Simple forum thing to allow survey / votes.
Would seem great to get feedback from users and for other users to see the data.
Like how long until your first device was detected (1 day , 2 day …)
What type of breaker box do you have (Siemens, GE, …)
Brands of TV not detected
Maybe this is just something moderators can do, but seems like good data for Sense , and interesting for users.
The Sense moderators and product marketing have the ability to do surveys and frequently use them when they do product planning. For instance, they hosted a solar survey before developing and rolling out the most recent solar features.
As for time to (first and next, …) detection, Sense probably have a very accurate view on this since already have the detailed data on detections. A few of us users have done the analysis here, although it’s more difficult to measure than you might think, unless you are accurately logging when new devices are actually added to the house…
As for the breaker box question, I suspect the type of breaker box plays a very minimal role in Sense “signal” noise - much more a function of other devices you have in the household. And surveying breaker boxes might be more complicated than it seems - I have 3 Square D breaker boxes in my house (1 main with 2 subs), populated with mostly Square D breakers, but a few other brands as well, with a solar backfeed breaker in my main panel. I’m betting that the topology of my main box WRT the subs and backfeed have greater effect on Sense “noise” than the make of the breaker box.
Cool, great pic showing an example detect timeline.
I think my head was in a strange place about the surveys a bit with your answer making lots of sense. I wonder though if a survey of devices that are not detected would be useful to help tune the auto detect algo.
Not sure what to think about the breaker box response. This is al 60hz and you can hear the noise… Screw in breaker systems are generally much quieter, but with all the AFCI and GFCI breakers these days the audio signature of breakers varies more. If you have screw in vs. snap in breakers I think the audio (hence electrical) noise is lower. Also, since the noise would be load / breaker dependent the noise varies and almost causes a phantom feedback. I would think screw in breakers would have a higher discovery / detect rate… but no evidence to that end.
Here’s a little better view on Days2Detect, removing all the smart plugs, which start “detecting” on the day you install the smart plug. And some of these aren’t exactly right because:
As for the breaker question, I’m no expert, but I suspect a 60 cycle hum is just related to the electromechanical nature of the thermal and current “sensors” in most breakers. If snap-in breakers have more play than screw-in models, they would be more prone to vibrate against the box causing a sound, but I think all electrical-mechanical breakers are prone to vibrate due to the electromagnet and bimetal strips usually used for the current and thermal sensing. And I don’t think a little bit vibration, even though it represents an little bit of an energy loss in the breaker, is going to cause electrical noise in the amplitudes and frequencies necessary to affect Sense detection. If anything, the buzz is a predictable, stable 60Hz loss of energy when looking at RMS.
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.
