Sense discovered a new device of type ‘light’ that I found to be the vanity lights in the master bathroom. There are 6 x 60W incandescent lights for a total of 360W. The lights are connected to a Lutron “Diva” DV-600P dimmer. When I turn the lights on, I see about 360W on the Sense Power Meter. So far, so good. Over the next few minutes however, the power consumption rises to about 1,100W. It’s not a gradual increase in consumption, see the attached graph. Notice that it drops back to the expected level, before rising again. I replaced the dimmer with a simple on/off switch and I see the expected binary on/off power consumption, but not in the Power Meter. I’m assuming that Sense recognized the dimmer’s electrical signature not the lightbulbs.
My question is: Is the dimmer at fault or is Sense confused about the actual power consumption?
You’re looking at the Device Power Meter so that’s the Sense-interpolated consumption (waveform) for that Device and is “learned” and does not, for sure, show the actual device usage.
Look in the overall Power Meter and so how things respond there as you switch things.
I don’t know what the spec on the dimmer results in but any dimmer is going to present a different electrical signature to the Sense than bulbs alone. Most dimmers I assume (especially on incandescents) are going to be inefficient so I guess you would see increased consumption.
On a “Sense can save” energy note: If you are replacing the dimmer you could switch the 6 bulbs to dimmable LED (with the right dimmer) and probably save at least 300W.
Thanks for the explanation. I didn’t realize that the Device Power Meter wasn’t reporting the actual real-time device usage.
These incandescents are next on the list to be replaced. The problem I’m having is finding LEDs that have a similar warm color temperature. Even those that claim to be warm seem to have a cold, green/blue tint to them. I’ve been through several sets of LEDs for other lights before giving up and compromising.
I have a fancy Sekonic C-7000 Spectrometer that I use professionally to analyze lighting but in domestic situations your eyes are better … and in most cases of switching incandescents out the results are pretty dismal.
I suspect you are using standard A9 60w bulbs … in which case the options are limited for that “warm” feel. Good spectral A9s are not even close to MR16 and PAR formats for some reason. That said, in domestic situations you aren’t aiming for any kind of “reality” … you just want it to feel right. Right?
I exploit diffusion and a bit of chaos. The more homogeneous things are at home the more the lights can irritate. A bad bulb can make an atrocious bulb look great!
In the warm-warm realm I’m happy (enough) with a Philips dimmable 2000K(!) 350lm 4.4w [MFG 9290012101]. Less than half the lumen output you’ll get from a regular 60w incandescent but if you have 6 bulbs you may have enough to please.
The bulbs are 60W T10 tubulars with frosted glass. They’re installed in long glass frosted tubes, so I’d need replacements that are long enough to have the same diffusing light effect. The T10s are 4" long. I think this is going to be challenging.
I like the Philips bulbs that allow you to adjust the color temperature. That 2000K is certainly on the warm end!
I’m not a real fan of the Philips Hues or even the fancy (Ketra) color-change bulbs because they aren’t capable of optimizing the spectral output for any given target. “The whites are wrong!”
Hmmm, so the bulb I mentioned is one of these newer “vintage” style (= old tungsten?) … Philips makes an equivalent-ish T10 2200K.
Good to know about the vintage Philips bulbs. Thank you. I’ll definitely check them out.
I have Ketra bulbs in my “studio” and hate them. I can never seem to get the right type of light, and the clunky Windows app you use to configure them is a real pain. Philips got that right with their Hue app. I’ll probably swap the Ketras out at some point, but they weren’t cheap!
You might also enjoy the Hue integration with Sense. I use IFTTT to trigger Hue bulbs on Sense activity … mostly just for fun and troubleshooting but there are some valid use cases.
Definitely a sense detection issue. Does the issue attributed to other deep at the same time? There is no way your dimmer is wasting 700w+ of power. You’d have a fire if that would be true.
