Synchronous clock loses time

[Hampus]

I have a synchronous wall clock that don't keep track with time. It seems to lose about a minute or two per day or thereabout, too much to be acceptable of course but too little to be an easy diagnose. I have several other synchronous clocks in my home and they keep time as they should.

A synchronous motor should not vary it's speed but this is a self-starting movement that can run up to speed unsynchronized and then lock to the mains frequency. So it is possible for it to run slower than intended. The movement is almost hermetically sealed and therefore the whole mechanism is oiled, wheels and all. It looked absolutely clean inside and the oil seemed fresh, despite this it was lagging so I dismantled it, cleaned and oiled it but no difference.

The mains cable is new. I have ran the clock in series with a meter while moving the clock in different positions and while tapping the coil to see if there were any anomalies in the current consumption which would indicate an intermittent coil. I have been observing the rotor under the light from a neon indicator too see stroboscope effect, I stared at it for a while (boring) but it kept perfect speed while doing this. I have removed the hour hand and the wheels that drives it so only the minute hand is left, to eliminate these parts from the fault finding.

The wheels inside the movement do not seem to bind, they can easily be rotated by hand and they all look like new and have a sufficient axial play. The pivots are shiny and oiled.

I am getting quite frustrated Any ideas?

[Nickthedentist]

It looks a good quality item in very good condition.

No ideas at all, I'm afraid, you've done almost everything I would have done already.

No doubt someone cleverer will be along soon with some ideas...

[broadgage]

Low mains voltage, that is within tolerance for your other clocks, but not for this one ?
A synchronous motor is not normally affected by voltage variations, but as this one can also run asynchronously might be a possibility.

Or intermittent low voltage on only the sub circuit that supplies the clock, due perhaps to say a washing machine motor starting.

Is the loss of time random or consistent from day to day ? Does it occur steadily throughout the day or concentrated at certain times of day ?

[Mike Phelan]

Originally Posted by Hampus

I have removed the hour hand and the wheels that drives it so only the minute hand is left, to eliminate these parts from the fault finding.

That only eliminates the hour hand - if that was the problem then the hour hand would not be showing correctly against the minute hand.

The wheels inside the movement do not seem to bind, they can easily be rotated by hand and they all look like new and have a sufficient axial play. The pivots are shiny and oiled.

Were that so, it would probably just stop.
Most clocks of this sort will run happily in much less mains voltage - say 150 VAC.
A couple of thoughts; does the centre arbor (thing holding the minute hand) seem too free?
Is the 'slowing down' regular or just at random?
Have you another electric clock you can connect to the same socket?

[Malcolm G6ANZ]

Just a random thought, is the minute hand secure to the arbour? If it is slightly loose then as the hand sweeps from 6:00 to 12:00 it could drop back slightly under gravity.
Malcolm

[kalee20]

Does this actually run up to speed unsynchronised and then lock?

I have seen small synchronous motors with lots of poles, that run at only a couple of hundred RPM and can accelerate to synchronous speed in a single cycle - so they keep turning. They won't run asynchronously. Loading the shaft stops them abruptly. But I have also seen motors like you say - basically a shaded-pole induction motor with a permanent magnet on the rotor - which will run asynchronously yet lightly-loaded can lock-in to mains frequency.

I can't help thinking that Malcolm might be onto something, and mechanical slippage is the culprit!

[Mike Phelan]

Originally Posted by Malcolm G6ANZ

Just a random thought, is the minute hand secure to the arbour?

That would make the hour hand keep time, Malcolm.

I still think the centre arbor isn't keeping time but the rest of the train is - ie the friction device, spring or whatever it is will be the culprit.

[dseymo1]

Do you have a camera which can be set to take pictures at, say, 20 minute intervals?
If so, set it looking at the face for 24 hours, then review the pictures, which will indicate either a gradual or a sudden loss of time.

[Hampus]

Thank you for the many answers. It was relieving to talk about it and to get some ideas. This is the kind of odd fault that make you wonder if it really is a problem or if it is yourself who are having a 'brain slip' and missed something obvious

Originally Posted by Nickthedentist

It looks a good quality item in very good condition.

Yes I think this is a quality movement, it is from a huge wall clock made by Westerstrand, one of swedens major maker of clocks for public places.

Originally Posted by broadgage

Low mains voltage, that is within tolerance for your other clocks, but not for this one ?
A synchronous motor is not normally affected by voltage variations, but as this one can also run asynchronously might be a possibility.

Is the loss of time random or consistent from day to day ? Does it occur steadily throughout the day or concentrated at certain times of day ?

The motor is rated for 220V which was the previous nominal voltage in Sweden. Now it is 230V and at my home it is usually about 235V. I have also tried the clock in different sockets on different circuits. I have tried to run the motor with a variable transformer and it do not seem to be picky at all regarding the voltage.

The clock has no second hand so it is difficult to see when or how it loses time. It is only when the error has accumulated some that it "shows up" on the minute hand. I will anyway try to monitor it and write down the error so I can get a more accurate estimate of the problem.

Originally Posted by Mike Phelan

That only eliminates the hour hand - if that was the problem then the hour hand would not be showing correctly against the minute hand.

I was thinking that since the hour hand is driven by the minute hand, any binding here would affect the timekeeping.

Originally Posted by Malcolm G6ANZ

Just a random thought, is the minute hand secure to the arbour? If it is slightly loose then as the hand sweeps from 6:00 to 12:00 it could drop back slightly under gravity.

The hand has a square fit so no slip possible there, but the hand is mounted on a "friction device" that fits on the arbour to make time-setting possible. See attached photo. This seems like a tight fit though.

Originally Posted by kalee20

Does this actually run up to speed unsynchronised and then lock?

I have seen small synchronous motors with lots of poles, that run at only a couple of hundred RPM and can accelerate to synchronous speed in a single cycle - so they keep turning. They won't run asynchronously. Loading the shaft stops them abruptly. But I have also seen motors like you say - basically a shaded-pole induction motor with a permanent magnet on the rotor - which will run asynchronously yet lightly-loaded can lock-in to mains frequency.

This motor is a bit different than the other clock motors I have seen. The rotor is made of aluminium with a smooth ring of steel in the circumference. The ring is not magnetic by itself. It may be a hysteresis synchronous motor..? See https://en.wikipedia.org/wiki/AC_mot...chronous_motor If so it has good torgue from start to full speed and can run slower than expected. Good for reliable starting but maybe not beneficially for timekeeping! The motor seems quite strong though so I would expect it to stay in sync unless much overloaded.

Originally Posted by dseymo1

Do you have a camera which can be set to take pictures at, say, 20 minute intervals?
If so, set it looking at the face for 24 hours, then review the pictures, which will indicate either a gradual or a sudden loss of time.

That is an interesting idea, unfortunately this clock has no second hand and that makes it a bit hard to spot deviation until it has accumulate some. But I agree that I should somehow log the deviation to get some accurate numbers of how much it is. I will try to reset it now and write down current time and check it regularly.

[The Philpott]

Amazing to think that that the lag is only around 0.14%, but noticeable because we expect major precision from relatively modern pieces.

I would repeat the mention of moving the clock to another power source to exclude any possibility of a 'dirty' supply causing intermittency. i would think that running the clock for a few days at a significantly lower ambient temperature might help to ascertain whether the problem is mechanical or electrical. (Shed?)

[Mike Phelan]

Aha! Your picture shows that the motion work is like those on French clocks.
The centre arbor does not slip, but the cutaway part seen on the cannon pinion on which the minute stands is the friction device.

Solution:
Attach something to the centre arbor to act as a temporary minute hand. If it keeps time then it can only be the cannon pinion slipping; if so a slight squeeze is needed.
A common fault in French roulants.
There should be no oil anywhere on this part of the clock.