Electrical help please

This is where I am up to with John Wilding's Large Balance Wheel clock. At the moment, I'm running it on a transformer as it eats batteries. The large 'D' cell batteries last for around 48 to 50 hours. According to 'The Book' the coils are wound with a resistance of 10 ohm, now, if my sums are correct, with a voltage of 4.5v this gives a 0.5 amp load which seems a lot and there is no wonder the batteries don't last long. It's not a problem, as it runs ok on the transformer, I was just wondering if anyone had any thoughts on the matter.

As always, Many Thanks,

David

Edited By Neil Wyatt on 15/04/2020 12:27:10

David, for how long does the contact feeding the coils remain closed? Presumably they are only momentarily energised by some sort of contact when required - it may be that the period is too long.

The problem with using a transformer is that you are dependent on the mains. On my version of the Synchronome I run the electronics from a 12V 1AH sealed lead acid battery which is fed from a cheap plug-top smart charger from Halfords. This gives me ~ 6 - 10 hours running if the mains fails. Depending on the voltage from your transformer you could wire a couple of diodes so that as long as the rectified transformer voltage is higher than the battery voltage (plus diode drop), the battery is isolated but if the mains fails the batteries take over.

Synchronous electric wall clocks consume about 2-2 1/2W, so that power does not appear excessive.

Make sure your transformer is an efficient one (new ones should all comply with grade V standard by now, I think) or run it from a switched mode supply.

Simply running the clock from an 18650 lithium ion cell might well be adequate? Being charged by a suitable regulated power supply, should suffice, I would have thought? (No different than a car battery being charged by an alternator?)

Could the problem be that the impulse is being applied too frequently? Perhaps due to excessive friction somewhere. I have no experience of that clock but I see that it uses ball bearings for the balance staff. They are usually supplied in a greased condition and for clock use should be run dry to minimise friction. Did you wash them out thoroughly? Just an idea.

Russell

Not seen the design but.

I think my approach would be to add series resistance to the coil until the clock stops then back off a little. Most conveniently done with a suitable variable resistor. Then you can work on making sure the clock runs a freely as possible. If you are moving in the right direction you shoud be able to increase the series resistance even further.

This is no different than reducing the weight on a weight driven clock to a minimum.

As I say I haven't seen the design but as John says the impules period needs to be no longer than neccessary too.

regards Martin

+1

What flattens the battery is the current flowing (0.45A) multiplied by the amount of time it flows.

Depending on make and chemistry the capacity of a 'D' cell will be somewhere between 10000mAh and 20000mAh but this is measured at a low discharge current, say 25mA. Drawing an amp or more can be expected to halve the battery's effective capacity because charge is wasted heating the battery's internal resistance.

So two rules for long battery life:

1. Keep the current drawn low rather than high.
2. Draw current in short bursts only.

Nothing can be done about the 0.45A load because the clock is designed for 4.5V with a 10 ohm coil.

However, drawing current in short bursts may be fruitful. Is it possible to adjust the mechanism so the contacts only close for the minimum time needed to keep the clock going reliably? ie they go 'clicky click' rather than 'click on, pause, click off'. Halving the ON pause will double battery life, or more.

Dave

I do not think adding a series resistor to reduce the current flow is the correct approach, as energy would be waisted in heating the resistor.

The correct operation of this clock would be to use a very short pulse of current as said above. As the coil is an inductor it will have impedance which for short pulses wil be much higher than 10 ohm. With an inductor, when a voltage is applied, the current starts at zero and then rises to the steady state current determined by the DC resistance. So if the pulse is short enough the current flow is minimal.

Adrian

Perhaps you could measure the frequency of impulses and ask on here and other clock forums for other makers to report the frequency of theirs for comparison.
You could also try reducing the voltage to find out how low it will go though this might interact with the frequency of impulses. Either way determine the power input. then having perhaps decided on a psu+battery as suggested above rewind the magnet coil to suit. People on here can assist with deciding on the necessary change in wire size and number of turns.

Yes, but SoD is a clever old sod, and has full diameter resistors that magically don't dissipate energy.

Andrew

Some energy would be wasted heating the resistor* but the current draw would also be reduced, put simply a 10 ohm resistor in series with a 10 ohm coil would halve the current drawn. Whether it would still work is another matter though. And yes shortening the 'live' period would be the proper way to do it.

Disclaimer, I know a fair bit about electricity but bu99er all about clocks

* just as some is wasted heating the coils, it's just physics and unavoidable

That resistor value will share the potential difference, supplied by the battery/power supply, with the clock mechanism resistance (in accordance with Ohm’s Law). Once known, the battery/power supply of a just sufficient value can be used for the clock, while dispensing with that resistor. Simple as that.

The frequency is presumably once per oscillation (or maybe twice)? That's baked into the clock I think.

NO! I see that this is based on the Hipp toggle giving intermittent impulsing - so I suspect there is something not set up right, possibly too much friction so it's havigg to impulse too frequently. In a discussion over on YouTube someone says their version impulses once per 16 cycles, each cycle is 4 seconds, so that's every 64 sec. A Synchronome impulses every 30s and I believe the standard ones will run for months on 2 or 3 D cells.

With electric clocks of this type both the mechanical and electrical efficiency must be good if the battery is to last.

As an example the Eureka clocks will typically run for a year of a D sized cell and this has a 30 coil resistance.

To check the mechanical efficiency just set the balance wheel in motion, without the battery connected, and see how long the balance wheel remains oscillating. On the Eureka clock I have in for service I released the balance wheel from 120 degree position from the at rest position and the balance wheel final stopped 90 seconds later.

For good electrical efficiency the timing of the energisation of the coil must coincide with the magnetic pull produced by the coil acting on the balance wheel at the correct position of the swing.

I'm not sure about the exact design of the Wilding clock but on the Eureka clock the contact closes 30 degrees before the pole pieces are at their minimum distance and the distance is about 3 thousands of an inch. The coil/pole piece arrangement forms a reluctance motor in that force is generated between the pole piece and coil as the length of the magnetic field in the air gap is shorted. If the air gap is very small the inductance of the coil can be very high even for a very low resistance coil. A high inductance has the effect of slowing the rate of rise of electrical current when a voltage is applied so the effective current drawn may be quite low provided the duration of the coils energisation is short.

So for good efficiency the air gap between the pole piece and coil must be short, the timing of energisation should be short and at the correct point in the balance wheel position to give the greatest mechanical impulse to the balance wheel.

I hope this helps.

Clive

I see that John says the clock uses a variant of the Hipp toggle to control impulsing of the coil depending on the swing amplitude rather than at every swing. As the amplitude decays the Hipp toggle is triggered and an impulse is applied to the balance wheel. As mentioned before if the mechanical or electrical efficiency are not optimum then the impulsing will be too often giving a short battery life.

The mechanical efficiency ( Q ) is easy to check by doing the oscillation check. The electrical efficiency can be checked by holding the balance wheel at its neutral position and energising the coil. Then move the balance wheel either side of the neutral position and see what force is felt acting on the balance wheel. There should be a sweet spot just before the neutral position that maximum force is felt. When you know this position then check that operation of the Hipp Toggle acts slightly before this point. The lead timing is needed so that current can build up in the coil. Usually impulse is given only one side of the neutral position (top dead centre in engine terms) as any current flowing in the coil after TDC could retard the balance wheel and reduce efficiency

Clive

Thank you all.  Yes, it's a Hipps Toggle arrangement. At the moment, I have a pulse every 20 seconds or so but I am able to change the timing etc fairly easily.

Lots to keep me occupied here. Quite looking forward to it actually

Many thanks, David

Edited By David Noble on 15/04/2020 15:49:37