4 Tap Test in large house

[DAlter01]

I have a large house that is a bit over 150' from end to end with a large install of about 140 devices.  In doing my 4 tap test I find that if I set up a LampLinc in the center of the house in test mode I can test good to within about 30' of both ends of the house.  However, no matter how many additional LampLincs I put in to extend the signal I can't get the 4 tap test to reach the far ends of the house.

 

If I stop that test and then set up the 4 tap test at either end of the house, each end tests fine and will test good to about the middle of the house but will not reach past approximately the middle of the house.  Similar to the center test from above no matter how many additional LampLincs I put in I can not get the test signal to extend past the middle of the house.  

 

Referencing back to my first test in the center of the house, that center section tested good to about 30' from each end so I don't understand why, when I test from the ends, I can't get the test signal to extend through that center section of the house.

 

It would seem that the 4 tap test emits a signal from just that beacon device which is set up to initiate the test and that the signal isn't repeated throughout the mesh network.  Am I interpreting this correctly?  In other words, it would seem that each beacon device set up to do a 4 tap test is a test from that device to see which devices are seeing that device?  

 

I would have thought that the 4 tap test set up each device to repeat the signal throughout the network so that one could find non-responding devices to determine where weak spots are within the mesh and not just weakspots from the beacon device set up to initiate the 4 tap test.  

 

Or, possibly, is there just a repeat limit on the 4 tap test so that once I've hopped the RF signal 5 or 10 times it doesn't continue the test?

 

Oh great minds of Insteon please shed some insight as to what the test is trying to tell me and what part of this I am misunderstanding.

[stusviews]

Dual-band Insteon device have two purposes. One objective is to ensure that the opposite legs of the split, single-phase electric supply are adequately bridged for the Insteon signal. If you can get any one pair of Inseon dual-band devices to indicate that (as shown byt the 4-tap test), then you've accomplished the first aim.

 

The other goal is the increase the range and reliability of the Insteon network. If you can get a distant device to communicate at all (via RF), either with a device on the opposite leg (green) or same leg (red) and the second device to communicate with a device closer the middle, etc., then you've accomplished the second aim.

 

In short, there's no requirement for every Insteon device to respond to the 4-tap test from one single location. If every device, scene, and program functions, then there's no need for concern.

 

BTW, even if the 4-tap test fails, the main means of Insteon signal propagation is the powerline.

[Teken]

I would also add this is why its best practice is to have a dual band device on all four corners of the home, level, floor, zone, of the home.

 

Doing so will ensure each area is within RF to power line reception.

 

I am a true fan of the plugin dual band relay module as it allows a person to disable RF vs Power Line. Having this ability allows more capability in testing a site location - It also offers one of the longest RF ranges when compared to other Insteon products. Other products which offer similar RF (Stated) distance is the dual relay outlet, range extender, and HUB II controller. 

 

Lastly, the dual band lamp linc is the only device which continues to emit the beacon test with out timing out. 

[DAlter01]

So, to make sure I and any other readers of this topic understand, in additional to confirming that an installation is properly bridged, the 4 tap test is a test from the device I'm putting into test mode to see the RF coverage "from that point".  It isn't really a test of the RF coverage of the entire mesh network.  To test the RF coverage of a large house and their entire mesh network it may be necessary to test from multiple points?

[Teken]

So, to make sure I and any other readers of this topic understand, in additional to confirming that an installation is properly bridged, the 4 tap test is a test from the device I'm putting into test mode to see the RF coverage "from that point".  It isn't really a test of the RF coverage of the entire mesh network.  To test the RF coverage of a large house and their entire mesh network it may be necessary to test from multiple points?

 

Yes, from the initiator device to another point within the home where there is a Insteon device within RF reception. The RF beacon test does not cause other Insteon devices to broadcast out to (repeat) form the mesh network. As you probably saw first hand devices which are installed in a metal junction box's offer limited RF reception.

 

This is why having the signal travel through the power line is such a great benefit to supplement the RF portion.   

 

Lastly, all the coupling / bridging in the world will not supersede the need to identify noise makers / signal suckers in the home and have them removed, replaced, or filtered out.

[DAlter01]

Thank you, that provides a lot of clarity for how the 4 tap test is supposed to function.  That being said, I have a lot of extra LampLincs for sale now that they are not necessary.  I was trying to get the network to extend via RF with the LampLincs thinking I had a dead zone.  In the meantime I have the strongest RF network one has probably ever seen.

 

Teken you have mentioned in multiple posts about how the RF will not overcome issues related to noisemakers and signal suckers.  And, consistent with this fact, I do have a couple of problem areas in my house and, as you suggest, my robust RF network still does not provide reliable performance in those areas.  I've attempted to do some trial and error to find these powerline signal issues.  However, this house has one main panel, and three subpanels and they are mostly full.  As such, there are roughly 110 breakers controlling hundreds of potential powerline noise issues.  Trials and error could literally involve hundreds of hours of effort with no guaranty of success as some of those noisemakers will be redundant (while I might turn off a noisemaker if another device is also making noise I will not notice the signal improvement).  

 

I've got to believe there is a methodical approach to identify these noisemakers/signal suckers.  I've played around with using my old HUB and Houselinc diagnostics to identify changes in the reliability of my 10 or so problem devices as I go through and turn off different circuit breakers.  However I don't see any fruits from that labor thus far.  Is there another method out there that can be reliably be used to identify the noisemakers/signal suckers?

[Teken]

Thank you, that provides a lot of clarity for how the 4 tap test is supposed to function.  That being said, I have a lot of extra LampLincs for sale now that they are not necessary.  I was trying to get the network to extend via RF with the LampLincs thinking I had a dead zone.  In the meantime I have the strongest RF network one has probably ever seen.

 

Teken you have mentioned in multiple posts about how the RF will not overcome issues related to noisemakers and signal suckers.  And, consistent with this fact, I do have a couple of problem areas in my house and, as you suggest, my robust RF network still does not provide reliable performance in those areas.  I've attempted to do some trial and error to find these powerline signal issues.  However, this house has one main panel, and three subpanels and they are mostly full.  As such, there are roughly 110 breakers controlling hundreds of potential powerline noise issues.  Trials and error could literally involve hundreds of hours of effort with no guaranty of success as some of those noisemakers will be redundant (while I might turn off a noisemaker if another device is also making noise I will not notice the signal improvement).  

 

I've got to believe there is a methodical approach to identify these noisemakers/signal suckers.  I've played around with using my old HUB and Houselinc diagnostics to identify changes in the reliability of my 10 or so problem devices as I go through and turn off different circuit breakers.  However I don't see any fruits from that labor thus far.  Is there another method out there that can be reliably be used to identify the noisemakers/signal suckers?

 

There is only one true method and that is to take the time over a course of days, weeks, to unplug every device you can reach. Anything that can't be unplugged you simply turn off the breaker as a quick test. The most common inconvenient appliances are fridge, freezer, stove, HVAC, etc.

 

Investing the time and effort to unplug all the items in your home and perform a *Scene Test*  along with watching the level 3 logs is the most factual thing you can do in terms of trouble shooting. Anytime I've had to do a large install I have simply broken the home into their respective zones.

 

Bedrooms, living room, kitchen, basement, etc . . .

 

One major thing I always do is remove the load from any fixture unless its a incandescent bulb.

 

Once everything has been unplugged or turned off at the breaker (large appliances noted above) I initiate a scene test and watch the actual room and the level 3 log. If everything is good to go - I simply add one single device that was unplugged and watch for the end results.

 

As tedious as this is its fool proof and needs only be done once . . .

 

It should be noted multiple scene tests and activations are done repeatedly to ensure it wasn't a *One time lucky* event. Doing this will 95% of the time result in locating and identifying that noise maker / signal sucker. The real problem comes down to what do to with something that is really expansive or that can't be replaced / removed.

 

I simply leave it up to the owner to decide but at least they know said device is the root cause(s) for a under performing Insteon network. The reality is a person isn't going to replace a furnace, HVAC, stove, fridge, if its found to be a problem device. There are various inline filters for smaller appliances like fridges, freezers, etc.

 

But there isn't a plugin filter for a stove  . . .

 

Lastly, I would encourage you to deploy those extra dual band lamp lincs through out the home as I noted up above. The only negative of doing so is the vampire draw these devices consume over time. If you started off with older Insteon products the average energy consumption for these devices were 1.00 watt.

 

With 100 devices in a home that is 100 watts you simply can not do anything about if they were hardwired . . . Those with 200 plus Insteon devices are consuming 200 watts 24.7.365 of standby power. Those living in expansive States / Provinces with high KWH rates will be penalized heavily each month! 

[DAlter01]

I have put the Lamplincs on all four corners as you advise and a few in the middle where there are "light" spots for the number of Insteon devices.  Those interior ones probably are not necessary but assuming they are not hurting, another 5 watts isn't going to be a budget changer.

 

I hadn't used Scene Test until you mentioned it in your previous post.   It appears to just give a success fail outcome.  I don't see a lot of diagnostic detail in my log.  Here it is from one of my larger scenes.  

 

Mon 02/27/2017 09:37:28 AM : [GRP-RX      ] 02 61 28 13 00 06 

Mon 02/27/2017 09:37:28 AM : [iNST-SRX    ] 02 50 24.86.09 43.A9.2C 65 13 28    LTOFFRR(28)

Mon 02/27/2017 09:37:28 AM : [std-Cleanup Ack] 24.86.09-->ISY/PLM Group=0, Max Hops=1, Hops Left=1

Mon 02/27/2017 09:37:28 AM : [iNST-SRX    ] 02 50 24.89.C4 43.A9.2C 65 13 28    LTOFFRR(28)

Mon 02/27/2017 09:37:28 AM : [std-Cleanup Ack] 24.89.C4-->ISY/PLM Group=0, Max Hops=1, Hops Left=1

<html><font color="red">----- All On-Off - Inside Test Results -----</font></html>

<html><font color="red">[Failed]</font> Main 6-Inch-Vanity - Master B (2B CA F3 1)</html>

<html><font color="red">[Failed]</font> Under Cabinet - Laundry (29 3 CA 1)</html>

<html><font color="red">[Failed]</font> Bar Under Counter Accent Ligh (28 ED 90 1)</html>

<html><font color="red">[Failed]</font> Inside All - Laundry 6 Butt (2A 17 78 6)</html>

<html><font color="red">[Failed]</font> Kichen Sink - Kitchen (25 57 2E 1)</html>

<html><font color="red">[Failed]</font> Inside Lights - Davids Pendan (44 2C C5 6)</html>

<html><font color="red">[Failed]</font> Dining Room Accent (24 71 A3 1)</html>

<html><font color="red">[Failed]</font> 6 Butt TV Room Main (2A 14 DD 1)</html>

<html><font color="red">[Failed]</font> Master Bed Fan Light - Collee (28 F1 B5 1)</html>

<html><font color="red">[Failed]</font> Overhead All Lights-Entry - M (25 39 E1 1)</html>

<html><font color="red">[Failed]</font> 6 Butt Pendant - Davids - Mas (44 2C C5 1)</html>

<html><font color="red">[Failed]</font> Fan 6 Butt - Master Lights- A (2A 17 DA 1)</html>

<html><font color="red">[Failed]</font> West Hallway - Kitchen (24 97 29 1)</html>

<html><font color="red">[Failed]</font> East Hallway - Master (25 51 7F 1)</html>

<html><font color="red">[Failed]</font> Main 6 Inch-Entry - Master Ba (2B C7 DD 1)</html>

<html><font color="red">[Failed]</font> Inside All - TV Room 6 Butt (2A 14 DD 6)</html>

<html><font color="red">[Failed]</font> Evening - TV Room 6 Butt (2A 14 DD 4)</html>

<html><font color="red">[Failed]</font> East Hallway LD - Fitness Roo (25 47 56 1)</html>

<html><font color="red">[Failed]</font> Evening - Laundry 6 Butt (2A 17 78 4)</html>

<html><font color="red">[Failed]</font> Dining Room Pendant LD (25 57 89 1)</html>

<html><font color="red">[Failed]</font> Dining Room Pendant - Kitchen (24 99 BF 1)</html>

<html><font color="red">[Failed]</font> Day - Laundry 6 Butt (2A 17 78 3)</html>

<html><font color="red">[Failed]</font> Inside-All - Entry 6 Button (29 50 B3 6)</html>

<html><font color="red">[Failed]</font> 6 Butt Laundry - Laundry (2A 17 78 1)</html>

<html><font color="red">[Failed]</font> Evening Scene - Entry Six But (29 50 B3 4)</html>

<html><font color="red">[Failed]</font> Bathtub Pendant-Entry - Maste (28 FE 68 1)</html>

<html><font color="red">[Failed]</font> Day - Entry Six Button (29 50 B3 3)</html>

<html><font color="red">[Failed]</font> Range Hood Lights (25 81 DC 1)</html>

<html><font color="red">[Failed]</font> Kitchen Under Counter (2F C6 C3 1)</html>

<html><font color="red">[Failed]</font> Chase Lounge Recept - Master  (2C 16 AB 1)</html>

<html><font color="red">[Failed]</font> Inside All - Hallway 6 Butt (25 7 31 6)</html>

<html><font color="red">[Failed]</font> Shower - Master Bath (2B D4 5E 1)</html>

<html><font color="red">[Failed]</font> East Hallway - Powder-Entry (25 58 39 1)</html>

<html><font color="red">[Failed]</font> 6 Butt Pendant Hallway (25 7 31 1)</html>

<html><font color="red">[Failed]</font> Disable Program (25 7 31 3)</html>

<html><font color="red">[Failed]</font> 6 Butt Pendant - Colleens - M (25 9 76 1)</html>

<html><font color="red">[Failed]</font> Kitchen Main - Near Dining RM (24 90 99 1)</html>

<html><font color="red">[Failed]</font> Vanity 4 inch-Entry - Master  (24 AF D3 1)</html>

<html><font color="red">[Failed]</font> Kitchen Main - Hallway (24 9F 12 1)</html>

<html><font color="red">[Failed]</font> Flourescent Over Bed - Master (25 7F 3C 1)</html>

<html><font color="red">[Failed]</font> Fitness Shower (29 59 9C 1)</html>

<html><font color="red">[Failed]</font> TV Room FP Accent (24 8A 4 1)</html>

<html><font color="red">[Failed]</font> Inside Lights - Colleens Pend (25 9 76 6)</html>

<html><font color="red">[Failed]</font> Pool Table Accent (24 8C 18 1)</html>

<html><font color="red">[Failed]</font> TV Room Outlet Table Lights (30 1 FE 1)</html>

<html><font color="red">[Failed]</font> Bathtub Pendant - Near Tub -  (29 5 79 1)</html>

<html><font color="red">[Failed]</font> Fireplace Accent LD - Living  (25 5D 2B 1)</html>

<html><font color="red">[Failed]</font> Bar Center Island (24 A8 C0 1)</html>

<html><font color="red">[Failed]</font> Vanity 4-Inch-Vanity - Master (25 78 73 1)</html>

<html><font color="red">[Failed]</font> Bar High Seating (24 9F C5 1)</html>

<html><font color="red">[succeeded]</font> Colleens Office Accent lights (24 89 C4 1)</html>

<html><font color="red">[Failed]</font> Shuffleboard - Hallway (24 90 9D 1)</html>

<html><font color="red">[Failed]</font> Kitchen Main - East LD (25 4C 4E 1)</html>

<html><font color="red">[Failed]</font> Fanlinc-Ceiling-Light - Maste (27 88 89 1)</html>

<html><font color="red">[Failed]</font> Bar Low Seating (25 55 70 1)</html>

<html><font color="red">[Failed]</font> Powder Room Lights (25 5C C6 1)</html>

<html><font color="red">[Failed]</font> Entry Area - Front Hallway (24 7F FE 1)</html>

<html><font color="red">[Failed]</font> Big Fixture Living Room (25 78 7A 1)</html>

<html><font color="red">[Failed]</font> Onyx Light (25 82 3B 1)</html>

<html><font color="red">[Failed]</font> West Hallway LD - Entry Hall (25 4A 14 1)</html>

<html><font color="red">[Failed]</font> Pool Table Room Main (30 42 22 1)</html>

<html><font color="red">[Failed]</font> Game Table Dimmer - Living Ro (2A 21 1A 1)</html>

<html><font color="red">[Failed]</font> Laundry East LD (28 E6 4F 1)</html>

<html><font color="red">[succeeded]</font> Shuffleboard Table - Shuffleb (24 86 9 1)</html>

<html><font color="red">----- All On-Off - Inside Test Results -----</font></html>

Mon 02/27/2017 09:37:36 AM : [iNST-ACK    ] 02 62 00.00.28 CF 13 00 06          LTOFFRR(00)

Mon 02/27/2017 09:37:36 AM : [Ext MH      ] Unexpected Ack imCmd=62 cmd1=LTOFFRR 0x13

Mon 02/27/2017 09:37:37 AM : [iNST-TX-I2CS] 02 62 2E 70 42 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0

Mon 02/27/2017 09:37:37 AM : [iNST-ACK    ] 02 62 2E.70.42 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0 06        (00)

Mon 02/27/2017 09:37:46 AM : [iNST-TX-I2CS] 02 62 2E 70 42 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0

Mon 02/27/2017 09:37:46 AM : [iNST-ACK    ] 02 62 2E.70.42 1F 2F 00 00 00 00 00 01 00 00 00 00 00 00 00 00 D0 06        (00)

 

Unfortunately there is a lot more "Failed" than "Succeeded".  However, this is a scene that is reliable most every time despite the high number Failed findings in the Scene Test.  The devices above are scattered throughout the interior of the house from end to end and include most of the interior lights.  

 

You mention doing scene tests by room/section.  Do you create a scene for each room and then test that room it until it "works" and then move on to the next room and eliminate noise makers by room?  This leads me to believe that noise makers tend to have a localized impact and mostly affect those devices that are in close physical proximity to the device this is failing.  Is that assumption accurate? 

 

I noticed in your post to not worry about incandescent bulbs.  However, I don't have any incandescent bulbs.  All of my bulbs are LED bulbs.  There are over 200 LED bulbs that are mostly in ceilng can lights.  Are you finding that LED bulbs are occasionally causing noise? I'm really hoping you say no as I can't go back to incandescent and disabling all of those bulbs.... wow, bunch of work.  

[Teken]

RE: Scenes / Local -> More specifically while I am in a room I activate any Insteon device at the switch and observe how responsive or the lack thereof. If the local paddle operates as expected I move forward to seeing if a linked scene is operating at the that local area fine.

 

Keeping in mind a scene can be local devices in that room vs linked devices through out the home as you noted you have like many do. If we just assume the scene is for that specific room obviously its easier to tell what's happening at that location. The benefit of using the Admin Console to initiate the scene test is it allows you to wonder the home for any large scenes and watch how responsive it is or the lack thereof. 

 

Again, starting off small if all is well after (successive) scene tests I add back just one electronic device and rinse & repeat the local / scene tests. 

 

RE: LED's fixtures -> Its safe to say quality vs Insteon friendly are not related or the same. If you have lots of LED light fixtures this is the *First Go Two* items I would review. You have a 50-50 chance that you will find the bulbs are just fine - whereas you might find them impacting the Insteon network.

 

The most common and *Extreme* results of LED fixtures causing issues is when you're able to turn something on but not off. This sort of problem is exasperated by the fact you can have in between gray zones of the very same. Meaning, that same LED bulb may not stop you from turning a load off - but it may be slow to react in terms of shutting off.

 

When you have multiple loads on one circuit this adds to the difficulty because the *Observed* problem isn't apparent until someone points it out to you. Which is for example say you have two separate rooms all being fed by the same branch circuit.

 

For a moment lets keep this simple and say this is only a lighting circuit and ignore any outlets in this example. Again, you have two rooms which are physically using the same wire run. In room (A) has the same LED lights and its on - In room ( the lights also use the same LED fixture.

 

So later on you enter room ( and decide to turn on the load.

 

You may notice the switch is slow to react or slow to turn off said load. In the meantime the room (A) was on during the same time and unbeknownst to you was adding to the slow response etc. Now, you turn off room ( and turn on room (A) you will quickly notice it turns on quickly with out fault or slow response.

 

Now imagine an entire home with all the same LED's bulbs . . . As each bulb starts to emit / suck the signal there is a cumulative sucking / emitting of noise on the power line.

 

Not surprisingly when you have a LED bulb which displays one of the following attributes its an indicator something is going to happen to power line device like Insteon:

 

Flicker, strobing, humming, buzzing, pulsing, etc . . .

 

A very simple test for *Extreme* and average RF noise is to take any hand held radio and tune it to the AM band. If you walk around the home and turn any light fixture and start to see radio issues from injected noise, hum, bird whistle, etc. This is a very common indicator those bulbs are lacking in proper filtering and bad design.

 

I always recommend people (name any brand) purchase 5-8 brands / models of bulbs and do the above basic tests. Once you have identified bulbs which offer smooth dimming with out flicker, pulsing, humming, buzzing. That you repeat the same tests by turning it on and place a AM radio next to the same bulbs.

 

If you don't hear any injected interference to the portable radio - You'e probably 90% further ahead in being able to say that bulb is less likely to cause Insteon issues. Keeping in mind this has no relations to the quality of the bulb and whether or not that same behavior will be seen over time.

 

As these same tests will show up the exact same problems 9 months down the line!

 

As has been seen in many of the Cree, Fiet, Home Depot (Econo) LED bulbs . . .

 

I had a very large install last year that had no issues until the 6-9 month mark. The customer noted all comms had gone dark in the entire home. Nothing was added to the home and every device that could be filtered, replaced, removed was. So what changed in that 6-9 months of perfect operations?

 

You guessed it - LED bulbs started to fail and emit RF and EMI into air and power line . . . I was lucky because some of the bulbs were showing exactly what I hoped for which was when dimmed started to flicker, strobe, or offer a slight hum / buzz. Whereas some did nothing at all but as soon as I turned on my AM radio the howling noise was incredible.

 

Those bulbs offered no visual cues as to RF leakage in the air but was readily present. After pulling out 25 bulbs all was fine but this left the home owner with deciding what to purchase for replacement. Return for warranty and hope for the best or to select another brand?

 

I don't know what the customer did and haven't heard from them either regarding this . . .  

[DAlter01]

Ok, so your scene test is more of a real world "function" test then using Scene Test within the ISY.  As I wrote in my post, the scene works fairly well but as you saw in the results from the ISY Scene Test it failed miserably.  As such, I probably don't need to obsess with trying to get Scene Test within ISY to have success with all of the devices as long as the scene is working in the real world without repeating problems.  

 

I'll try the AM radio trick, that makes tons of sense.  Regarding LED bulbs, I had the insteon dimming wall switches installed already when I put in the LED bulbs.  I found that some brands did not dim well (flicker, light levels bouncing around, etc.).  I ended up trying several brands before finding some the dimmed well.  I did not have good luck with CREE from Home Depot.  I do have them in a few circuits where I don't dim the bulbs.  And, each year I assume CREE changes the design to some degree but I'm not a big fan of the CREE bulbs at least the ones from three years ago.  They may be fine now.

 

And, like you, I found that price point had no impact on how a bulb would function with the Insteon wall dimmers.  Expensive bulbs were just as likely to fail or do well as cheap bulbs.  I also found that different wattage bulbs from the same brand could react completely differently.  I suspect this may be either a generation progress of technology as they release the next generation of bulbs using newer technology or it could be that they bulbs are manufactured by different suppliers and the brand name is putting their name on different bulbs that have different internal technology that come from different suppliers.  Frankly, I think it is the latter.

[Teken]

Yes, the scene test is more of a factual data point which as you noted may not even be seen in the real world. But it should be noted if the scene tests always fails and you observe slow and inconsistent comms from the hardware. Its more likely the network is just hovering above the threshold where it starts to be noticed by the wife.

 

We all know the WAF is king . . .

 

Like you I never saw price, brand, model, ever dictate as to what I would see for Insteon.

 

Surprisingly, I've had great success with what I consider mid tier branded bulbs and have never had anyone complain using the same brand and model LED fixtures. Also, as you noted this whole LED technology is still very much under way so expect to see lots of variance and trial and error.

 

I've tested quite a few brands and models in my home and right now am conducting a Arctic weather sub zero temperature tests for a few makers. Thus far, I am extremely impressed with the two brands under trial and so far both have passed extreme heat and cold while offering perfect lighting.

 

RE: Cree, like you there seems to be a period of time where this brand offered some of the best value and performance bulbs. Whereas now it seems there is a huge variability of life expectancy / Insteon induced problems. One site has had more than 60 Cree bulbs when they were very expansive and continue to operate with out fault. Whereas some other sites purchased current 2015 - 2016 bulbs and have seen early failures to lots of Insteon comm issues.

 

 

This isn't unique to Cree alone as we have seen more issues with the Feit branded bulbs found at Costco. Sadly, lots of folks won't ever see the mythical 25 years service life many tout as being the case. 

[jtara92101]

A very simple test for *Extreme* and average RF noise is to take any hand held radio and tune it to the AM band. If you walk around the home and turn any light fixture and start to see radio issues from injected noise, hum, bird whistle, etc. This is a very common indicator those bulbs are lacking in proper filtering and bad design.

 

 

If you do have a radio capable of tuning BELOW the AM radio band, it would be useful to listen closer to the Insteon powerline frequency (131.65 kHz).

 

While I realize it's an inaccurate description, I can't help but refer to the below-pictured radio as my Signal Sucker Seeker...

 

 

I just did a quick scan, and chief culprits in my home are refrigerator, plasma TV, and - the biggie - LED bulbs. My ceiling can conversions are the loudest, and project over quite a range in the room! My computer setup is quite quiet, as the bulk of the equipment is inside a metal rack cabinet, and all of the power strips (and honking UPS) are behind a multi-stage filter. I pick up very little from the power lines themselves.

 

The real test is, of course, the latter (noise injected into the line). The ceiling cans don't seem to inject much into the line (I held the antenna near the wire to the dimmer that control them), but they sure blast some RF into the room. (which shouldn't hurt Insteon).

 

If you have something that can tune to 915mHz, you can test your RF signal strength, and you can hear the Insteon traffic when you press a switch. I get full bars everywhere.

 

BTW, the above is an old model long ago discontinued. It is a VERY handy tool to have if you can find one! I checked eBay, but could only find one not working for parts for $20. It covers a very wide range but not useful for modern public service, etc. as neither trunk-tracking nor digital capable. But for listening to anything from Radio Moscow to an airport control tower to AM/FM radio to Insteon signals to use as an interference locator, it is ideal!

 

It even has a built-in spectrum analyzer. You can sick it on a band segment, and it will display a little graph of the signals found. Move a "cursor" to match, and then you can listen to signals of interest.

 

 

(Imagine Teken is now rushing off the set up an eBay watch! But won't be surprised if he already has something like this...)

[Teken]

This is the little old school radio that has proven to show case the RF noise from bulbs to various electronics

 

 

These are just a few of the back up radios our team uses to supplement our satellite and long range comms.

 

.

 

I don't have a photo handy on my phone but when things get serious I pull out a dedicated RF counter which tracks all in band / out of band RFI / EMI

 

 

=========================

 

The highest calling in life is to serve ones country faithfully - Teach others what can be. Do what is right and not what is popular.

[stusviews]

Re; the problem areas in your home and a systematic approach.

 

Are they scattered or concentrated to one or two areas? If the latter, are the problem areas on the same panel?

[DAlter01]

A Radio/Electronics untrained guy like me is started to see behind the wall of secrecy to see how all of this works....  I like the idea of getting ahold of a radio that will pick up these other frequencies.  I imagine they are not cheap but neither is time.  I did play around with a little battery AM radio earlier today.  The noisiest things in the house are a couple of drivers for under counter LED tape lights.  They really hum.  I need to track down some wire-in filters it seems.  

[jtara92101]

OT, Teken, what on earth is that abominable snowman with the maple leafs on it in the front? Some kind of Canadian-themed emergency weather radio?

 

----

Keep in mind that RF emitted from devices near the Insteon powerline frequency of 131.65kHz are not inherently harmful to Insteon. It is only if those signals are actually injected into the power line.

 

If you know where the power lines run inside the walls (you should be able to take a good guess, and fancy stud finders have a mode for finding them) check AWAY from the device itself, but along the power line. You might even introduce some grounded metal shielding between the radio antenna and the device in question. If the signal is injected into the power line, then the power line will radiate it. If it is a plugin device, probe along the cord that plugs into the wall, but away from the device.

 

It is not the RF radiation from the device itself or even from the line that is the problem. It is the noise on the line. RF emitted by the line can by used as a simple proxy for examining the actual line voltage, and is quick and easy. The best way, of course, would be to look at the line voltage directly with an oscilloscope or spectrum analyzer.

 

So, for example, probe your refrigerator. If it's a fancy one, it will probably be a noise maker. Now, pull it out from the wall, stand between the fridge and the wall (in hopes of blocking the radiated signal), and probe the cord going to the wall.

[DAlter01]

 

stusviews, there are two problem areas with multiple devices in those areas having problems.  The odd part is those devices are far away from what I would think are noise generators (appliances, electronic equipment, etc) .  One of the problem areas is the master bedroom suite.  I've unplugged everything in there and that seemed to have no beneficial impact.  Lights on (All are LED) or lights off seems to have no impact.  However, the master suite is very near the main panel.  I got to thinking the noise may be coming from a distant noise maker on a circuit that doesn't have Insteon devices and then coming through the main panel and polluting the circuits to the master suite.  But, in reading the posts in this forum where it seems that noice/signal suckers cause a localized effect maybe that theory will not hold up.

 

[Teken]

A Radio/Electronics untrained guy like me is started to see behind the wall of secrecy to see how all of this works....  I like the idea of getting ahold of a radio that will pick up these other frequencies.  I imagine they are not cheap but neither is time.  I did play around with a little battery AM radio earlier today.  The noisiest things in the house are a couple of drivers for under counter LED tape lights.  They really hum.  I need to track down some wire-in filters it seems.  

 

As noted my go to tool of choice has been a old school AM / FM radio as listed up above. Its very basic and has very few solid state electronics and thus is able to capture and *Hear* noise not typical in integrated circuits unless designed to do so. The telescopic antenna enables me to literally point and shoot at something and if its strong enough it will be picked up by the radio. One tip to note is often times you will see great success simply tuning into a empty channel.

 

This dead band obviously produces noise . . .

 

But, this mid range also enables you to hear - listen to those very faint RF disturbances too! You can not adjust a solid state radio to sweep and track this dead band area because the PLL crystal disallows this.

 

NOTE: As pointed out in the scene test don't get too caught up with this sort of thing. Its merely intended as a method to help isolate and track *Possible* culprits in the environment. You will quickly notice lots of this is very forgiving to other systems in the home. But when your intent is to reduce your homes RFI / EMI contamination like I its invaluable to assist in this effort.

 

My home has probably lowest RFI / EMI output of any other and I like to keep it that way.

 

I don't want to break my tin foil hat too soon - LOL

[Teken]

OT, Teken, what on earth is that abominable snowman with the maple leafs on it in the front? Some kind of Canadian-themed emergency weather radio?

 

LOL - Its a adjustable (rubber body) clock, timer, alert beacon . . .

[stusviews]

dalter2, distance doesn't matter, circuits do. Are the two problem areas on the same panel? What are the specific problems? Which particular devices are affected?

[DAlter01]

stusview,

 

The problem areas are on the same panel, actually same subpanel.  I've switched off all the breakers on that affected subpanel except the three circuits that have the problem insteon devices.  Then, on each of those three affected circuits I've turned them each off to check individual circuits and the devices on those circuits (using Houselinc Diagnostics).  

 

My system no longer uses the HUB and Houselinc as I'm controlled now by ISY.  However, I fired up the old HUB and Houselinc to use its diagnostics routine.  The problem seems to remain in that Houselinc Diagnostics returns 50-80% success on those problem devices.  I've attached the HUB for running the diagnostics to a receptacle on that same circuit to make sure turning off the breakers didn't shrink my network so much that my signal would not reach.  The problem devices are dimmable Switchlincs and 6 button keypads.

 

Since I switched off all of the breakers on that subpanel I was thinking that the noise was coming from the main panel or one of the other subpanels through the main panel and polluting the insteon circuits on that subpanel.  The problem is sometimes a device will not turn on or off for the Switchlincs when being told to turn on/off by a scene.  With the 6 button, sometimes when using one of the buttons to trigger a scene or a program the command isn't heard and I have to trigger it again before the scene or program command is received.  I've converted a few of my scenes from having controllers within the scene to program calls with buttons triggering a program to run which in turn turns on/off the scene.  I didn't seem to make any significant impact on the reliability of the running of the scene by making this change.

 

I did have three filters which I just deployed today on items that made some noise as evidenced by the AM radio test.  I have another 4-5 hotspots to address once I get some more filters.  A couple of items I need to filter are hardwired.  They are LED drivers for LED tape lights.  They are the noisiest items in my house based on the AM radio test.  I don't see an insteon hard wire filter.  Will the X10 hard wire filters do a reasonably good job?  

[DAlter01]

Teken,

 

You mentioned LED life.  You are right on, it will be the rare bulb that goes 25 years.  Not only will many fail before 25 years but they get a little dimmer each year like fluorescents do, just not as much degredation/year as fluorescents.  People will want to replace them before their lifspan is up just so they can see again.  I've had mine for three years now.  And, maybe it is my eyes getting worse (I'm 51 and they are in fact getting worse), but I had to reset my scenes at a higher level recently because the house was dimmer than what I think it was when I set it up.  

 

I ended up using the CREE bulbs on the outside patios, in closets where they would only be turned on 100%, etc.  I ended up using mostly Phillips, Soraa, and another brand I can't recall that was the house brand from either Lowes or Home Depot for the rest of the interior lights.  There are a few other specialty LED bulbs thrown in from Ebay for unique situations.  So far on failures I've had one Phillips fail, probably 8 CREE bulbs fail (out of probably 50-60) and a few of the Ebay bulbs.  To be fair, some of those CREE bulbs have been in outdoor can fixtures so they are living a tougher life than the ones that get nice conditioned air all day and night.  However, even on the interior I probably have 25-30 CREE bulbs and have had to replace 3-4 of those.   

[stusviews]

Are the SwitchLinc dimmers dual-band? What about the KPLs? What is controlled by each SwitchLinc? Is the problem with the KPLs restricted to the load button or are only secondary buttons affected or both? If it's secondary buttons, are they controllers for the scene with the errant dimmers? Do you have any incandescent bulbs that you can test with?

 

BTW, it's highly unlikely that the problem originates on a circuit other then those that the exhibit a difficulty.

[DAlter01]

stusviews,

 

Yes, the Switchlincs and KPLs are all dual band.  And they are all vintage of late 2013 or later (v.41 or later).  All of my affected Switchlincs control LED bulbs.  I've had issues on the KPLs of the loads not responding to scenes and the secondary buttons not triggering scenes.  I've had the secondary button trigger scenes (via program) and the KPL load not respond on the load side within the same KPL.  I don't think I've had a secondary button within a KPL fail to control the load on the same KPL when the secondary button of the KPL was a controller of the scene (as opposed to a program trigger).

 

I don't have any incandescent bulbs.  I did notice today with the AM radio test that I have some Ebay corncob LED bulbs (G4 base I believe) in a pendant fixture that are pretty noisy in that part of the house and that is one of my worst performing devices.  Plus, the load on that device goes back to a box that has four other switches and I don't believe all of those switches are on the same circuit.  As such, there is a fair chance I'm introducing some noise into two circuits and the subpanel through those LED Ebay bulbs. Those bulbs were not nearly as noisy as the LED drivers for the tape lights at the other end of the house.  However that other end of the house works pretty well so maybe RF noise isn't always indicative of noise that will affect the insteon network...

[stusviews]

Very few RF devices operate near the frequency used by Insteon devices, 915 MHz. Two of the more common devices are baby monitors and older cordless phones. If you don't have any RF devices near 900 MHz, then you can rule out RF interference. In any case, as I've mentioned earlier, the primary cause of communication difficulty is something on the powerline which is also the main means of Insteon signalling.

 

LED bulbs are known to cause difficulties with HA dimmers, all brands, not just Insteon. The only real test is to observe the result with incandescent bulbs.

 

BTW, why are you using programs to control scenes instead of including the trigger device as a scene controller?