X10 Community Forum
🖥️ActiveHome Pro => ActiveHome Pro General => Software Problems & Bugs => Topic started by: pondo on March 03, 2011, 07:59:53 AM
-
This is about an X10 apparent problem that I’ve had since 2004 off and on.
What could be corrupting the program of timers and macros in the CM15A so certainly every 3 to 7 weeks?
I bought and installed the CM15A, as soon as they came out with it in 2004 or whenever.( I needed more storage memory )
Sometime after installation, I noticed that after a few weeks or months that the CM15A would begin starting macros and timers too early/late or not at all, and it was obvious the CM15A’s time of day had been significantly changed, by some unknown process.
Since I knew nothing had been done to the CM15A since the last download of macros and timers, I would connect up the CM15A via the USB and clear the memory and re-download the last dated timers and macros. I would then disconnect the USB cable and CM15A from my downloading computer and plug the CM15A back into a 120 vac wall socket near to the main panel. In a few weeks or months the same thing would occur again where the CM15A performing like its program of timers and macros had become corrupted.
I bought another CM15A and the same thing happened to it.
For several years now I have had to re-download to CM15A the same program every few weeks. And alternated from one CM15A to the other by re-download- all with the apparent corrupted outcome in a few weeks.
I’m of the opinion but I don’t know anything about it that there is nothing being received, by the CM15A except maybe STATUS REQUESTS AND SOME POLLING .
We have few lightning storms and surges are not much problem because I live in a salt marsh region. Power company surges are more rare than power failures, and they corrected their phantom ground faults several years ago.
1. What could be corrupting the program of timers and macros in the CM15A so certainly every 3 to 7 weeks?
2. What can be done to cause the CM15A to maintain the integrity of its downloaded program of timers and macros?
-
I know the early CM15As had many problems.
Some of us even volunteered to get special engineering versions. That could archive data and then have it sent back to X10. For evaluation.
How old is the second unit?
I know my old date coded 2004 units have a different revision of firmware in it than the later ones do.
What revision AHP are you using?
From what I read. You are not leaving the USB cable hanging off of the CM15A when you moved it.
-
the first CM15A was bought November 2005, and a year later I bought the 2nd. unit CM15A in Dec. 2006.
I am running AHP ver 3.306 rel 2-14-11, but I have always updated AHP to the then most current version since 2005, and the updates did not fix the corrupting of the down loaded program of macros and timers,
The Cm15 A has been disconnected from the computer and the usb cable immediately after the down loads, and the cm15a relocated to an ac socket near my main panel.
Then in 2 or 3 or as much as 5 weeks sometimes the program is corrupted where it was fine the day before.
-
Have you tried not disconecting the CM15A and see what happens? I've always had the CM15A (original one from 2005) connected to the PC and never had a problem with " corrupting timers and macros". Also are any of the macros "run from the computer"?
-
no, none of the macros are run from the computer, and the disconnect of the usb and Cm15A is done correctky. I'm a computer experienced person going way back 40 years, and I'm a wire head too havin retired from the Bell System as a telephone engineer.
Keep in mind that my program downloaded into the Cm15A runs correctly for 2 or 3 sometimes 5 weeks and then on a single day is corrupted.
How often do you have to re-download your program of macros and timers in order to renew its accurate performance as you initially set it up?
-
Sounds like you are technically inclined.
There have been reports {2005} on timers and macros going bad. Along with some unit appears to be not responding.
Many times the remove battery, power reset and download timers and macros would correct it for a awhile.
It seems X10 left a few bypass capacitors; most electronic engineers would normally use; out of the design.
Could be noise corrupting things on a very random basis.
Now I know X10 has redesigned many modules to use more easily to find parts. So latest CM15A could be different as my latest one is a Date Code 2008 if memory serves me.
Not sure this will help but you may want to read andre's fix in this message.
http://forums.x10.com/index.php?PHPSESSID=dn0kub8g1fg2shiri210g5tbc7&topic=4181.0
The schematic and internal photos of the CM15A are in the FCC database.
http://www.fcc.gov/oet/ea/fccid/
Grantee: B4S
Product Code: CM15A
In the details section.
-
Thanks Brian, I have ordered a new and 3rd CM15A It was just the 50 bucks and if that new one continues corrupting the program of timers and macros down loaded, I will rewrite the program from scratch on with out any connection from my present file of Timers and macros that I've down loaded.
I don't want to spen a couploe of days rewriting the program, and then tweaking and debugging all the errors and oversights that i will create in the new oprogram from scratch.
Do you know any way that' s easier to get a purely fresh file program, Or is it likely even necessary to write a new program. Will it be any different from the present one.
-
I think your terminology is a bit wrong. The .ahx files AHP creates are not programs, they are files, like what a word processor does.
-
There have been many reports of the timers getting off-track when the CM15A is disconnected from the PC. This is due to the design of the time-keeping circuit in the CM15A, which uses the 60Hz power to keep the time, instead of using a real-time clock chip. Fluctuations in the AC power can cause the clock to start to "drift," or throw it off completely.
When connected to the PC, the CM15A syncs the time every two hours, so there is less chance of power fluctuations causing the time to drift enough to notice.
If you are able to leave the CM15A connected to the PC (with the PC running, of course), you should be able to determine if that improves the stability of your system or not.
-
I'm a wire head too havin retired from the Bell System as a telephone engineer.
Then you should be good at running wires! rofl
Many users have ran a extention from the HA PC to the CM15A thus keeping it plugged into the USB port.
As Noam pointed out, There are several threads on the forum about the time loss.
Your Macros and timers are not getting corrupt it is the clock in the CM15A.
The issue is even more noticeable around the old & new daylight savings time settings
>!
-
Tuicemen -
To be fair, you can't prove that it is "clock drift," but that seems to be the most likely culprit. Especially since he has the same problem with two separate CM15A units, purchased at different times.
Connecting the CM15A to a PC for as long as it normally takes his CM15A to "corrupt" itself would be a good way to test that theory. If it still gets corrupted, then there might actually be something wrong with his CM15A.
I didn't have much luck with extending the USB connection between the PC and the CM15A. I ended up running a long extension cord between my CM15A (near the PC), and my XTB-IIR (near by breaker box).
I keep the PC on 24/7, connected to the CM15A.
I think this issue has become less common recently, once X10.com started introducing more plugins that require a 24/7 connection to a running PC.
-
True, although it is the most likely issue!
For two CM15As to suffer the same fate makes it even more likely!
The Fact that every one that runs a CM15A disconnected from the PC for a good length of time experiences this is another point!
The fact that all these posts happen around daylight savings time is another( thought not related here)
The fact that it is timers out of wack
You your self stated the issue with timers!
I'm just adding up the stats. ;)
-
Agreed. "clock drift" seems most likely. I read back through the original post, and he is reporting that the timers are "off" (either early or late) after a few weeks. That certainly does fit with "clock drift." From what I've read over the years, "clock drift" normally seems to show up as a slow drift over a period of hours/days/weeks. You might only lose (or gain) a few minutes a day, but that can add up over a few weeks. It might take that long to notice the drift.
Some users have created timers to monitor this, having a chime go off every hour, and comparing that to the same clock every time. If the timer goes off at 12:00, then 1:01, then 2:02, it is a clear sign that there is some drift in there. (that's an exaggerated example. It might appear to be on the hour for a whole day, before it starts chiming early or late).
It may also be possible that the 60Hz power coming from the utility is *mostly* clean, but does have some fluctuations from time to time. Enough of those can cause the clock to drift, and possible a noticeable amount in a single day!
Keeping the CM15A connected so it can sync every two hours should keep that problem in check.
-
Its been a long time, and I'm still wrestling with the corrupting or clock drift problem.
Can you describe for me if you know, how the CM15A being connected via USB to a 24/7 running pc gets updated and corrects the CM15A's clock to agree with the PC's clock, by merely being 'wired together" with a USb cord or even with ActHmoe Pro running and doing something that is not obvious.?
I see no descriptio, in ActHmPro software, of the every 2 hour update, and the activity monitor does not seem to reflect any polling or mutual contact.
-
You have something else that somehow blocks the CM15A sending signals. I mentioned before that I have been running AHP and the CM15A since 2005 without any clock drift or corruption of the .ahx files (what AHP makes when you save the file you just created). Are you 100% sure ALL your modules are correctly established as "old style" (no Soft-Start) or "Soft-Start modules depending on what you have.
Check your house to make sure no other electronic equipment is blocking X10 powerline signals. In 2011 every house has some kind of X10 noise due to that electronic equipment. Computers are a prime noise source, every one in my house has a filter on it and my CM15A is on a circuit all by itself.
-
Can you describe for me if you know, how the CM15A being connected via USB to a 24/7 running pc gets updated and corrects the CM15A's clock to agree with the PC's clock, by merely being 'wired together" with a USb cord or even with ActHmoe Pro running and doing something that is not obvious.?
I see no descriptio, in ActHmPro software, of the every 2 hour update, and the activity monitor does not seem to reflect any polling or mutual contact.
I got that information straight from the developers. They were helping me track down a very strange USB communication problem I was having at the time.
-
Maybe it is buried in X10nets.exe as I believe it starts on boot up.
-
Maybe it is buried in X10nets.exe as I believe it starts on boot up.
Yes, X10nets runs as a background service, so it starts when the Pc starts. However, it is possible to shut it down, or to plug the CM15A into a Pc that doesn't have it installed at all.
In either of those cases, the time syncing won't happen every two hours.