ergodic Posted December 7, 2012 Posted December 7, 2012 I'm not entirely clear on your configuration, except that I think it may be a bit too "busy" to allow for diagnosing this problem. Normally a Leviton whole-house blocking coupler would only be installed at the main panel where utility service connects. I'll confess I don't understand the reason for having two. Subpanels by themselves are no issue for Insteon, but I'm also presuming this isn't 3-phase service? I believe the Leviton X10 whole-house is an active device insofar as blocking incoming X10. But it is just a passive filter to couple between the inside phases. If the Leviton is wired in then it could still generate X10 blocking offset voltages. If the hot lines are disconnected it likely can only couple without blocking, but you can't really be certain of that without knowing exactly what's inside. And I have no real idea what it may do with Insteon signals that it sees, especially if it is only partly wired in. Frequency-wise they are close to X-10, but not identical and the zero-crossing is handled differently from X10 also. Regardless the Leviton just wasn't designed for Insteon and at least for diagnostic purposes I'd want to remove them, along with any other X-10 devices, couplers, etc. Basically just try to simplify the configuration. Right now I think you have too much going on to be able to reason through the problem. Once that's accomplished you can try putting X-10 back in if you want it and see what happens. Assuming this is a typical residence with standard 2-phase service, I believe you should be able to get to a fairly reliable Insteon install using a configuration with your 2413S benind a filterlinc, just 2 APs on the opposite phases (preferably nearby to the PLM). Which panel the APs are on shouldn't matter as long as you take out all the X10 stuff and the panels are properly wire-connected without any filters, couplers, blockers, etc. You also should make sure you don't have any RF-generating devices near to the PLM or APs. This would include things like portable phones, wi-fi routers, microwaves, and the like. Nothing else really should be needed overall to make this work, except filterlincs on individual loads that either generate interfering HF noise, or HF loading causing too much Insteon signal loss on that branch or phase. These are usually pretty easy to identify.
IndyMike Posted December 7, 2012 Posted December 7, 2012 Hello nowandthen, I agree with ergodic - I don't understand the use of Leviton 6284's on subpanels. Are you trying to isolate (segregate) the panels? Additionally, some X10 couplers (passive) inverted signals across the phases while others were in phase. I'm not sure which technique the 6284's use. Insteon uses non-inverted coupling. I've played with both in/out of phase coupling with Insteon and really haven't seen much in the way of degradation - but I'm only using one device. You also mentioned the presence of X10 repeaters. True repeaters shouldn't interfere because they will try to validate a X10 transmission prior to repeating it. The chances of them being fooled into thinking that a Insteon communication is X10 is very low. There are X10 "boosters" that can pose problems. This includes boosterlincs and boosterlinc enabled modules. These devices attempt to determine if a transmission is X10 "real time". They can be fooled and will step on Insteon transmissions as a result. Since X10 transmission are much slower than Insteon, this can take out multiple Hops of a Insteon transmission. Given the fact that your RF devices (including the PLM) appear to prefer RF, a booster type X10 device fits the problem description. It would not interfere with RF and could effectively prevent your devices from repeating the RF communication on the powerline. IM
tedbusen Posted December 8, 2012 Posted December 8, 2012 Hello Ted, Glad to hear that you are having success with your (substantial) efforts. The post regarding passive couplers passing noise may have been mine. I believe in listing all the pros and cons of a device/method. Depending on the model coupler you are using, I believe that the pros of a passive coupler outweigh the cons. I've been using a passive coupler since the beginning of Insteon as my only means of coupling. While I do not recommend this for everyone, I see benefits in having the coupler with little risk of problems. Could you please post back your coupler model. There are some differences between models/manufacturers. Finally got a chance to check the couplers. Have two due to adjacent boxes. However, I removed one since the two boxes are close together. The Model number for both is Signal Linc V2 2406 H v 0171.
tedbusen Posted December 8, 2012 Posted December 8, 2012 Discovered another problem during my explorations of Insteon performance. I had 2 LL 2486d v.36 and 2 Keypads Lincs B2457D2 v.41 on the same wall of the Family room (same circuit) that would be very intermittent in adding them to scenes or other maintenance type functions. From a control and status perspective they worked 98% of the time. I finally realized I had purchased some new "night lites" that have the flashlight and charger as a two piece plug in. These are Eco-Lite multi-function 3 Led Power Failure Lights. Home Depot carries them. They use electromagentic induction to charge the flashlight. Capstone Lighting is the manufacturer. Says on the label that they may not cause harmful interference. Well, once I remove these from my outlets, everything works fine again. Seems the label may not be stating the truth Never had a label lie to me before, I feel violated. hehe Made in China, so.... Summary, if you are having issues with various devices, verify that you have isolated anything that can create an electrical noise on the ciruit. Use the filterlincs and other solutions as I mentioned in my post in this topic on Sat Dec 01, 2012 9:13 am. Long time Insteon user, but still learning something new everyday.
nowandthen Posted December 23, 2012 Author Posted December 23, 2012 Can't say that the access points improved reliability. Can't recommend buying them. I swapped out all non dual band lamp modules in the problem area with dual band lamp modules. Seems to work just fine now. Sucks that you have to replace perfectly good modules with dual bands but that's what it took to get my foyer Xmas lights to work. Maybe the access points improved communication between the 2nd floor and basement sub-panels, but can't recommend other s to buy access points. I think with enough dual band devices you will be ok. I moved my PLM down about 2 feet from where it was. It'sn a closet with a lot of other communications equipment, but most isn't active. Maybe the HVAC xfr's or the Router was interfering. Seems to be better now.
SunSentry Posted January 15, 2013 Posted January 15, 2013 Teken is absolutely right. Without access points I think you are bound to have issues. I don't believe SmartHome even supports a configuration without APs on both legs. I can't say for certain this problem is due to that of course, but you have to start there. I have two APs dropped directly down from the panel. I also have a passive coupler between them. The coupler is a leftover that's too hard to remove, but - like your experience - it makes no difference to Insteon whether it's connected or not. My PLM is near the panel, but plugged through a Filterlinc. This forces it to communicate to the APs via RF only, and that configuration has been much more reliable overall. Now, if I pull one of the APs, I get pretty much exactly the situation you describe. Communication on that side of power becomes problematic. Oddly, it's mostly the devices that are closest to the panel. (I know this because it actually happened a few weeks ago. The breaker on that AP tripped when I was working on something else and I didn't realize it.) I have a total of about 80-some devices overall, a fair number now dual-band. But still, without those two APs operating, things don't work well. For an average-sized home, 2 APs are all you need. This is rather interesting information. I have no APs or any of those types devices, yet everything is working beautifully in our 2 level, 4,0000 sq foot home. All we have are dual band switches to pass along the information, and yes they are on both legs of the home AC. Even with plasma tvs, computers and other nasty noise devices around the house, we have not been at the point where we needed anything else. Maybe we are just lucky.
nowandthen Posted January 16, 2013 Author Posted January 16, 2013 I'm not entirely clear on your configuration, except that I think it may be a bit too "busy" to allow for diagnosing this problem. Normally a Leviton whole-house blocking coupler would only be installed at the main panel where utility service connects. I'll confess I don't understand the reason for having two. Subpanels by themselves are no issue for Insteon, but I'm also presuming this isn't 3-phase service? I believe the Leviton X10 whole-house is an active device insofar as blocking incoming X10. But it is just a passive filter to couple between the inside phases. If the Leviton is wired in then it could still generate X10 blocking offset voltages. If the hot lines are disconnected it likely can only couple without blocking, but you can't really be certain of that without knowing exactly what's inside. And I have no real idea what it may do with Insteon signals that it sees, especially if it is only partly wired in. Frequency-wise they are close to X-10, but not identical and the zero-crossing is handled differently from X10 also. Regardless the Leviton just wasn't designed for Insteon and at least for diagnostic purposes I'd want to remove them, along with any other X-10 devices, couplers, etc. Basically just try to simplify the configuration. Right now I think you have too much going on to be able to reason through the problem. Once that's accomplished you can try putting X-10 back in if you want it and see what happens. Assuming this is a typical residence with standard 2-phase service, I believe you should be able to get to a fairly reliable Insteon install using a configuration with your 2413S benind a filterlinc, just 2 APs on the opposite phases (preferably nearby to the PLM). Which panel the APs are on shouldn't matter as long as you take out all the X10 stuff and the panels are properly wire-connected without any filters, couplers, blockers, etc. You also should make sure you don't have any RF-generating devices near to the PLM or APs. This would include things like portable phones, wi-fi routers, microwaves, and the like. Nothing else really should be needed overall to make this work, except filterlincs on individual loads that either generate interfering HF noise, or HF loading causing too much Insteon signal loss on that branch or phase. These are usually pretty easy to identify. Hello nowandthen, I agree with ergodic - I don't understand the use of Leviton 6284's on subpanels. Are you trying to isolate (segregate) the panels? Additionally, some X10 couplers (passive) inverted signals across the phases while others were in phase. I'm not sure which technique the 6284's use. Insteon uses non-inverted coupling. I've played with both in/out of phase coupling with Insteon and really haven't seen much in the way of degradation - but I'm only using one device. You also mentioned the presence of X10 repeaters. True repeaters shouldn't interfere because they will try to validate a X10 transmission prior to repeating it. The chances of them being fooled into thinking that a Insteon communication is X10 is very low. There are X10 "boosters" that can pose problems. This includes boosterlincs and boosterlinc enabled modules. These devices attempt to determine if a transmission is X10 "real time". They can be fooled and will step on Insteon transmissions as a result. Since X10 transmission are much slower than Insteon, this can take out multiple Hops of a Insteon transmission. Given the fact that your RF devices (including the PLM) appear to prefer RF, a booster type X10 device fits the problem description. It would not interfere with RF and could effectively prevent your devices from repeating the RF communication on the powerline. IM Sorry I missed these posts. Blocking X-10 at each panel was recommended by my local Home Automation reseller. Then bridge the panels to circuits on "the inside" of the filters. Not sure why we didn't block at the main panel??? Maybe disconnecting the neutral of the main feed made me too nervous! I will simplify my system (remove all X-10 stuff) I'm pretty much free of X-10 now. Thanks for the suggestions! Yesterday I moved my router, to improve the range. Sounds like a side benefit is getting it away from the ISY. Too soon to know if that helped with the occasional light that does not turn on or off. The problem lamp modules worked well once I replaced them with dual band lamp modules. $$$
switches Posted January 17, 2014 Posted January 17, 2014 Not sure if this helps or hinders but I have around 50 Insteon devices in my home (mixed dual-band, RF only, powerline only, and no APs) and have had few communication issues. However, recently I could not make changes to a Remotelinc 2 when it was downstairs (where my dual-band PLM and many other dual-band devices are) but could upstairs (also where many dual band devices reside). I tried unplugging anything I could think of that would cause powerline or radio issues and plugging and unplugging dual-band Lamplincs in various spots and on both phases to try to isolate the issue. After reading this post I determined that if I plug the PLM into a Filterlinc I could update the Remotelinc when downstairs although doing that did create communication issues with other devices. So, if I want to update my Remotelinc I either have to go to a different floor away from my PLM or plug the PLM into a filterlinc (forcing it to be radio only). Puzzling.
ergodic Posted January 17, 2014 Posted January 17, 2014 IndyMike has suggested based on his experimentation that PLMs are susceptible to powerline noise at Insteon frequencies causing them to basically shut down in the presence of even a small amount of that noise. It makes a lot of sense and I think he's dead-on as for this being at least a good part of the explanation for the Insteon communication "X-factor" problems. My one problem is I haven't yet been able to reproduce this myself. For one thing, I have no measurable noise in the 130KHz region showing when I run a spectrograph on my powerline (and I'm not even gating on the zero crossings) -- yet still I get noticably better reliability when the PLM is behind the filterlinc. But in then end I think most of the problem getting his results is me: probably the way I'm trying to couple my function generator and I need to build a better line filter. I haven't gotten back to it in a while, but in the meantime I'd accept his theory as the explanation. I'd be looking into why you have communication problems when the PLM is behind a filter. Fixing that might well fix your other issue. Putting APs on both legs would be an excellent start IMO.
switches Posted January 17, 2014 Posted January 17, 2014 I have many dual band devices including lamplincs on both phases. I understood all dual band devices' radios transmit at maximum allowed power so an AP shouldn't provide an advantage over a lamplinc? Am I wrong?
ergodic Posted January 17, 2014 Posted January 17, 2014 They don't have the same power level as the APs. Other people can tell you more (the older APs are actually best in that regard, though I believe the power level was likely reduced to work better with the newer firmware and dual-band devices.) Purely empircal, YMMV: I have almost 100 devices, quite a few now dual-band, but if I pull either AP that drop from the two legs at the panel, I immediately see odd comm. problems.
switches Posted January 17, 2014 Posted January 17, 2014 Seems strange that they'd make them transmit less power. I wonder if it's just the old APs that transmitted more for some reason? Out of curiosity have you tried swapping lamplinc I the same spots as your APs?
ergodic Posted January 17, 2014 Posted January 17, 2014 Haven't tried that. You should post to IndyMike - he's done an incredible amount of testing and investigation on this type of thing.
Brian H Posted January 17, 2014 Posted January 17, 2014 The older hardware revision one and newer revision two Access Points are different. Revision one units have the older Powelinc III main board and a RF daughter card. Revision two units have the RF portion on the main board and no RF daughter card. In my particular installation. From my Living Room. A hardware 1.6 Access Point will connect with my 2413S PLM in the communications tests and the newer revision 2.5 one will not even make the PLMs LED flash. I can't say if the power level is different as there are no test results for the revision one Access Points in the FCC Database. It could be as simple as different antenna position. The hardware one has a pair of completely vertical PC Board etched copper runs. The hardware two models have a pair of wires from the back of the main board. Horizontally then bent at a ninety degree bend with the rest being vertical. Some of the Dual Band Modules have their antennas in the back of the module. Not the ideal place for wall switches. The 2457D2 Dual Band Lamplinc has its antenna on the front side of the case. So it may not be as influenced by location as some of the other Dual Band Switches. In further tests I found an odd quirk. One of the original 1.0 refurbished Access Points I got in the 2442 to 2443 swap offer. Depending on when the communications test was started. Would give me a same phase flashing RED LED or Opposite Phase flashing Green LED. On my PLM from the exact same position in the Living Room.
Recommended Posts
Archived
This topic is now archived and is closed to further replies.