Why is this?

[Michel Kohanim]

Hello IndyMike,

 

I would have to defer this one to Chris as he's the genius behind our message handler class.

 

With kind regards,

Michel

 

Michel,

 

Edited - original question was unfair to the ISY team

 

Can you comment on how the ISY implements RS-232 flow control, or am I stepping on proprietary toes?

IM

[Guest]

Well its been several weeks with the new PLM (actually maybe 6 or more) and there have been ZERO lockups. A+++++

 

Device to device communications are nearly 100% (I actually cant say I have pushed a button on a keypadlinc and not had the other linked device work but I dont want to swear to 100%).

 

Status lights on the keypadlincs are not always 100% but not to bad.

 

Communication failures with the PLM on a query with the ISY are about 7 or 8 devices on average having a communication faliure on a system query. Some other forums have people mentioning a similar occurance.

 

Steve L. are there any updates from SH/SL on the beta PLM's that also have the newest firmware thats not yet released?

[Guest]

After kicking this problem around a lot with a few coworkers (yes we are a bunch of engineers with nothing better to do at lunchtime when its raining out) we came up with an idea.

 

I wired a Switchlinc relay in a 2 gang box with an outlet. The switchlinc turns the outlet on and off. Attached a line cord to this. I linked a controlinc to this switch.

 

I unlugged the PLM and plugged in the switch and outlet. I then plugged a radio into the outlet and cranked it up.

 

Then for the next hour and a half I went from outlet to outlet in my house and used the controlinc to turn on and off the switchlinc thus turning the radio on and off (drove everyone nuts).

 

No problems!!!!

 

This seems to indicate that the signal is getting to the outlet where the PLM is normally plugged into. So maybe the PLM needs a hearing aide?

 

So why?????

 

Maybe the micro or something else in the PLM is causing a signal antenuation or noise. If you piggyback another device the same thing would in theroy still happen???

 

Now what???

 

So maybe the PLM needs to be filtered and then a device with a receive sensitvity of 1 mv piggybacked after the filtering?

 

Thats the next experiment!

 

 

[upstatemike]

I don't know if piggybacking is the answer since the close proximity of the PLM and the other insteon module will load the line and reduce the signal available to both devices. I would put the PLM on an isolated circuit with no other devices. I would put an insteon device of some sort at, or very near, your power panel on a different circuit that is on the same AC leg as the PLM circuit.

 

This will ensure nothing reduces the signal that actually reaches the PLM. It will also ensure that all signals are repeated right at the breaker box so the PLM circuit will have the maximum possible level for all signals entering or leaving the dedicated PLM circuit.

 

I think anybody trying to make a PLM work on the same circuit where a PC or other device that can load the line is plugged in, is going to have ongoing problems with PLM communications.

[Guest]

I don't know if piggybacking is the answer since the close proximity of the PLM and the other insteon module will load the line and reduce the signal available to both devices. I would put the PLM on an isolated circuit with no other devices. I would put an insteon device of some sort at, or very near, your power panel on a different circuit that is on the same AC leg as the PLM circuit.

 

This will ensure nothing reduces the signal that actually reaches the PLM. It will also ensure that all signals are repeated right at the breaker box so the PLM circuit will have the maximum possible level for all signals entering or leaving the dedicated PLM circuit.

 

I think anybody trying to make a PLM work on the same circuit where a PC or other device that can load the line is plugged in, is going to have ongoing problems with PLM communications.

 

Close proximity of devices in the same box should normally not be a problem. I have 2 keypads and 2 switchlincs all in the same box at my entrance way and they communicate with other devices with no problems.

 

The PLM for the past month has been on the same circuit as one high hat with a switchlinc controlling the high hat. Funny thing is that even that switch comes up as a comm failure at times.

 

I have 4 accesspoints (2 on each leg) and a phase coupler. No difference on device to device communications without the extra accesspoints or if I shut off the breaker to the phase coupler. One set of accesspoints seems fine but I had to try and fix the comm failures so I followed the SH motto of "buy more".

 

Again if the signal gets to the outlet where the PLM is plugged into then why cant the PLM hear it? A switchlinc hears it. (PLM has min 20 mv signal requirement where Switchlinc has 1 mv requirement).

[upstatemike]

But aren't you reinforcing my point? Multiple switchlincs in the same box work because they only need 1 mV of signal so the loading does not impact them much. The PLM needs 20 times the signal strength so any load (even a single high hat) could impact it a lot. Try the PLM on it's own circuit and see if it works. Even if SmartHome eventually makes a PLM that is as sensitive a switchlinc, it will still be happier on a dedicated AC circuit. And if the sensitivity is somehow related to the old heat problem, it may be quite awhile before you see a high sensitivity PLM offered (maybe in an aluminum shell instead of plastic?)

[Guest]

But aren't you reinforcing my point? Multiple switchlincs in the same box work because they only need 1 mV of signal so the loading does not impact them much. The PLM needs 20 times the signal strength so any load (even a single high hat) could impact it a lot. Try the PLM on it's own circuit and see if it works. Even if SmartHome eventually makes a PLM that is as sensitive a switchlinc, it will still be happier on a dedicated AC circuit. And if the sensitivity is somehow related to the old heat problem, it may be quite awhile before you see a high sensitivity PLM offered (maybe in an aluminum shell instead of plastic?)

 

The highthat and switchlinc are way down the line.

 

If you are saying that the PLM needs a dedicated breaker with nothing else I can try that in a day or two. Worth a shot!

 

If thats the case the website and installation instructions should state that.

 

Thanks

[upstatemike]

I'm just suggesting it is worth a try. That is what I do and I don't see the same issues that others have.

[IndyMike]

Digger,

 

First and foremost, I agree wholeheartedly with Upstates' suggestion to move your PLM to a dedicated circuit next to your breaker panel. This configuration has worked for me with both X10 (20 years) and Insteon.

 

I took this one step further by adding a passive coupler to my panel (basement), and locating my signalincs on the second floor. My thinking here is that I will save a hop with the coupler while improving communication at the most remote locations on the second floor.

 

Past that we may be focusing a bit too much on the receive level differences between the PLC (10mv), PLM (20mv) and Switchlinc (1mv Insteon). The complete specifications for the SL, Icon, and KPL are below:

 

1) SL 1mv Insteon, 10mv X10

2) Icon 1mv Insteon, 10mv X10

3) KPL 1mv Insteon, 20mv X10

 

It's interesting that the PLC/PLM spec's don't distinguish between Insteon/X10 - a combined spec perhaps? I find it difficult to believe the SH has incorporated different hardware in the two devices. Different firmware - definitely. In reviewing the schematic for the PLM (don't have one for the PLC) it would appear that the AGC is firmware controlled. This might account for the difference between the two devices.

 

Stepping back for a moment, the X10 standard has existed for 30 years. During that time a 100mv signal level was considered a minimum for reliable reception. These devices had similar output levels and relied on a single point transmitter (possibly coupled with a active repeater at the panel) to broadcast across the entire house.

 

In contrast, Insteon uses multiple repeaters to keep the signal level high, active RF coupling that can be placed at the users discretion, and a far more robust communication protocol.

 

Sounds great don't it. Well, in your case one of those links is broken. However, given the "repeating" nature of Insteon, I would not think that a 20mv receive level at the PLM would be the problem. If you ever had X10 operating in you home (I believe you did) this just doesn't seem possible. To be short, if you're getting less than 20mv receive at your PLM you have other problems.

 

To be honest, I view a 1mv min receive level as a marketing ploy - can't be achieved in nature. The AGC (SH has told me the switches have AGC) circuit will quickly dial the min receive level upward above the ambient noise.

 

I remember some posts indicating that you had run tests with a ELK-ESM1. Did you check for receive levels at your PLM? I just did, I'm getting roughly 2V as transmitted from a second floor switch (repeated of course). The min receive level registered by the ELK is 100mv.

 

I don't think I've ever asked -

1) How old is your home?

2) Copper Romex or Aluminum?

3) Single or multiple panels?

 

So, I've written a bunch and haven't given any suggestions.

1) Check your panel for loose/oxidized connections. NO MULTIPLE connections on breakers or neutrals.

2) If you can isolate to a circuit, check box terminations.

 

If your into the extreme, here's a link (cleaning breaker panel connections) -

http://idobartana.com/hakb/x10alignment.htm

 

Let us know how things are working out,

IM

[Guest]

Digger,

 

First and foremost, I agree wholeheartedly with Upstates' suggestion to move your PLM to a dedicated circuit next to your breaker panel. This configuration has worked for me with both X10 (20 years) and Insteon.

 

I took this one step further by adding a passive coupler to my panel (basement), and locating my signalincs on the second floor. My thinking here is that I will save a hop with the coupler while improving communication at the most remote locations on the second floor.

 

Past that we may be focusing a bit too much on the receive level differences between the PLC (10mv), PLM (20mv) and Switchlinc (1mv Insteon). The complete specifications for the SL, Icon, and KPL are below:

 

1) SL 1mv Insteon, 10mv X10

2) Icon 1mv Insteon, 10mv X10

3) KPL 1mv Insteon, 20mv X10

 

It's interesting that the PLC/PLM spec's don't distinguish between Insteon/X10 - a combined spec perhaps? I find it difficult to believe the SH has incorporated different hardware in the two devices. Different firmware - definitely. In reviewing the schematic for the PLM (don't have one for the PLC) it would appear that the AGC is firmware controlled. This might account for the difference between the two devices.

 

Stepping back for a moment, the X10 standard has existed for 30 years. During that time a 100mv signal level was considered a minimum for reliable reception. These devices had similar output levels and relied on a single point transmitter (possibly coupled with a active repeater at the panel) to broadcast across the entire house.

 

In contrast, Insteon uses multiple repeaters to keep the signal level high, active RF coupling that can be placed at the users discretion, and a far more robust communication protocol.

 

Sounds great don't it. Well, in your case one of those links is broken. However, given the "repeating" nature of Insteon, I would not think that a 20mv receive level at the PLM would be the problem. If you ever had X10 operating in you home (I believe you did) this just doesn't seem possible. To be short, if you're getting less than 20mv receive at your PLM you have other problems.

 

To be honest, I view a 1mv min receive level as a marketing ploy - can't be achieved in nature. The AGC (SH has told me the switches have AGC) circuit will quickly dial the min receive level upward above the ambient noise.

 

I remember some posts indicating that you had run tests with a ELK-ESM1. Did you check for receive levels at your PLM? I just did, I'm getting roughly 2V as transmitted from a second floor switch (repeated of course). The min receive level registered by the ELK is 100mv.

 

I don't think I've ever asked -

1) How old is your home?

2) Copper Romex or Aluminum?

3) Single or multiple panels?

 

So, I've written a bunch and haven't given any suggestions.

1) Check your panel for loose/oxidized connections. NO MULTIPLE connections on breakers or neutrals.

2) If you can isolate to a circuit, check box terminations.

 

If your into the extreme, here's a link (cleaning breaker panel connections) -

http://idobartana.com/hakb/x10alignment.htm

 

Let us know how things are working out,

IM

 

My breaker panel is 10 years old and all wiring is 10 years or less old since the entire house has been rewired (originally about 6 circuits and now over 20). All wire is 12 AWG copper romex with the exception of a few 10 awg and of course the stove line (cant remember the gauge).

 

Yes X-10 worked fine here. Insteon works fine here as well from device to device )I press a buttom on a keypadlinc and the scene works). PLC works fine here. PLM does not work fine here. I have a passive coupler (Insteon) at the breaker panel. I have 4 accesspoints.

 

I have not had time to add another circuit to the panel as I have been remodeling the master bedroom the past few weeks. Almost done (one more closet door, 2 heating vents, etc etc).

 

Using a switchlinc, an outlet, and a radio as a test fixture (where the PLM is currently plugged) I went to about every outlet in my house with a controlinc and I was able to control that switch.

 

 

With the ELK meter the signal level is pinned most of the time.

[IndyMike]

Digger,

I'm back on board again. I had forgotten some of the details of your long and winding road.

 

I find it strange that you're seeing a >5v indication on your ELK. I've tried this with the PLM, PLC, CM15a, and individual modules. My levels come back a ~2.5V (actual level should be 3.2V).

 

I did encounter one situation where I received levels around 5V. It was when my Boosterlinc was fooled into boosting Insteon signals. This only occurred during times of "high Insteon activity" and only after my system had grown past 25 devices. Replacing the Boosterlinc did not correct the issue. Failures were sporatic and not limited to any particular electrical leg or phase.

 

Using the PLC, I can't create the same situation. I'm using powerhome and don't have anywhere near the complexity that I have loaded into the ISY/PLM. Hence the level of traffic is reduced and I don't see the problem.

 

I don't know if anyone has performed a "throughput" comparison of the PLM VS PLC. I'm raising the possibility that the PLM is capable of putting more information on your powerline in a given period of time, thereby showing up a problem that can't be observed with the PLC.

 

Sorry - still no solutions,

IM

[Guest]

Digger,

I'm back on board again. I had forgotten some of the details of your long and winding road.

 

I find it strange that you're seeing a >5v indication on your ELK. I've tried this with the PLM, PLC, CM15a, and individual modules. My levels come back a ~2.5V (actual level should be 3.2V).

 

I did encounter one situation where I received levels around 5V. It was when my Boosterlinc was fooled into boosting Insteon signals. This only occurred during times of "high Insteon activity" and only after my system had grown past 25 devices. Replacing the Boosterlinc did not correct the issue. Failures were sporatic and not limited to any particular electrical leg or phase.

 

Using the PLC, I can't create the same situation. I'm using powerhome and don't have anywhere near the complexity that I have loaded into the ISY/PLM. Hence the level of traffic is reduced and I don't see the problem.

 

I don't know if anyone has performed a "throughput" comparison of the PLM VS PLC. I'm raising the possibility that the PLM is capable of putting more information on your powerline in a given period of time, thereby showing up a problem that can't be observed with the PLC.

 

Sorry - still no solutions,

IM

 

The meter is pinned when I do a query etc. with one or two exceptions. Its really not designed to measure the Insteon signal so its not accurate.

 

In a few days I should have time to put the PLM on a dedicated circuit next to the breaker box. We will see if there is any improvement.

 

For the moment its not a big deal since everything basically works except at times the status lights on the KPL's are out of sync and the GUI in the ISY is out of sync. That and a ton of comm failures popping up in the GUI when a query is done.

 

But the device to device communications seem fine. I press a KPL secondary button and the lights work etc. (I only have 8 KPL's installed at the moment if I remember correctly) and 3 way circuits work fine as well.

[upstatemike]

If you have a plug near your panel maybe you could simulate the dedicated circuit using an extension cord to see if anything changes before doing a big rewire job.

[Guest]

I reread the "White Paper" today and noted that the min sensitivity for power line communications for an Insteon Device is 10 mv. The PLM has a spec of 20 mv. That in part is probably why people have comm failures.

[linuxguy]

I have just finished browsing through three pages of this thread. I noticed at least two or three places where Steve L was asked questions (by name). I found no answers from him. Perhaps I missed them. If he is the moderator of this forum, why doesn't he pay attention and respond to questions?? About four months ago I made a post in another thread (quoted in this thread), asking him a number of specific questions. Dead silence. What are we to think of SH?

 

Linuxguy

[Guest]

I have just finished browsing through three pages of this thread. I noticed at least two or three places where Steve L was asked questions (by name). I found no answers from him. Perhaps I missed them. If he is the moderator of this forum, why doesn't he pay attention and respond to questions?? About four months ago I made a post in another thread (quoted in this thread), asking him a number of specific questions. Dead silence. What are we to think of SH?

 

Linuxguy

 

I was not expecting an answer. SH policy is not to answer questions like that. They cannot admit there is a problem until they have a fix. I gave up asking SH tech support the status of the fix over a month ago.

 

I still think its interesting that the PLM does not meet the white paper spec. and I believe that is part of the problem with the comm failures.

[linuxguy]

Digger,

 

Thanks for the reply. I guess I have been naive enough to expect more from SH. Perhaps I've become spoiled by the fantastic response we get from UDI. Too bad. Its a real loss for SH.

 

Linuxguy

[Guest]

Digger,

 

Thanks for the reply. I guess I have been naive enough to expect more from SH. Perhaps I've become spoiled by the fantastic response we get from UDI. Too bad. Its a real loss for SH.

 

Linuxguy

 

UDI is saving SH. The ISY products are making up for a lot of shortcomings in Insteon both in the products and the service. The ISY-26 is well worth every penny I spent and I have been telling many people they should look into buying it. A few are waiting on the UPB version.