Herding cats

[ryarber]

I'm coming to the conclusion that solving communication issues is like herding cats. I'm about to just buy about 30 filterlincs and put them on everything in my house.

 

I think my freezer is a problem. I already have filterlincs on my home entertainment centers, my refrigerator, my computer UPS, and maybe even a couple more of them. I didn't put one on the freezer before. When I unplug the freezer, there is a noticeable improvement. It is also right next to my breaker panels and that noise may be spreading throughout the whole house. I've also removed all my CFL bulbs.

 

What about these devices....?

 

laptop power supplies

aquarium filters

cell phone chargers

power tool battery chargers

central heat/air units

[Brian H]

Be careful with the filter you pick for the Freezer.

Take the compressors starting current into account.

The FilterLincs do have a fuse in them that is not easily replaced.

I would use the 10 amp FilterLinc at the very least. An ACT120 15 amp maybe needed for the freezer.

 

Laptop Power Supplies; Cell Phone Chargers and Power Tool Chargers if they are the newer switching type. Are known to be noise makers or signal suckers. Not all problem devices make noise. Some absorb the signals on the power line.

[oberkc]

Most people here feel at least some of your pain. In addition to TVs and the typical devices mentioned by BLH, I have had known problems with an Roland electric piano and an on-demand water heater (gas fired). My personal sense is that anything COULD be a problem, but nothing is ALWAYS a problem.

 

Like you, I have wondered about my central heating and cooling system. I think, too, that the problem is additive. One device by itself may not be a problem, but put several on a house power system, and you may have problems.

 

I DO use CFLs, however, but they are either not a problem, or not enough to notice. There are several posts on this topic, and it sounds like a lot of people believe it is dependent on brand. Some are worse than others.

 

I don't have a UPS, but understand these to be known problems. Even though yours is filtered, have you tried removing it to see if that was a contributor? Is your PLM powered by uninterrupted and filtered power or is is on a separate supply of power from your computers? If the latter, you should try separating your PLM from your computer system.

[ryarber]

I've tried my PLM in several different places. I'm about to move it for the 3rd time in as many weeks to see if it will help.

 

Is it possible that I have a bad PLM? My issues are that the PLM isn't hearing what is on the network. The commands being sent by the PLM are practically never failing. But the PLM is having a difficult time "hearing" what is going on over the power lines.

 

I just added 4 new devices tonight (I/OLinc's to control garage doors). I can control the garage doors just fine with the ISY. I have set up the I/OLinc's on scenes so that I can control them and monitor their states with a KPL. The scene commands are never failing to trigger the desired actions, but the PLM just doesn't know anything is happening. The sensor state isn't changing, even though the scene that the sensor is involved in is triggering fine and changing the state of the KPL button like it is supposed to. The KPL that is part of the scene is in a distant part of the house, so I know I have a good signal there.

 

How would I check to see if the PLM is the problem here? I'm going to try to move it one more time before I call smart home and have this puppy exchanged.

[ryarber]

BTW, one other thing I've tried is plugging an access point directly into the pass-through outlet on the PLM. Even this didn't help.

[dkaleita]

Are you guys talking about Insteon signals or X10 signals? I thought FilterLincs only helped X10 and not Insteon.

[ryarber]

I'm speaking of insteon.

 

My understanding is that insteon and X10 use similar frequencies so filters that work for one should work for the other. Smarthome markets them that way as well.

[brad77]

According to the Wikipedia page on INSTEON, the powerline signals operate at 131.65 kHz. X10 operates at 120 kHz. Apparently they are sufficiently close enough to each other that FilterLincs will work for both. Additionally, the AF120 advertises signal attenuation for all signals between 30 and 200 kHz.

 

Someone (I believe it's Brian H) had re-calibrated (re-tuned?) a FilterLinc to operate at 131.65 kHz and reported that the effects weren't significantly different than stock FilterLincs.

 

I use both. I can attest that they have made a big difference in the reliability of my INSTEON network. If you have signal problems, this is a very necessary troubleshooting step IMO.

[oberkc]

I use smarthome filterlinks exclusively, and believe they work well on my primarily insteon home.

[oberkc]

I use smarthome filterlinks exclusively, and believe they work well on my primarily insteon home.

[brad77]

You can say that again!

[IndyMike]

Hello ryarber,

 

There are situations where nearby noise sources could prevent the PLM from hearing transmissions while remote devices can hear the PLM transmission. It would, however, be extremely unlikely to have this condition in multiple locations.

 

How would I check to see if the PLM is the problem here? I'm going to try to move it one more time before I call smart home and have this puppy exchanged.

 

1) Check the links in your scene devices to make certain they programmed correctly. It's possible they do not have a link to the PLM (if multiple scenes are having this problem this is unlikely).

2) Power cycle the PLM (~ 1 minute off) and retry your scene. I think I can remember locking the receiver up at one point due to an "event" (memory isn't what it one was).

3) Secondary test: program a lamplinc/appliancelinc as a scene member and plug it into the PLM. If the LL/AL responds and the PLM does not, I'm leaning toward a bad receiver in the PLM.

 

If the LL works in test #3 (and the PLM doesn't register) we still can't pronounce the PLM "bad". We'll need to check with Michel on how to verify that the PLM has the correct links (lost that memory as well).

 

Side note: In general, I don't like plugging other devices into the PLM (your reference to plugging in the Accesspoint). Older PLM's had problems with running hot. Although the newer versions are greatly improved, I don't like the additional heat load of the added device.

 

IM

[oberkc]

If the LL works in test #3 (and the PLM doesn't register) we still can't pronounce the PLM "bad". We'll need to check with Michel on how to verify that the PLM has the correct links (lost that memory as well).

 

Does not one use the "show device links table" under diagnostics? Once done, use the "compare" function to show what the PLM/ISY thinks that device should have. If any show up in error, it will be so noted.

[ryarber]

If the devices were first put into the ISY and then the scenes created from the ISY, shouldn't the PLM be linked?

[oberkc]

If the devices were first put into the ISY and then the scenes created from the ISY, shouldn't the PLM be linked?

 

If the links were established in times of communication difficulties, it is possible that errors may have occured. I think, too, that extra or missing links on a device can include links beyond the PLM and cause those flashing lights on your keypads. It is definitely worth checking out. I had a few of my own not too long ago. They can definitely happen.

[IndyMike]

If the LL works in test #3 (and the PLM doesn't register) we still can't pronounce the PLM "bad". We'll need to check with Michel on how to verify that the PLM has the correct links (lost that memory as well).

 

Does not one use the "show device links table" under diagnostics? Once done, use the "compare" function to show what the PLM/ISY thinks that device should have. If any show up in error, it will be so noted.

 

Hi oberkc,

 

The "show device links table"/"compare" function is what I was indicating for test #1 (I stated things rather poorly). The compare function checks the device against the ISY.

 

For #3, I was alluding to missing links in the PLM itself. I don't believe there is any way to "compare" the plm against the ISY. Inspection of PLM links would be manual (the PLM would be listed as a group responder to the scene).

 

IM

[IndyMike]

If the devices were first put into the ISY and then the scenes created from the ISY, shouldn't the PLM be linked?

 

Hello ryarber,

 

The operative word is "should". As oberkc indicated, communications problems can lead to missing/incorrect links in devices. These can be detected with the "show device links table" followed by a "compare" (compares links in the device to the ISY).

 

The PLM link problem, that I was attempting to describe, is far less common. I've created this problem when I was doing mean things to my system (stress testing) and basically crashed the PLM. Recovery would presumably be through deleting/re-adding the scene.

 

IM

[ryarber]

Here is what I have done since last update:

 

1. Moved ISY again

2. Put filterlinc on my freezer

3. I had ge surge supressors in each of my breaker boxes. I've removed those today

4. Checked the links on a couple devices and checked them against what ISY thinks they should be

5. Added 4 new devices to control garage doors (I/OLincs)

6. Created the debug scene with lamplinc as you suggested.

 

I probably shouldn't have added new devices until I get all this worked out.

 

The links match up fine with ISY.

 

When I first set up lamplinc scene, it worked fine every time. The switch controlling the scene and the PLM/Lamplinc were all in the same room on the same circuit. It was all working fine in that the scene triggered correctly every time and the ISY always knew the correct status of the lamplinc. I left it plugged up that way overnight, then yesterday when I log onto admin console, it tells me there was a communication failure with the lamplinc. After a query, the scene works perfectly again. This is the sort of intermittent behavior I'm seeing. The lamplinc is plugged into the PLM and there shouldn't be any problem communicating with it.

 

in the meantime, I have set up my garage door controllers. I set them up pretty much just like you describe in the wiki. The PLM knows when the sensor is on, but never hears the "off" signal even though the scene that turns my KPL lights off when the doors close works every time. So the PLM always hears sensor on, but not sensor off. The scenes I have set up with the KPLs hear them correctly every time. Come to think of it, those are the commands I'm having problems with the PLM hearing almost every time. It is always either an off or a fast off that the PLM isn't hearing. The "on" commands seem to be heard more often than the "off" or "fast off."

 

Now every source of noise on my powerline should have a filterlinc except a few flourescent light fixtures that I cannot do anything about (but those stay turned off almost all the time anyway). My whole home surge supressors are taken out, my device lincs match fine.

 

I'm gonna be doing more testing in the next few days. It is frustrating because my problems seem to only appear at times and then things get better. So I'll have to keep observing for a while to see if things continue to act better.

 

Next, I'm going to try to factory reset my I/O Lincs and readd them to the ISY/PLM and see if I can get this sorted out. If that works, I may try to do it on some of the newer devices I've added and see if that will fix my issues that I'm having with them as well. I also may try to move my access points again and see if I can get some better signals.

 

I think I need to hire a consultant

[Michel Kohanim]

Hi ryarber,

 

I have one question and one statement:

Question:

 

Do you have any SwitchLinc v35s in your network? If so, is it the case that they exhibit most of the communication errors and, specifically, not responding correctly to scenes?

 

Statement:

 

IOLincs do not ever send their off state (which I cannot fathom why). Furthermore, on IOLincs versions 31 and below, IOLinc always returns the status of the relay for the status of the sensor. Please ensure that yours is not v31.

 

With kind regards,

Michel

[ryarber]

I have looked at the version number in the ISY for the SWLs and KPLs. None of them say v.35. Most of them have a number and a letter eg 2D or something like that. My IOLincs are brand new. I don't remember the version number but I saw the issues with v.31 on the forums and checked it out. Mine was not that version or earlier.

[IndyMike]

Here is what I have done since last update:

When I first set up lamplinc scene, it worked fine every time. The switch controlling the scene and the PLM/Lamplinc were all in the same room on the same circuit. It was all working fine in that the scene triggered correctly every time and the ISY always knew the correct status of the lamplinc. I left it plugged up that way overnight, then yesterday when I log onto admin console, it tells me there was a communication failure with the lamplinc. After a query, the scene works perfectly again. This is the sort of intermittent behavior I'm seeing. The lamplinc is plugged into the PLM and there shouldn't be any problem communicating with it.

 

The communication failure was likely generated during the ISY system query (3:00 AM query program). Did you have a load connected to the Lamplinc? Regardless, it sounds like you've eliminated the PLM as a problem (it is hearing the devices).

 

It's a bit curious that you are able to reliably program devices, but have problems at different times with scenes. As you noted, you may have an intermittent noise source that is kicking in and causing problems (HVAC?).

 

The ISY scene test would be a good tool for investigating weak areas since it uses a limited number of hop counts. You would need to wait for a "problem time" and then being running the scene test to confirm communication faults. Then start opening breakers until the scene test recovers (not a very "family friendly" procedure).

 

Next, I'm going to try to factory reset my I/O Lincs and readd them to the ISY/PLM and see if I can get this sorted out. If that works, I may try to do it on some of the newer devices I've added and see if that will fix my issues that I'm having with them as well. I also may try to move my access points again and see if I can get some better signals.

From your previous description, the scene was initiated by the KPL (scene controller). If that is correct, I don't believe that resetting the I/O Lincs will help. The PLM/ISY should be responding to the KPL group command.

 

Another question - is the KPL in question on the opposite phase? If so, and you are using Accesspoints, it's possible that one or more AP's are located in a "noisy area" and are not coupling the phases properly.

[ryarber]

The communication failure was likely generated during the ISY system query (3:00 AM query program). Did you have a load connected to the Lamplinc? Regardless, it sounds like you've eliminated the PLM as a problem (it is hearing the devices).

 

No, I was just looking at the LED and querying the device after the scene change.

 

It's a bit curious that you are able to reliably program devices, but have problems at different times with scenes. As you noted, you may have an intermittent noise source that is kicking in and causing problems (HVAC?).

 

Is there a filter you can put on the HVAC to see if that will help? I have 4 units (20 tons) on the house. Curiously, it seems that it seems to be worse at a time when the units shouldn't be running. eg. after the ISY is supposed to have turned the units down for the night (after the kids are in bed).

 

The ISY scene test would be a good tool for investigating weak areas since it uses a limited number of hop counts. You would need to wait for a "problem time" and then being running the scene test to confirm communication faults. Then start opening breakers until the scene test recovers (not a very "family friendly" procedure).

 

It is difficult for the reasons you have mentioned. It is not very family friendly to troubleshoot this. And it often happens when the kids are in bed. My job also cuts down on my time at home to work on this and often it isn't very convenient for me to be in there flipping breakers to get this ironed out. That is why it is taking me so long to work on it.

 

From your previous description, the scene was initiated by the KPL (scene controller). If that is correct, I don't believe that resetting the I/O Lincs will help. The PLM/ISY should be responding to the KPL group command.

 

Actually, the scene that is being initiated from the KPL seems to be heard just fine by the ISY (ie. the scene controlling the relay). It is the return scene (sensor scene) that doesn't seem to be heard by the ISY even though it is heard by the KPLs. That the sensor is on the same breaker box, but a different circuit. When I set the scene up and I was installing the I/OLincs, I know the ISY/PLM was hearing the on/off command just fine because I was watching it turn on and off on my iPhone while I was figuring out how to hook up the wires to the normally open/normally closed contacts. That was before I moved the ISY again.

 

Another question - is the KPL in question on the opposite phase? If so, and you are using Accesspoints, it's possible that one or more AP's are located in a "noisy area" and are not coupling the phases properly.

 

The last time I moved the ISY, I was careful to put it on the same breaker box that my other problem devices are on so I don't think that is a problem. I have 4 access points (it is a big house). Today I'm going to take them out, do some testing, and try putting them back in at different points than I have them now. I'll try putting them closer to my problem devices.

 

I have a light schedule today so I should be able to get a lot done today. Just trying to formulate a gameplan before I get back home.

[ryarber]

my I/O Lincs are version 36

[brad77]

Is there a filter you can put on the HVAC to see if that will help? I have 4 units (20 tons) on the house.

I use the AF120 15-amp filter on my furnace. I would imagine that you could do the same without blowing it out assuming that your units run on a standard 110-volt outlet hooked up to a 15-amp breaker.

 

I'm no HVAC specialist, however, so I'm having a hard time picturing what a setup such as yours should look like. Take this for the grain of salt that it's worth, but I'd filter those units pronto. I can image that four A/C units could generate plenty of noise.

[IndyMike]

Is there a filter you can put on the HVAC to see if that will help? I have 4 units (20 tons) on the house. Curiously, it seems that it seems to be worse at a time when the units shouldn't be running. eg. after the ISY is supposed to have turned the units down for the night (after the kids are in bed).

 

4 heat pumps? That is a good sized house. Fortunately, the motors in the heat pumps/condensers are not typically a problem. Variable speed blower motors (variable speed fans) for the evaporator can be a problem depending on how they are implemented.

 

It's very interesting that you see degraded performance when the temperature is turned down. More on this below.

 

It is difficult for the reasons you have mentioned. It is not very family friendly to troubleshoot this. And it often happens when the kids are in bed. My job also cuts down on my time at home to work on this and often it isn't very convenient for me to be in there flipping breakers to get this ironed out. That is why it is taking me so long to work on it.

 

Understand completely. My family takes a dim view of "service" interruptions as well. When power actually does go down (rare for this area) the first words out of my mouth are "it wasn't me".

 

The last time I moved the ISY, I was careful to put it on the same breaker box that my other problem devices are on so I don't think that is a problem. I have 4 access points (it is a big house). Today I'm going to take them out, do some testing, and try putting them back in at different points than I have them now. I'll try putting them closer to my problem devices.

 

Be advised that sub-panels have two phases as well. Said differently, all of your devices could be fed from the same sub-panel and still be on different phases.

 

I beginning to suspect that your AP's are not adequately coupling the phases. If your system indeed functions better when one of your heat pumps are running, it may indicate that the device is performing the phase coupling while on. If you have a 220V stove or dryer, you can test this by checking communications when either is on (they will resistively couple the phases). This may not tell the whole story as you may have long wire runs to these devices that will attenuate the signal.