clock running fast

[jfaska]

I uprgaded to Active Home pro at the begining of the summer and all was going well.
I had been using the older software/hardware since 2000.

Suddenly the clock in my module cm-15 is running fast.

Support suggested pulling the batteries and reseting the device.
I did that and it ran "OK" from sunday to some time over the weekend picking up a minute every day or two.
(I added a nightly event so I can see what time it is, cause AHP won't tell me what time the interface thinks it is)

THe clock was OK on Friday night, but by Monday night I found that it was 30 minutes fast.

I just cleared the memory (via the program) and reset the clock.

Tonight it was 15 minutes fast again.

Any sugesstions?

Thank You
Thanks

[murphy]

The clock is based on the 60 Hz line frequency.  You have something in your house that is injecting noise of sufficient amplitude and frequency to make the clock run fast.

[jfaska]

thanks, two questions

1 - How do I figure out what?

2 - Why did it work fine for a month or so (and why does the amount of noise/drift seem to be variable)?

[MANOWAR©]

thanks, two questions

1 - How do I figure out what?

2 - Why did it work fine for a month or so (and why does the amount of noise/drift seem to be variable)?

What have you added in the last month that takes power? Finding it is as simple as unplugging all the new stuff then adding it back one piece at a time until it shows up again, then put a filter on that item or if it's like a cell phone charger use a plug strip for it.

[tjosan]

The clock is based on the 60 Hz line frequency.  You have something in your house that is injecting noise of sufficient amplitude and frequency to make the clock run fast.

I'm not so sure that will help. I had the same happening to my controller and the only way
to correct the problem was to buy a new controller. The old one ran for about a year or so
without any problem and suddenly it started to add minutes. I could reset it, but after a few
days it was back to the old behaviour.

[Dan Lawrence]

You are incorrect.  What was posted by : MANOWAR© was the correct thing to do.  There are so many electronic devices that make noise and can alter the electric power frequency slightly.  The biggest noise producers are cell phone chargers and number one is the PC, desktop and laptop.

[dave w]

You are incorrect.  What was posted by : MANOWAR© was the correct thing to do. 

Maybe not. tjosan corrected his time problem by replacing the CM15A, which points to a problem other than line noise, otherwise the new unit would likely exhibit the same problem. The CM15A uses a crystal clock which could be drifting.

[Brian H]

The CM15As controller chip. CY7C63723 has an input for a crystal but in the schematics it looks like X10 chose to use the chips internal one that can be calibrated though the USB Port. I wounder if they are getting reset and are out of calibration.

[Alan V]

The CM15As controller chip. CY7C63723 has an input for a crystal but in the schematics it looks like X10 chose to use the chips internal one that can be calibrated though the USB Port. I wounder if they are getting reset and are out of calibration.

If the circuit uses zero crossings of the 60Hz AC line for the clock/calendar, then I wouldn't think that the processor internal clock calibration would matter.

[Dan Lawrence]

The problem in this thread is that the OP (jfaska posted only twice and never posted any more. We don't know what he did to solve his CM15A clock problem.

[dave w]

The problem in this thread is that the OP (jfaska posted only twice and never posted any more. We don't know what he did to solve his CM15A clock problem.

Yes, good point. Would be nice to know if the problem was fixed and how.

[dave w]

The CM15As controller chip. CY7C63723 has an input for a crystal but in the schematics it looks like X10 chose to use the chips internal one

Yeah Brian, your are right. When I did my antenna mod, I thought I saw a "watch xtal" on the board. But I just checked and conclude it was hallucination (or wishful thinking). So the clock for the micro P pretty much free runs?

[tjosan]

You are incorrect.  What was posted by : MANOWAR© was the correct thing to do.  There are so many electronic devices that make noise and can alter the electric power frequency slightly.  The biggest noise producers are cell phone chargers and number one is the PC, desktop and laptop.

As someone pointed out, when I switched for a new one the same problem would have existed, right. After changing for a new
CM15 it's been running fine for a year or so.

[jfaska]

x10 sent me a new one it was ok for about a week

last night it picked up a few minutes

I'll have to watch and see what happens (and see if anyone remembers pulgging in anything last night)

[GeorgeM]

Here is some parallel information.  A couple of years ago, my electric utility upgraded to remote meter reading.  I knew the then chief engineer of the utility and he gave me a thumbnail description of how the system worked.  It modulates the AC zero crossings digitally.  And so, all my old digital clocks started running fast because there were now additional zero crossings.  I started opening them up, and discovered that the RC low pass filter shown on the chip data sheets was absent in every bloody one of them.  Even though I found 3 different clock chips, none were in circuits recommended by the chip designers.  All different makes and models of clocks, including some clock radios.  Adding the recommended resistor and capacitor fixed the problem.  My CM 15 runs fast for this same reason, and I will eventually get in and filter the optical coupler output which provides the 60 Hz signal to the CY7C63723 chip. 

The real problem which arises here is that any filter added externally to the CM15 will also remove the output signal.  I may end up bringing the output out on a separate set of conductors, plug the CM15 into a UPS, and feed the output directly into a power outlet.