Improper Programm Behavior Since Firmware Update.

[IndyMike]

Hi Lou,

 

From our previous conversations, I've got a fair idea of the size of your home and installed equipment...

 

I have to say here that I'm in Lee's camp on this one - I'm surprised your system is functioning.

 

The current Smarthome 2406 hardwired Insteon Phase coupler is an X10 coupler. It previously carried a different model # and was advertised as an X10 coupler - the internal components are identical.

 

Some X10 couplers are a bit different - the ACT CP-000 uses twin tuned transformers to couple the signals across phases. Elegant, but it actually performs a bit worse than the Smarthome model.

 

Your X10 coupler should help (assuming it's passive). Were it me, I would try locating it as close to your PLM as possible. I say should because I don't understand how things are functioning now.

 

IM

[Brian H]

If you coupler is a XPCP coupler. It also has a set of twin tuned transformers.

Not sure how narrow it coupling range is.

[apostolakisl]

Hi Lou,

 

From our previous conversations, I've got a fair idea of the size of your home and installed equipment...

 

I have to say here that I'm in Lee's camp on this one - I'm surprised your system is functioning.

 

The current Smarthome 2406 hardwired Insteon Phase coupler is an X10 coupler. It previously carried a different model # and was advertised as an X10 coupler - the internal components are identical.

 

Some X10 couplers are a bit different - the ACT CP-000 uses twin tuned transformers to couple the signals across phases. Elegant, but it actually performs a bit worse than the Smarthome model.

 

Your X10 coupler should help (assuming it's passive). Were it me, I would try locating it as close to your PLM as possible. I say should because I don't understand how things are functioning now.

 

IM

 

What exactly makes you think my system shouldn't work. The only piece I am missing is a phase coupler. I had assumed my rf signalincs were doing that, but it would seem that they were superfluous based on my normally function system without them.

 

My system works at least 99% of the time. I have one icon dimmer that consistently fails, and by that I mean about twice per month it doesn't turn on or off as it should on its program. I get occasional "failed to communicate" messages on the ISY, but the thing is that it did communicate and everything worked. Perhaps that is failed ACK message.

 

My only real problem is the thing that started this entire thread. Which for mysterious reasons is much better now only happening about 1/20 tries.

 

 

I dug out my x10 coupler. It is a xpcr

[LeeG]

Apostolakisl

 

Although some of the symptoms have been categorized as acceptable such as the ICON switch not responding and the comm error messages, it may be useful to add them to the mix when evaluating the SwitchLinc to/from PLM symptoms. These may be separate independent problems but they may be related to something more common. Analyzing the entire picture may reveal something that would not be apparent when evaluating a single problem. I normally pick one symptom and track it down as too many variables can confuse the issue. This may be the exception where looking at the overall pattern may be helpful.

 

There is no track of which devices were in the repeat (hop) path to get from source to destination. Insteon addresses are not changed along the way so even if monitoring the powerline at a low level that information would not be available.

 

Lee

[apostolakisl]

Any comments on the xpcr. I pretty much figured that a hop trace would be impossible, but it was worth asking.

 

My command central room is behind a utility room which has a 220 drier outlet. I was thinking of drilling through and pulling a 220 wire into my communication room (actually part of the attic, but my attic is conditioned space), and putting the plm on one leg of the 220 and putting xpcr linking the two legs together.

 

Having a dedicated line for the plm might be nice, but except for when the drier is actually on, I think this should be pretty much the same thing.

[IndyMike]

 

What exactly makes you think my system shouldn't work. The only piece I am missing is a phase coupler. I had assumed my rf signalincs were doing that, but it would seem that they were superfluous based on my normally function system without them.

 

My system works at least 99% of the time. I have one icon dimmer that consistently fails, and by that I mean about twice per month it doesn't turn on or off as it should on its program. I get occasional "failed to communicate" messages on the ISY, but the thing is that it did communicate and everything worked. Perhaps that is failed ACK message.

 

My only real problem is the thing that started this entire thread. Which for mysterious reasons is much better now only happening about 1/20 tries.

 

 

I dug out my x10 coupler. It is a xpcr

 

I don't understand how you are coupling the phases. I believe my house is a fair amount smaller than yours. There are times when I can couple through the utility transformer well enough to communicate across phases. This is not reliable for long durations.

 

Simply put, I would expect your performance to be worse than mine (with no coupling) given the increased size of your home.

 

The XPCR is an X10 active repeater - it will not work for insteon. The repeater monitors X10 one one phase and then repeats it on the opposite phase. Since the repeater does not understand Insteon, it will not repeat the signal.

[apostolakisl]

So, here is what I think would be a first place to start to try to cut down on hops.

 

1) Buy a 2406h http://www.smarthome.com/2406H/SignaLin ... red/p.aspx

 

2) Drill through the back of my dryer j-box and pop out the wall near my HA stuff.

 

3) Put a double gang j-box and pull the 4 wire 240 line in.

 

4) Connect the 2406h to the 2 legs of the 220 and the neutral

 

5) Connect a 120 receptacle to one leg and the neutral plus ground

 

6) Plug the PLM into that outlet.

 

Would anybody suggest that this is not the most direct way to get both phases connected to the PLM. It is effectively a dedicated circuit with a straight shot back to the breaker box. It would only have other activity when that dryer is on, and dryer's should be quiet as far as noise is concerned.

[ELA]

 

My system works at least 99% of the time. I have one icon dimmer that consistently fails, and by that I mean about twice per month it doesn't turn on or off as it should on its program. I get occasional "failed to communicate" messages on the ISY, but the thing is that it did communicate and everything worked. Perhaps that is failed ACK message.

 

My only real problem is the thing that started this entire thread. Which for mysterious reasons is much better now only happening about 1/20 tries.

 

 

Hello apostolakisl,

I have tried reading your post from beginning to end a few times with hopes that I could offer some help. The post above seems somewhat inconsistent with the rest of the thread? I have to question when someone says their system works at least 99% of the time. From the help others have offered and your responses I believe your system has some (possibly several) very real issues that you need to address if you want to attain reliable operation.

I understand it can be very difficult without test equipment to make progress. I do want to throw out a couple of thoughts in hopes it might help some.

 

If I read correctly It appears that Ack messages from the PLM to several devices are not making it though on 1 hop. Your desire is to increase the system performance such that 1 hop is all that is required.

 

1) You stated that you have a 2412S PLM. My 2412S had a rather weak transmitter. That may have been only my one device but I beleive the hardware design of the 2413 has a stronger transmitter. In addition the newest 2413 I purchased also has new firmware that appears to have changed the hop assist protocol. Normally the PLM would not participate in simulcasting its own 1st hop. My newest device does.

I hate to suggest that you invest in a new PLM and find that it did not improve your situation but you might consider it. Without ability to monitor you signal levels it would only be a guess.

 

2) You talk about removing old signallincs and others have questioned how you might be coupling your phases. I think you need to be serious about getting a good intentional coupling whether that be via a hardware coupler or newer Access points.

 

3) Are you using any filters? Have you read others posts about devices they have had to add filterlincs to? In my experience there are lots of home appliances that can degrade your communications and it is likely that at least some of these may need to be filtered.

 

As a side note:

I also read where you used a 100ft extension cord when testing different positions for your PLM (as a diagnostic tool).

I recently did a test to compare 175 ft of 14/2 romex against a 75Ft 14/2 extension cord. This test showed that a long extension cord (dependent upon its design) can be way worst than romex for Insteon signals. In romex the conductors are spaced farther apart and that results in a lower capacitance (there are other factors as that affect this as well but this is the obvious one).

While the romex added very little measureable attenuation the extension cord added a fair amount. Using the comparative system system I have been previously the romex was null whereas the extension cord was greater than 1 Standard Insteon load.

 

The point here is when doing a diagnostic test, where you move the PLM, use as short of a cord as required.

 

Hope this did not sound too harsh. I did hope to help.

[apostolakisl]

 

My system works at least 99% of the time. I have one icon dimmer that consistently fails, and by that I mean about twice per month it doesn't turn on or off as it should on its program. I get occasional "failed to communicate" messages on the ISY, but the thing is that it did communicate and everything worked. Perhaps that is failed ACK message.

 

My only real problem is the thing that started this entire thread. Which for mysterious reasons is much better now only happening about 1/20 tries.

 

 

Hello apostolakisl,

I have tried reading your post from beginning to end a few times with hopes that I could offer some help. The post above seems somewhat inconsistent with the rest of the thread? I have to question when someone says their system works at least 99% of the time. From the help others have offered and your responses I believe your system has some (possibly several) very real issues that you need to address if you want to attain reliable operation.

I understand it can be very difficult without test equipment to make progress. I do want to throw out a couple of thoughts in hopes it might help some.

 

If I read correctly It appears that Ack messages from the PLM to several devices are not making it though on 1 hop. Your desire is to increase the system performance such that 1 hop is all that is required.

 

1) You stated that you have a 2412S PLM. My 2412S had a rather weak transmitter. That may have been only my one device but I beleive the hardware design of the 2413 has a stronger transmitter. In addition the newest 2413 I purchased also has new firmware that appears to have changed the hop assist protocol. Normally the PLM would not participate in simulcasting its own 1st hop. My newest device does.

I hate to suggest that you invest in a new PLM and find that it did not improve your situation but you might consider it. Without ability to monitor you signal levels it would only be a guess.

 

2) You talk about removing old signallincs and others have questioned how you might be coupling your phases. I think you need to be serious about getting a good intentional coupling whether that be via a hardware coupler or newer Access points.

 

3) Are you using any filters? Have you read others posts about devices they have had to add filterlincs to? In my experience there are lots of home appliances that can degrade your communications and it is likely that at least some of these may need to be filtered.

 

As a side note:

I also read where you used a 100ft extension cord when testing different positions for your PLM (as a diagnostic tool).

I recently did a test to compare 175 ft of 14/2 romex against a 75Ft 14/2 extension cord. This test showed that a long extension cord (dependent upon its design) can be way worst than romex for Insteon signals. In romex the conductors are spaced farther apart and that results in a lower capacitance (there are other factors as that affect this as well but this is the obvious one).

While the romex added very little measureable attenuation the extension cord added a fair amount. Using the comparative system system I have been previously the romex was null whereas the extension cord was greater than 1 Standard Insteon load.

 

The point here is when doing a diagnostic test, where you move the PLM, use as short of a cord as required.

 

Hope this did not sound too harsh. I did hope to help.

 

 

Thanks for the info. And by 99% I mean not including the issue with that program that requires one hop. That is really an "on the fringe" use of Insteon in my opinion. I would say that when given the chance for 3 hops, damn near everything works every time.

 

My 2412s PLM is 2 years old and I replaced it at that time to try to get to 100%. My previous one would be about 5 years old now (forgot the number). Frankly, I am not sure anything changed 2 years ago with that trade out. The newest ones may be more of a change. I suppose it would be worth a try.

 

The 100ft cord I was using was 12 gauge industrial grade (makes your arm muscles burn as you coil it). And it didn't matter where I plugged it in or if I didn't use the cord at all. I couldn't find any change in my scene test results.

 

I do have 2 Insteon noise filters. The one is filtering a known problem with a low-voltage light transformer. This was obvious. The other filters all of my HA stuff (except the PLM of course). This includes all of my audio/computers/Elk panel/video cameras/sattelite receiver/other sundries. It isn't clear that installing this changed anything, but it is there. There is no association with anything in my home being on/off and the extra hop count. If I sit there and hit the same switch over and over and over, it will just every once in a while take 2 hops instead of 1. Not like something is turning on and screwing it up unless it just flips on for a second, which I just don't think is happening. I think it is just a situation where it is slightly on the fringe of enough signal strength and every now and then it just isn't quite enough.

[ELA]

Hi apostolakisl,

I have done a lot of communications monitoring and found it very common to see the number of hops required change on subsequent send/receive attempts, even when other loads are not changing states. This has been true when the communications are marginal (on the fringe as you said).

 

Regarding your post about adding the coupler:

I would not recommend the installation as you laid it out. Adding a device to an existing 240V dryer circuit as you detailed sounds like it would be an electrical code violation.

I would recommend you install it at the service cabinet with circuit breakers as the installation guide says to do.

 

Then address any remaining issues once this is complete. Best of luck to you.

[IndyMike]

Hello Lou,

 

I was all set to chime in and agree that paralleling the passive coupler and Line to Neutral outlets off the 240V dryer branch was a definite code violation. Now I'm not at all that sure.

 

The above qualifies as a multiwire branch circuit and is covered by Article 210.4 of the NEC code. There appear to have been a number of revisions to this section in recent years.

 

1999 NEC Code, 210.4 Multiwire Branch Circuits.

• 
  [*:30rmpb4w](A) General. Branch circuits recognized by this article shall be permitted as multiwire circuits. A multiwire circuit
  shall be permitted to be considered as multiple circuits. All conductors shall originate from the same panelboard or
  similar distribution equipment.
  FPN: A 3-phase, 4-wire, wye-connected power system
  used to supply power to nonlinear loads may necessitate
  that the power system design allow for the possibility of
  high harmonic neutral currents.
   
  [*:30rmpb4w]( Devices or Equipment. Where a multiwire branch circuit supplies more than one device or equipment on the same yoke, a means shall be provided to disconnect simultaneously all ungrounded conductors supplying those devices or equipment at the point where the branch circuit originates.
   
  [*:30rmpb4w]© Line-to-Neutral Loads. Multiwire branch circuits shall supply only line-to-neutral loads.
   
  Exception No.1: A multiwire branch circuit that supplies only one utilization equipment.
  Exception No.2: Where all ungrounded conductors of the multiwire branch circuit are opened simultaneously by the branch-circuit overcurrent device.

 

Your current dryer installation is covered under Exception #1 (Line to Line load - 240V). You are proposing to add Line to Neutral loads to this branch. This may be covered under Exception #2 if your breaker opens both legs of the circuit simultaneously due to an overcurrent in either leg.

 

Code Updates - Found the following on the Mike Holt site (http://ecmweb.com/nec/code-basics/electric_branch_circuits_part/

 

Multiwire branch circuits. Multiwire branch circuits are circuits that have more than one ungrounded conductor sharing a common grounded (neutral) conductor. These circuits are very beneficial in that they use less material, result in a lower circuit voltage drop, and ultimately result in cost savings. They do, however, have some specific Code rules that can't be ignored.

 

To prevent inductive heating and reduce conductor impedance for fault currents, all multiwire branch-circuit conductors must originate from the same panelboard or distribution equipment [210.4(A)].

 

Multiwire branch circuits must supply only line-to-neutral loads [210.4©].

• 
  [*:30rmpb4w]Exception 1: A multiwire branch circuit can supply line-to-line utilization equipment, such as a range or dryer.
  [*:30rmpb4w]Exception 2: A multiwire branch circuit can supply both line-to-line and line-to-neutral loads if the circuit is protected by a device (multipole circuit breaker) that opens all ungrounded conductors of the multiwire branch circuit simultaneously (common internal trip) under a fault condition.

 

If the continuity of the grounded (neutral) conductor of a multiwire circuit is interrupted (open), the resultant over- or undervoltage could cause a fire and/or destruction of electrical equipment. See 300.13( for the requirements relating to the continuity of the grounded (neutral) conductor on multiwire circuits.

 

From that above NEC code revision (date??) - you are allowed to combine Line to neutral and Line to Line devices on a multiwire branch circuit. You'll need to determine the following:

 

1) What revision of the code applies to your area. My area uses 1999 base code with pieces of recent code. While I believe it could be installed safely, I would expect a lot of flak from my local inspector.

2) Whether there is a code exclusion preventing the use of dedicated appliance circuits (dryer) with 120V line to neutral branches. I don't have a full copy of current code to verify this.

3) Whether your current installed breaker/wiring will support the addition of 120V legs. This is a multi-part question -

• 
  [*:30rmpb4w] Your total loading on the conductor and breaker cannot exceed NEC requirements. This was a dedicated circuit to your dryer. You are now changing the application and different rules will apply.
  [*:30rmpb4w] Most importantly - you will need to ensure that your 120V (Line to Neutral) branches will trip the breaker in the event of a Line to Neutral overload. This is covered under 210.19 (A) (4)

http://ecmweb.com/nec/code-basics/branch-circuits-nec-requirements-20110201/

In summary, I can't say this would be a violation. On the flip side, I can't say that it would be code compliant either. It depends on your local code and your particular installation.

 

My advice would be to call someone knowledgeable in local code and installation - have any neighbors in the business?

[ELA]

Assuming the dryer is a 30amp or larger dedicated circuit you cannot simply tap a 15A receptacle from one leg and neutral.

 

At a bare minimum you would have to add a overcurrent device rated for the receptacle. Certainly recommended that the coupler be protected at a current lower than 30A as well ( as per the installation instructions). There was no mention of this in the planned install.

 

A dryer is a dedicated circuit intended for that load and not intended as a distribution point from which to feed other loads. In addition consider that the dryer has a rather large motor that may produce large transients that will now be coupled very easily into your PLM.