oct 1860 : 58,756 days (less 121 days)
    3) October 1929 to 1st Jan 1931 : 58,496 days (less 260 days)
    The ball has dropped a maximum 58,496 times, made less by maintenance.
    Therefore at 60000 days will occur on 13th June 2021 (excluding maintenance.)

    Apologies for these silly errors.

  • Great story about this great Sydney Harbour icon. Some general comments.

    Interestingly, H.C. Russell said in 1888: “It was required that the time-ball should be visible from the harbour generally, and that the Observatory should be within easy reach of the other Government departments ; practically these conditions decided the question of site, as the only suitable piece of Government land was on Flagstaff Hill. For purely scientific reasons it would doubtless have been better to place it outside the city bounds to avoid the dust, smoke, and vibration; but these considerations were made subservient to the practical purposes for which the Observatory was established.”

    This does seems to infer that the time-ball was the principal reason above everything else, and I suspect it was the main selling point to establish an observatory! This is likely true, as when Government Astronomer James Nangle in 1930 said: “From 1847 until 1855 New South Wales was without an official observatory, though proposals to establish a system of time determination with time signals by the dropping of a time ball were constantly under consideration.

    The truth and reasoning of this appeared in the Sydney newspaper “Empire” on 27th June 1855, when William T. Denison wrote on 31st March 1855 [1]: “In the first place, provision has already been made for tho erection of a building to contain the machinery of a time ball, and for tho purchase of the machinery ; but the time ball will, in point of fact, be worse than useless, unless there are means for determining the time correctly – that is, unless there are proper clocks and proper instruments for determining the time, and these instruments are in the hands of the observer responsible to the Government for their proper application. I say that a time ball would be worse than useless without those, for as the time ball is established for the purpose of enabling captains of vessels to rate their chronometers properly, any error in the time given by the ball has the effect of deceiving the captain as to the quality of his chronometer, and as to the daily rate at which it either loses or gains; and a very trifling error in the rate, accumulating daily, will, in the course of a month, amount to a serious error in time, and a still more serious one in longitude.”

    Also Russell further said “…[The Observatory was finished in] June, 1858 ; it is a handsome stone structure; the eastern front comprises the Astronomer’s residence. On the south or principal face there is a stone tower 58 feet high, upon the top of which is the time-ball, 5 feet in diameter, and having a drop of 10 feet.”

    So from the start date on 3rd June 1858* (the observatory’s ball dropped at noon until 1st September 1858, then at 1 pm.) would mean the time-ball has been dropped to the 1st May 2017 at 1pm a maximum of 58,042 days.

    However, apparently the ball was not dropped on Sundays, which end on 1st October 1880 (8156 days, 1165 Sundays), making the maximum now 56,877 drops. Yet Nangle also said that in 1930: “The time ball at the top of the tower is dropped automatically, through apparatus in the Mean Time Clock, exactly at 1 p.m. every weekday. The wireless time signals are transmitted, also automatically, from the Mean Time Clock, through VIS Radio Station at Pennant Hills, from 12.55 p.m. to 1.00 p.m., and from 8.55 p.m. to 9 p.m. each day.”

    This being on weekdays, I believe was for nearly years between October 1929 – 1st January 1931, and was held back due to the Great Depression and the lack of Government financing to employ people during the weekends.

    So the start date would mean the Observatory’s time-ball has been dropped to the 1st May 2017 for a maximum of 56,877 days. Less 1929-1931, or about 130 days, suggest about 56,747 drops since 3rd June 1858. (excluding repairs.) Note: If you want an Anniversary, 60,000 drops would occur on 25th March 2026, but due to the uncertainty, the Equinox would be equally fine.

    Q) How was the actual knowledge about the timeball’s existence passed onto foreign or interstate ship’s so they could synchronise their chronometers?

    Comment: Another interesting fact is according to Harley Wood in 1958: “In 1871 Russell arranged for a time ball to be dropped at Newcastle as Smalley had recommended. This was afterwards placed on the Custom House building and dropped for many years at 1 p.m.”

    Do you know if this other timeball still exists? It also leaves the question, how did they know when 1pm occurred in Newcastle?

    As a final comment, there was some interesting account about this by Govt. Astronomer William Scott, where on 24th July 1860 Sydney Morning Herald, where he apologised that the ball dropped 7.7 seconds too late on the 23rd! Lucky for him, the day had “Strong gales and rain” all day, so no one probably would have missed it!

    *Smalley said the start date was the 3rd June 1858 (Thursday), though the MAAS website says 5th June (Saturday) [2]. Difference is trivial. The first reported ship to use this was the 37-ton passenger schooner “Christopher George” that set sail for Wanganui, New Zealand on the morning of the 6th June.

    • Hi Andrew, Thank you, as always, for your thorough historical commentary. I shall mark the date of 13 June 2021 in my long-term calendar!
      But where do you get the date of June 3 for the first time ball drop? Scott’s first annual report dated 2 December 1858 describes the buildings on site, including “The Time Ball Tower, which was so far completed as to be made use of on 5th June last.” Based on this we celebrate the Observatory’s “birthday” on June 5th – 2018 will be our 160th birthday! And the date of Sept 1 when Scott switched to a 1pm drop?
      The Newcastle time ball apparently still exists, although I haven’t seen it myself. There are plenty of links to it including Fort Scratchley and Newcastle Customs House. The timing of the ball was controlled from Sydney Observatory via the telegraph system as Russell indicates in this letter dated 1901. Given that Russell says the ball was dropped at “1pm Sydney zone time” ships captains would have corrected their calculations for the difference between their actual longitude at Newcastle and the longitude of the time zone, which is 150-degrees east of Greenwich.

      • Andrew Jacob stated: “But where do you get the date of June 3 for the first time ball drop?”

        This appears in Kerr’s ‘Sydney Observatory’ pg.23 (2002), where it says:
        “On and after the 3rd of June, the Time Ball will be dropped daily (Sundays excepted) at mean Noon. The Ball will be elevated 5 minutes before Noon, and its dropping will be the signal for firing the gun at Port Macquarie.” It is referenced to ‘GG, 1.6.1858’

        I also have this in my notes, but I think they did trials to work out how to organise the time ball drops. Apparently, there were important discussions between the Signal Master Joseph Moffit and Smalley. (Two Signal Master existed: One in Sydney (Fort Phillip / Observatory Hill) and one at South Head. John James Jones first occupied this position in 1843, Joseph Moffit in 1854-c.1862, and by 1869, these were George J. Moffit (son of Joseph) and South Head: J. Graham. )

        As Wendy Thorpe said in ‘Historical Context : Observatory Hill” (Jan 1997) [A]: “In January 1858 an electric telegraph was connected between South Head, the city and Liverpool. The station master then became responsible for morse reception and transmission as well as the manual signals which continued in use.” (and doing flagstaff & semaphore.)

        The telegraph to South Head is important for shipping, and I’d assume that knowing the time would be important to relay information and recording regarding incoming and outgoing ships. (It was also important for military defence of Sydney and the Colony by invasion. Although South Head couldn’t see the time ball, the time could instead be relayed by telegraph.) [Again Kerr’s ‘Sydney Observatory’ partly discusses the signalling processes this pg.10-12]

        However, other than the letter you quote, the 5th June was also noted by Harley Wood (” Sydney Observatory :1858 to 1956″) “The time ball was at first at mean noon (beginning on 1858, June 5) but by the time of Scott’s in December, 1858, it was being dropped at 1 p.m.”

        Oddly, Wood also says; “The first recorded meridian observation is of a transit of the Sun on 1858, July 5.” You would think you would need to have this in action to know when to drop the time ball?

        Scott had also moved into SO on 11th April 1858, no doubt to get the Observatory up and functioning.

        After some further thought, the complexity of determining the number of drops gets more interesting.

        Sydney Observatory’s EO, Geoffrey Wyatt was reported in the SMH 31st August 2013 to say: “The bright yellow time ball, weighing about 120 kilos, has been dropped without mechanical failure an estimated 55,845 times at 1pm since 1858.” [1]

        Accordingly, between these dates (03/06/1858*-31/07/2013) are the maximum of 56,702 days, and subtracting (121+260) 381 days, finds 56,321 days. The difference between Geoffrey and me here is roughly 476 days.

        This is further complicated, where the Maudslay, Sons and Field timeball was damaged in 2003 and had maintenance to produce a new part, and this caused “a more than two-year break.” So if we know the breakage period, then the timeball drops must be reduced by this duration as well.

        Again assuming that he didn’t count these non-drops over the years, finds that 476+381= 857 days or 2 years 4 months 6 days – which is about the same duration as the Geoff states for the repairs.

        Repairs were also made by Russell (Letter to Todd 21 March 1899) in the 1870s, where they replaced the original ball, replaced a wooden shaft, and modified the trigger mechanism. On other occasions, the teeth were stripped by incorrect releasing of the ball. (See Kinns, R., “Time-keeping in the antipodes: a critical comparison of the Sydney and Littelton time balls”, Journal of Astronomical History and Heritage1440-2807), Vol. 12, No. 2, p. 97 (2009). [2]

        These modifications would have been only a few days at best. When the time ball mechanism went electric, there would have been another few days there it would have missed. Combining these together makes about 7-10 days. Even adding a different starting day makes (as I’ve said) that “difference is trivial.”

        • I should have also said that the haste of the instigation of the time ball was part of a important maritime safety upgrade of Sydney Harbour by the Government and authorities. This included harbour markers [1], especially the Hornby Lighthouse (red and white), stated that ; “Hornby Light (also known as South Head Lower Light), at the entrance to Sydney Harbour, was built with the lightkeeper’s cottages in 1858 immediately after the tragic wreck of two ships, the Dunbar and Catherine Adamson, both in 1857, on South Head with the loss of many lives. It was constructed near the outer gun battery on South Head which had been commenced in 1854 but remained uncompleted.”

          The battery was in fear of possible invasion by sea, which was still prevalent in 1858.

          I also found this interesting quote which describes the origin of a harbour time ball here Ridgeway’s ”A Story of Time’ [2] : “In 1824 Captain Robert Wauchope RN proposed that local time could be conveyed to all the ships leaving port by an exact time-ball in a prominent place. He suggested a sphere, which slides up and down a vertical mast and which can be abruptly dropped at an appointed hour – it is raised halfway up the mast at 12.55pm, to the top at 12.58pm and drops at 1pm precisely. The first time-ball was erected at Portsmouth in 1829. After Portsmouth another one was installed in 1833 at the Greenwich Observatory by Astronomer Royal John Pond which has dropped at 1pm every day since then. Around 150 public time-balls are believed to have been installed around the world after the success at Portsmouth and Greenwich.”

          Finally, this recent source told me something I didn’t know. In NSW Transport site under a pdf entitled ‘Running on Time: Clocks and Timekeeping in the NSW Railways’ pg.6 (Feb. 2016) [3] it says about the railways;

          “The time ball was connected to the railways via a telegraph wire. At first guards set their watches by the station clock at the start of each journey, with no apparent method of synchronising the clocks once the train left the station. With only a single line between two stations, this was adequate for safe working of the system. ”

          pg.8. also. says: “As the network spread, so too did railway time. It was often the case that a town had more than one source of public time, with clocks at the railway station, a court house, post office or town hall, each showing a variation in the local time, causing confusion and irritation.
          The railways, however, enjoyed the advantage that all stations were connected back to the Sydney terminal via the telegraph, so that, as the time ball at Observatory Hill sent a signal to the Sydney station at 1pm, that signal was then relayed to all the stations connected to Sydney.
          Keeping Sydney time was then written into the Rules and Regulations of the NSW Government Railways. Rule 44 1 (a) stated that “Sydney time must be observed by all Stations, and clocks and watches must be regulated accordingly””

          NOTE: Funny too, this railway article talks about the Windsor Church (St. Matthew’s) Clock in Windsor NSW put up in 1821, that was donated by King George IV. This was formally open by Rev. G.E.C. Stiles in 1822. (the article wrongly says the opening was by the son Rev. Henry Tarlton Stills (1808-1867), but he was only 13 years old then, but he did marry John Tebbutt to his wife Jane Pendergast in 1857.) James Dunlop at Paramatta Observatory made the time set up for this clock. Someone would arrive (probably the Tebbutt’s primary school teacher Edward Quaife) with a handheld watch from Windsor, get the time from the PO clocks, then go back to Windsor – probably done by Tebbutt’s theology teacher. Once this observatory had closed, John Tebbutt then provided the astronomical observations for this very same clock!

        • Ah ha! Kerr’s book was right under my nose! And the wonders of the internet and Trove show us the Government Gazette for June 1, 1858. And this entry may explain the apparent discrepancy regarding the first transit obs of the Sun being in July. I must search more of these Govt. Gazettes!

          I think estimating the number of time ball drops is not going to be easy – for example, it was out of action for about 4 months a couple of years ago when a gear broke!

Leave a Reply

Your email address will not be published. Required fields are marked *