Question Re: Quartz Frequency, and the 1-Second Standard
Like Tree10Likes

Thread: Question Re: Quartz Frequency, and the 1-Second Standard

Page 1 of 2 1 2 LastLast
Results 1 to 10 of 16
  1. #1
    Member Everett464's Avatar
    Join Date
    Nov 2015
    Location
    Eugene, Oregon
    Posts
    882

    Question Re: Quartz Frequency, and the 1-Second Standard

    Hi All,

    Me and a buddy have a little podcast re affordable watches, and last week we talked about quartz movements. We're total quartz newbs, and the information was largely gleaned from what we could learn on the internet, including, heavily, the very good sticky (here) called Thermocompensation: Methods and Movements.

    Anywho, we got a question after posting our episode this week, and I don't know the answer. I thought some of the brains here may be able to help me. The question is basically this:

    Why is 1-second the standard output for quartz movements?

    For context, and to perhaps avoid some surface explanation, I know that standard quarts oscillates at 32,768 hz/s. That number is important because it is a power of two, and, thus, allows, through a series of 14 electronic "flip-flops," conversion of the output signal to 1 beat (or really two - tick and tock) per second.

    Ok.

    So why not, instead of flip-flopping 14 times (converting 2^15 to 2^1), flip flop some other amount of times to make a "sweep." theoretically, couldn't one flip flop 11 times (converting 2^15 to 2^4) resulting in an output of 8 "ticks" per second? That would practically replicate the the standard 28,800 standard of mechanical movements. I think that most mechanical lovers are generally of the consensus that more beats is better, so why not here? I assume that there MUST be a pragmatic reason for the decision - I just don't know what it is.

    I know that Bulova does something like this with the precisionist movements, but I am made to understand that they do this by an exceedingly fast initial oscillation. Why does the initial oscillation have to be so fast?
    Last edited by Everett464; August 2nd, 2019 at 23:11.
    Co-Host of 40 and 20, a weekly podcast and discussion of affordable watches, food, drinks, life, and other things we like.
    Available anywhere you get your podcasts
    Instagram: @40and20

  2. #2
    Member
    Join Date
    Jun 2015
    Location
    southern New Mexico
    Posts
    8,754

    Re: Question Re: Quartz Frequency, and the 1-Second Standard

    In mechanical watches, this type of action is called deadbeat seconds. It is, in fact, a fairly high-end complication, altho never a common one. (Now, of course, there's strong negative connotations by association with a quartz movement.) The point was that the lower movement rate is easier to set and read more accurately, quickly. Also, moving the hands requires power. It's not linear, in that you don't need as much power to move a second hand a tiny amount versus a larger amount, but it does decrease the inherent power reserve of your energy cell.

    32K came about, I believe, because quartz crystals can be cut in many ways. The cut used is the XY cut. It's got fairly good properties, and it's easy to cut to get good accuracy. The XY cut oscillates in a low(ish) frequency range including 32K, so...that's what's used. Going higher frequency like 256K was possible with the XY cut, I believe, but that would be more complex. The first Seiko, and some early Swiss quartz clocks, ran at 8K, but they weren't that good...about 5 seconds a day. My gut says, the engineers stepped up slowly to try to get the performance they wanted. You don't really want to jump, when things are so new.

    That said, there were some MHz-range movements, but power consumption was an issue.
    The truth is rarely pure and never simple.

    Oscar Wilde, The Importance of Being Earnest

  3. #3
    Member Everett464's Avatar
    Join Date
    Nov 2015
    Location
    Eugene, Oregon
    Posts
    882

    Re: Question Re: Quartz Frequency, and the 1-Second Standard

    Quote Originally Posted by gangrel View Post
    . . . moving the hands requires power. It's not linear, in that you don't need as much power to move a second hand a tiny amount versus a larger amount, but it does decrease the inherent power reserve of your energy cell.

    32K came about, I believe, because quartz crystals can be cut in many ways. The cut used is the XY cut. It's got fairly good properties, and it's easy to cut to get good accuracy. The XY cut oscillates in a low(ish) frequency range including 32K, so...that's what's used. Going higher frequency like 256K was possible with the XY cut, I believe, but that would be more complex.
    So, do I understand you to say that the major issue is power consumption? Because that would make sense, but it seems counterintuitive. Especially given the dead-beats example.

    Regarding the XY cut, am I correct in my belief that the XY cut is simply a common tuning fork orientation? That versus something like an AT cut, which is not a tuning fork at all, but solid. Is there something about the XY cut, in and of itself, that affects the analysis? Why would we need to be faster at the outset, when (again, in theory) we could just step the beat down fewer times? Start with 32K and end with 16 (nay, 8). Sorry for being dense!

    Although not really on point here, the AT cut, which was used historically (70s and 80s) in some insanely high-end stuff, but was recently brought back by Citizen in its 0100 movement, seems pretty dang cool.

    Oh! and deadbeat seconds. What a trip! I learned about them recently. Neat stuff.
    Last edited by Everett464; August 2nd, 2019 at 23:16.
    Co-Host of 40 and 20, a weekly podcast and discussion of affordable watches, food, drinks, life, and other things we like.
    Available anywhere you get your podcasts
    Instagram: @40and20

  4. Remove Advertisements
    WatchUSeek.com
    Advertisements
     

  5. #4
    Member
    Join Date
    Jun 2018
    Posts
    594

    Re: Question Re: Quartz Frequency, and the 1-Second Standard

    Quote Originally Posted by Everett464 View Post
    So, do I understand you to say that the major issue is power consumption? Because that would make sense, but it seems counterintuitive. Especially given the dead-beats example.

    I found this quote in a web search, which is as good as any answer you're going to get outside of tracking someone down who designs oscillators for a living.

    "Practically, in majority of the applications, particularly digital, the current consumption has to be as low as possible to preserve battery life. So, this frequency is selected as a best compromise between low frequency and convenient manufacture with market availability and real estate in term of physical dimensions while designing board, where low frequency generally means the quartz is physically bigger."

    With digital circuits the power consumption (at a fixed voltage) is generally proportional to the operating frequency so if we used 64k instead of 32k the power consumption would double, and it might not provide any real timekeeping benefit. If you wanted to go to 16k instead of 32k the quartz crystal would be longer, and maybe more expensive??? but would use half the power. I'm not sure how important the base frequency is relative to absolute accuracy, I think the drift over temperature is always proportional, so having more bits of resolution wouldn't necessarily be of any direct benefit there. But it might make compensation more accurate by being able to adjust in finer steps, dunno. So, from the quote it appears that 32k happens to be the sweet spot in the tradeoff between cost, size and power consumption.

  6. #5
    Member
    Join Date
    Jun 2018
    Posts
    594

    Re: Question Re: Quartz Frequency, and the 1-Second Standard

    Oh yeah, with respect to your other question about why use 1-second intervals instead of dividing by a smaller number and have multiple pulses per second, I suspect that has to do with using more power to actuate the second hand multiple times per second. I happen to have one of the Bulova accutron II models with the sweeping second hand, but haven't owned it long enough to comment on battery life other than saying that the battery inside is quite large compared to other analog quartz watches.
    Everett464 likes this.

  7. #6
    Member
    Join Date
    Jun 2015
    Location
    southern New Mexico
    Posts
    8,754

    Re: Question Re: Quartz Frequency, and the 1-Second Standard

    https://www.electronics-notes.com/ar...t-bt-sc-ct.php

    We do know of one 256K implementation...Bulova. The accuracy is better; I believe it's around 5 seconds/month. The battery life is OK...I believe they say 3 years...but they're using a CR2016 battery. Capacity's 90 mAh. By comparison, the ETA F06.412...the large (30 mm) TC, PreciDrive movement quotes 68 months for a 40 mAh formulation, 94 months for the 55 mAh. So almost double the life on half the power. If you know the Precisionist, it ticks at 16 Hz, IIRC, so ok, you have the combination of higher power for the higher-frequency crystal, and more power for driving the motor so much...but you are talking ~ 4x the drain, assuming that both batteries' voltages would drop too low at the same percentage of total capacity. That might not be true...if the voltage drops based on absolute capacity, then the Bulova's drawing even more, relative to the ETA.

    EDIT: note the size difference. CR battery numbers are size numbers...20 means 20 mm diameter, 16 means 1.6 mm height. The 371 is 9.5 mm by 2 mm. 20 mm is similar in size to most movements, which means it probably can't be integrated. That means the movement can't be all that thin...I can't find how thick the Precisionist movement is, but it's not matching the F07's 2.5 mm.
    Last edited by gangrel; August 3rd, 2019 at 00:45.
    The truth is rarely pure and never simple.

    Oscar Wilde, The Importance of Being Earnest

  8. #7
    Member BarracksSi's Avatar
    Join Date
    Feb 2013
    Location
    Washington, DC
    Posts
    16,240

    Re: Question Re: Quartz Frequency, and the 1-Second Standard

    Quote Originally Posted by dwalby View Post
    Oh yeah, with respect to your other question about why use 1-second intervals instead of dividing by a smaller number and have multiple pulses per second, I suspect that has to do with using more power to actuate the second hand multiple times per second. I happen to have one of the Bulova accutron II models with the sweeping second hand, but haven't owned it long enough to comment on battery life other than saying that the battery inside is quite large compared to other analog quartz watches.
    That’s the gist of it. The stepper motor driving the hands is the most power-intensive part of the watch’s operation. More motor pulses means faster battery drain. The Bulova Precisionist’s 16-steps-per-second movement uses a battery big enough to drive a digital watch for ten years, but it lasts for about three years.

  9. #8
    Member Everett464's Avatar
    Join Date
    Nov 2015
    Location
    Eugene, Oregon
    Posts
    882

    Re: Question Re: Quartz Frequency, and the 1-Second Standard

    Quote Originally Posted by BarracksSi View Post
    Quote Originally Posted by dwalby
    Oh yeah, with respect to your other question about why use 1-second intervals instead of dividing by a smaller number and have multiple pulses per second, I suspect that has to do with using more power to actuate the second hand multiple times per second. I happen to have one of the Bulova accutron II models with the sweeping second hand, but haven't owned it long enough to comment on battery life other than saying that the battery inside is quite large compared to other analog quartz watches.


    Posted by BarracksSi
    That’s the gist of it. The stepper motor driving the hands is the most power-intensive part of the watch’s operation. More motor pulses means faster battery drain. The Bulova Precisionist’s 16-steps-per-second movement uses a battery big enough to drive a digital watch for ten years, but it lasts for about three years.
    This is the single nerdiest thing I have ever engaged in in the watch world. I thank you all for humoring me.

    I think there may be two separate things happening here (they are separate things, no(?); see below), and I appreciate that I am likely getting into territory that is way above my (or most everyone's) pay-grade. With that said, I'm hoping to parse this out a bit more.

    Thing One, is the power to drive the base crystal oscillations. Bulova's Precisionist uses a lot of power to thwack its crystal really fast. the resonant frequency of the crystal is way way faster, and that takes more power and more frequent thwacking to keep it going. I think I get that, at least at a surface level.

    Thing Two, is the power required to turn the second hand more frequently, although proportionally less distance each turn. Not being a physicist, I am in no place to know which direction the efficiencies are.

    It sounds like there is a suggestion that it takes more power to turn the second hand more frequently, but I'm not sure we are all talking about the same thing. Each time I think someone is going to commit, they transition to talk about more frequent crystal oscillations, which is not an apple(!).

    My question/hypothetical/challenge is to know whether, using an XY crystal, at the standard 32.768 kHz, but stepped down fewer times, such that the stepper motor turns the second hand 8 times per second (but only 1/8 as far), is practical. Again, I figure it MUST not be, but I'm not sure. To the extent that we've already figured it out (it takes more energy to go 1/8 as far, 8x as fast), but I'm just too dense to see it, I apologize. lol
    Last edited by Everett464; August 3rd, 2019 at 02:34.
    Co-Host of 40 and 20, a weekly podcast and discussion of affordable watches, food, drinks, life, and other things we like.
    Available anywhere you get your podcasts
    Instagram: @40and20

  10. #9
    Member
    Join Date
    Jun 2015
    Location
    southern New Mexico
    Posts
    8,754

    Re: Question Re: Quartz Frequency, and the 1-Second Standard

    Go back to Barracks' comment. Turning the motor is simply expensive. No engineer, but there's a minimum energy that has to be used to overcome inertia, no matter how far the hand is gonna move. So, yeah, the energy to move it 3/4 of a degree (8 ticks per second, 1 second is 6 degrees of movement) is less than the energy to move it 6 degrees, but not 1/8 of the cost. If we assert the Bulova draws at about 4x the rate of the ETA? What Barracks is saying is, figure driving the movement might be 75% of that. So, that fixed cost of getting the second hand to move, is just too high.

    How much reduction in battery life, or in the case of solar, power reserve, would you accept to gain smooth sweep? (Oh, BTW, the second hand moves at HALF the movement rate. 28800 movements tick at 4 Hz, not 8.) A major draw of quartz is the casual way you can treat it...it runs accurately without any attention. Shorten the battery life, and you reduce that. Force more battery changes, increase the risk of a screw-up, of not re-sealing it properly or doing some other Bad Thing.

    One can't ignore the possibility that inertia's a big factor. Most people just don't care about sweep, judging by sales. Like they're not overly concerned about 20 SPY or better accuracy. KISS is just fine, and KISS says, keep things as they are. It's a totally mature technology, the processes are highly developed. It's these factors that make us so surprised by the Caliber 0100.
    Everett464 likes this.
    The truth is rarely pure and never simple.

    Oscar Wilde, The Importance of Being Earnest

  11. #10
    Member Everett464's Avatar
    Join Date
    Nov 2015
    Location
    Eugene, Oregon
    Posts
    882

    Re: Question Re: Quartz Frequency, and the 1-Second Standard

    Quote Originally Posted by gangrel View Post
    Go back to Barracks' comment. Turning the motor is simply expensive. No engineer, but there's a minimum energy that has to be used to overcome inertia, no matter how far the hand is gonna move. So, yeah, the energy to move it 3/4 of a degree (8 ticks per second, 1 second is 6 degrees of movement) is less than the energy to move it 6 degrees, but not 1/8 of the cost. If we assert the Bulova draws at about 4x the rate of the ETA? What Barracks is saying is, figure driving the movement might be 75% of that. So, that fixed cost of getting the second hand to move, is just too high.
    Thank you! This is what I was looking for. That makes a ton of sense!

    Quote Originally Posted by gangrel View Post
    (Oh, BTW, the second hand moves at HALF the movement rate. 28800 movements tick at 4 Hz, not 8.)
    So, since we're on the topic, this is another thing I struggled with as I learned about what was happening. I'm not sure if I'm missing something, or if I'm just using bad terminology.

    First, in a mechanical movement 28,800 is 4hz, but 8 ticks per second, no?

    Second (and maybe this is related to the question above), I struggled to understand the 14 step-down of hz, versus beats. So, we start with 32,678 (2^15) hz. Then we step that down 14 times, each step-down reducing the exponent by one. But 14 step-downs only reduces the exponent to... 1. And 2^1 is 2! But we all know that a quartz movement ticks exactly once per second, give or take something like 20 seconds per year ;). I think, and this may be where I'm wrong, that the reason for the apparent discrepancy is that each movement of the second hand requires 2 hz (a tick and a tock). Yes, no?

    Why then do we get the opposite phenomenon with mechanical movements? each hz gives us two motions of the second hand.

    Quote Originally Posted by gangrel View Post
    A major draw of quartz is the casual way you can treat it...it runs accurately without any attention. Shorten the battery life, and you reduce that. Force more battery changes, increase the risk of a screw-up, of not re-sealing it properly or doing some other Bad Thing.

    One can't ignore the possibility that inertia's a big factor. Most people just don't care about sweep, judging by sales. Like they're not overly concerned about 20 SPY or better accuracy. KISS is just fine, and KISS says, keep things as they are. It's a totally mature technology, the processes are highly developed. It's these factors that make us so surprised by the Caliber 0100.
    This seems like a totally plausible explanation.

    I think I'll leave here with the following explanation for why 1-second movements are the standard: It's a mixture of (a) power loss from more frequent turns; (b) the resulting potential for more frequent servicing, and, thus, less durable movements; and (c) the maturity of the technology and a general reluctance to fix what is not broke.
    Last edited by Everett464; August 3rd, 2019 at 03:54.
    Co-Host of 40 and 20, a weekly podcast and discussion of affordable watches, food, drinks, life, and other things we like.
    Available anywhere you get your podcasts
    Instagram: @40and20

Page 1 of 2 1 2 LastLast

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

    Posting Permissions

    • You may not post new threads
    • You may not post replies
    • You may not post attachments
    • You may not edit your posts
    •