Why do all my clocks run slow?

[greypilgrim]

Hi.
I noticed that all electronic devices in my household that also tell time eventually lag behind, except the ones that get synchronized by radio signal or internet. Most of them are battery-powered, except my alarm clock (which runs slow as well).
Why does none of them run too fast? Deliberate design (why)? Wrong temperature for quartz crystal? Decreasing battery voltage?

Or just a coincidence?

 

[.Scott]

If you have an alarm clock that runs off wall current, it is almost certainly getting it pace from the power company. It will never run significantly fast - but if it becomes unplugged or you lose power now and then, it could appear slow.

Quartz oscillators are generally designed to be isolated from power source voltage fluctuations, but temperature is different issue.

Temperature is another matter. I have looked at a couple of sources and they say the same thing - that crystal-driven clocks are designed for room temperature (20C) or, for wrist watches 25C. Varying above or below that temperature makes the watch less precise - and that's as far as it goes. I found no mention that "less precise" favors either a lower or higher clock rate - but they do say it is the same in both directions (hotter or cooler).

 

[Baluncore]

Why does none of them run too fast?

Each clock has particular characteristics and needs to be discussed in isolation. Maybe you only notice slow clocks because you miss things, rather than get there a little early.

Clocks were once designed to be corrected after a stop by moving the clock hands forwards in time. If you turned the hands backwards, they came undone. To set a clock backwards you had to stop the clock, wait until it was correct, then start it again.

A clock is advanced in many steps. In general, it is easier to miss one step than to take two.

Some clocks were deliberately designed to run fast, then they are held waiting at a defined point, until a master clock pulse arrived, to release their gear train, hence the term "waiting train" clock.


https://waitingtrain.blogspot.com/2008/08/wt-pendulum-hipp-toggle-and-driving.html

 

[greypilgrim]

If you have an alarm clock that runs off wall current, it is almost certainly getting it pace from the power company.

I don't think so, as it runs off a wall adapter that outputs DC.

Clocks were once designed to be corrected after a stop by moving the clock hands forwards in time. If you turned the hands backwards, they came undone.

Interesting. Some of my devices have so few buttons that setting the correct time is quite tedious as you can only move the hours and minutes forward. So in order to correct a clock that is 1 minute in advance, you'll have to press a button 59 times (some devices also allow press and hold).
Correcting such a clock that is 1 minute behind is significantly easier; so maybe it IS part of the design to make the clocks tend to be a little slow?

 

[JT Smith]

Could they all be slow by coincidence? I haven't noticed that with our clocks even though I'm particular enough that I adjust them regularly. In fact I made a couple of clocks for our home just to avoid the problem (one uses a GPS and the other a crystal oscillator and temperature probe). The GPS clock is always right but of course has to be able to at least occasionally see one satellite. The other clock drifts less than 1s per year). From that sort of monkeying around I have learned that a typical watch crystal has an inverted parabolic frequency curve. At temperatures both above and below a certain temperature (e.g. 25°C) the crystal will be "slow". I would imagine that crystal clock designers would also know this and design the circuit so that it runs a little fast at 25°C and hence averages out with typical temperature fluctuations. But maybe not.