Like many of you I’m trying identify devices that make up the Always On mystery bucket and looking for help from the group.
On a typical summer day, with two IT professionals working from home, our typical power usage is 800 to 1,200 watts or 40 to 55 kWh per day. What’s driving me nuts is the 625 to 700 watts of Always On, of which 220 watts is identified (freezer, cable modem, wi-fi router, multiple computers and monitors).
According to my wife, I may have over done the smart plug installations with 6 Kasa HS300 power strips and 5 Kasa HS110 single plugs. Did you know a CPAP machine draws 3 watts, a Sleep Number bed draws 6 watts, a wireless headset charger draws 1 watt, and an Amazon Alexa Echo Dot draws 2 watts all day long? Lots of gadgets have been donated after realizing how much power they consumed and the rest put on switch power strips.
In the pursuit of knowledge (aka driving my wife nuts), I’ve turned off most of the circuits in my house, unplugged just about everything else and still had a phantom 400+ watts of Always On usage that I cannot identify.
Here’s what Sense or my smart plugs have identified:
All TVs, Blu-ray players, Apple TVs, etc.
All computers, monitors and printers
Air conditioner, furnace, air exchanger, sump pump, sewage lift pump
Just because Sense identifies an on-off cycle for a device, it doesn’t mean it is identifying the Always On component for that device.
Garage door openers - we have 3 that all have 8W of Always On - radio listening for the open signal. Sense only picks up the open/close part unless you have on a smart plug.
Furnaces - once again, unless you have them on a smartplug you won’t see the 6-8W they use constantly (part of that is the 2W that go to the thermostat).
The Always On measurement provided by Sense is one of their worst efforts in my opinion. My accurate always on is about 300 watts but Sense shows it at 600 and sometimes even 900.
Their are many discussions here and even some explanations from Sense about how it is calculated where you can get an idea of why your numbers look different than you’d expect.
Speak for yourself, I find always-on to be a great way to help track phantom loads. Combined with smart plugs always-on is a great asset.
OP for anything hard wired it will be more of a challenge. I haven’t had much luck with LED lighting or ceiling fans but security lights have been detected over time.
I’d like to hear how to track a phantom load using Always On. Being as you can go turn things off and it won’t have an immediate effect to the Always On calculation. That’s the problem with it, it’s more a wild guess than an accurate measurement.
Study how they make this calculation and you’ll see. Just because they use the terminology “always on” and it sounds good, it does not make it accurate.
When my detected devices (fridge and freezer) are being used in the Always On calculation even though they are not truly “Always On”, they have work to do.
Smart plugs help break out always-on power usage. You should try it out, I’ve been able to reduce my always-on dramatically over the last 6 months with this insight.
@senseinaz, you are right about the smart plugs and I currently use 2 HS300’s and a single HS110. The HS300 are pretty much just for my Always On devices.
That is the only accuracy so far with it.
I had a situation where my Always On was high, I think about 600. My fridge and freezer were both on at this time and are detected. Sense was note showing these as on. @HilarioAtSense confirmed that if Sense would have shown these as on in the device list at that time, the total wattage would have been more than was actually being used. Think always on 600 plus 100 each on fridge and freezer equals 800. But total consumed was 600, so what does Sense do? Instead of adjusting the always on down that 200 watts, it just hides the devices it knows are on so that it’s calculation appears to be correct.
Saw your post on FB as well. Someone there mention the importance of looking at the real time power meter when hunting phantom loads. Your response wasn’t clear to me if you are, or if you are just basing it off of the device list.
Always On is not a live bubble. It is a stats based result looking at your usage over time. So if you have a 25w always on device and you unplug it, you won’t see that change in the Always On for up to 48 hours.
So just in case there was a misunderstanding - make sure you are watching the live power readout on the bottom right corner of the bubble or timeline view in the app. (circled in magenta in the below photo).
This updates in real time and works well when power hunting, especially if you are able to isolate breakers.
So for instance. Turn off all the breakers in the house except for the Sense. You should see very little power draw.
Turn on a single breaker that you know. Say bedroom outlets. See how much the power goes up. I suggest giving things a minute to settle as its possible items when turned on will go to a “full on” state and then return to idle.
Then you can go to the bedroom and start to unplug devices 1 by 1 and watch the power go down again. Once everything you can think of is unplugged, if there is still a lot of power, you can try to figure out what else may be on that breaker.
Later, Rinse and repeat.
Things get trickier on breakers that have hard wired devices, but you should still be able to narrow down where you are looking.
@ben’s post is spot on to what I do in my own home, except I prefer the Power Meter view. I actually went through this process over the weekend, since my AO has gone up 15W in the past two months. I didn’t quite figure out everything, but was able to figure out that my furnace is drawing 10W constantly and my Alexas each use around 2W.
The big surprise was my soldering station pulls a constant 3W, even when fully powered off. Not entirely sure why on that one…
One more thought - post / look at a couple days of your Power Meter history timeline. The real question is what undetected devices are still on when your house is at its quietest (maybe 5AM, but useful to see the full cycle). There’s likely a cycling of some devices even in the wee hours, but look at the lowest points.
That’s the point I was making. Let’s say at 5 am your house is the quietest. This is the lowest use of power but some detected or undetected device is on at that time, that device just got used to calculate your AO.
That’s also the time to try the breaker flipping experiment, leaving only your Sense, router/modem/access point/range extender on, plus allowing for 5-8min for Always On devices to ramp up.
We have had our sense installed for about 4 months now. It’s found few things, but nothing substantial in the energy use dept. As of now we are still showing 1064 watts of always on and 2312 watts of other.
Our electric bills are on the verge of bankrupting us so i splurged on the sense in the hopes it would help us figure out where the vampire is.
Unfortunately so far - nothing.
We have gone around and flipped breakers, unplugged appliances etc - no significant drops.
We are in an 1800 sq ft house with one teen during the summer and during the school year it’s usually just two full time working adults.
I’m desperate to figure out what is pulling 3400 watts of power that isn’t what has been idemtified as a fridge, dishwasher, washer/dryer sump pump well pump.
Turn off everything in your breaker box except for the breaker Sense is attached to plus the breaker your router/modem/wifi is plugged in to, while watching the live Power Meter.
If you have still lots of power showing in the realtime display on your Power Meter (500W or more), then you’ll have to look closely at the two circuits still live, plus the possibility you have a sub panel still using electricity that you are unaware of.
If the Power Meter goes to a very small value, then you can start flipping breakers on, until you see a big jump. But you’ll probably need to add breakers very slowly allowing 5-8 min between each one to allow for the startup of devices (things like furnaces, refrigerators, a hidden server all take time to reach operating power)
You might not see a big jump - it might be lots of little ones. That just means you’ll have to look more carefully at each breaker. Of course there’s also a chance that you have one big consumer that is slower than 8 minutes to start typical consumption, or one that needs to be turned on.
Let us know how that process goes…
For High Always On.
Please post a couple of 24 hour cycles of your Power Meter. That could help put a finger on at least when you should be looking for your big Always On.
Do the same breaker flipping experiment during the time of day/night when your power is at the lowest in the house, per your 24 hour Power Meter.
One more thought - I am wont to suggest more power monitoring hardware, but the 20$ purchase of an TP-Link HS110 would give you a nice tool for looking at consumption of individual plug-in devices. The best part is that you can link the HS110 to Sense and get 24 hour usage traces for all you plug-in devices. I have one in my house called General Purpose HS110. It has been very useful in looking at the usage patterns of all my electronic devices, even when off or in standby.
You mention some devices there. Are those the devices that have been identified by Sense in your home?
Everyone on the forums might be able to help more if you can give us more background on your situation - what’s been detected so far, what major appliances do you have that haven’t been detected, etc.
For instance:
What kind of water heater do you have? Is it an electrical resistance storage tank water heater or an electric tankless on-demand water heater? Or maybe a hybrid heat-pump water heater?
Do you have a swimming pool?
Do you have central air? What is the cooling load in your area of the country? If you care to, you can be specific and that can help understand how much of your electric bill is heating/cooling. If you just have window units, how many do you have and how often do you run them?
What’s the total monthly kWh used on your electric bill?
As others have said, when seeking out loads before the Sense AI detects them, the real-time meter is your friend. And the real-time meter on the website version is your friend for more easily posting data to this forum:
@Trish welcome to Sense land and just letting you know, as a regular on the community, that I can feel the palpable excitement among the Watt-zapping troops in response to your desperation.
I don’t have much constructive to add at this point because the others have covered most of everything … though I will also point you to the Settings > My Home > Sense Monitor > Signals window. It will give you the most precise measure of your instantaneous usage (including voltage and the usage on your 2 power legs [phases], which can in some cases be helpful). The total usage is the sum of the 2 Mains readings. It will look like this and be changing in real-time:
[@RyanAtSense & @kevin1: On a meta note, it seems like there should be an official checklist go-to document for these cases, culminating with “Consult Support or the Community”, because they are common and the work taken to reiterate methods is significant and, as we see in this case, people don’t ask for help until they are desperate!]
Interesting,
Electronics makers have been on a tear to reduce their standby energy consumption, so newer stereos, TVs, etc. are typically much lower when in idle, and even when on. Not so true for plug-in appliances with big motors in them, like washers or garage door openers.
ps: I can’t look at the standby (compressor not running) power use of my fridges, but I’m betting they use 5W+ each in that mode.
Neither of those are surprising…most modern devices have electronic controls that run 100% of the time, and usually a few watts. However, if you think of a 5 watt draw for the
8,760 hours per year, that’s just under 44kwh. At my $0,18/kwh, that amounts to $7.92 annually, which isn’t a big concern.
Much more interesting would be the top 5 or top 10 most powerful/most used devices, many of which Sense wasn’t able to detect (or not detect accurately) in a year and a half.
