Marine: Hamilton Model 22, Military Chronometer

Discussion in 'Chronometers' started by zambranonyc, Jan 26, 2009.

  1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  1. zambranonyc

    zambranonyc Registered User

    Jan 10, 2009
    30
    0
    0
    freelance language interpreter
    New York City
    Country Flag:
    Region Flag:
    What timekeeping standard was specified, if any, for the Hamilton Model 22 military chronometer? (like the 30-seconds-a-week for RR watches).

    Did these 21-jewel 1940's watches come with the Elinvar Exta hairpsring?

    Thanks. oz.
     
  2. Jim Haney

    Jim Haney Registered User
    NAWCC Member Donor

    Sep 21, 2002
    6,288
    859
    113
    Male
    Working the farm, Garden,horses, goats, chickens,
    Decatur, TN.
    Country Flag:
    Region Flag:
    Oz,
    They were equiped with the Elinvar Extra hairspring. The model 21 had the huge Elinvar Extra Helical hairspring. that is over 1 inch tall.

    The timing specification were what the gov. contract specified and Hamilton could time it down to the second, if specified.
     
  3. zambranonyc

    zambranonyc Registered User

    Jan 10, 2009
    30
    0
    0
    freelance language interpreter
    New York City
    Country Flag:
    Region Flag:
    Thanks, Jim.

    It seems, then, that even with the smaller model 22, Hamilton was able to meet a more strict time-keeping standard than 30 seconds a week. That is quite an achievement.

    oz.
     
  4. Don Dahlberg

    Don Dahlberg Registered User
    NAWCC Member

    Aug 31, 2000
    3,425
    12
    38
    #4 Don Dahlberg, Jan 26, 2009
    Last edited: Jan 26, 2009
    The model 22 was not a chronometer, it was a chronometer watch and as such had less stringent standards than the model 21 chronometer.

    Rates were measured at 55 degrees F, 72 1/2 degrees, 90 degrees and back to 72 1/2 degrees in a relative humidity of 40 percent. Each test was done for five days and the largest deviation of any two days within the five was measured and a mean deviation measured. Recovery was a comparison of the two 72 1/2 degree periods.

    The model 22 was regulated to within +/- 2.00 seconds per day. Temperature compensation was withing 2 seconds for a 17 1/2 degree difference in temperature and 3 seconds for a 35 degree difference in temperature. Recovery was less than 1.50 seconds per day. The largest deviation within any two days of a five day period was 2.0 seconds and and the mean deviation had to be within 0.75 seconds per day.

    There were no positional standards because the watch was to be kept dial up.

    Don
     
  5. zambranonyc

    zambranonyc Registered User

    Jan 10, 2009
    30
    0
    0
    freelance language interpreter
    New York City
    Country Flag:
    Region Flag:
    Thank you, Don, that is very helpful.

    I imagine that the Model 22 being a Hamilton watch, and sharing the Elinvar Extra with the 992Bs and the 950Bs, it may also have proved amenable for an expert to adjust it positionally as they did the RR's. Apparently, those Model 22s were sometimes carried around off deck by servicemen.

    oz.
     
  6. Don Dahlberg

    Don Dahlberg Registered User
    NAWCC Member

    Aug 31, 2000
    3,425
    12
    38
    The model 22 came in two forms. One was in a weighted brass tub that was gimbled in a box like a chronometer. It was used as a chronometer in minor ships or as a deck chronometer in major ships.

    The other form was in a large pocket watch style case, but this was still supposed to be kept dial up at all times. It was kept in a flat sandwitch style wooden box. It was used as a more mobile chronometer. During the war there were 11,954 mounted versions made and 10,235 unmounted versions made. 420 of the latter were used by the Army Air Forces, but I do not know if they were used on airplanes or just as a primary timekeeper at the air stations.

    The following is taken from the prepair manual: One or more mounted model 22 chronometer watches is alloted to minor fighting ships (destroyers through submarines) and one or more model 21 chronometers. The chronometer watches were to serve as an additional check upon the readings of the chronmeters. For safety, these instruments were located in different stations aboard ship, according to the discrestion of the ship's Commanding Officer. Seagoing ships (such as tenders, supply ships and hospital ships) and non-seagoing ships (like patrol craft, yard craft and LST) are provided with chronometer watches, and may be alloted chronometers.

    I suspect that many of the latter were the unmounted types.

    It goes on to say that for the exact navigating timepiece allowance on a particular type of vessel, see the latest "allowance list".

    Hamilton continued to produce model 22 chronometer watches after the war. Many were sold to jewelers as a primary timepiece to set and time watches.

    Don
     
  7. zambranonyc

    zambranonyc Registered User

    Jan 10, 2009
    30
    0
    0
    freelance language interpreter
    New York City
    Country Flag:
    Region Flag:
    Don, this is very interesting information, and I thank you. It suggests that through the "allowance list", the war personnel in charge of those decisions assigned to specific vessels greater or lesser stringency in timekeeping.

    I imagine the criteria for such assignations, as well as the specific numbers of the assignations themselves, might have been classified information back then, and may still remain so today.

    Where in the world could I check that out?

    On a related issue, I gather that the Model 22 had to be an exceptionally accurate timekeeper since jewelers kept it as the master time teller from which to set their other watches. Also, the fact that the 22 was used as backup, and sometimes as counter-check for the larger, more complex and accurate Model 21, indicates that the smaller 22 could not have been that far off the 21.

    Until I read your information above, I thought a good Hamilton 950B and a good Model 22 would be almost the same in timekeeping accuracy. But now I think that with Model 22, Hamilton went into another class of timekeeping beyond the 950B.

    oz.
     
  8. Don Dahlberg

    Don Dahlberg Registered User
    NAWCC Member

    Aug 31, 2000
    3,425
    12
    38
    Comparing the model 22 to the 950B or 992B is a little apple and oranges.

    The railroad grade watches had to keep good time over five or six positions. Otherwise the requirement was 30 seconds per week or nearly five seconds per day. The emphasis was on keeping good time over varied positions and varied temperature. Same is true with the military Master Navigational Watches 3992B, 4992B and the model 23.

    In the model 21 and 22, they did not have to worry about positional error. The timepieces were gimbled or otherwise kept dial up. They could concentrate on keeping good time in one position. They were also larger than the 16 size watches. It is easier to adjust a large timepiece. Of course, the model 21 had other advantages, with the detent escapement, helical hairspring, fusee mainspring, and others. The model 21 was also kept in a much more controlled environment midships. I do not have the standards for the model 21 with me right now, but they were another big step above the model 22.

    The allotment did not seem to be much of a secret. Timely Topics, the Hamilton Employee magazine/newspaper, often had short articles on how the military was using their products. They might show a picture of an aircraft carrier and have a short article on the fact that they carried multiple model 21 chronometers below deck and model 22 chronometer watches in the bridge and other locations. I have not found the allotment list for WWII, but I have found it for battleships for about twenty years ago through a Google search.

    Don
     
  9. Don Dahlberg

    Don Dahlberg Registered User
    NAWCC Member

    Aug 31, 2000
    3,425
    12
    38
    I was just thinking that it would be great to hear from any former military men who used any of these great timepieces in the service. This includes anything from the chronometer to comparing watches. I would love to hear how you used them and for what purpose.

    Don
     
  10. itspcb

    itspcb Registered User
    NAWCC Member

    Aug 6, 2006
    540
    0
    16
    Retired
    UK
    Country Flag:
    Region Flag:
    Just a point or two to add to Don's excellant input on the Model 22.

    The non-gimbal version was gradually removed from service and the movements placed in gimbals when they came in for service. So there may be far less non-gimballed versions about today than Don's numbers indicate.
    This info was obtained from the official manual which is an excellant read and can be obtained from several sources.
    Again a question for that former military man Don would like to meet, why was the non-gimballed version withdrawn?

    Also I read, but can't recall where, that the 22's actual performance in a well set up watch can rival a 21.

    The 22's are lovely things, and they came into official use in the UK too, (So are the 22, 4992 etc!) and I'm pleased to have one. They are very nice to work on and parts are available.

    Peter
     
  11. zambranonyc

    zambranonyc Registered User

    Jan 10, 2009
    30
    0
    0
    freelance language interpreter
    New York City
    Country Flag:
    Region Flag:
    Don, I've noticed these facts:

    1) Hamilton's classic RR PW, the 992B, has 21 jewels, and 6 positional adjustments, and temperature adjustments, and a small seconds hand.

    2) Both the 4993B and the 3992B have 22 jewels, and 6 positional adjustments (and temperature adjustments?), and a sweep seconds hand.

    Questions:

    ** Why the "extra" jewel in the 4992B and the 3992B?

    ** Does the numerical similarity of all these models (the number 992B in each of them), mean that they all share in common a 992B movement, but with minor modifications in each case?

    ** Are the 4992B and 3992B also size 16?

    ** Would the 4992B and 3992B also have met the RR standards?

    oz.
     
  12. Kent

    Kent Registered User
    Gibbs Literary Award NAWCC Fellow NAWCC Silver Member

    Aug 26, 2000
    18,278
    1,466
    113
    Country Flag:
    #12 Kent, Jan 28, 2009
    Last edited: Jan 28, 2009
    For the top end of the sweep second hand.


    Yes. In fact, the last of the 992Bs built used 4992B pillar plates, carrying the 4C serial number prefix.

    Yes.

    No. Both are pendant-set and the hack mechanism actually stops the movement, which if used, degrades its ability to hold accurate time. When the U.S. government needed watches for the Army railway battalions (who ran the Army installation railroads and captured railroads in wartime Europe), they ordered 992Bs.
     
  13. zambranonyc

    zambranonyc Registered User

    Jan 10, 2009
    30
    0
    0
    freelance language interpreter
    New York City
    Country Flag:
    Region Flag:
    Thank you much, Kent.

    I have seen "pendant set", and "hack", and "detente", but I have not yet found pictures to understand what those terms actually mean in the watch movement.

    I know many RR watches had a small lever by 1 o'clock which needed to be pulled out for resetting the time. I also know some Hamilton cases had a cross bar over the winding crown to avoid accidentally changing the time while winding the watch.

    Perhaps military-use Hamiltons had to allow servicemen to synchronize their watches in quick re-settings of time, without having to unscrew the bezel.

    oz.
     
  14. Kent

    Kent Registered User
    Gibbs Literary Award NAWCC Fellow NAWCC Silver Member

    Aug 26, 2000
    18,278
    1,466
    113
    Country Flag:
    Hamilton's "bar-over-crown" cases were designed in the early 1920s for the company's line of railroad standard watches (which were all lever-set to meet railroad time service requirements). They did not "... avoid accidentally changing the time while winding the watch ..." because the crowns on the "bar-over-crown" cases did not pull up, and if they did, the standard watches were not pendant-set, thus the time could not be changed that way. Apparently, the "bar-over-crown" was a marketing gimmick.

    Pendant-set means that you pull out the crown (winding knob) to engage it to the hands to set them. This occasionally allows the crown to work its way up, engaging the hands and placing added load on train, changing the watch's rate. Or, as you mentioned, the could be accidentally changed when winding the watch.

    The hack feature was just as you described, used to synchronize the watch. It actually stopped the balance so that the watch could be set to match time to the second with other watches, chronometers, or other standard. Although this would seem useful in railroad time service, I have never seen anything to indicate that it was used that way.
     
  15. Don Dahlberg

    Don Dahlberg Registered User
    NAWCC Member

    Aug 31, 2000
    3,425
    12
    38
    I was reading the standards for the 4992B and 3992B. Although they are essentually the same as the 992B, they had different time standards.

    For five days dial up the average daily rate was not to exceed 5 seconds. The average variation from the average daily rate should not exceed 2 seconds.

    The difference in the rate for the first 12 hours should not differ from the rate after 24 hours by more than 1 second.

    The difference in the rate at -4 F and 41F should not be more than 16 seconds.
    The difference in the rate at 41F and 95 F should not be more than 10 seconds.

    For three days dial up the average daily rate was not to exceed 5 seconds. The average variation should not be more than 2 seconds. The average daily rate in this test should not differ from the original test by more than 4 seconds.

    Notice no test are done in the other five positions. Again it is assumed that these watches would be used dial up.

    The 4992B and 3992B were mainly used for navigation in planes. They only had to keep good time during the duration of the mission, less than one day.

    Don
     
  16. zambranonyc

    zambranonyc Registered User

    Jan 10, 2009
    30
    0
    0
    freelance language interpreter
    New York City
    Country Flag:
    Region Flag:
    Which one came first, the 4992B or the 3992B?

    What was their main difference?

    oz.
     
  17. Don Dahlberg

    Don Dahlberg Registered User
    NAWCC Member

    Aug 31, 2000
    3,425
    12
    38
    The 3992B has a 12 hour dial like most watches, but the 4992B has a 24 hour dial. This required a differen minute and hour wheel to make the hour hand go around half as fast (once in 24 hours). The British liked the 12 hour version, but the US military wanted the 24 hour version. They were in production at the same time.

    Many 4992B were converted to 12 hour versions after the war. This involved changing the hour and minute wheel and the dial.

    Don
     
  18. zambranonyc

    zambranonyc Registered User

    Jan 10, 2009
    30
    0
    0
    freelance language interpreter
    New York City
    Country Flag:
    Region Flag:
    Thank you, Don.

    From my basic understanding about adjusting a model 22 for one position, and adjusting an RR watch for several, it is not a matter of using a different regulator, or regulating system.

    It is rather a matter of "averaging out" the error in the various positions for the RR watch, and then setting its regulator at the point of best "compromise".

    With the model 22, in contrast, it's about maximally reducing the error in one single position (face up), thereby eliminating more of the "compromise" in that one position, than would ever be possible to achieve in the same face-up position of an "averaged out" RR watch.

    This is the concept I currently have of watch adjustment. Please correct it where necessary. I need to get it right.

    oz.
     
  19. zambranonyc

    zambranonyc Registered User

    Jan 10, 2009
    30
    0
    0
    freelance language interpreter
    New York City
    Country Flag:
    Region Flag:
    Will anyone take a stab at my question above, regarding whether I understand correctly the adjustment of a watch, in one, or multiple positions?

    oz.
     
  20. Jim Haney

    Jim Haney Registered User
    NAWCC Member Donor

    Sep 21, 2002
    6,288
    859
    113
    Male
    Working the farm, Garden,horses, goats, chickens,
    Decatur, TN.
    Country Flag:
    Region Flag:
    Oz,
    Your assumption is wrong. The regulator, while a major tool in keeping the watch adjusted, is not the way the time pieces are adjusted.

    You may find a link when doing a search here to help you, but in short form, the major tool for adjusting the time piece are the screws around the balance wheel. Some models have the meantime screws and others you have to use timing washers under the screws or shave some material off of them or use lighter or heavier ones, etc. Use any of those words for a full explanation on timing.
     
  21. Don Dahlberg

    Don Dahlberg Registered User
    NAWCC Member

    Aug 31, 2000
    3,425
    12
    38
    Adjusting of these watches is mostly a question of labor and time. Yes, in some situations, there is compromise, but mostly it times and labor.

    When adjusting a watch, there is a great deal of trial and error. For example, if you are trying to adjust a watch for isochronism, then you might make a small change in the shape of the hairspring overcoil. Then you have to test it. Then you make another change, and test that. There were only about a dozen people at Hamilton who could shape a hairspring. They were paid by how many hairsprings they shaped. If a hairspring did not meet specification, there were only a couple who then corrected them. It was an art learned over many years.

    There are about 20 reasons for a watch not keeping the same rate dial up and dial down. There are another 20 reasons for a watch not keeping the same rate pendant up relative to the dial positions. Finally, one needs to perfectly poise the balance and have a perfectly shaped hairspring to eliminate the difference in time in the three major pendant positions. There will always be a 6th position (pendant down) that will be off more than others. If you want to reduce this, it will be at the cost of the other three. It is a great deal of labor to check a watch over all the positions. If it fails, someone then has to go over the watch to find the problem.

    Finally, the smaller the allowable error in any respect, the more work it is to reduce the error observed in a watch.

    To get an idea of the labor involved in adjusting look at the 1915 price of a grade 974 (adjusted for temperature), the 978 (adjusted for temperature, isochronism and three positions) and the 972 (adjusted to five positions). They were identical in most other respects. The 974 sold for $10, the 978 sold for $12.50 and the 972 sold for $15.50. This was at a time when labor was cheap. As time went on, labor became more expensive relative to materials.

    Which cost more to make, the railroad grade 992B or the military 992B? The wholesale price of the 992B was about $26.00 in a gold filled case. The 4992B was about $29.75 in a base metal case. In this case, I am not sure the price reflected the cost of production.

    Don
     
  22. zambranonyc

    zambranonyc Registered User

    Jan 10, 2009
    30
    0
    0
    freelance language interpreter
    New York City
    Country Flag:
    Region Flag:
    Holy cow! Adjusting a watch is an enormous endeavor!! Thank you for the thorough information, and for framing the issue so usefully.

    The art of the stone cutters of old who built the great cathedrals, died with them. Is the art of the watch maker--and watch adjuster, I should add--alive and well today?

    oz.
     
  23. zambranonyc

    zambranonyc Registered User

    Jan 10, 2009
    30
    0
    0
    freelance language interpreter
    New York City
    Country Flag:
    Region Flag:
    Thank you, Jim. That, right there, was a major revision of my previous idea of what adjusting the timing was about.

    oz.
     

Share This Page