IBM My first IBM master

[D.P.]

Hello All,

Posting my IBM master I just picked up yesterday. I've never worked on one before so it's a interesting project. It's probably the most complicated clock I have so far with all the extra equipment it came with. I have it cleaned and running right now without power. Needs more adjustment to run perfectly, but it hasn't stopped so far.

There's a lot of information on these clocks it boggles the mind, but mine has a feature I haven't seen posted yet(unless it has. I apologise if it's been covered already). I noticed it has a telechron rotor with contacts attached above the movement which is possibly connected to a correction mechanism?? Not sure what it is. Have any of you seen something like it before?

The clock looks to have it's own transformer installed. Its marked for 52, 40, C, and 24vDC. Not sure what the movement coil takes as it had no markings. Planning on disconnecting the un-needed equipment and running on a wall adapter or a pulser described in another thread. Am thinking of running a few secondary clocks with it some day though.

 

[Toughtool]

Nice looking clock.

Planning on disconnecting the un-needed equipment and running on a wall adapter or a pulser described in another thread.

You seem to have all you need to run the clock without a pulser. The coil voltage is most likely 24 volts DC and all you need to do is power the mains.

The unit at the lower third (see my photo) is probably an accumulator feature, although it is a little different than the one in my photo. It's job is to bring all secondaries back on time after a power failure. It may be different because the spring wound master movement will not run long without power.

Although I've never see one, the" telechron rotor" and contact unit could be a duration feature to blow a horn, whistle, or bell for a period of time. Not sure. The 52 V and 40 V voltages are for powering Bells, etc. The C is probably the 24 Volts for the secondaries, with A and B lines taken to ground when active.

The silver looking unit in the middle left looks like a programming unit, programming the picking of the relays at the bottom to ring bells, etc.

As for the extra solenoid on the upper right, I don't know.
This coil is a 60 Hz AC coil, voltage unknown. It is possible this clock may be a slave master to a "master" master clock. Look and see if the armature of this solenoid contacts the escapement wheel or the verge. I have read about this configuration but not seen the mechanism. If it is a slave master it would have been adjusted a couple of seconds fast and the escapement would have been held up until the top of the hour. Interesting non the less. Joe

 

[D.P.]

Hello and thanks for the reply!

All the extras have been removed from the movement for the time being and it's running in the case by itself. But I took pictures before hand to make sure I had an idea of what wires went where.

The solenoid is as you describe. When I removed it I did notice that the arm did contact the verge so it likely was used like you said. Previous owner said it came out of a Air force base so that would make sense.

I'm tempted to try and use the existing hardware to run it, but I don't trust the wiring and don't know much about electricity. I only see one broken wire so far(pictured) though and all else seems intact when looking it over in the back. I was careful not to stress any of the old wire when taking the movement out.

 

[D.P.]

Also, in regards to the accumulator, it definitely looks a bit different from yours. Couldn't quite figure out how it works, but I noticed the wires went to the movement coils and from there to the set of contacts to the left of the movement. I'm assuming it was actuated when the winding coils were. The little post is marked A, B, and C so I guess power went directly to there, and then to the movement contacts?

 

[Toughtool]

OK D.P., It looks like I was right about this being a slave master. The base probably had slave masters in more than one location. You will notice the gear, mounted on top of the escapement wheel in your photo. The armature would drop in one of these teeth and hold the escapement wheel from turning, making the movement wait until the top of the hour. I would suggest you leave all the hardware but disconnect the circuit so it does not pick. The clock will run as a stand along master just fine.

The accumulator works this way. For every minute the clocks advances, the gear on the hour hand shaft will move the accumulators gear one minute of rotation. Every minute, when the minute impulse takes place, This large disk (with a notch at the bottom in my photo) is brought back in the opposite direction by one minute. The net result is zero rotation while the clocks runs. You can think of it as a secondary and it has an A, B, and C connection. When the power is lost there is no voltage to C (always +24 volts) and no secondaries nor the wind coils operate, and the clock runs on it's mainspring, advancing the disk. When power is restored the circuits will revert to an advance mode and provide an impulse every other second until the disk is brought back to the zero position. This also advances all the secondaries, the program unit and the movements wind coils to wind the mainspring. This auto-advance does not happen on masters without an accumulator. Those masters must be advanced manually with the Run/Advance switch.

Every IBM master clock is built for that specific customer so schematics for a particular system probably will not exist. Here is a schematic that is close and if studied should provide you with the information needed for the critical parts of your system. Such as winding the mainspring, the Run/Advance circuit, the accumulator and any secondaries attached. The schematic is one I drew for Jims clock and be aware that some of your contact wiring may differ slightly. The main thing about the wiring shown on my schematic is that everything on the left side of the terminal strip shown is contact wiring on the master movement. Everything to the right of the terminal strip is wiring in the relay logic. Both sides of the terminal strip wiring may contain differences in wiring. The terminal strip, usually found on top of the cabinet, is a standard for most IBM masters, either with relay logic in the clock's case or relay logic and power supply in a different cabinet. That is why there is no "schematic that fits all" for these clocks. If you can understand my schematic you should be able to translate it to your clock. My schematic does not include the program unit, or the bell and whistle logic relays and switches. Just the minimum logic needed to wind and run secondaries.

 

[ibm clock]

the device at the left, under the movement, looks to be a modified 803 program unit. they were used in the later 91 Series masters. perhaps one of the added bits is to advance the 803 program drum. all the rows of newer relays on the bottom, the pushbuttons, and the switches is looking like they added bell timing circuits sometime after the master was installed. the 803 programs units usually used 115v ac in the bell circuits. the advance coil in the 803 could be a 115v ac coil and i suspect one of the added timing relays may be using 115v ac. hopefully the advance coil in the 803 i marked to clear that one up.

 

[D.P.]

Thanks again Toughtool! The schematic seems to mostly line up with everything I have. So I'll ask another possibility dumb question. To power the clock, would I use some type of power supply or power pack (like for a computer) in 24vdc? Or does the large black metal box(transformer possibly) do the same thing? If so it probably just connected mains power to the terminals at the top which goes to the AC and AC+ marked spots on the box? Then it's the correct voltage is converter from there?

ibm clock: The clock came with a small baggie of metal clips that seems to snap into the drum on the program unit. I guess I'll unhook it so it's not powered. Some more photos added to give an idea of things.

 

[ibm clock]

My opinion is the telechron rotor and contact might be the minute impulse trigger for the 803 program unit. It "might" use a 115 volt AC impulse coil, so the contact is also seeing 115v AC. Really hard to tell just by looking at pictures.

 

[Toughtool]

Thanks IBM Clock. About the baggie of metal clips that are used to program the drum, do you know if the clips will also work on the traditional programming disks, usually found on these IBM master clocks?

To power the clock, would I use some type of power supply or power pack (like for a computer) in 24vdc? Or does the large black metal box(transformer possibly) do the same thing?

D.P.,
Your first photo shows the mains wiring to power the clock, connected to the A.C. terminals (upper left and upper right) on the transformer. I notice there is a brown wire and a black wire also connected to these two terminals so something else is being powered in this system with mains power as well. Maybe the 803 programming unit that IBM Clock mentioned. There may be a fuse somewhere between the building wiring and these terminals. Mine is on top of the clock in a ceramic block. Depending on country of installation, This could be 115 or 230 VAC and 50 or 60 Hz. We already know your clock works on 60Hz line frequency from the printing on the solenoid. I can not read what the values are on the voltage and frequency tag in the second photo.

You will be OK to connect these two terminals, A.C. and A.C., to a 120 volt cord and measure with a meter, set to AC on a scale greater than 240 volts, and measuring the 52 VAC terminal (Red Lead) and the +/- terminal (Black Lead) and see if it is 52 V or 26 volts. If the voltage is 26 volts then the primary will be 230 VAC, not 115 VAC. If you do not have a meter, they can be purchased for $6.50 at Harbor Freight.com or better ones at Amazon, Lowes, HD, etc. The $6.50 one works and there are plenty of videos on the web to show how to use them. Just be careful, working with 115 volts can be lethal.

Based on the style and the square face, I'm pretty sure your clock was made after World War II. Your question about using a power adapter is a valid question. IBM's power supplies were rated at 2 amps maximum, and that is intermittent, and they are not regulated or filtered. This places a peak voltage of 29.7 volts DC across the secondaries and wind coils. I prefer a regulated 24 volt adapter, but that means the clock will not be historically correct. You get to choose.

Allaboutcircuits.com has a very good basic electronics course for free. I recommend you at least take the courses in Vol I at: Vol. I - Direct Current (DC) - Electronics Textbook On their main web page click on the Education tab at the top. Be sure to read the sections on electrical safety and how to use a meter in this course material.
Joe

 

[ibm clock]

the easiest way to wind the clock is to just bypass everything. get a clock winder from kensclockclinic.com. you can disconnect the wind coil wires and connect the winder right up to it. all it will do is just wind the clock. none of the master clock features or even the correction cycle will work. but it will at least wind the clock. he has a winder that can wind the clock and support 2 slave clocks. again, will just be straight minute impulse, no correction. on the slave clocsk if you use kens winder, you also have to jump the A and B wires, i think. i forget wich two wires are for the impulse circuit and which one is common. if you dont, when the slave switches for the correction cycle, the slave would stop.

 

[Toughtool]

the easiest way to wind the clock is to just bypass everything.

I think the easiest way is just connect the 120 volts to the transformer as mentioned in post #9, and see if the clock works as it is. These things are pretty reliable, even after 60 years or more. I have also developed a one pulse per minute pulser using a MOSFET transistor and a Trinket processer ( Adafruit Trinket - Mini Microcontroller - 5V Logic ) that cost around $10.00 to build.

If you only want to wind the clock just connect the minute impulse contact as in my third schematic
(post #5) and use a 24 volt power adapter like this one: https://www.amazon.com/Generic-Power-Adapter-Supply-Black/dp/B00J8CG86G/ref=sr_1_52 for $7.00 with free Prime shipping.

If you tie the A and B wires together in the secondaries and use the one (1) pulse per minute unit from Kensclockclinic.com , then you loose the correction feature these secondaries have. However, if you want to run ITR and IBM secondaries accurately, you can build one of the electronic masters I discuss in my article on the Encyclopedia forum; Collaborative Articles ; Clock Articles.
Here is the link: Development:“A Computer Based Master for ITR & IBM Minute Impulse Secondary Clock Movements”
The second schematic shows the electron current flow through the circuits.

 

[ibm clock]

Well, im using the seperate winder on mine so I dont have that transformer plugged in contantly. It sits there and hums, plus i have peace of mind knowing im not keeping 70 year old wiring constantly plugged in. Plus, it also eliminates the correction cycle relays clanking 20 times each hour. Only need to hear it once a minute, for one second instead of two.

 

[D.P.]

A bit of an update to make sure I'm doing things right before moving on ahead.

So far I've unhooked and removed the program unit. Took pictures of everything before doing so to make it possible to put it back later. It's seal bagged and set aside for now. Accumulator was treated the same. Removed and set aside for later. Unhooked the 60hz rotor and removed wires from to terminals. Left it installed. Plan on seeing what it does later if possible. Carefully reattached the wires to the contacts on the left of the movement. Cleaned the dirt out of the terminals on top and attached a power chord to the terminals connected to the transformer(marked a.c. 115 and a.c. 60cy 174watts). That the correct spot? S1 & S2 were for the telechron rotor thinggie and was attached to the empty side of the 115ac terminals.

I've used the pulser from Ken's clock clinic before on a clock project for someone else. I was planning on getting one for a Standard Electric clock I have that was put together from parts. I also ordered a 24v power adapter just because. I'm perfectly happy if I can make it work with all the hardware clacking along just to see it do it's thing then swap to a modern power source for everyday use. I might just leave it as is if I feel comfortable enough with it.

Pics of the terminals and control panel for reference.

PS: I just thought of it. Should I use some type of resistor across the coil leads to help protect the contacts?

 

[Toughtool]

Plus, it also eliminates the correction cycle relays clanking 20 times each hour.

You are absolutely right about the noise! Mine is in the room above the garage and I can hear it in the house, and I need hearing aids.

Cleaned the dirt out of the terminals on top and attached a power chord to the terminals connected to the transformer(marked a.c. 115 and a.c. 60cy 174watts). That the correct spot?

That looks about right to me. If you trace the two wires already attached there and down the hole, see photo, they should go to the transformer's A.C. terminals. Is that right? Joe

 

[D.P.]

Toughtool: Thanks again for all the help so far! Traced the wires again and it went to the transformer. It's the only other spot with a porcalin tube luckily. I was looking for signs of former hookup to mains power, usually a piece of cut wire, but only found some cut portions on a different terminal that also had a wire attached to the 115ac ones so that confused me a bit. I've never seen something like this run with all the extra equipment before, so I'm expecting a much louder *ka-chunk* than my SE master in the basement.

 

[Toughtool]

Should I use some type of resistor across the coil leads to help protect the contacts?

I think you mean Diode. At this point I would wait. I think your problem on the other clock was that the clock came with a 24 volt power supply instead of the 10 volt it was supposed to have, causing more than twice the current through the contacts and accelerating the eventual damage to the contacts. I suspect the minute impulse relay will be the contact set that actually operates the wind coil and any secondary connected to the systems A, B, and C lines. If the minute impulse contact on the movement is used, it still may be a good idea but my circuit using the MOSFET transistor will not need one as it had a protecting diode built into the transistor. As far as the Ken's clock clinic unit, I am sure it has the protection to handle the counter EMF generated as it is a commercial unit and has been on the market for some time.

OK on removing the accumulator. Just insulate the three wires to the terminal strip (A, B, and C,) and the wires to the run/advance circuit and the clock should be happy.

You can actually leave the escapement inhibitor solenoid in place since you do not have the pulses coming from the Master master clock to activate it.

Let us know if the clock winds itself. You can expect one pulse per minute and 21 extra pulses during the 59th minute because the wind coil is attached to C (always at 24 volts) and A, which is grounded 80 times per hour.

The ceramic insulator is for the 115 volt AC circuit which is a lethal voltage. 48 volts DC and below is generally considered non-lethal.

 

[D.P.]

Another update: Success... mostly.

It startled me quite a bit but it powered on no problem. I just wasn't expecting a super loud and strong *ka-chunk URRRRRR* as the solenoid, wind coils, and transformer made their noises all at once. So on the electrical side everything is working like it should.

The next problem (other than the noises that I'll have to live with) are the arm that winds the mainspring isn't going forward far enough to reach the end of the ratchet teeth, so it's not winding. I've moved the one pawl that prevents the ratchet from turning backwards forward to it's maximum and that helped a little, but the main pawl still doesn't go forward enough to grab the teeth. Any advice on adjustments to correct it?

 

[Toughtool]

You should have the minute hand installed with it's nut or the contact cams won't rotate. Now make sure the spring is not fully wound. If the spring is fully wound, the pawl will not drop into the next tooth. To verify, just unplug the clock and start the pendulum swinging and let the clock run for an hour, then stop the pendulum. You should now be able to manually press the armature while watching the pawl assembly and see what it is doing. There are a couple of different styles of wind pawl assemblies. The mainspring requires 60 pulses per hour to keep it wound, but it gets 80 pulses per hour so that it can rewind an unwound spring after a long power outage. When the spring is fully wound the drive pawl will not grab the next tooth, until it runs a while.

Page 6 of the ITR Service Instructions 222 at: https://www.ibm.com/ibm/history/exhibits/cc/pdf/cc_2407MCE1.pdf (A lousy copy) show one of the styles. See also pages 12 ad 13.

I have noticed that IBM has added a couple of new documents. I wish they would post some of the service instruction PDFs I sent them. Photos in mine are sharp and clear. They told me there was no money to update the site. Joe

 

[Toughtool]

Oops, Page 11 says I am wrong. The clock will only run for 51 to 53 minutes from fully wound. So only run it about thirty minutes and then do the adjustments from pages 12 and 13. Notice the reference to the stop pins from my screen shot of the PDF. That is probably why you are not seeing the clock wind. Joe


click on this for a better view.

 

[ibm clock]

his clock should run about 12 hours if it has the original mainspring. those old ITR instructions are just that, OLD, for older products. Plus the section you got that 55 minutes from is headed with Old Style Spring Driven Clock. Some older clocks used a coil spring and not the flat spring steel mainspring, like his 25 should have. The flat spring steel type will run for about 12 hours.

 

[Toughtool]

OK, I wondered about that. I mean, why have a 12 hour accumulator for a movement that would only run for less than an hour on spring power. I referenced that PDF because it was on IBM's site and obviously offered to the public. I'll look at my book but my book is an April 1, 1938 version, much older than his clock. The PDF they have, Service Instructions #222 was retyped and they used blurry photos that are almost unreadable. I would offer scanned copies here but not sure of the copyright laws. The 80 years was up in 2018 but you never know.

 

[D.P.]

After some tinkering with the movement coils position I managed to get the clock to wind. It's been going for a few hours now. The loud buzz takes some getting used to though. As far as the mainspring goes, it is the flat coil type. I wound it up by hand before work and it was still going when I came back home.

The Amazon prime ninjas delivered the power supply today so I'll give it a try some time. If done right, I should be able to return the clock back to its original configuration without problem if I want.

Thanks everyone for your help getting it running!

 

[fdew]

You have a very unique clock with all the accessories. You have received great advice.
If you want to get rid of the CLUNK buzz winding you can look into a AC winder Simplex/IBM offered one that mounter where the coil is but they are impossible to find, but many people have built one. The idea is to use a 1 RPM motor to slowly pull the lever that the coil pulls and then release it.
You can search for IBM Winder to get some ideas.
I made one using a 1 RPM motor and a scrap gear with a hole drilled off center. I mounted a RC car tie rod end using the off center hole. A spring was connected to the tie rod end and a wire went from the other end of the spring to the winding lever. If the off center hole is placed right the spring is not needed. Crude, but quiet.

 

[D.P.]

Thanks for the reply and advice. I've considered looking into making a version of the wonder. So far It's running off of a variable dc power supply wired directly to the contacts and coil. Much quieter and seems to run reliably off of 16vdc. Still clunks but I'm used to it so it doesn't bother me luckily. We just had a power outage and it kept going for most of the time.

 

[spclock]

Great clock, pretty rare with both an accumulator and the synchronizer. I collect US military clocks and I currently have 2 WW@ era model 35 IBM masters that were US military, one USN and one Army Air Corp (a recent Ebay find). Generally, of the military marked masters I've seen, all (before yours) were WW2 era and weight driven. The Army Air Corp clocks I've seen all used a Mercurial pendulum and had 13 thru 24 numbers overprinted in red on the standard dial. otherwise they were a standard Model 35. The one I bought recently was built as a very accurate wall clock, with no contacts or wiring to drive a slave system. The Navy clock uses an Invar pendulum, and the weights and pendulum were finished in grey wrinkle paint. otherwise it's a standard model 35 set up to drive a slave system.
I'm building (slowly) a web site documenting some of these clocks, would you mind If I used a few of your pictures of your clock ? It's the only post war (1952) military master I've seen so far. Also, are you in the states or in Europe ? Most of the accumulator equipped clocks I've seen were in Europe, they are very uncommon in the US. There's a ton of good info available here, but I I can help with anything, let me know...

John

 

[Toughtool]

I collect US military clocks and I currently have 2 WW@ era model 35 IBM

How about a couple of photos of the clocks. Joe

 

[D.P.]

spclock Sure, you can use the photos if you wish. I'm located in the U.S. I was actually trying to get a IBM master in my area that looked the same as my current one but was also weight driven with polished nickel weights and mercury pendulum. Figured the weight driven ones to be more accurate. Owner said it came out of a Steel Factory. Unfortunately another buyer got to it first. Oh well. Also post pic!

On a somewhat related note, the local Airforce Base near me has a Standard Electric master clock in their collection.

 

[spclock]

Hi Joe, I'll be glad to post some pictures, it might take a day or two till I get back to my shop ( The Navy Model 35 is over there..) I'll dig out some other IBM stuff and post some pics also ...

 

[spclock]

Here's a few (lousy) pics of the Navy Model 35 I had on my phone... This clock came from one of the Naval Aircraft Factory buildings in the Philadelphia Navy Yard when the building was closed down in the 1960's. It's a normal Model 35 except the weights and Invar pendulum are finished in silver crinkle paint instead of chrome. The little plate at the bottom of the door says U.S. Navy Property and has a property number stamped. The weights are both dull silver, I don't know why one photographed so bright.

 

[spclock]

Here's 2 slaves that I know were military, The 24 hr dial clock is Property U.S.Army Air Force stamped on the back and has a 24V coil, The 12 hr dial clock is Property US Navy marked on the lower front edge and has a 48V coil The AAF clocks show up occasionally on Ebay. The Navy marked one was a recent Ebay find but it's the only one I've seen.

..

 

[Toughtool]

Nice looking clocks. It doesn't look like D.P.'s clock, which had the logic and power supply inside the clock. Did you get the logic cabinet with your clocks?

I like the 24 hour secondary. IBM made a 12 Volt, 20 Volt, 24 Volt, 48 Volt and a 120 Volt DC. The ITR instructions list all but the 20 volt models but I happen to have a 20 volt model, so they made one of those too. I like the extra hour hand that they set to GMT. This would come in handy for logging amateur radio transmissions.

Too bad it is a 48 volt clock, but there is a fix for this. You may try to make a converter like I did for converting a 24 volt secondary to work on a 12 volt master clock system with something like this.
https://www.amazon.com/Yeeco-Converter-Adjustable-Transformer-Stabilizer/dp/B074J9D278/ref=sr_1_8

If we start talking about a different subject you (or me) need to start a new thread and not hijack D.P.'s thread. So if you are interested, let me know. Joe

 

[spclock]

I'll get some more/better pics together and start a new thread, check back in a few days .....

 

[Toughtool]

Just for clarity, I looked at the minute impulse secondary voltage list and the 120 volt I mentioned is actually 115 volts DC and there is also a 230 volt DC voltage. Please don't mistake the 115 volt DC and 230 volt DC models for an AC version and plug it in to the mains. You could get some smoke. There are 115 volt AC secondaries and these usually have a red sweeping hand. The Impulse secondaries do not have a seconds hand.