Insteon now or wait?

Bill,

I'm a long time X10 user, and also use the Ocelot. We have our share of electronics and compact fluroescents, but have virtually 100% reliability. It does take some work to keep a "friendly" X10 environment. Filters must be used on major signal suckers and certain brands of compact fluorescents. If you have a large house, you might consider the XTB. It added a significant signal level margin here, and I can't recall any X10 failure since installing the first prototype.

Jeff

I would wait on Insteon, at least until they get their flicker problem fixed (supposedly it is fixed, but the fix is not yet being manufacturered).

If you KNOW it is due to your CFL lights (and this should be easy to confirm), then the only solution is to install filters (or use a different brand of CFL).

In addition, you could install an X10 repeater (Smarthome has several different solutions for this that are pretty good).

Consider these $$-free preliminaries:

1) Review your Ocelot CMAX code for bugs/unintended instructions 2) Check under the sofa cushions for misplaced remotes ;-) 3) Make a schematic (= map; = circuit diagram) of your home's AC wiring.

With an accurate circuit diagram in hand, you may find that you can partition your powerline control (PLC) devices (X-10, A-10, INSTEON, UPB) in such as a way as to

1) isolate and test issues such as 'incompatible' CFL's 2) improve performance by finding signal-degrading components. 3) isolate X-10 on one phase and (eg) INSTEON on the other.

Since you presumably have non-dimmable CFL's, you can safely disregard the chatter in this newsgroup about flickering problems if you switch to INSTEON wall _switches_ (not dimmers). The switches use relays, not semiconductors and don't flicker. They are in stock and cost no more than $24.99 each

So if you start by replacing the CFL's with INSTEON switches, the discussion about INSTEON dimmers is not a reason to wait (IMO). It should also be noted that if you limit the use of the existing supply of INSTEON dimmers (which come with either two- or seven-year warranties in any case) to 200 watts or less, based on all the evidence I've seen including Smarthome statements, and on my own experience, you won't have flicker problems.

HTH ... Marc Marc_F_Hult

First thing to try is Jeff's XTB. It may be the best solution. If the CFL's are at fault, try a different brand and/or install in-line filters. I have switched to Insteon and am using Smarthome's 2414X translator which is a beta model that never went into production. While the Insteon signaling protocol is far more robust than X10's, it is much more cumbersome to setup and manage. If I were in your home I'd stick with X10 and an XTB till a better translator is available, which, BTW, is coming soon from Dave's workshop.

I think you need to re-read my posts re Insteon. I was the first person to post a "hands on" review of Insteon to CHA and I gave it very good marks (but with the disclaimer that SmartHome and the engineer who designed Insteon have poor track records). I have reason to believe that Bruce and some of the others who have tried Insteon have done so based on my positive review.

I still have a positive opinion of the Insteon concept and technology but also still worry that SmartHome and the designer will screw it up by cutting corners to cut costs. The current flicker problem experienced by some users looks like they may have done just that.

I'm currently working on a DIY translator that will do what you want (and much more). It is probably a couple of months away.

However, I'd suggest you first find the root cause(s) of your X-10 woes. Insteon will have trouble with CFLs that output noise and will also be bothered (perhaps to a lesser degree) by the same signal sinks. The Insteon carrier is 131.65kHz as opposed to X-10's 120kHz so they will be affected by the same things. If you don't already have an ESM1 meter, my advice is to get one. It can eliminate most of the guesswork when troubleshooting.

As for hardwired motion sensor controlled lights acting up, that would seem to indicate your problems are not all related to X-10.

As Bruce can tell you, once you reach a critical mass of Insteon devices, your X-10 performance will deteriorate because the two-way Insteon devices act as signal sinks to the X-10 signals.

I'd wait until the flicker problem is resolved.

"Bill Stock" wrote:

Yes, but the decision not to turn on is controlled by some Ocelot code and a variable. My first thought was a errant clock, but the Ocelot checked out OK. So I reloaded the Kernel and reprogrammed anyway. It seemed OK for a day or two, but it's acting up again now. Strangely the TOD variable is being set properly.

I've tried a couple of different brands, with no difference. I've been tempted by the inline filters, but I'm not too sure they will fit in the junction box. I've also got filterlincs on all the plug-in fluorescents and UPSs.

I already have an XCPR, but I'm not sure that it isn't part of the current problems. eg. Send G2 Off get G2 Off AND A2 On.

Are you using more than one housecode in your house? It sounds like you're having collisions. There are lots of potential causes, but the ActiveEye series of RF PIR motion detectors seems to head the list. Are there any X-10 devices that can transmit at the same time in your house? Have you reviewed your transformational (IF X, then Y) Cmax code for errors? (What joy!) One of the most pervasive problems with CMax seems to be "dropping through" the ladder logic so that conditionals are not properly evaluated. It's not a fault of the system, but of people bringing expectations learned from other programming languages to CMax.

Elsewhere you say you have an XPCR repeater. Does the problem still occur when you power this unit down?

I doubt the X-10 mode of the switch will perform any better than what you've got. You've got demons in your house and only a meter will properly excise them. I'm betting you've got at least two or three completely undetected but severe signal "sinks" going. They can be as small as a plug in AA battery charger and as big as a whole house AC unit. Only a meter will tell you the extent of their impact on your X-10 setup. It will probably enable you to redeploy your existing filters to items that really need them. Not all fluorescents or UPS's are X-10 unfriendly.

If you filter some of the massive sinks, the rest of the system improves immediately. Throw an XTB or two into the mix, use X-10 friendly CFL's where you can't install series filters and I'll bet you'll see a dramatic change.

They'll work their kinks out - eventually - but for now, you're probably better off trying to fix what you have since it seems that you have some serious money invested in X-10. Give Insteon some more time to work the kinks out and to get the redesigned switches approved and out into the field. In the meantime, an ESM-1 meter and Jeff's XTB will do a lot for you.

Even with those two tools, it may take either $100 worth of filters or finding non-interfering CFLs to clear up the issues caused by the CFLs, depending on how many you have and how they're wired.

X-10 *should* sell X-10 friendly CFL's on their site. I'm surprised HA vendors that frequent CHA have not bought an inventory of X-10 friendly CFL's to resell. It would be nice if someone designed a simple "button" screw-in filter that you could put between the bulb and the socket that would fix errant CFL bulbs, X-10 wise. But X-10 friendly bulbs *are* out there. They're just not 5 for $10 at Walmart. :-(

I believe it appeared briefly in beta form which could lead one to believe the operational problems were so enormous that development ceased. Something spooked them, although it could be as simple as a prediction that demand would not cover manufacturing costs.

I'll leave it to Dave to describe all the potential gotchas making two less-than-perfect protocols behave together perfectly.

-- Bobby G.

That sounds an awful lot like CMax, especially if you're using both those addresses in your program. The XPCR shouldn't add codes. The worst it should do is garble a command. I'll bet if you posted your CMax code to Guy Lavoie over at ADI he'd figure it out in a heartbeat.

-- Bobby G.

Housecodes A-Primary, C-Stanley Garage door sensors and G-Secondary. Don't use any RF motion sensors. I found the ActiveEyes useless, but the DM10s weren't bad at all. My first thought was an Ocelot (code) issue, but this same code has been running without issue for three years. The Ocelot code is pretty much restricted to the hardwired motion sensors and I use HomeSeer for the more mundane stuff, because it's so much easier to set up.

Did a walk test of the sensors today and the lights did not come on, which is as it should be. Although they seemed to come on in the past with visitors arriving (anecdotal). It could still be an intermitent problem. I've unplugged the RF transmitter temporarily to see if that's the culprit.

Haven't tried it yet, but I have considered it. The weird thing is that this does not happen when the event runs in HomeSeer, only when I launch it manually from HomeSeer.

Yes I got FilterLincs on the aquariums and AF-120s on the larger UPSs.

My XTB s/b in the mail soon I hope. :)

I've had an ESM for a while and recently added a TesterLinc. But I can't say the TesterLinc showed any noise, so it must be a sink issue.

I was wishful thinking about the button filter today. But I imagine it would be difficult to make it thin enough.

Things that work fine with one or two people moving around the house often get dicey with more. Whether that's because they block RF paths that are normally clear, activate things simultaneously or what I couldn't say for sure, but I had similar problems with the whole house becoming haunted when it's filled with people. Not as embarrassing as flaming birds dropping down the chimney during Thanksgiving dinner, but embarrassing none the less as bathroom lights kept going on and off without reason - while occupied.

Sounds like it's something in HS otherwise it would happen all the time. I'm betting a variable is properly initialized in one case, but not in the other. I remember discussing building an LED panel that shows some of the more basic assumptions HomeSeer was making as well as switches to correct any misimpressions it might have (i.e., thinking it's day when it's really night or a weekend when it's really a weekday) or to temporarily override system settings.

Do you know which outlets or lights are at the end of the circuit branch? I'd set up a test transmitter and take readings at each terminus to see if one of them was way lower than the others although I think your unreliability is less signal strength related than I previously thought.

It will certainly make a difference. I still marvel out how much it boosted the overall reliability of the system.

One bad thing I discovered in trying to hunt down sinks is that some of them are intermittent. My cheap UPS was fine while idling but blew out that entire circuit when it was charging, emitting both horrendous noise and sinking the signal. When the float charger kicked out, it was fine again. I had let it charge up before I tested it for X-10 nastiness and it passed. I'd take the Testerlinc, mount it on a 100 ft cord, set up a test transmitter and then go to the basement and do the sequential shutdown test. That's how I found the drill charger, the ultranoisy UPS and a few other felons. It *should* be a lot easier to find X-10 killers, but I haven't found a better way yet. I have some projects in planning to at least monitor the system so I know when something recently acquired causes a profound loss of signal but they're not top priority.

Who knows? They've come out with SAW (Surface Acoustic Wave) devices that are incredibly small compared to the components they replace. It would be nice, though, to have a button filter, especially for situations like yours where there's no room to easily install a series filter. In that case, I think all that's left is to play Indiana Jones and hunt down compatible CFL's.

-- Bobby G.

No I have never mapped things out. The house is about 30+ years old, so there are too few circuits for today's toys. I swear that half the upstairs is on one 15amp circuit. Plus the house was added on to, so that an outlet at the back of the house may be on the same circuit as the lights at front of the house. It's also a mixture of original Aluminum wiring and Copper in the basement subpanel and additions. I suspect my TW523 is in a bit of a backwater, as it did not work at all in one outlet in the same circuit. It would likely help to move the TW523 closer to the fuse panel, but I'm not sure if there is any limit on the distance to teh Ocelot?

OK, I will!!!

Error

Please visit another page.

Contact a representative today.

That can't be good....

The page cannot be found The page you are looking for might have been removed, had its name changed, or is temporarily unavailable.

--------------------------------------------------------------------------

Please try the following:

a.. If you typed the page address in the Address bar, make sure that it is spelled correctly.

b.. Open the

home page, and then look for links to the information you want. c.. Click the Back button to try another link. HTTP 404 - File not found Internet Information Services

--------------------------------------------------------------------------

Technical Information (for support personnel)

a.. More information: Microsoft Support

That's a real confidence builder..... maybe if you charged crazy prices you could have a website that actually worked.

Regards,

Neil

Their site works OK from here.

Works for me, to, but I wonder why a Noo Yawk vendor is talking about UK X-10 model numbers:

-- Bobby G.

[snip]

I think I need to respond directly to the charge that I'm an Insteon "naysayer".

When Insteon first released a description of their planned technology, I reported (I think I was the first or second to mention Insteon here.) that it sounded very good but, given SwitchLinc's track record, one had to wonder whether they could implement it as planned. The main attraction was that it was a two-way protocol and that all commands were acknowledged (positively or negatively) which makes for a robust system. I believe I wrote something to the effect that if they could pull it off it would become the system of choice for low to mederately priced home automation.

When they first started shipping Insteon (about a year later) I asked a dealer to loan me an Insteon Starter Kit so I could review it. By and large, my X-10 gear has been fine and I think I know how to keep it working that way so I had no plans to switch to Insteon. After a couple of weeks of hands on testing, I published a series of reviews/comments here. My reviews were positive and accurate. But I again questioned whether the same people who had designed SwitchLinc could implement the Insteon system without problems similar to those that had made the term "SwitchLinc" anathema in CHA.

I recommended that anyone just starting with automation should take a serious look at Insteon. I did not recommend that those who already had a large investment in X-10 (and presumably had learned how to keep it functioning) should switch to Insteon.

If you search CHA on "SwitchLinc address", you will find reports about them losing their X-10 address and other major flaws that go back to 1999. I never bought any of their earlier products but did order their PowerLinc

1132B as soon as it was announced. I found it overly sensitive to low level out-of-band noise. I had lengthy email correspondence with a manufacturer who was trying to use it instead of a TW523 (in quantities in the thousands) who saw exactly the same problems. Later, I ordered their LampLinc 2000STW as soon as it was announced (incurring a bit of heckling in CHA from Tom Morgan at Worthington). I found it was also overly sensitive to low level out-of-band noise and would also lose its X-10 address during power glitches (i.e. brownouts).

Some of the current problems being reported by Insteon users sound hauntingly familiar. Many (e.g. losing an X-10 address stored in EEPROM during brownouts, sensitivity to out-of-band noise) are obviously design flaws that should have been fixed long, long ago. Given this history and given the costs (and hassle) associated with replacing Insteon switches, my current recommendation is to wait until the flicker issue is resolved.

But my view is still that the concept and protocol are solid and I hope that SmartHome will get it sorted out. If I did not think the concept valid I would not have spent the past few months trying to develop an embedded DIY project that will (among other features) translate between the TTL (TW523 protocol) output of controls like an Ocelot, JDS, Stargate, HomeVision, etc. and Insteon, UPB and other PLC interfaces.