At the moment I am a little baffled!!!!
I have been asked by an electrician friend why a standard "Obis" time switch is GAINING time. For about 5 years it set automatic door locks at a residential care home and had been working perfectly until recently when it started to gain about 25 mins every 24 hours.
There is nothing special about it, no "electronics" just a disc magnet with pole pieces mounted in a little iron frame with a mains operated coil for excitation.
there is the usual gear train and cam system to operate the switches and it really could not be more basic.
An identical one is in another block and works perfectly.
Now I can understand how for various reasons it could run slow, but as I see it, the only way it could gain time was if the "mains" frequency (50Hz) increased which would then affect the other clock as well.
Any ideas anyone??

 

I'm puzzled as well.
Do you have an isolated oscilloscope that you can view the main's waveform with?
It's slightly possible that something is creating harmonics on the line that is causing the motor to run fast.

 

I'm puzzled as well.
Do you have an isolated oscilloscope that you can view the main's waveform with?
It's slightly possible that something is creating harmonics on the line that is causing the motor to run fast.

Well I had thought of that so I brought the timer to my home and I am running it from my isolated and filtered bench supply which is a good clean 50 Hz 230v rms supply and it is doing the same thing here. it has gained about 3.5 mins in 4 hours. Very weird!!! The manufacturers spec (Orbis alpha) says accuracy depends on line frequency and I have checked it several times and it is spot on. It really does not make any sense at the moment.

 

Is there any capacitor connected anywhere near the motor?

 

I'm puzzled as well.
Do you have an isolated oscilloscope that you can view the main's waveform with?
It's slightly possible that something is creating harmonics on the line that is causing the motor to run fast.

In an archive of ham-radio magazines I recently read about an engineer employed by the South African railway to do the electrification.

Having a difference of opinion with his employer - he somehow contrived to arrange the overhead lines to be an open ended resonant line at the mains frequency.

The first time it was powered up, all the fuses blew. So they checked everything for shorts and tried bigger fuses - then the generators blew up.

 

It is possible that some maintenance was done on the system and that parts were replaced with incorrect parts?

 

I'm not sure why they can do that but I know they can.

Years ago I worked at a office supply store as a copier service tech and our main time clock was one of those odd ball half mechanical half digital units from the late 70's or early 80's that ran the primary mechanical time stamping mechanisms from a synchronous motor like you are talking about but then had a digital clock readout that would trip a solenoid that resynchronized mechanical system to the digital time at noon every day.

Well anyway it was one someone has returned since the mechanical system had the exact same problem you describe and would run it self a good half hour fast every 24 hours so the company owner put it in as the company timeclock to save money.

The cheating bastard had the two times set so that it ran about 10 minutes fast on the time stamp in the morning so you has to punch in at 7:50 AM to get your timecard stamp to read 8:00 AM but the digital system was set slow so that at the end of the day you had to be there until around 5:10 PM to get your time card to punch out as 5:00 PM thus getting us to work an extra 20 minutes of unpaid time every day.

It's been about 15 years since I quit working there but I bet the cheap SOB still has that ancient time clock in use.

 

I'm not sure why they can do that but I know they can.

Years ago I worked at a office supply store as a copier service tech and our main time clock was one of those odd ball half mechanical half digital units from the late 70's or early 80's that ran the primary mechanical time stamping mechanisms from a synchronous motor like you are talking about but then had a digital clock readout that would trip a solenoid that resynchronized mechanical system to the digital time at noon every day.

Well anyway it was one someone has returned since the mechanical system had the exact same problem you describe and would run it self a good half hour fast every 24 hours so the company owner put it in as the company timeclock to save money.

The cheating bastard had the two times set so that it ran about 10 minutes fast on the time stamp in the morning so you has to punch in at 7:50 AM to get your timecard stamp to read 8:00 AM but the digital system was set slow so that at the end of the day you had to be there until around 5:10 PM to get your time card to punch out as 5:00 PM thus getting us to work an extra 20 minutes of unpaid time every day.

It's been about 15 years since I quit working there but I bet the cheap SOB still has that ancient time clock in use.

Seems a bit hard to believe that anyone could have gotten away with that for any length of time at all. It would only take one person filing a grievance with the state labor department to start the ball rolling.

 

You must be new to this planet.

I have yet to ever work for a company that was 100% honest and ethical 100% of the time with everyone 100% of the time. What they say and what they do tend to be quite a bit different at times.

The honest person may be in the right but if they are not in the majority it doesn't do them the least bit of good.

 

I must admit that it's got me beat at the moment, it is such a simple electro-mechanical device that really either works or does not. my mains supply is very good and I am testing it from my isolated and filtered bench supply. The place it came from is nearly 50 miles away so the chances of having the same supply are remote.
The history is this :- The apartment block has two entrance's with time controlled door locks, both control panels are in the same cupboard on the same supply. Both were running for several years without problem, until a few weeks ago when one started to lock/unlock ahead of the other by about an hour.
The engineer reset the clock but within a few days it was again gaining time, so he replaced it with a new one which is working correctly.
As it is strange for a synchronous clock to run fast, he asked me to see how this could happen. I looked inside to see if anything was obviously wrong
but all appears fine. there are no capacitors or other components, just the extremely simple coil and mechanically poled magnetic armature as is used in millions of similar devices.
I ran it now for 23 hours and it has gained approx 24 mins.
Mains wave form is really clean, and spot on 50 Hz.
I think I may send it to the manufacturer for investigation if I/we can't come up with an answer.
Thanks everyone for the input

 

Seems a bit hard to believe that anyone could have gotten away with that for any length of time at all. It would only take one person filing a grievance with the state labor department to start the ball rolling.

No one in the company owned a watch?

 

I must admit that it's got me beat at the moment, it is such a simple electro-mechanical device that really either works or does not.

My theory is that the tiny magnetic poles on the rotor loose enough of their strength over time that they go from being fully synchronous to having a slightly variable speed due to lack of magnetic pole strength to keep them properly synchronized.

No one in the company owned a watch?

You had to work there to understand.

Sales people were treated like gold because the business owner was a former sale people and only looked at how much money they brought in not what it cost have them there. The service department people were just something to be tolerated and every dollar spent on parts was a dollar that didn't go into his pocket because we weren't selling the customer a new machine every time their old one broke down.

 

My theory is that the tiny magnetic poles on the rotor loose enough of their strength over time that they go from being fully synchronous to having a slightly variable speed due to lack of magnetic pole strength to keep them properly synchronized.
.......................

But it would seem that would cause it to lose speed, not gain it.
I don't see how a motor can run at higher than synchronous speed.

 

It's sort of like how a double fed (powered stator and powered rotor) motor can run in super synchronous modes by cheating the magnetic field rotation ratios between the rotor and stator but in a tiny permanent magnet rotor a few of the poles are so weak that they either have too little or no magnetic field strength left and can at times allow for the ratio of rotor poles to stator poles to jump from being equal to being off by one or two favoring the rotor to have a few less poles than the stator thus allowing it to jump ahead a few degrees every so often.

A 20 pole non phased stator turning a rotor that has ~18 or less weakly defined and uneven spaced poles can do odd things to the synchronization effects. One being allow for odd jumps (forward stepping) ahead in rotational phase alignment if the rotor poles are unevenly spaced.

 

Okay, so in effect, it's reducing the number of effective poles, which leads to a higher synchronous speed.
Sounds problematic, but it's the only theory so far, that would explain the observed anomaly.

 

I had started to think along those lines, it's about the only possible explanation for it running fast. I just had a thought ! (dangerous at my age)
I can hear that the motor is not rotating at an even rate, this may sound odd, but the rotor is making a whirr - click sound, but another similar motor has a continuous whirr sound, maybe the click is it missing a pole and jumping faster in the gap. Just as tcmtech suggested. Anything at this point is possible I guess

 

Have you tried lubricating the mechanism?

 

I would guess mechanical, something happened to the large reduction gearing that these units have..
Especially a fixed amount of error.
Max.

 

You have my permission to open the motor. I am betting on a gear with a bad tooth.

 

I will open it up again and check, but I have already looked at the gearing and they all seemed fine and rotated easily with no binding or bad teeth.
I am now more inclined to think along the lines of tcmtech's post 14.