IndyMike

Congratulations wrj0, I believe yours is the first report of a Rev 1.1 sensor reporting its low battery status. Curious that your sensor date code indicates that it's actually older than my unit(s) (0909 date code). Can you post what type/brand battery gave you the "low battery" confirmation?

6 Button KPL Dimmer Power Consumption (ES model Beta unit with I2 and "beep" function) Configuration 1) Load unused (capped) and turned off (off LED is illuminated) 2) Device not linked (factory reset configuration) Instantaneous measurements: Voltage - 122 V Current - 0.02 A Power - 2 W Accumulated measurements Power - 0.68 kWhr Time - 339 hours Average power consumption - 2.00 W Observed Power factor - .82 Lower overall consumption and better power factor than the previous V.2d unit.

Rand, You've astounded me again. We're about the same age. Somehow I had formed the impression that anyone doing iron work must be much younger. Now I feel like a real wuss having just been informed that I managed to destroy my shoulder while working on an exhaust system (6 months ago). How in the world were you able to survive setting doors, windows, and stairs all those years? Hello Michel, That's an extremely interesting data point. I wasn't involved with the ISY in the early days. I waited for you to add the X10 capability prior to jumping in. At that point it was already fairly mature. If I understand things correctly, Rand was critical of your initial offering. You took that criticism, analyzed it and used it to improved the product. Eventually, you recruited Rand as part of the team. Gotta say, I love your methods as well as the team.

6 Button KPL V.2D Power Consumption Configuration 1) Load unused (capped) and turned off 2) Button 2: Toggle (Linked to my security system) 3) Button 3: Toggle (unlinked) 4) Button 4: Non Toggle ON (always lit) 5) Button 5: Non Toggle ON (always lit) 6) Button 6: Load off button (on) Instantaneous measurements: Voltage - 122 V Current - 0.08 A Power - 3 W Accumulated measurements Power - 0.21 kWhr Time - 88.82 hours Average power consumption - 2.36 W Observed Power factor - .24 Next up - 6 button KPL dimmer (ES model). This unit appears to have power factor correction.

Brian, This doesn't surprise me at all given your background. I actually suspect that you may also disassemble units to look for component and layout changes between revisions (I've done it). While it's interesting to look at how the modules have changed over the years, it does create a bit of a problem for returns. That's most likely why I have a number of "spare units" around. Rand, You amaze me guy. During the day you work the high iron, and then come home to mess with trains, Insteon, and software. That's pretty diverse. As a side note, I used your "Group commander" with the PLC in the early days of Insteon. It really helped me to understand the protocol and the device linking process. Thank you!

Hi Zick, Sorry for the delay - I had missed your question earlier. The LampLinc simply plugs into the UPM (or the Kill-a-watt). For the KPL, I wired a three prong power cord (Power tool replacement cord - 2 wires + ground) to the KPL. The red wire (load control) is capped. I actually use the power cord to "bench test" every unit prior to installing them in their final location. That way I know I have good links in the device and the ISY. If I encounter a problem at the final destination, it's due to a problem on the local circuit. OK, so I'm a bit anal. The first step to recovery is admitting...

Hi Mark, Your numbers on the Dual-band LL are intriguing. Since the device is essentially still active (transmitting/receiving RF) I'd expect some increase in the avg power. However, 0.7 W is quite a bit. Either we have a difference in the measurement capability of our devices (UPM vs Kill-a-watt) or the dual-band units are significantly less efficient. What does the power factor look like on these units? Edit: Just noted that the 2456D3 is not a dual band - even more interesting. My unit is an old 2856D2 Icon dimmer.

Background - the power display on the UPM and Kill-a-watt only have a resolution down to 1 watt. This isn't sufficient to display the actual power being consumed by many automation modules. When you have 50 (or 100) of these devices installed, it makes a big difference if the unit is consuming 1 or .5 watts. By performing a long term kWhr test you can significantly increase the resolution of the measurement. I've accumulated 167.33 hours of power consumption on my Lamplinc using a UPM EM100: Instant current: 0.08 A Line Voltage: 122 Vrms Instant power: 0 W Run time: 167 hours 20 min Accumulated consumption: 0.03 kWHr Avg power = (0.03 kWHr * 1000)/167.33 Hr = 0.179 Watts Uncertainty ( the 0.03 kWHr could range from 0.025 to 0.034) @ 0.025: 0.149 Watts @ 0.034: 0.203 Watts Lamplinc is a V1.0 (rev 1.2) and was run without a connected load. The above is quite a bit less than what I was anticipating for the LL (I expected something in the 0.5 W range). Next up - 6 button KPL.

Hi Ergodic, An inductive power meter (clamp style) may have problems measuring the low current draw of your Inseon devices. Also the 30% power factor number - is that a typo? Most HA devices consume power (off power) with power factor well below that (15 - 20%). If your intent is to measure circuit currents to major devices and lighting loads, these items may not be a concern.

The old electro-mechanical meters were a work of art (spinning dial meter). Fortunately, with modern microprocessors, we don't need a strong physics/e-mag background to do this anymore. Most of the current devices simply sample the Voltage and Current waveforms at a high rate. Average power is then a simple multiplication of the two over a time period (T). Note: I have seen my UPM get "confused" in the presence of voltage spikes (CFL's, etc). This normally occurs at low power consumption/near 90 degree phase shift ( PF near 0). It will sometimes alternate between 0 and 8W on my LL (0w being the correct reading). Adding a filter to the input of the UPM typically corrects this fluctuation. The Dave Houston link that I provided earlier lists measurements for a number of devices. I trust both Dave's methods and the numbers he provided. Since there wasn't much data on current Insteon devices, I've just kicked off a long term (kWh) test on a spare LampLinc I have. I'll report back in about a week. IM

ergoic, Sorry to hear about the Kill-a-watt. It should handle a resistive load easily. Don't mean to be insulting but - you aren't connecting the Kill-a-watt to to output of the dimmer are you? The UPM EM130 does appear to display power factor (My older EM100 does not): EM130 Manual

Mark, Thank you for the confirmation. My understanding was that the Kill-o-watt was pretty good in determining device consumption and PF. At one time I actually considering buying one as a backup. IM

Hello ergodic, I'm a bit surprised that your Kill-a-watt gets confused with dimmers. I had understood it to be equivalent to, or better than, my UPM EM100. Are you measuring incandescent or CFL loads? If CFL, the Kill-a-watt may have some problems with the poor power factor on some CFL's. I can say that my UPM appears to measure both incandescent and dimmable CFL loads under full-on and dimmed conditions. My dimmable CFL is a Sylvania 15W. It measures 19,52 VA and 15 W (PF ~. with the UPM. Hopefully someone with a Kill-A-Watt can comment on their use with dimmers. IM

Hi Illusion, You're absolutely correct about commercial applications. Power companies penalize "big users" with poor power factors (causes problems for their distribution systems). It's interesting that the dual band LL's are showing better power factors. There may be "incentives" for oem's to improve power factors on small devices as well. Most older CFL's exhibited horrible power factors. I was surprised that this was allowed. Obviously a "green" trade that was made at the time. My recent dimmable CFL's are showing PF's of ~ .9 so things are improving.

Hello Ergodic and Illusion, The current measurements that you've observed appear in line with what I've seen in the past. The good news is, you're not being charged for this. This is because the devices are drawing power "out of phase" - it's referred to as apparent power. Your utility company only charges you for the "real" component (in phase portion) of this power consumption. In order to perform this measurement you need a device with power-factor correction. It measures both the current and voltage (or current and phase) to compute the real component of the power. Wiki: Power Factor As an example, I have a lamplinc plugged into my UPM energy meter. Current consumption is 80 ma. Corrected power consumption is 1 W. This device is drawing current nearly 90 degrees out of phase from the voltage waveform. Here's a list of devices that Dave Houston baselined some years ago. Dave improved the accuracy of his meter through long term tests: http://davehouston.net/x10-power.htm

Thanks for elaborating. I did not realize the Insteon signal was more complex and therefore more vulnerable to noise. Makes more sense as to why I have some reliability issues. Hello Intellihome, Insteon signaling is more complex, but I would not say that it is more vulnerable to noise. Quite the contrary, much of the complexity comes from embedded CRC codes (for error checking) and the command/response/retry functions in the protocol. I believe the complexity that UpstateMike was refferring to was the actual programming of devices. This is complex and lengthy. The same protocol/error detection is used, but since the programming can take minutes and you are effectively writing to many units, there is a greater chance that it will be upset. Nothing like this existed in the X10 world. I also use X10 units alonside my Insteon hardware. Since you appear to be having problems with Insteon units in the same location as your old X10 hardware check for common mode problems: 1) Location of your PLM - If different than your X10 controller, it may be near a noise/absorber 2) Phase coupling - The Accespoints are wonderful little devices that are absolutely required for RF coupling (motion sensors and remotelincs). When used for phase coupling, remember that they can also be affected by local noise/absorbers. For this reason I prefer the X10 method of using a passive coupler at the panel. 3) Presence of V.35 Switchlincs - Michel has indicated that these can cause intermittent failures. IM

Hi Mitch, I have a number of the V.2C SWL relay units. I haven't had a problem with these, nor do I remember problems with this firmware revision being reported. They are older units, but that does not mean they shouldn't perform well in your system. I'm beginning to understand your log below, and I believe you're making progress. You appear to be using the event viewer in mode 1. Please switch to mode 2 or 3 for additional information. In your log above you turned on devices 1,2 and 3 manually from the ISY GUI. When you use the GUI to turn on a specific device the ISY uses "Direct communication mode" which requires a response from the device. When you turned on device 4 it did not respond to the ISY and you received the "unable to communicate with xx device". This may have been due to one or more of the following: 1) Turning on devices 1 - 3 created noise on the powerline and prevented device 4 from responding. 2) Turning on device 4 generated noise (additive) preventing a response. The following shows the typical command/response from the ISY to a device when using mode 3 in the event viewer. I really believe this has to do with your loads generating noise. Since you can reproduce the effect from the GUI, try turning off all four devices, then activate only unit 4. Hopefully you can narrow things down to a particular load (CFL). Be careful with your conclusions. All CFLs generate some noise. The noise could be additive. IM

Sorry, Saw the following post after I had replied. This may actually make sense if you are turning on noisy loads with your KPL (CFLs). Please try the scene test again with all of the KPL scenes off. If you get good results, you might try turning on certain loads and rerunning the scene test to get a feel for the noise environment.

Not exactly the night and day results that I had hoped for. I would have liked to have seen a significant difference between the on and off conditions. We might have been able to attribute the difference to the CFL's generating noise. Are these results typical for other scenes or is this a problem area? Are all of these SWL on one circuit, or spread across multiple?

Mitch, Thank you for the replies. I am not proficient with the Climate module (I don't have the module) but understood this to be network based. As such, I would not expect it to cause Insteon communication problems as you described previously. I will, however, defer to the ISY masters in this regard. The log that you posted indicates that the ISY is receiving the communication from you SWL when turned on. The remaining ST entries are, I believe, the ISY predicting the status of the scene based on the communication. In other words, there is only one communication and the ISY is showing the status that "should" be attained in the scene. Please try the following tests using the ISY diagnostic "scene test". You can find the scene test under the menu item Tools\diagnostics\scene test. There's a description of the test here: Scene Test 1)Try the scene test on your Porch scene with all of the devices off. Hopefully everything will pass. 2)Turn your devices on using your SWL "mini scene" and retry the scene test. If we've both lead clean lives, you'll get some failures here. I'm mostly depending on you here since my life can't be construed as clean. 3)Turn the devices off, then back on via your KPL. Retry the scene test again. Please report by your results. Hopefully we can come up with an explanation to fit the data. IM

Hello Mitch, In a word - NO. You are not violating anything by having four controllers within a scene. Please answer the following: 1) What type of load is attached to each of the four switches? 2) Is your relay unit a V.35? 3) Do any of your fixtures have photo-electric cells (light sensors)? 4) Rand's question from above - do you see a lot of Insteon communication when you activate the "Porch Light" You can view Insteon activity addressed to the ISY/PLM in the event viewer (Rand linked to instructions above). Another method to get a feel for total activity is to watch the led on a lamplinc or Accesspoint. For a four unit scene, flashing of the led for over a couple seconds would be excessive. The reason for questions above are we are trying to separate electrical noise from communication/programming problems.

Hello Mitch, From your description, I can't see that you have violated any rules with your "scene within a scene". I have numerous examples of this in my home, including a whole house scene that encompasses most of my devices. I'm hoping that you may be using a dimmable CFL in one of your Switchlincs (porch light perhaps). If so, it's possible that the scene on level or ramp rate is different than your KPL scene. Dimmable CFL's like to be turned on quickly and at full brightness. After they have stabilized (temperature and frequency) you can dim them down (30% in my experience). They will generate more noise in the dimmed condition. Check your ramp and level setting between the two scenes and at the device itself. If you don't find any differences (or aren't using dimmable CFL's) please post back and describe your scene devices and loads. IM

Hello Steve, To the best of my knowledge, scenes that are initiated by the ISY are predictive. In other words, the ISY "assumes" that the devices have responded properly and does not issue "cleanup" commands to verify the status of the receivers. This normally works well and dramatically increases the system speed. I'm not sure whether your "backup" program is any different. I believe the ISY treats devices in programs in the same "predictive" form and assumes that they have responded properly (I'm trusting the ISY experts to set me straight if I'm incorrect here). If the above is true (?), then devices that did not respond to the initial scene command should have a the same problem with the "backup program". So why does the backup work? By turning on devices you are changing the impedance of your wiring (communication system). Simply turning on a 60 watt bulb can dramatically change the noise/signal level on a length of wiring (branch circuit). In your case, you may have a situation where a "noisy" device is preventing device "B" from hearing the ISY. If you turn on device "A", a load is added to the circuit and device "B" can now hear the ISY. The opposite is also a possibility - turning on a "noisy" device on a circuit may prevent others from hearing communications. To diagnose this, try the following : 1) Scene test with all the "scene" devices off (this should produce the same 2 failures on your problem devices). 2) Scene test with all the "scene" device on (I'm hoping this will show 100% - not sure, because this is a tough test). If test 2) works, you could try issuing a second scene "on" command within your program (quick fix). If item 2) doesn't work, we'll need a bit more info including a description of the types of loads that you're activating and whether they are on common circuits. Disclaimer - V.35 devices have been reported to have problems with scenes. I have very little experience with V.35 units. If some of these devices installed, please indicate that and consult the ISY experts. IM Edit - After reading the above, it gives the impression that the Insteon communications are "fragile". That is not at all the case. I would characterize my home as average size (4500 sq. ft.) with some above average problem devices. The devices include a Grunfos variable speed well pump (documented problem for UPB), six PC's/servers (signal absorbers), 30+ cfl's, 3 a/v systems (soon to be 4). I am currently using 1 filter (Toshiba laptop), and 1 accesspoint (hardwired X10 coupler at the panel).

Sorry Brian, Although I quoted you, the comments were intended for others. I know you are a knowledgeable X10 and Volp follower as well. Didn't mean to imply otherwise.

Sounds like you're in business. Please let us know how things perform for you. The 240V PLM is a great idea for signal integrity, but it might not have much of a market. There aren't a lot of people that have a 240V plug near a lan connection. Adding an outlet near the panel is an expense that many people would not want to incur. The additional components and 240V ratings would also drive up the price. The closest item that I've seen to a true 240V amplifier are the line of X10 repeaters (X10, ACT, XTB). The XTB-II is unique in that it boosts X10 levels to ~20Vp-p. While the increased signal level is great for X10 (1 way communication) simply boosting the PLM output won't work Insteon (2-way). While an increase output level from the PLM would help with scenes, the responding devices would need to be boosted as well ($$$ again). I'm sure someone has done a cost/benefit analysis here... Given the repeating nature of Insteon, this shouldn't be required for most installations. The current work around is to use the passive coupler (with listeners/repeater in range) or the Accesspoints. I'm a big proponent of the passive coupler. Exactly correct. Jeff Volp (the XTB developer) worked to specifically ensure that the XTB line of devices would co-exist with Insteon. As mentioned above the XTB-IIR is a 240V device that wires near the panel and provides ~20Vp-p X10 on both phases. Jeff provided me with a XTB-R (120V plug in) that I baselined at 32Vp-p on my system. Since the device is 120V (single phase) a passive coupler is required. To my knowledge, this is the only plug-in X10 booster that is compatible with Insteon (the boosterlinc is not - I have data to prove it). If you can't tell, I'm a fan of the Jeff Volp devices as well... IM