ergodic

Just for anyone searching this topic, in my case I found by trial and error that Kaspersky Internet Security 2016 was causing exactly this behavior with the admin console when either the "Web Antivirus" or "Private Browsing" functions were enabled. It didn't matter how I activated the ISY console, either browser or direct Java did the same thing. After upgrading to the most current KIS (16.0.0.614 d), I can leave both of those features enabled, but I have to set the Web Antivirus to Security Level "Low". Private Browsing does not seem to be an issue. YMMV

I've been slowly replacing my older KPLs with the newer DB devices. For the most part these are devices that seem to have developed minor comm problems, a couple of spots with occasionally spotty comm, or devices where I just want to speed up the programming. I had one of the new ones die shortly after installation. I disassembled it just to see if there was anything going on. While I couldn't see (or smell) anything obvious, I did notice that SH is still using Fujicon RK caps which are 1000-2000 hr rated components (ie crap caps). So especially after reading some of the SH reviews on the new 2334-2 KPL, and having this one fail, I'm a little concerned about longer-term reliability of these things I'm buying. Wondering if anyone has any experience or thoughts on this?

Teken: It is about 50/50, new vs. old devices. But the problem happens on both the older and newer DB controllers. All the lighting loads are halogen incandescent. This is definitely not a "flicker", and it isn't limited to just to the initial switchover. It continues as long as it's on the generator. I guess I would term it a "flash off". Maybe about 1/4 second off, then back on. Sometimes one flash, sometimes 2 or 3 in fairly close succession. It is peculiar (to say the least). I used to have the PLM on the UPS But as I've replaced controllers with the newer DB units, the performance seems good either way, so it's been cabled over at the panel. I did move it back on to the UPS for testing while the generator was running and it made no difference. I'm going to replace the UPS tomorrow since it's the only thing I can see that's different between utility and generator power. The generator output is just as clean as utility power. (There is a minor and expected frequency wobble when on the generator of about .5hz, but no reason that should matter.)

I'm posting this in case anyone has seen this behavior - it has me baffled. I have an Insteon-only system with about 100+ controllers. It all works fine on mains power and no comm issues. I also have a whole-house 20KW Generac for when the power goes out. Which it did on Friday. When the generator was running, the lights on the Insteon controllers would flash off for a split-second, and then back on. Sometimes once, often two or three times in rapid succession. They will occasionally do it on their own, but I can make it happen about 75% of the time by toggling any of the controllers off or on (whether that controller is for that light or not.) Now I watched the Generac power with a DVM, a digital scope and a spectrum analyzer. Absolutely nothing is going on with the AC when this happens. Voltage stays rock-steady with freq. from 59.5Hz to 60.2Hz . The generator is supposed to be pure sine wave, and indeed there seem to be very little 60Hz harmonic and almost no energy at all in the 100KHz and up area where the Insteon signalling would be. There is a tiny spike about halfway up the waveform (about 2V), but I also see that same thing when on mains power, so it's a bit of backfeed from something in the house or whatever. So I'm left with -- it has to be the controllers or the signalling or something that is only happening when the generator is active. But what? The network equipment and ISY are on a UPS to avoid power loss during the generator transfer switchover, which takes about 15 sec. It seems to be working perfectly fine - the battery is only about 1yr old and it self-tests fine. But since it's the only difference I can think of I've ordered a replacement anyway. I'm just wondering if anyone has run across anything like this??

LEDs and the dimming electronics that drive them require a minimum voltage and available net energy to be operational. So until the 120V dimmer gets up to at least that point on the A/C curve, an LED is going to stay dark. So dimming for LEDs represents something of a compromise to the existing deployed technology of 120V edge triac dimmers. One way you could work around this is to "bottom-set" a dimmer. Something that's done all the time in theatrical lighting. Unfortunately Insteon dimmers offer a top-set ("on" level), but no bottom-set ("off" level), though you emulate something along those lines with an ISY program - not sure how successful that would really be. The hysterisis and non-linear dimming of dimmable LEDs is a somewhat different issue and seems to vary a lot from model to model. What's needed are pulse-width dimmers. These would work for both incandescent and (non-dimmable) LEDs. But you also then run into a problem that the electronics imbedded in most "dimmable" LEDs are not compatible with PWM dimming. It's a complicated situation. It will take a few years to sort out I suspect.

It works for Ergodic Double fixtures is an option I suppose, but it increases the visible 'clutter,' and with high-quality fixtures would get quite costly. I'm not sure I see how it simplifies installation since you have double the number of connects to stake.

There's a "LED Brightness" button at the bottom of the screen when you've selected any KPL button. It brings up a dialog to set the on and off state brightness levels. You can set it in steps from 0..7. You can also set the backlight level in a program.

Bummer. I would first try a second restore on the new PLM to rebuild its link tables. Be sure to do a factory reset on it first. [To do that, unplug it for a minute, hold the set button in while you plug it back in and wait for the long beep.] Then do a restore PLM from the ISY. This should go pretty quickly since the hardware address isn't changing and so the ISY knows (or thinks) it doesn't need to rewrite all the device PLM links. Assuming that restore runs without errors, try a compare links with a few of your marginal devices to see if their link tables match up with what they should be. If not, do individual restores and see if that gets them working. That will show if the PLM is able to communicate. If it still doesn't work, and we know the network itself was OK (you didn't change anything else, did you?) -- I'd suggest you might have a bad PLM. It happens. Get a warranty exchange and/or recap your old one if you are able to do that.

For incandescent lighting I just use the Malibu magnetic transformers you find at Home Depot, etc. I oversized them about 50% and feed them off Switchlincs. (Although I don't think the SLs are officially spec'ed to be balanced, they're fine for this.) If you use the Foster transformer you won't find any noticable hysteresis at the low end of dimming curve. The output waveform from it is a sort of weird, spiky PWM, so it dims LEDs uniformly, which of course is what it is designed to do. I use the warm white LEDs and honestly you can't tell any difference between them and a regular incandescent bulb on a dimmer. I've done nothing special as far as dimming programs etc.

I have dimmers on all my landscape lighting. The newer Insteon dimmers seem to be reasonably symmetric and don't put out much DC bias. So for the incandescent stuff I just used regular landscape transformers (admittedly somewhat oversized just to be extra-cautious). In 6 years I've never had a problem with any of it. When I built a BBQ area on my patio a couple of years ago I put in all 12V LED, and for that I put a couple of Foster 16192 transformers which allow LED dimming with standard triac dimmers and it works great. At the time that was the only unit like it available, but there may be others now. If you scope the output waveform of the thing it's really weird, but it does work well. I like having dimming on the landscape, it looks really nice. I keep the lighting low (about 50%) and use a motion sense to bring it up full when there's activity. FWIW, if you're looking for LED and incandescent landscape fixtures, I can highly recommend the Volt stuff from landscapelightingworld.com. The stuff is all solid bronze, very high-quality, they have a huge variety with modular stems, and at about 1/3 the price of the equivalent from a fixture store. (I'm not associated with them in any way, just my 2 cents.)

You've asked a great question. And the answer depends on what you're expecting. I gutted and remodeled a house in 2007 before we moved in. Everything electrical in and around this house is now Insteon. I have 82 active devices in the ISY console, as I've added on over the years. And for the most part I'm very happy with it all. (The ISY is the only thing that makes this workable BTW.) I just ran a report and for my Insteon/ISY stuff it totals about $7K. Included in that are some spares, occasional replacements over the years, and some fooling-around-with items. By comparison, I could have installed, say, a Crestron system. Of course it would have cost about 10x that, taken up a wall in my garage, required special wiring and I'd need an authorized dealer to do much of anything with it. There have been some reliability bumps. I had a batch of controllers with bad firmware that SH replaced under warranty, though they weren't able to replace all the time I spent installing them, troubleshooting the endless weird problems, pulling them out and putting the new ones in. I've had 3 PLM failures, and I now keep a recapped spare on hand. I've had a few KPLs that inexplicably died. And I've just now been replacing a few old KPLs that have become unstable with poor communication. The new dual-band units work better anyway. My overall impression is that the newer products will turn out to be much better on reliability, just based on what I see internally and with my own testing, but I'll really have to reserve final judgement on that for a few years. I will say that I have a whole-house generator that came on one night during a power failure and after a couple of hours went nuts due to a bad stepper motor. Lights dimming and brightening up all over like a haunted house. 48-65Hz on the scope, and voltage all over the place before I finally just shut it down. Not only did it not cause any controllers to fail, but the entire system continued to work just fine. (I do have the ISY and PLM behind a UPS). The suggestion to invest a few dollars in trying out the various technologies that are of interest to you is very good advice, and you may find through testing that the hybrid approach works best for you. But by and large, and trying to be objective, I believe that the ISY/Insteon solution remains the best overall 'DIY' solution. The larger question is whether Insteon as a proprietary technology will thrive. That's a tough one, and of course it's important for a lot of reasons if you're making a big initial investment. As long as the HA market remains fragmented, they'll do fine. But as things consolidate where Insteon will fit in is harder to predict. I think they're doing some of the things they need to: updating the product line, better design, revamping the developer program, improving quality, signing up big retailers etc.

I've been upgrading quite a few devices (KPLs and SLs) recently. I've not run into any big issues with replacing an older model device with the new equivalent device (in the 994ir/pro running 4.0.5) But I have found that quite frequently the ISY will report during the replace that it "lost communication" with the device or a similar error. Restarting the admin console, requerying the device's status and then doing a full restore always seems to get it working fine and I see no comm errors after that.

If you have fluorescent lighting in the garage then you probably need to put them through a filterlinc. If the garage is on a subpanel then you may need to do something there. If not, I agree with the other posts: try range extender(s), put one near the start of the branch that goes out to the garage. If it still doesn't work then even try two with the other in the garage itself. Whatever, since you have a reasonably new dual-band unit and you're having comm issues then it's either poor signal, signal suckers, or line interference from something, and you really want to find out what that something is. No amount program finesse will really ever fix it. Assuming it's only the detached garage it shouldn't be too hard to sort out. It really helps to have a laptop or smartphone with remote access in order to access the ISY console while you're out there testing. Another idea might be to try setting MR=3 inside the ISY configuration via telnet as a test (this sets the max number of ISY retries.) I suspect UDI will say - with good cause - that's not a great idea, but that's how I have mine configured and it does help with some issues (in my case mainly getting successfully through endless restores on the older non-DB devices when there are a lot of links.) As a last resort, you can of course write a program with no conditional clause and a "repeat x times" that does a query & wait, and call that from your other program. But from experience I can tell you it'll never be very satisfactory.

No fan then no switch . A lot of gas fireplaces (like mine) have a fan unit built-in to blow the air out. I'm not sure I see any role here for anything like a 2477D or 2477S, or for that matter any 120V switching. But maybe I'm missing the idea. The best, simplest, and most certain solution I believe is the battery thermostat like a z-wave, that can be linked to the HA system and won't require line power. You could put this in a nearby recess box with a matching cover and it really shouldn't be obtrusive. A lot depends on the construction and finishes on the wall and your talents in that area. The next best solution I think is to an IOLinc or simplehomenet L/V I/O relay unit and put it in a recess box with a cover and wire power for that to the outlet and put the igniter through that relay. I don't know the distance limits on the igniter line, but given the very small voltages I suspect it isn't much. So the basement thing if that works better and doesn't get the leads too long. Plan to dry-test whatever solution with cords and such before committing to tearing into the walls. All this requires unpleasant drywall/paint work, and if you go the relay route you need a relay unit where the contacts are isolated from mains neutral, and I don't know the internals of either of those devices in that regard. But I'm sure somebody here does. Also I need to just note that you can't run the LV igniter line (a "class 2" circuit) within a live 120V conduit or into a standard J-box with 120V (a "class 1" circuit). That's code and there's a good reason for it.

Likely you have a thermocouple-driven gas spark igniter. Same sort of thing used in gas water heaters, it allows it to work even if power is out. These work on very low voltages and don't have anything like the energy needed to power any sort of relay or switch . They also have to be isolated from mains power and everything else. Do you possibly have a fireplace fan switch close by?

That seems unlikely. More likely is you could be seeing the same failure mode as with your PLM at about the same time frame, or you have corrupt link tables somewhere. Though I personally wouldn't put any money on any of that just yet. I'm not reading that you've actually replaced your PLM. Until you do there's absolutely no point in doing any ISY-based diagnostics that rely on it. Madness lies down that road. And no re-linking or restoring either. A wonked-out PLM can corrupt link tables in device controllers. A depressing syndrome I've witnessed first-hand. There also was one wretched iteration of switchlinc (V.35?) firmware that had the same magical capability to corrupt unrelated controllers at random. And corrupt link tables anywhere naturally open the door to all kinds of irrational mischief anywhere else. (Attention SH: how about a damn CHECKSUM?) You can of course manually reset and link device controllers to test their basic function with no PLM or ISY. (If you do that you'll need to restore them once you get your new PLM.) If you do a hard reset and manual link and they still don't work, very likely something's failed. But as I said I don't think you can lay that one on the PLM. (Almost anything else. Just not that.) I would probably also do a few device link comparisons after you replace your PLM - you may be surprised. Or shocked.

Before giving up, you might double-check your replacement capacitor polarity. It's easier to get wrong than you might guess. The board has + marked for each. If you do find one backward, it has to be replaced.

The rule of thumb for wet electrolytics is every 10°C decrease from the max operating temp doubles the lifetime. Ripple current and operational voltage also play into it. If you really need the thing to last forever (more or less) use film caps for the lower-voltage ones, and an MLCC or organic for the 250V input filter. But that will cost. I wouldn't have quite as big an issue with PLM failures (I'm on my 4th now) if SH would just make the @!#&!!** hardware address user-settable or override-able in some way.

I recapped a PLM that died (and now is reborn), and I just bumped into this thread. So I thought I'd add a couple of notes: 1) I drilled 1/8" holes in the bottom and back top of the PLM. Heat is the big enemy of wet-dielectric capacitors (good or crappy). And mine sits at the top of my garage where summertime temps occasionally get over 100F. Hopefully the venting will reduce the inside case temp a little. 2) My PLM main board was covered in rivers of crusted, dried-up dielectric that had leaked out. I suggest wiping down the board with denatured alcohol. From all the residue I saw I'm wondering if this another instance of the so-called "capacitor plague" of the last decade. Even for crummy electrolytics 2-3 years is too short to start leaking like this, and also the failure timing people are seeing seems suspiciously consistent.

What I think - and hope - is the last word on this: my PLM was definitely defective. I replaced it this afternoon (a V99 with a new V9B) and was then able to run the ISY "replace PLM". A few devices failed to respond on the PLM reprogram, all were on the same side of the powerline. So I replaced the access point on that leg and everything was 100% again. Still no clear idea what happened here though. Obviously it's too much of a coincidence for the PLM to have failed at exactly the same time I did the ISY firmware upgrade. But it also seems unlikely that an ISY firmware upgrade by itself could cause a PLM issue. So, having had numerous problems with PLMs in the past I'm leaning to the idea that the two aren't directly related: that the ISY firmware upgrade failed for whatever unknown reason, and then one of my efforts to power-cycle or reset the PLM to correct that induced a PLM failure. But that's just a guess. I've also noticed that response time for wireless events like remotelinc clicks is noticeably faster now (my wife noticed it also). Don't know if that's the new PLM or the AP.

As a follow-up, I've discovered I still had a problem - none of the scenes were working. I've tried to reset/restore the PLM several times but all the scene tests still failed. And now after several resets/power-cycle/restore efforts I now seem to have a system that can't even read out device links. At this point I have to assume I have a failed PLM. Possibly the upgrade exposed a problem with it I wasn't seeing. Or it created a failure mode in some way, though I have no idea how that would be possible. Anyway I'm going to get a replacement today and see if that fixes this. Not looking forward to a restore with a different PLM address, but not much choice.

I'm using http://<lan-address>/admin , but the UDI URL gave me the same issue. After a long afternoon of randomly rebooting and power-cycling the ISY and the PLM in various combinations I was finally able to get into a working admin console. The ISY showed no status for any device, even though I no longer got the "safe mode" warning dialog and the ISY diagnostic showed it was connected to the PLM. I took that window of opportunity to restore 4.0.5 firmware back and everything is working fine again. Think I'll stay there for a while. Thanks very much for the suggestions. Wish I had more information to offer. My guess (and only a guess) is that the PLM got itself stuck in some weird condition as a result of something during the upgrade. Looking back, possibly I needed to leave the PLM unpowered for a minute or two to fully discharge everything - something I never specifically tried. One interesting note is I'm now able to link the leak detectors just fine. Go figure.

I got two leak sensors and was trying to link them. I put one into link mode, ran the new device link and the ISY would send three TX packets out with no response and then report it couldn't determine the engine version. Both sensors did the same thing. So I decided to try upgrading from 4.0.5 to the current firmware (4.2.10.) That backup/upgrade went fine and I cleared the Java cache. But when I logged in I got a pop-up that the ISY was in "Safe Mode" and unable to communicate with the PLM. After a couple of tries I went and power-cycled both the PLM and the ISY. So now when I start the admin console, I just get the three-tab blue admin screen with nothing on it, and no login prompt. What now?

It sounds like something is triggering the micro on/off - the scope will show the waveform so that's probably going to be the most helpful.

Have you scoped the voltage to see what's happening? Normally the Generac is clean power and I'm wondering if there's maybe a problem with the little micro or noise on the circuit.