Print

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

[bkenobi]

I have a Powerflash module installed with a motion sensor at the front of my garage.  The setup has worked fine over the last few months, but I'm noticing issues with the setup.  The sensor sees movement, triggers a relay, activates the reed switch on the PF which sends the "ON" command.  The sensor is in "Test" mode, so it does not stay on for any duration and works both day and night.  After ~5 seconds, the sensor goes off, releases the relay, releases the reed switch on the PF, and the PF sends "OFF".  For some reason, I'm not getting every "ON" or "OFF" command at the CM15A (nothing in the AHP log).  I know that the motion sensor is going on/off because I have a light bulb installed as an indicator inside the garage.  I have tested the continuity on the relay and it's correctly actuating and releasing.  I have tested with a different PF module, so I don't think that it's a PF issue, but that seems to be the best place to look.

I am using a custom code written in AutoHotKey to control this motion sensor via the SDK.  I have saved logs of script variable values as well as X10 commands sent/received.  I can see that the sensor seems to be working fine generally, but it does have issues once in a while.

So, what do I do now?  I don't have access to an X10 signal tester (which was my first thought).  I thought that with the Arduino and other proto boards out there, I could perhaps build something to interface with a laptop, but I don't know where to get started.  I see that there's a user here that has a unit he built, but for an assembled unit, it's a bit pricey.  Granted that it's a LOT less than the professionally built units, but still a bit steep if this is my only use for it (hopefully).

Any suggestions?  I'm open to any thoughts on how to fix this at this point.  I guess I could rip it out and go with an X10 sensor, but the whole point was to increase the range and angle over what X10 sells.

 

[Noam]

It sounds to me like a signal/noise issue on the circuit somewhere in your house.
It is possible that something in your home is creating signal-blocking noise, or might just be absorbing enough of the signal to prevent it from getting to the CM15A.
Have you plugged in anything new recently?
Have you installed or replaced any CFL or LED light bulbs?
Is the CM15A on the same phase as the PowerFlash, or are the on opposite phases?
Do you have any signal-bridge devices in your setup at all?

[bkenobi]

My fear is that the new electric furnace/heat pump is causing these issues.  I had that installed in December and that's around the time that I started NOTICING the issue.  I don't know if this has been going on for longer than that though.  I did just have a CFL burn out last night, but when I removed it, the issue is still present.

Is there any way to filter noise off a furnace?  The unit is on 220 and draws LOTS of power when starting the HP or running the electric strip heater.  Perhaps I should note that I have noticed a spike in power when the furnace turns on which flickers the lights slightly.  The installer didn't seem interested in hearing about it, so I sorta let it pass since it's a transient thing.

My initial thought was that I need to figure out a way to reliably fail the motion sensor so that I can test things.  Once I know where it's failing, I can know why and hopefully how to fix it.  I really didn't want to spend the $100-200 on a DIY signal tester or $300+ for a pro model.  I was hoping this would be a one time thing where I won't need it once I get things fixed.

[bkenobi]

Here's an example of the log of PF module actions:

Code: [Select]

1/16/12   07:16:19 RecvAction: RecvPlc, o3, On, , ,
1/16/12   07:17:02 RecvAction: RecvPlc, o3, Off, , ,
1/16/12   07:21:42 RecvAction: RecvPlc, o3, Off, , ,
1/16/12   13:11:38 RecvAction: RecvPlc, o3, On, , ,
1/16/12   13:12:07 RecvAction: RecvPlc, o3, Off, , ,
1/16/12   13:13:39 RecvAction: RecvPlc, o3, On, , ,
1/16/12   13:14:30 RecvAction: RecvPlc, o3, Off, , ,
1/16/12   13:16:32 RecvAction: RecvPlc, o3, On, , ,
1/16/12   13:17:05 RecvAction: RecvPlc, o3, Off, , ,
1/16/12   13:19:33 RecvAction: RecvPlc, o3, On, , ,
1/16/12   18:10:43 RecvAction: RecvPlc, o3, Off, , ,
1/16/12   18:18:57 RecvAction: RecvPlc, o3, On, , ,
1/16/12   18:19:50 RecvAction: RecvPlc, o3, Off, , ,
1/16/12   18:40:10 RecvAction: RecvPlc, o3, On, , ,
1/16/12   18:40:47 RecvAction: RecvPlc, o3, On, , ,
1/16/12   23:49:25 RecvAction: RecvPlc, o3, On, , ,
1/16/12   23:49:29 RecvAction: RecvPlc, o3, Off, , ,
1/17/12   01:02:01 RecvAction: RecvPlc, o3, On, , ,
1/17/12   01:02:04 RecvAction: RecvPlc, o3, Off, , ,
1/17/12   01:53:31 RecvAction: RecvPlc, o3, On, , ,
1/17/12   01:53:35 RecvAction: RecvPlc, o3, Off, , ,
1/17/12   03:20:21 RecvAction: RecvPlc, o3, On, , ,
1/17/12   03:20:24 RecvAction: RecvPlc, o3, Off, , ,
1/17/12   05:02:22 RecvAction: RecvPlc, o3, On, , ,

Based on how this is set up, every on MUST be followed by an off.  Any time there is a mismatched pair, there was an issue with the setup.  As such, my initial statement that it's working mostly okay is obviously WRONG!

[dave w]

Is there any way to filter noise off a furnace?  The unit is on 220 and draws LOTS of power

It may be the control board creating noise and not the high current HP or electric furnace. If so you might be able to filter the power to only that board with an inline filter. I had to do that on an old Lenox Pulse unit. Obviously voiding your warranty would be a prime consideration.

[bkenobi]

I'll have to figure out if this is an issue with the HP or the main unit (assuming that's the source).  I would first have to figure out when each is turned on.  Actually, I can say for sure that the HP is NOT the source since it's covered with several inches of snow today.  So, this is either because the furnace needs a filter, or there is an issue with how it's installed/set up.  I talked to the supplier yesterday and they mentioned that the installer could have set things up such that the furnace is "back powering" the HP.  I'm not sure what that means, but it sounds like a potential source of noise.

[Noam]

You mentioned that you know the PF is working, because you have a light in the same room, and it always works. I'm assuming it is on a lamp (or appliance) module, and set to the same code as the PF.
Have you tried moving that to the same location as the CM15A, and seeing what happens? If the signal is having trouble getting through, then I would think the lamp would have trouble if it is located in the same place as the CM15A (although different devices have different signal thresholds of what they can hear).

The blower motor might be capable of putting out lots of noise, too.

As a test, you might want to try turning off the furnace (at the breaker), and seeing if you can replicate the problem.
If you can't, then turn the furnace back on, and then test again. You'll probably need to test both with the system on (but idle), and also with it running.

The problem of a weak signal path from the PowerFlash to the CM15A might be remedied in other ways, as well.
If they are on separate phases, you might try moving one of them to another phase, and see if that helps. You might need to extend the wires from the motion sensor to the PF to be able to move it, but that is probably easier than re-wiring the 110V in the house (especially if you only do it temporarily as a test).
Without a signal bridge, the signals generally have to travel out to the pole transformer and back to get from one phase to the other. That can seriously reduce signal strength. There are a number of ways to bridge both phases in your home. If you have a electric clothes dryer, there is a simple-to-install plug-in signal bridge, that plugs into the 220V outlet, and has a pass-through for the dryer to plug into.
As a test, you can try turning on your electric dryer, and see if the problem goes away when the dryer is running (it acts as a signal bridge to cross the signals from one phase to the other).

I have Jeff Volp's XTB-IIR, which is certainly more expensive than the plug-in coupler. For me, however, it was worth the money.

Another possibility is that the new furnace/heat pump isn't the only cause, but rather a "straw that broke the camel's back" type of thing. You might have other noise-generating devices in your home (computer power supplies and cell-phone chargers are big ones), and the combined noise of those things AND the new furnace is just too much.
Since the problem seems to be intermittent (some signals get through, others don't), that might point to something that isn't always on (like a furnace that cycles on and off throughout the day).

You could try to do a "breaker-box test", where you selectively turn off one breaker at a time, and see if you can find the breaker that contains the noisy device(s).

[bkenobi]

I didn't do a very good job of answering the original questions about my setup.  I'll add on so that it is more clear.

I have a 4000 sqft home that I'm working with.  When I moved in, all of the X10 was done via TM751 units.  I got rid of those and installed a CM15A with an upgraded antenna (posted in another thread) and a XPCR phase coupler-repeater.  I am able to send/receive signals to/from the shop which is 150ft away from the house to control lights without issue.  I have several ActiveEye motion sensors that control a couple exterior lights and those seem to work fine (though the battery life is apparently not all that great when deer keep tripping them).

I also have a motion sensor set up for my garage/front door lights.  This is area I'm worried about right now (and the only one I'm currently noticing issues with).  The schematic is attached to show how I have things set up.  The motion sensor is a Zenith 5716 which is hardwired to power.  In order for this unit to work, it must be connected to a 40W+ resistive type light bulb, so I installed it inside the garage to act as an indicator that the sensor has been tripped.  I originally was going to swap it out for a resistor of appropriate size, but I couldn't figure out the right resistance and it's nice to know it's working anyway.  The output of the sensor is a switched 120VAC line.  Since the PF can't take that, I use an AC relay to trip a NO contact.  The motion sensor is in test mode so it will work all day/night and it releases quickly.  This allows me to use it with a chime to tell me if there's motion any time of day and turn lights on at night.

As can be seen in the log from my previous post, the setup seems to be losing commands.  I don't know how long this has been an issue, but I only noticed it at the end of December which corresponds to when I got my new furnace/HP installed.

Based on your suggestions, I think the best bet is to trip the motion sensor repeatedly with the furnace on to make sure I see missing signals.  If so, I'll turn off the furnace at the panel and repeat.  If I don't see problems, it could be the furnace on its own, or it could be other things AND the furnace.

For reference, I did find a power strip that caused an issue in one location (where I was going to use a lamp module with a xmas tree).  I did not remove that, so that's probably a better first step!  I'll review this thread and will also post results as I can get them figured out.

Thanks!

PS, I wonder how my wife would feel about turning the furnace off tonight...with a major snow storm coming through.   

[Brian H]

Yes a 40 watt light bulb is better than a 360 Ohm 40 watt resistor and easier to find.
May not be easy to tell, but when the On is missed does the chime also not sound.

[kenrad]

I didn't do a very good job of answering the original questions about my setup.  I'll add on so that it is more clear.

I have a 4000 sqft home that I'm working with.  When I moved in, all of the X10 was done via TM751 units.  I got rid of those and installed a CM15A with an upgraded antenna (posted in another thread) and a XPCR phase coupler-repeater.  I am able to send/receive signals to/from the shop which is 150ft away from the house to control lights without issue.  I have several ActiveEye motion sensors that control a couple exterior lights and those seem to work fine (though the battery life is apparently not all that great when deer keep tripping them).

I also have a motion sensor set up for my garage/front door lights.  This is area I'm worried about right now (and the only one I'm currently noticing issues with).  The schematic is attached to show how I have things set up.  The motion sensor is a Zenith 5716 which is hardwired to power.  In order for this unit to work, it must be connected to a 40W+ resistive type light bulb, so I installed it inside the garage to act as an indicator that the sensor has been tripped.  I originally was going to swap it out for a resistor of appropriate size, but I couldn't figure out the right resistance and it's nice to know it's working anyway.  The output of the sensor is a switched 120VAC line.  Since the PF can't take that, I use an AC relay to trip a NO contact.  The motion sensor is in test mode so it will work all day/night and it releases quickly.  This allows me to use it with a chime to tell me if there's motion any time of day and turn lights on at night.


With the size of your house I would definitely look into an xtbr I also have a big house like yours 5000+ sqft as well as a 4 car garage that I run on all x10.  I was running the XPCR and had huge issues with signal and noise issues.  I installed Jeff Vlops Xtbr and also the XTB-ANR which really helps with the noise issues and my system went from troubled to reliable over night.  If you have alot of noise issues the XTB-ANR will help with this issue extensively.

  Ken
As can be seen in the log from my previous post, the setup seems to be losing commands.  I don't know how long this has been an issue, but I only noticed it at the end of December which corresponds to when I got my new furnace/HP installed.

Based on your suggestions, I think the best bet is to trip the motion sensor repeatedly with the furnace on to make sure I see missing signals.  If so, I'll turn off the furnace at the panel and repeat.  If I don't see problems, it could be the furnace on its own, or it could be other things AND the furnace.

For reference, I did find a power strip that caused an issue in one location (where I was going to use a lamp module with a xmas tree).  I did not remove that, so that's probably a better first step!  I'll review this thread and will also post results as I can get them figured out.

Thanks!

PS, I wonder how my wife would feel about turning the furnace off tonight...with a major snow storm coming through.   

[bkenobi]

No, the chime does not work either.  I dug out a socket rocket and put it in the garage in a lamp.  I did not see any issues with the motion sensor with the testing this evening.

I then moved the PF module to the same socket I had issues with for the xmas tree lights.  Turns out that the surge protector was an issue in blocking signals.  But, interestingly, so is the laptop power supply.  I'm not sure how to deal with that one other than getting a few filters.  Looks like I need to go around and see what does and does not block signals around the house.  I wish I could rent an X10 signal analyzer, but I don't think anyone does installs locally.

[Noam]

Chances are the surge protector is a "signal sucker," and the power supply is a "noise generator."
Each interferes with your X10 signals in different ways, but the end result is the same - the device is not able to properly detect the signals on the line.

Yes, filters are probably in your future. I've had good luck with the SmartHome FilterLinc ones (I don't have any experience with the X10 ones, but I hear they can get warm and start to smoke if they are pushed close to their rated limits).

I keep telling myself that I'll buy an XTBM the next time I run into problems. From what I've read, it really is a very useful tool, and well-worth the money (from what others have said about it). I have an XTB-IIR, and I know that Jeff does very high-quality work, and puts a lot of thought into the products he creates.

For now, it sounds like you have at least *some* method for testing. Just like you found the problematic surge protector and power supply, you can test each device the same way, and see what causes a lot of noise and what doesn't. It isn't the same as having a meter to give you actual numbers, but it is much less expensive.

[bkenobi]

I think I may have found the problem based on my findings yesterday.  I have found that my UPS (APC BackUPS 725ES) which I have been using with my main desktop machine and the file server is dramatically underrated for what I've been using it for.  The battery is dead, but I think it's causing noise/signal sucker issues.  What makes this one so bad?  Well, it's in the outlet right next to the CM15A.  I'm going to have to figure something out, but my guess is I'll be getting a set of filters for at least those 2 locations.

[kenrad]

I bet that is a big issue.  I find that the UPS in my house not only created line noise but it also seemed to amplify the line noise from the PC.  In my testing if I had just the UPS plugged in it would cause intermittent issues but then when I connected the PC it completely crushed my x10 setup. I added a plug in filter and the xtb-anr I mentioned In my last post and I haven't had noise issues since    I would recommend giving that a try,

  Ken

[bkenobi]

I'm looking at picking up a 1500VA UPS from Fry's tomorrow (assuming the come back in stock).  I'll have to get a filter as well, but hopefully the new UPS will at least make things less bad?