|
Other Vintage Household Electrical or Electromechanical Items For discussions about other vintage (over 25 years old) electrical and electromechanical household items. See the sticky thread for details. |
|
Thread Tools |
5th May 2021, 9:01 pm | #1 |
Nonode
Join Date: Mar 2008
Location: Aberdeen, UK.
Posts: 2,838
|
IBM Factory Clock
I'm venturing into a new field for a pal. It's a Bakelite factory slave clock with a 12" dia. dial. I'm guessing 1950's. The Bakelite is a bit grubby but is responding to some Farecla G6 finishing compound. Glass is just a wee bit scratched.
Guess a quartz fitment is needed, but I've never removed big hands off their spigots before and need advice as to sourcing the correct battery mechanism replacement, please. Regards, David. |
5th May 2021, 9:42 pm | #2 |
Dekatron
Join Date: Jan 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 8,171
|
Re: IBM Factory Clock
Hi David, it looks like an impulse driven type, but others will know better; why not make your own "pulser" (locked to mains frequency if possible)
Ed |
5th May 2021, 9:50 pm | #3 |
Nonode
Join Date: May 2007
Location: Preston, Lancashire, UK.
Posts: 2,510
|
Re: IBM Factory Clock
Sorry, David, but replacing a slave clock mechanism with a quartz mechanism is just plain bodgery in my book!
A bit like the old joke 'if I wanted to get to Dublin I wouldn't start from here'. If your pal wants a quartz clock, buy a new one, don't destroy an old one. I don't have any knowledge of IBM clocks (Synchronome and Gents Pulsynetic are my interest), but if you want to run the slave without a master, check if there are any electronic slave-driver modules available for the IBM system. What form of pulses does an IBM slave work on? I'll do some digging, maybe ask on the electric clocks group. Andy |
5th May 2021, 10:20 pm | #4 |
Nonode
Join Date: May 2007
Location: Preston, Lancashire, UK.
Posts: 2,510
|
Re: IBM Factory Clock
OK, Googling for IBM Slave Clock Impulser gets lots of hits, for example:-
https://www.piexx.com/imp2/imp2.html Andy |
5th May 2021, 10:31 pm | #5 |
Dekatron
Join Date: Feb 2005
Location: West Cumbria (CA13), UK
Posts: 6,118
|
Re: IBM Factory Clock
I agree that it would be a great shame to replace its mechanism with a quartz movement - assuming that the impulse-driven mechanism is in (or can be brought to) working order.
Presumably the first thing to discover is what voltage is required, though I suppose this could be determined experimentally by starting at the bottom and working up until a response is achieved. I assume that manually pulsing using a battery making momentary contact across the pulsing terminals should a) determine whether it's working and b) how far the hands move and therefore the pulsing interval needed (most likely 30 seconds). Interesting, however, that there are three wires - though maybe this is obvious to those familiar with pulse clocks (which I'm not).
__________________
Mending is better than Ending (cf Brave New World by Aldous Huxley) |
6th May 2021, 3:00 am | #6 |
Dekatron
Join Date: Apr 2012
Location: Worksop, Nottinghamshire, UK.
Posts: 5,549
|
Re: IBM Factory Clock
After looking at the link in post#4 it has made me think it is 6 volts pulsed at one minute with a far eastern board that looks better made than a quartz module that is very unlikely to work with the original hands.
The third wire is likely to be an earth. There will be little difference in cost for a pulser board and it will almost certainly be adjustable for differing pulse rates. |
6th May 2021, 6:49 am | #7 |
Nonode
Join Date: Jan 2010
Location: North Somerset, UK.
Posts: 2,129
|
Re: IBM Factory Clock
AFAIK these types of clocks were rated by CURRENT and not by voltage.
An original installation would contain numerous clocks and long cable runs, maintaining the correct voltage at each clock would be a challenge, but a constant current fairly easy. Clocks of different sizes intended for use on the same system would all have the same operating current, but probably a different voltage drop. For a new installation, the total resistance could be measured or calculated and the battery specified to drive the required current through this resistance. A lead acid battery on continual trickle charge was the traditional power source. The original master clock was generally a good mechanical clock, corrected regularly by radio time signal or astronomical observations. I agree that this clock would be better restored to original working condition, useing a modern electronic unit to provide the pulse signals. |
6th May 2021, 8:26 am | #8 |
Guest
Posts: n/a
|
Re: IBM Factory Clock
As far as I can recall the IBM ones are 24V wired in parallel, the third wire is for an hourly/daily sync. just in case it missed an impulse.
|
6th May 2021, 8:33 am | #9 |
Octode
Join Date: Nov 2011
Location: Manchester, UK.
Posts: 1,872
|
Re: IBM Factory Clock
I use a master and slaves on this system. It's a 24v pulse, once per minute, which lasts a couple of seconds (though a shorter pulse will work). There are three wires because the system does something odd on the hour - there is a little pip on the back of the minute hand wheel which lifts a switch contact in the slave, and changes which wires are connected to the coil (a semiconductor diode inside the slave is also involved on mine, which may be newer than this one). Also on the hour, some contacts change over in the master, and the thing applies about 10 pulses with a 2 second period (one on, one off). I half remember reading somewhere that the point of this was to bring all the slaves into synchrony once an hour, in case any had dropped a minute here or there - perhaps once that pip makes contact, any further synchonising pulses are redundant.
The master makes quite a clunk when it applies the pulse (via an electromagnet which pulls over a carriage containing a mercury switch) - and 10 big clunks on the hour was too much for a domestic environment! So I slipped some paper between the relevant switch contacts in the master clock, and connected mine up as a 2-wire system. I needed to think about what that little pip was doing inside the slave on the hour, to make that work - and my two slaves generally stay in concert. I have some more movements but no more faces etc - those bits tend to cost money these days! Hope that's useful. Nice clock - it would be good to keep it original. M (oops - crossed with Merlin) (Looking again - mine only have one electromagnetic ratchet in them, and yours appears (perhaps) to have two. This might have been the way to deal with 'normal' pulses vs 'synchronising' pulses, before semiconductor diodes were available.)
__________________
"The best dBs, come in 3s" - Woody Brown Last edited by mark_in_manc; 6th May 2021 at 8:42 am. |
6th May 2021, 9:33 am | #10 | |
Dekatron
Join Date: May 2003
Location: Edinburgh, UK.
Posts: 3,273
|
Re: IBM Factory Clock
Quote:
Peter |
|
6th May 2021, 9:45 am | #11 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
|
Re: IBM Factory Clock
The small 'clonk' every thirty seconds is part of the experience!
David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
6th May 2021, 11:10 am | #12 |
Nonode
Join Date: Mar 2008
Location: Aberdeen, UK.
Posts: 2,838
|
Re: IBM Factory Clock
Thanks guys for all the info. As Andy says, it would be athsetically preferably to keep the original pulsed mechanism, I was merely guessing that a quartz mechanism was the only option these days. The "Piexx" conversion looks the bee's knees, but guess its hellish expensive sourcing from the USA. Wonder if there is a UK supplier of a similar kit ?
The existing mechanism doesn't look too grubby or worn, and a gentle tap on the solenoid arm clicks the minute cog one tooth past the pawl. But hey - I ken sfa about these things, but hope to learn. Mark - yep that wee switch contact had me puzzled - I'll try & take a decent macro picture of it. How are the hands secured to their spigots ? By wee knurled nuts or a tapered press-fit, or what ? The securing nut for the whole mechanism is a bit loose beneath them. I wont attempt any removal until "I'm telt fit ta de". Having banged-on in the past about inexperienced folk poking about inside AVO VCM's - here's me delving into heffing big clocks, eh ! Mind you, I used to service & repair Sangamo Weston time clocks years ago, but I'm treading very carefully with this big thing. Regards, David Last edited by David Simpson; 6th May 2021 at 11:11 am. Reason: spelling |
6th May 2021, 11:36 am | #13 |
Moderator
Join Date: Dec 2010
Location: Exeter, Devon and Poole, Dorset UK.
Posts: 6,823
|
Re: IBM Factory Clock
You might be able to do something with this.
https://sound-au.com/clocks/arduino.html There are several forum members pretty handy with these. Cheers Mike T
__________________
Invisible airwaves crackle with life or at least they used to Mike T BVWS member. www.cossor.co.uk |
6th May 2021, 12:13 pm | #14 |
Heptode
Join Date: May 2010
Location: Warminster, Wiltshire, UK.
Posts: 682
|
Re: IBM Factory Clock
I have several Gents slave clocks round the house driven from a master clock.
After all these years i never get tired of looking for the clocks to pulse. I know the sound is there but most of the time I just blank it out. My partner recently got a new set of hearing aids. After about half hour she said I don’t now what it is but but these hearing aids are picking up a slight knocking sound every so often. Took me ages to work on it’s the clocks every 30 secs. I agree . Keep it original It looks a lovely clock . Andy
__________________
I bet that car doesn't have a suppressor. |
6th May 2021, 1:08 pm | #15 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
|
Re: IBM Factory Clock
There are several approaches you can take:
1) Track down and buy the type of master clock which once drove it. 2) Build a crystal oscillator and a chain of counter ICs to divide the frequency and make a pulse every 30 seconds. This can be done on veroboard. Almost any frequency of crystal could be used as a starting point, but a round binary number of Hz is easiest. Learn something new, if you've not done this sort of thing before. 3) Do a crystal oscillator and program up a PLD (prog logic device) to do the counting 4) as above but program a small microprocessor like and Arduino or raspberry pi. But the crystal oscillator needs to be accurate. 5) Buy a ready made board. I think the switch contacts allow a group of clocks to be synchronised. Imagine they were all scattered in the times they showed. All clocks would be given a fast series of pulses like a few per second, with the switches configured to stop the each clock when it reached a known setting, say 12:00. Then the whole group have their switches disabled and are fast stepped with enough pulses to bring them from 12:00 to the current time. It can save a lot of carrying ladders around! I once was at a meeting at the German post&telephone HQ. The room was part of a modern architecture tower with one full wall all glass window. In the middle of the glass was a box driving the hands of a big clock outside. Every 30 seconds it moved the hands with a motor. Clonk, off came the brake. ratch-ratch-ratch-ratch went the gearbox. Clonk, on went the brake. You could really sense the passage of time in that meeting! David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
6th May 2021, 4:17 pm | #16 |
Nonode
Join Date: Mar 2008
Location: Aberdeen, UK.
Posts: 2,838
|
Re: IBM Factory Clock
Hey David, you've out-guessed me, as I've been in touch with the owner, and am hoping that he's got contacts at the old redundant factory(CP- Consolidated Pneumatic). The original main office block is divided up between the local water board & an engineering business. The same business also has taken over some of the huge workshops for storage, other sheds have been taken over by wee local businesses.
Mike T's link might do for someone with 21st century micro-processor/logic skills. Me - I'm just an old analogue fart. I do have some old SW mains powered timeclocks somewhere, but very much doubt if the gearing is in any way compatible. The wee switch is a cam/lever driven SP/2 way. The rear cam moves the centre contact down for minutes 7 to 60, and the front cam moves it up for minutes 1 to 6. Shall delve further. I still need advice on safely removing the hands, please. They seem to be secured by 6BA sized knurled(non-hexagonal) nuts. And the main securing nut seems to be 0BA sized. Regards, David |
6th May 2021, 4:26 pm | #17 |
Nonode
Join Date: May 2007
Location: Preston, Lancashire, UK.
Posts: 2,510
|
Re: IBM Factory Clock
BA nuts on an American clock?
Andy |
6th May 2021, 5:39 pm | #18 |
Octode
Join Date: Nov 2011
Location: Manchester, UK.
Posts: 1,872
|
Re: IBM Factory Clock
If they are the same as my slightly later ITR (which appear to be the same as IBM) clocks, then the minute hand sits on a square shaft, held on by that knurled nut. So when the hand goes back on, you align it so that pip lifts the switch on the hour. The hour hand on mine is a light push fit on a circular shaft, so this can go back wherever you want it.
My master clock for the slightly later ones is this: https://picclick.co.uk/ITR-Master-cl...652357108.html and for this one, it will probably be this: https://www.ebay.co.uk/itm/401944903...cAAOSwTlddvvQ1 or earlier still, perhaps this https://www.ebay.co.uk/itm/183938555...kAAOSw8EhdbXiq Regarding 'American' - ITR seems to have been a name used by IBM in the UK, and my clocks claim to hail from London W6. cheers Mark
__________________
"The best dBs, come in 3s" - Woody Brown |
6th May 2021, 5:58 pm | #19 |
Hexode
Join Date: Jan 2016
Location: Ramsgate, Kent, UK.
Posts: 296
|
Re: IBM Factory Clock
This thread brings back memories, about 40 years ago my brother, now deceased, asked me to make something to drive a clock he had bought that was very similar to this one, it needed a short 12v pulse every 30 seconds IIRC.
CMOS dividers were newish kids on the block then so I used them, a few CD4018s to divide down the mains frequency to get the required pulse if memory serves. As the unit was to be mains powered and placed under a wooden floor to drive the clock situated in the room below I went to great pains with the design to make sure failure of components could not start a fire. It worked fine but his wife could not stand the 'clonk' every thirty seconds so it went where I know not! Sorry I don't have the circuit, sands of time I'm afraid.
__________________
Andrew Illegitimi non carborundum |
6th May 2021, 6:04 pm | #20 |
Nonode
Join Date: Mar 2008
Location: Aberdeen, UK.
Posts: 2,838
|
Re: IBM Factory Clock
Andy - I said "sized", meaning not exactly, but of a similar size! I wont know the exact size & type until I remove them. So - what's the best method of safely loosening them(as they've been on for 60 years or so), and what thread will I find ? Are the diddy shafts splined ?
Of the three coloured wires - Red goes straight to one end of the coil, and the other end of the coil goes to the pole of the 1P/2W micro-switch. Down connects it to "White" for 52 minutes, and Up connects it to "Black" for 8 minutes. Eg. 1.51 to 2.43 = Down, 2.43 to 2.51 = Up, then 2.53 to 3.41 Down again & so on - - . All operated by the front & rear 1hr Cams. Simple me thought that this change-over would occur "on the hour". Coil = 800 ohms & 3.32H. Regards, David |