Lowest Always-On Wattage?

Wow! Most of you ha e great numbers. When I get down to 600 watts I’m pretty happy.

Mine started out at 253w when we first installed Sense but over time we have identified numerous things that do not need to be on, plus we put things like TV’s and printers on wifi plugs that completely cut power and we are down to 89w.

Taking about 150w/hr of consumption out of our AO reduces our daily by about 3.6 kWh, or about 1.35 MWh a year, about 40 days of median Solar Production.


Since the start of 2020, my Always-On value maintains itself consistently, within an 84-89kWh range! However, this may change - due to our adult daughter’s long-term return!

Going to the airport now and only time will tell!

Apartment-dweller here. My Always On average (since installing Sense in early February) is around 42W, bringing my total Usage to date to 36kWh.

Mine is 467.

I have a QNAP Nas, a router, SmartThings, 24 port switch, 3 alexa, 2 soundbar, 3 wifi access points, 5 security cameras.
2 PC with 2 Monitors. Plus the usual appliances.

I would like to better understand how to lower it.

Does all that stuff have to be on 24/7? Does the QNAP have an auto sleep function, my NAS’s do. We put everything on CE SmartPlugs which we can voice control through GH. You can also schedule each plug independently so maybe somethings don’t need to be on in the middle of the night?

We put our big ass TV with sub and bar on a SmartPlug, it uses 225W when it was on, and 55W off. Now when we turn it off via SmartPlug it draws 0W for the 20 hrs a day it’s not on.

Same with a color laser printer. Use it a couple times a day for 30 seconds, but it is on 24/7 burning 50W in standby. Put in on a SmartPlug, turn it on twice a day, do my printing, shutoff via printer power button, cut power to it.

Have to change your way of thinking too. Everything today is designed for instant gratification, to be instant-on, to be at our fingertips. Putting traditional hard switch electrical devices on voice command added an extra step, but changed the way we interact with them.

Between changing our habits, SmartPlugging virtually every device in our home, and changing to 100% LED lighting we have dropped our daily power usage from about 28 kWH/day to 12 kWH.

Sorry about being long winded!


Long or short… 12kWh per day is outstanding! Is this within an apartment, condo, or single-family home?

I’ve maintained a daily goal for 15kWh.

1 Like

Detached bungalow with 2 adults. I’ve only recently become The Power Nazi when we got our Sense.

Before that, ignorance was bliss, except when the utility bill came. Looking back at power bills 10 years ago, with 2 teenagers living at home, full incandescent lighting and a home office CGI render farm running 24/7 we were using 50 - 75 kWH a day!

Our bills are 1/4 of what they used to be, plus our solar grid exports wipe out not only our electricity bill, but also our gas bill for 8 months out of the year.


My average over the last 4 months has been 200W of Always On. During the summer this is higher because air conditioner fans run all night, even though the compressor inside shuts off periodically.

I notice that everyone reports their number in Watts, so I have done the same. However, it seems to me that reporting it as a percentage would be more fair. A large household and/or one with many family members will use more Watts, both day and night. For me, Always On runs about 10% of total usage. Is that number normal?

Great point Jeff! I think 10% of total usage is fantastic. My always-on is 21.3%, and I’m up to 194W now (was 139W when I started this thread almost 2 years ago, haha). My house has slowly accumulated “smart” devices which tend to be vampires.

@jefflayman 10% is great! I’m hovering around 17.5% YTD average Always On, improving from about 25% at the beginning of the year :slight_smile: .

111W (8%), but will increase Wattage when I will plug the TV entertainment system (I moved to a new house this summer and did not plug it yet!). However, percentage will decrease in the next months as winter comes and house heating will jump the kWh to the sky.

2600 sq ft condo, avg always on 425 watts

Just installed Sense a few days ago and shocked at how low my Always On is: 39w for the last 24h and 42w currently. Rural 1600 sq. ft. single family house. Sonos, two air purifiers, tv, laptops, etc. We are millennials and have devices…?

Keep thinking this could change as more devices are detected and assigned. But a lot of the time the reading is that low at night? And it doesn’t seem to work that way?

This is great! But I guess I won’t have such an easy time lowering my electric bill. :face_with_monocle:

Confession: the 10% Always On that I reported two months ago could be more like 25%, depending where you look. This is because if you have devices with an AO component, then the downloaded .CSV data is different than what is visible in the app. The graph below shows the two different values of AO in my case, and how the two diverged at week 25 when I got my first AO device. At that time I wrote technical support to report the issue. They told me it was a known bug that their Engineering team is working to correct.
AO compare
This graph was generated in Excel. The blue line shows the downloaded data. The orange line is a manual modification of that data which approximates what the app shows. Disclaimer: my algorithm does not make a perfect match with the app.

The area under the blue curve is 26.6% of my usage to date, while the area under the orange curve is 9.7% over my time with Sense. My post from two months ago reported the second value without clarifying how it was measured. I am sorry for any confusion.

1 Like

Interesting comparison. I’m trying to figure out where the data for your Raw AO comes from. Just looking at my AO in the app vs. export and I’m seeing a small difference - 400W vs 419W. Maybe you can explain the derivation fo your blue line results a bit more ?

Here’s the two view on my Always On -

The Web app (same as iOS app) - 400W

Exported hourly Always On data for today - 419-424W

Thanks for your interest, @kevin1. The data for my blue line was from daily export, whereas your hourly export has a higher resolution than I am used to seeing, but I think the principle still applies. The difference you see between app and export of 400W vs 419W should be the sum of any listed Always On devices. Per your screen capture, your AO devices total (5w + 3w + 1w = 9w), which is not quite the same as the 19w you calculated. Do you perhaps have other AO devices that didn’t fit on the screen when you captured?

In my case the difference between the two lines is huge because at week 26 I installed a ~300w monitored device that runs 24/7. Without that significant draw, I probably would not have picked up on the difference between app and export. During weeks 45 and 46, that device was off for maintenance several times, so the two lines drew closer together.

I’m not sure your interpretation is correct… If I look at the definitions, the total Always On for monitored (smartplug) devices isn’t subtracted from the whole house Always On.

You are right about my AO list from the web app being truncated. The total actually adds up to the number on the bottom, 171W.

You inspired me to look at my daily AO for yesterday (today’s isn’t cooked yet), just to see if there are significant differences between daily an hourly export… Yesterday’s daily AO, not so different in terms of average wattage - 428W.

And if I look at hourly, it was a little higher than today’s numbers. So fairly well in line.

Kevin, it appears your experience is different from mine. I’ll give one possible explanation, but if that doesn’t satisfy you, I give up. I know that my algorithm is not perfect, but it is acceptably close to my own data. Your observations have given me something to think about, and direction for fine tuning my algorithm. Thanks again for your input!

My guess for the difference between your experience and mine is consistency. I suspect that your AO devices are quite consistent in their usage. Based on the numbers you report, some 90% of your wattage remains when exported (only 19w of 171w did not). My list of AO devices is overwhelmed by a ~300w device that can best be described by a picture:

Notice how the wattage is different at almost every sample point. The components inside this device cycle about twice a minute. Sense looks for the minimum value in all this mess and tags that as the Always On portion. Right now that value is 263w.

If I watch the “Now” screen for a minute, here is what I observe. The size of the bubble for this device quivers as updates arrive from TP-Link. The size of the Other bubble quivers, too. The bubble for Always On has a fixed size until it suddenly grows about twice as large and stays that way for a couple seconds. The device bubble is still there during this time. Since the total amount of bubbles is suddenly more, it appears that my total usage has spiked. It has not, as can be confirmed by the number in the top corner. Rather than more usage, this behavior occurs as the device goes through its minimum. The show is quickly over, and the bubbles continue their quiver dance.

Based on this behavior, I assume that behind the scenes Sense is subtracting 263w from the computed AO except when the device current usage matches what it considers that device’s AO portion. If your devices are consistent, they won’t be subtracted because that exception is mostly true. For me it is mostly not, so my app value of AO is quite different than my exported AO.

@JustinAtSense, maybe you can get resolution on this one. The Always On definition says nothing about subtracting off individual smart plug Always Ons. But @jefflayman’s experience seems to indicate that Sense is subtracting these values. Any clarity on this ??