Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[narib]

   Hi all

I have the latest version of the software and 12 modules that have worked fine for over a year.  In the last few days 5 of the modules have turned off at 9:00 pm when they were instructed to turn off at times ranging from 10:20 to 11:00.  I have changed nothing and they simply began to act differently.

I cleared the interface and changed the off times by a few minutes.  I saved the changes and downloaded to the modules.  Did not help at all.

Any Ideas??

[narib]

So to be clear, all the modules still come on as scheduled but each day they turn off at 9:00.  When they turn off, they are all react at the same time.  I have not added any lights or noise makers that I know of and no repairs were done to equipment in the house.  We are the only house on the transformer  and we have not been able to identify an event at 9:00 that could be creating the noise

[Brian H]

Your description sounds like AHP but you have not specified it is AHP.
Is AHP the problem and are you running it on a CM15A, CM19A or CM11A?
Are you running AHP 3.318 as it is the latest?
Do you have any added plug in AHP programs?
What module are the problem?
If Lamp and Wall Switches. Are they Soft Start, where they ramp On and Off?
Do you have any Motion Sensors?

[dave w]

 Any Ideas??

Maybe.

Something is sending an "All Off" command at 9:00PM. Are you sure an "ALL OFF" didn't sneak into your AHP program?

[Dan Lawrence]

Can narib post his timers here so we can see how the timers were built?

[cjhallx]

After one of the units turns on, change the House Code. 
If this one device does not turn off at 9 with the rest then you don't necessarily have the cause but you will have the solution.
Someone sharing my transformer out at the street was using the same House Code and we were both causing each other similar problems.
Luckily they agreed since I was using the House Code first they would change theirs.

[narib]

First, thank you all for your thoughts and help.  I have been out of town and have just returned.

I am using AHP 3.304 and  CM15A plugged into one leg of the power.  I have a bridge at the hot water heater (ELK9100).  I am not running any plug in programs that I know of.  There are no motion sensors in the present system.  The modules that are acting up are Transceiver tm571 running a series of LED lights in the entertainment center (code a - nothing plugged into it) - appliance module am466 for more led lights in the great room (code a10)- Lamp module lm465 for one small light in the entrance hall - transceiver tm751 for an led picture light in the great room( the light is plugged into the unit)

I looked at the activity monitor and did not see an all off command.  Is that the place to look for it?  I am confident this is the problem as the entire system did shut down

Here is the strangest part of the problem.  Without warning the problem came on and was there for 14 days in a row.  On the 15th day, while I was out of town, it began working again and has now operated for 2 days.  The modules are responding to the program perfectly.  javascript:void(0);

So now I wonder what was the cause of this and will it return later on?  Like so many other things, we have become accustomed to the functions of the system and really miss it when it turns off all the lights in the house during a dinner party. 

[Brian H]

Are you using the TM751s to cover areas with signal problems?
Is the CM15A sending an RF signal to them to control them? As a TM751 has no power line receiver in it and can not be controlled by the CM15As power line transmitter. Sometimes multiple TM751s on the same House Code can be a problem if more then one decides to send a power line signal at the sametime.

3.308 is not the latest AHP 3.318 is. AHP will not tell you there is a later version if you have it look for Updates. It is a bug that seems to never get corrected. Not that 3.318 would be your cure.

Maybe you have a neighbor that decided to mess with you and has a X10 Remote sending an All Off to A at 9:00 PM.

If it comes back. The suggestion to try a different House Code maybe a good way to start. A is the default and many don't bother to change it.

[narib]

I am using the TM 751s to cover signal problems and I did not know that it can not get a power line signal.   I am using two of them as insurance and they are both in the same room.  About 3 years ago I converted to LED lighting in the room and we really like the lighting improvement from an appearance standpoint.  The LED conversion gave lots of problems with the lamp modules so I found that the tm753 and the am466 could handle the LED and I "worked with it" and got it going.  This was not done with a lot of knowledge and trial and error is seldom the best method.

It is working and I am gaining energy to understand the system better and clean up the redundant features.  I would like to add other functions after I understand what I have running.

Thanks for your help

[dave w]

I am using two of them as insurance and they are both in the same room. 

Although this will not cause an "ALL OFF" to be sent at 9:00 pm, it can cause problems. If both TM751s are in the same room, they are likely on the same circuit and same power phase. The TM751s are lovingly called "impolite" transceivers because when they receive an RF command they immediately transceive it to the homes powerline without checking to see if there might be other X10 commands on the powerline at the same instant (as opposed to the RR501 and CM15A which will check and wait for a clear period before transceiving). So, being in the same room and likely on the same circuit and phase, the likelyhood of both transceiving at the same time seems high IMHO. When two commands are placed on the powerline at the same time, they corrupt each other and (typically) the command never gets to the target lamp, appliance, or CM15A module. Multiple TM751 are used successfully to ensure successful pick up of RF signals (i.e. a TM751 at each end of the house) or successful transceive of an RF command on both phases of the homes power. I would remove one of the TM751s. 

[dhouston]

I am using two of them as insurance and they are both in the same room. 

Although this will not cause an "ALL OFF" to be sent at 9:00 pm, it can cause problems. If both TM751s are in the same room, they are likely on the same circuit and same power phase.

I hate to contradict someone as helpful as dave w but multiple TM751s interfering with each other is an urban legend. I did extensive testing a few years back without being able to generate a collision. Years and years ago, the TM751 (if that's what the old brown ones were called) waited for a rising zero crossing to transmit and this guaranteed that those on separate phases would always collide. That hasn't been the case for years. RF travels at the speed of light so all transceivers will hear the signal within a few microseconds of each other and all now transmit to the powerline on the next zero crossing, irrespective of polarity, so its nearly impossible to generate a collision. You'll see more collisions from the supposedly polite transmitters which are guaranteed to collide if their numbers exceed the number of random slots available. Also, many of the supposedly polite systems used TW523s which delay reports for 11 powerline cycles, long after the collision took place.

Several years ago there was a thread in comp.home.automation where I calculated how many miles apart two transceivers would need to be to see a significant difference in RF reception time. Of course, in reality X10 RF seldom makes it past 30-40 feet let alone several hundred miles.

What does cause collisions is multiple motion detectors that see the same motion a few seconds apart. People, pets, varmints, etc. do not move at lightspeed so this type of collision is common.

[dave w]

I hate to contradict someone as helpful as dave w but multiple TM751s interfering with each other is an urban legend.

I pretty much consider it a compliment to be contradicted by a legend like you. However in view of the OPs description of two TM751s in the same room, (so I assume on same circuit and within a few feet of each other) I don't see what would stop collisions, unless a "simulcast" condition occurs (which actually makes sense)   

[dhouston]

Most RF remotes send 5-6 copies of the RF code. Each takes 108mS so the total is on the order of .5-.6 seconds. Transceivers will send to the powerline as soon as they receive a valid RF code. Standard PLC codes require about .4 seconds and the transceivers are deaf to RF while sending PLC. In nearly all cases the transceivers will simulcast to the powerline.

There is one condition that can cause collisions. If one transceiver sees a stronger RF signal that another, it may get a valid code sooner than the other which may need 2 or more copies of the RF to bias the AGC before seeing a strong enough signal to be valid. This is unlikely to happen with transceivers in the same room but its not impossible. It would likely happen more often with widely separated transceivers but the short range typical of X10 RF tends to prevent that.

[JeffVolp]

There is one condition that can cause collisions. If one transceiver sees a stronger RF signal that another, it may get a valid code sooner than the other which may need 2 or more copies of the RF to bias the AGC before seeing a strong enough signal to be valid. This is unlikely to happen with transceivers in the same room but its not impossible. It would likely happen more often with widely separated transceivers but the short range typical of X10 RF tends to prevent that.

I was going to respond to your prior post to point out this situation, which is what I have seen here.  I normally use a TM751 at the opposite end of the house to generate a weak signal for testing.  Communication with that unit is marginal with all the ductwork running down the center of the house.  If I put another TM751 on the same housecode in my office/lab, I will occasionally see a collision.  It is not frequent, but it does happen.

We had transceivers at both ends of our last house.  We used wireless remotes at both ends, but not in the center (kitchen/dining).  That configuration worked flawlessly, probably because only one transceiver picked up the signals.

Jeff

[dhouston]

I should have noted this earlier but all the recent trips to the hospital distracted me.

One on my web pages illustrates why even the type of collisions Jeff & I are talking about are infrequent. See...

• http://davehouston.org/rf-noise.htm

All of the standard* X10 RF codes begin with a pulse of ~8.8mS which serves as a marker, biases the AGC circuits and charges the data-slicer capacitor. This serves to both raise the signal level and raise the data-slicer threshold (which acts to reject weak signals). The result is that even marginal signals like the one shown on the referenced page (captured purely by accident) tend to result in a valid digital output, lessening the number of collisions.

That was captured using a third party superregenerative receiver but the internal superregenerative receiver used in the older TM751 operated similarly. I haven't thought through how the newer TM751s with superheterodyne receivers might handle similar signals

When I learned electronics in the late '50s, superregenerative receivers were considered long obsolete but still covered (very briefly) for historical completeness. 50 years later there were billions of these obsolete designs in use in applications like X10, garage door openers, ceiling fans, etc. It's amazing what one or two transistors, an op-amp and a few passives can do.

*The CR14A uses a shorter start pulse (and shorter codes).