Ceiling fan signal noise

[BCreekDave]

Been seeing issues with programs not running at all or partially. An example is  a program the turns outside lights off at sunrise. This morning while in bed I checked on Mobilelinc and sure enough the lights were still on. Tried manually running the then statement in the program and no luck. Tried several times. Then noticed the ceiling fan running.  I remembered that motors can cause noise on the power line.  Shut it off and tried the then statement again it ran perfectly. Lights went off. Now the question. How to filter the noise on hardwired fans?  The outside lights are controlled by 2466SW togglelincs so no RF, but have been rock solid for years and still work fine when directly controlled through the admin program or Mobilinc or at the switch so I think they are ok. It’s just the ISY 994 programs. I did have to replace the PLM a couple of months ago. One other  thing. The ceiling fan has a Hunter remote control setup installed. In the ceiling mount. Is that a known cause of noise?  

[paulbates]

Insteon offers 2 wired-in filters for sale on their site to combat noise making devices. Hopefully one fits in the fan's j-box. I don't know which one to recommend. Maybe ask on the insteon forums

Also, X-10 sold the xpf filter which I think still can be found. I've used it effectively on noisy transformers... But it can be a little bit of work to get it in a j-box

Edited June 15 by paulbates

[BCreekDave]

Thanks Paul.  I may have jumped the gun a bit.  Doing some more in-depth analysis on this at the admin console by running the then statements on the off-at-sunrise and on-at sunset programs and it still isn't consistent, even with the ceiling fan off.  Going through the breakers one at a time trying to isolate the noise, if it is in fact a noise issue after all.  Thought I had it narrowed down to bedroom 2 by flipping the breaker off, but on about the 20th cycle if the then statements it reappeared that one of the outside lights didn't respond correctly.  Very tedious and I am getting my walk time in going back and forth to the switch panel!  

[paulbates]

Yep. Years ago when my kids were still at home, they were recruited, or volun-told, with a walkie talkie 🤪😂

[PinchRoller]

On 6/15/2024 at 5:55 AM, BCreekDave said:

Now the question. How to filter the noise on hardwired fans?  

Paul's X10 xpf filter is a solid choice.  Here is a smaller one at Mouser that should be fine:

https://www.mouser.com/ProductDetail/Qualtek/851-05-006?qs=258DH3jk4VfO5HUmapCcvQ%3D%3D&mgh=1&utm_id=17222215321&gad_source=1&gclid=CjwKCAjwp4m0BhBAEiwAsdc4aMRi42v0n0pJAH0YhzNfTO5VCuYk5RSiNsiC8KYkMLC97IvEsaQxMxoCyfwQAvD_BwE

Pinch

[Brian H]

4 hours ago, PinchRoller said:

Paul's X10 xpf filter is a solid choice.  Here is a smaller one at Mouser that should be fine:

https://www.mouser.com/ProductDetail/Qualtek/851-05-006?qs=258DH3jk4VfO5HUmapCcvQ%3D%3D&mgh=1&utm_id=17222215321&gad_source=1&gclid=CjwKCAjwp4m0BhBAEiwAsdc4aMRi42v0n0pJAH0YhzNfTO5VCuYk5RSiNsiC8KYkMLC97IvEsaQxMxoCyfwQAvD_BwE

Pinch

One thing you have to take into account.

That filter will also absorb X10 and Insteon power line signals. So it could effect other modules on the same power feed.

 

[BCreekDave]

Turns out like most of these issues the ceiling fan was a red herring. The true fix was having to move the 994 and PLM to another location due to noise in the location that had worked previously for 10+ years. It is now physically a foot away from the electrical panels in the garage and is rock solid in reliability and most all hops are now 3-3. A few 3-2’s but still very reliable. I wish I could get it to work where it was (network wiring closet, about 10 feet from the panels) but there are so many devices in there tracing down the noise maker may be impossible. I had most everything there, except for the PLM and a ezio 2x4, plugged into a UPS and it was plugged into a filterlinc. Do those ever go bad and start letting noise through?

[Brian H]

The PLM and ezio 2x4 where connected to a UPS outlet?

The AC power line conditioner in most UPS units Will remove the Insteon power line commands as noise. In that case the PLM was using Insteon RF to communicate with the system. The exio 2x4 also on an UPS outlet may have been controlled by the PLM power line commands if both where in an UPS outlet.

I have seen a few Filterlinc issues reported here. Most where poor solder joints on the connections or dead if the internal fuse was blown. Though soldered to the PC board you can replace them. Hint the 5A ones looked the same to me except the rating of the fuse and some have used 10A replacement in the 5A ones with success.

[BCreekDave]

6 hours ago, Brian H said:

The PLM and ezio 2x4 where connected to a UPS outlet?

The AC power line conditioner in most UPS units Will remove the Insteon power line commands as noise. In that case the PLM was using Insteon RF to communicate with the system. The exio 2x4 also on an UPS outlet may have been controlled by the PLM power line commands if both where in an UPS outlet.

I have seen a few Filterlinc issues reported here. Most where poor solder joints on the connections or dead if the internal fuse was blown. Though soldered to the PC board you can replace them. Hint the 5A ones looked the same to me except the rating of the fuse and some have used 10A replacement in the 5A ones with success.

I may have not been very clear in my explanation.  The PLM and the EZIO were plugged in directly to the wall outlet.  Everything else were plugged into the UPS which was plugged into the wall through the filterlinc (filtered).  This setup worked well for 10 years.  My original (recapped) PLM failed about 3 months ago (all on events, programs not running) so I replaced it and at first it worked well.  Then after a week or two programs started failing until eventually they wouldn't run at all.  This is when I thought it may be noise like from the ceiling fan.  I went through the breaker box breaker by breaker trying to isolate the noise with no luck.  I never shut off the power to the wiring closet as that would have killed the PLM too.  Then I moved the PLM and ISY 994 to another outlet and Bam, it instantly just worked.  So the noise (or signal sucker) is coming from something in the wiring closet but nothing has changed in there in over a year (added a fiber optic ONT).

I will add that my test method was to run the "then" part of the If-Then-Else statements in programs and identified certain programs-devices that were especially trouble prone.  Many times the program would run and the progress bar would appear and cycle through several times and eventually fail with a "couldn't connect to device etc" and then the device would have an exclamation mark next to it.  If I then did a query on the device it would sometimes clear it and sometimes stay with the exclamation mark.  Hops would vary from no hops (3-3) to 3-0. This is I think a classic example of noise.  Now when I do the same thing there is no progress bar and it instantly responds.

The last thing to add is that the new PLM, V2.7, seems to have very poor RF performance.  When trying to link to motion sensors it just would not connect, even with the motion sensor laying directly on top of it.  I then took the same motion sensor and placed it next to an 2443 access point and it connects perfectly.

I guess I am replying back to add to the forum info database as noise seems to be a big issue and to ask if the filterlincs ever fail and start letting noise through.  It looks like no to that.

Edited July 14 by BCreekDave

[IndyMike]

On 7/14/2024 at 10:03 AM, BCreekDave said:

I may have not been very clear in my explanation.  The PLM and the EZIO were plugged in directly to the wall outlet.  Everything else were plugged into the UPS which was plugged into the wall through the filterlinc (filtered).  This setup worked well for 10 years.  My original (recapped) PLM failed about 3 months ago (all on events, programs not running) so I replaced it and at first it worked well.  Then after a week or two programs started failing until eventually they wouldn't run at all.  This is when I thought it may be noise like from the ceiling fan.  I went through the breaker box breaker by breaker trying to isolate the noise with no luck.  I never shut off the power to the wiring closet as that would have killed the PLM too.  Then I moved the PLM and ISY 994 to another outlet and Bam, it instantly just worked.  So the noise (or signal sucker) is coming from something in the wiring closet but nothing has changed in there in over a year (added a fiber optic ONT).

Tracking down communication issues using the breaker method was far easier in the X10 days.  With the repeating capability + RF of Insteon devices, there is normally not a single offending device.  Instead it's a combination of signal absorbers + noise that causes issues. 

One rule that normally applies: If you are having problems with a few devices - inspect the circuit.  If you have having problems with many devices, inspect the PLM circuit or the PLM itself.

Moving that PLM to the breaker panel is optimal.  The panel will normally be the lowest noise point with best electrical access to all points of your system.  A RF coupling device installed on the opposite phase near the panel is also highly advisable (or passive coupler).

On 7/14/2024 at 10:03 AM, BCreekDave said:

I will add that my test method was to run the "then" part of the If-Then-Else statements in programs and identified certain programs-devices that were especially trouble prone.  Many times the program would run and the progress bar would appear and cycle through several times and eventually fail with a "couldn't connect to device etc" and then the device would have an exclamation mark next to it.  If I then did a query on the device it would sometimes clear it and sometimes stay with the exclamation mark.  Hops would vary from no hops (3-3) to 3-0. This is I think a classic example of noise.  Now when I do the same thing there is no progress bar and it instantly responds.

Using a program to test communication is valid as long as you are not turning on/off scenes (which you obviously are not doing).  The ISY does not test the status of devices in a scene on/off command. 

You can use a program to query a scene.  You can also query scenes from the Admin Console by right clicking the scene.

 

On 7/14/2024 at 10:03 AM, BCreekDave said:

The last thing to add is that the new PLM, V2.7, seems to have very poor RF performance.  When trying to link to motion sensors it just would not connect, even with the motion sensor laying directly on top of it.  I then took the same motion sensor and placed it next to an 2443 access point and it connects perfectly.

I guess I am replying back to add to the forum info database as noise seems to be a big issue and to ask if the filterlincs ever fail and start letting noise through.  It looks like no to that.

Your poor RF performance is interesting.  Have not seen that reported previously.  Is it possible that you had a 900Mhz interference device nearby?

If this is true, your problems may have stemmed from the fact that you were not coupling the phases due to poor RF performance. 

You could retry linking a RF device with the PLM near the panel to confirm.  If the poor RF persists (and there are no obvious interference sources) I would consider replacing (returning if possible).  There are workarounds, but you really do want your PLM to communicate/couple via RF.

[BCreekDave]

Thanks Indymike for the tips and comments. I’ll do some more testing in the near future, but for now after spending the last few months getting reliability back into it I think I will just bask in the glow of success for awhile. 

[IndyMike]

Months of troubleshooting would wear anyone down. 

Hope that glow of success lasts you a good long time.