MCM Smiths Sectronic Clock MK1, help please

[dvdberg]

Hi,

I want to restore a nice Smiths Sectronic Clock with the original MK1 mechanism. It was not dirty at all and last night it ran for 20 minutes, stopped, and then ran again for another 20 minutes, stopped and now won't run for more than a few seconds. The battery voltage is 1.57V, could that be an issue since vintage batteries might not of had the same initial output? Any other ideas because I have none and am not at all savvy with electronics. I did check the voltage such that I know the battery connection is good.

[Phil G4SPZ]

Are the battery contacts pristine-clean? The slightest trace of corrosion from an old leaky battery will add resistance and could lead to intermittent operation. Use a miniature wire brush or fine emery paper, remove any traces of rust or tarnish. If you have any switch contact cleaner, a light application on a cotton bud will hopefully prevent further problems.

Good luck!

[Silicon]

There could be some corrosion hiding under that crimped eyelet on the black cable.

The die-cast chassis/structure may have distorted over time.

[dvdberg]

Thanks for those suggestions, I did them and no change. Still 1.5V being supplied.

[Robsradio]

Usually it is dried out oil that is the problem, they will either not run for more than a couple of seconds or will run for less than a minute with the balance wheel amplitude slowly reducing. As yours has run for a couple of 20 minute periods, it’s unlikely to be dried oil.
I would suggest it will be a failed electronic component, or an intermittent component or coil solder joint. As for battery voltage, I’ve seen them still working at near 1v.
Hope you get it sorted as it looks in nice condition.

[paulsherwin]

To reinforce what Phil said, battery connector problems are a very common source of trouble in old battery clocks, and faults are difficult to pin down. You need to look closely at the surfaces that make contact with the battery, but also at the other ends where the contacts join the PCB.

Of course, there are other possible causes - the mech may simply be worn if it's been running continuously for 50 years.

[dvdberg]

Oddly, it runs for 20 minutes and then won't run again for more than a few seconds. Then, waiting 24 hours, it will once again run for 20 minutes and then not again after that. I understand there to be a capacitor and transistor as the only electronic components. I also have read that capacitors fail, not wear out. But could this capacitor be doing just that, wearing out? Also, I have looked hard for an equivalent capacitor but cannot find one except from a single retailer and it would be a lot of money to replace a part I do not know whether is responsible for the clock's incapacity (pun intended) to work.

[blabbersnitch]

I have fixed a few balance wheel clocks over the years. Most were from old Datsuns and so they operated at 12V. A different circuit to your clock, but for what it's worth, I found either capacitors or transistors (or both) were failing in those clocks, and I would generally replace them all and then they all would work as good as new.

But one clock I fixed was a 1.5V wall clock from the 70's, possibly with an oscillator like yours. One transistor and once capacitor driving the balance wheel. I can't see a capacitor or transistor in your photos. Can you identify either?
If you are not savvy with electronics, the capacitor will be a component with two legs, and the transistor will have three. A 70's era transistor might be in a metallic 'can'.
The capacitor will be easy to replace. It should be pretty easy to find an equivalent (and a precise value match won't be necessary).
The transistor could be tricky. I expect it will be a germanium transistor at that voltage.
I was able to find one for my wife's clock. I traced out the circuit and found a germanium pnp transistor on eBay for a few bucks.
And she's been ticking away since.

Can you get a better photo of the electronic components? Can you identify what is printed on them?

[dvdberg]

Hi Blabbersnitch, the capacitor is a .02, 10V. The transistor is a Texas Instruments 2G300, D345 824. Thanks for your input and help!

[blabbersnitch]

Hi dvdberg,

For the capacitor, .02 would be 0.02 microfarad, or 20nF. It looks like ceramic and the tolerances and temperature coefficients would have been huge. What I am getting at is that you don't need to be too precise in finding a replacement. I suggest a standard value polyester 22nF capacitor, any voltage >10V (it will be hard to find one less than 50V rated anyway).
Something with long legs, something like this.
Pick the 22nF option, pack size, and the 100V rating is fine.

The transistor is a bit trickier. I will attach my notes from when I fixed a similar clock. You will see a note in pencil referring to "Junghans ATO clock". If you google "Junghans ATO clock circuit" and view images, you will see some similar circuits.
So... I think it would be a fair assumption its a PNP germanium transistor. I couldn't find a reliable equivalent for my 13x just as I can't find a reliable equivalent for your transistor. The AC188 worked (and continues to work) well for me, so why not give that a try, pretty easy to get, for example here.
But you will have to trace out your circuit to identify which leg is the emitter, collector and base. Trace out the circuit and see if you get a circuit like mine (maybe you should do that first). Then you can replace with the new transistor (you will see in my notes which leg is which when viewing the transistor from underneath).
Be very careful when soldering around the tiny tiny wires from the sense/drive coils. If they break... fixing gets a lot harder. Hopefully you won't have touch them or their solder pads.
Let us know how you go, or if you have any questions, drop them here.
Eric

[dvdberg]

Thanks for sharing your knowledge Eric. Is it worth trying to just replace the capacitor first and see whether the clock works or might a faulty transistor hurt a new capacitor?

[blabbersnitch]

Yes, give it a go. Try the cap first. Good luck.

[Robsradio]

I’ve just bought and repaired a couple of Sectronic clocks and thought I would sketch out the circuit in case it is of interest here.

[paulsherwin]

Towers suggests an OC44 as a sub for this transistor, but any small signal PNP type is likely to be OK in such a simple circuit, even silicon.

[Robsradio]

Originally Posted by paulsherwin

Towers suggests an OC44 as a sub for this transistor, but any small signal PNP type is likely to be OK in such a simple circuit, even silicon.

The early 1962/3 production movements had an OC series fitted, it was probably part of a cost cutting exercise to use the 2G300 or maybe supply issues of the OC devices in later units, the clocks I’ve just repaired are from ‘65 ish and have the latter device.

[dvdberg]

I give up! I've spent over $100 on this clock mechanism repair and have given up the hope of repairing it. Anyone want the clock mechanism? I still think it is a transistor issue, not a clock works issue. Just pay the shipping from Edmonton, Alberta, Canada and it is yours.

[Robsradio]

That’s a shame you’ve come to a dead end with this clock, they do turn up on eBay and charity shops here, but I guess not such a common sight where you live.