Check where every watt goes in your house

[mswlogo]

I got solar a few years ago along with a whole new HVAC with central air. We never had central air when the system was sized. When they started to install it I had them increase the size a little. 8kw. That’s pretty small by what I’ve seen posted here.

Turns out it’s plenty enough and we have a 23k pool as well. To make sure we had enough I went on a watt finding binge and reviewed my whole house and saved probably another 10%.

So we have about 2 megawatt excess a year. Well that will cover like half the unplanned EV model 3.

I decided to have another review and bought a nice amp meter. I already had a whole house meter that’s part of solar system. And a Kilowatt meter. I recently had purchased a nice handheld clip on Amp meter for some hobby work.

My whole house standby is 300 watts, which I was pretty proud of.

I started to go through the circuits and for the heck of it measured the central A/C unit. 1 Amp at 240 Volts !!! I had loaned my monitoring tools to a friend so he could review his house and found that his heat pump was pulling 200 watts on standby. Well that made some sense and it turns out he does not use his heat pump in winter. Just for A/C. So he now turns it off in winter.

I never thought to measure my A/C which is a super “high efficiency” variable speed compressor with a DC motor. Turns out these have heaters when it’s below 21C. I might have not noticed it either if I reviewed the whole house when it was warm out.

That waste alone might cover the balance of charging my model 3.

If you do decide to turn off your A/C or heat pump for winter read think below. The heaters are in there for a reason.

AHE - Home Electricity Reduction Solutions: How To Reduce Your Air Conditioner's Standby Power

Make sure you know where all your watts are going. And how necessary they are.

 

[mswlogo]

I just helped “audit” a friends house for “watt leaks”.

In about 90 minutes we found he had an older Dewalt Charger (18V NIMH) running with an almost dead battery in it.

It was pulling 2.0 Amps at 122 Volts.
That’s 244 watts 24/7 365 days a year.

That’s 2.1 megawatts a year. 1/4 the total years capacity of his solar system. That’s about $534 worth of electricity at our rates

Equates to about half a year of charging his Model.

Do you know where all your watts are going?

 

[TW14 9LF]

Can one of you post a "how-to" describing how to make these measurements? I'd kind of like to do this.

Keith

 

[KevinWC]

There is a good energy monitoring system from Sense.com that provides great details on where your energy is going. It provides real time information. As you move around your house and turn on and off devices it will tell you the energy used. It takes some time, but it will learn all your home devices and provide information on how often each one is on and the energy used. The sense units are not cheap, $300-$350 depending if you have a solar system. It you confortable with electrical system, it takes less than 1 hour to install. I have had mine for 2 months and love it.

 

[swaltner]

• Buy an amp meter along the lines of Commercial Electric LCD Digital Clamp Meter-MS2033C - The Home Depot
• Turn off every device that you can in the house
• Take the from panel off the breaker panel
• Measure the current on the feed lines coming into the breaker panel. This current in Amps times 240 Volts is the total idle power draw in your house
• Now measure individual circuits (single pole breakers get amps times 120 V for Watts; double pole breakers - your A/C, clothes dryer, store, EVSE - get the current * 240 Volts for Watts)
• Find the circuit that has "too much" stray phantom power draw and start unplugging stuff on that circuit until the current drops

Anything connected to power that doesn't have a mechanical on/off switch is a potential source of phantom power draw. Some sources of phantom power draw are hard-wired to power (the daily timer for my outside lights embedded in the switch, the countdown timer for the exhaust fan in the master bedroom, ....)

Edit: And if you want to monitor this on an on-going basis, get a Sense system like @KevinWC or a TED (The Energy Detective) system like I have or any number of other alternatives that get hardwired into your breaker panel.

Edit 2: How to use an infrared camera to find overloaded circuits shows a clamp meter measuring the power on a circuit in a breaker panel. In that case, it's showing a current of 22.97 Amps (on a 20 Amp breaker). It's a single pole breaker, so roughly 2760 Watts (2.6 kW) is flowing through that one breaker.

 

[Kant.Ing]

Last month when we were away, I checked from my PW2 gateway reading for the house's consumption, which was 4.2kWh in a day, or 175W in average. The numbers should make sense as I also used Belkin Insight to measure some appliances' idle consumption.

 

[mswlogo]

Can one of you post a "how-to" describing how to make these measurements? I'd kind of like to do this.

Keith

There are a few tools and a few ways to do it.

You can buy a whole house meter. Like the one I linked below. These are pretty easy to install but you need to be confortable going into the main panel. No "wiring" is needed but you need to clip the coils on the MAINS. Sometimes wires might be in the way. I'd say if you don't know how to change a light switch, don't do it and hire someone. It should cost probably less than $50 (minimum house call) because it would take all of 10 minutes.

Warning: If you have Solar back fed through a breaker most "whole house meters" will not read correctly when Solar is producing. A whole house meter is still useful as tool to monitor things but it won't give accurate usage over time, or when solar is producing. My whole house meter is integrated with Solar and knows how to measure pure load. If your solar is back fed on the Mains you want to monitor the pure load side.

You should start with a properly labeled circuit panel. Common Sense Applies to figure that out

I turn Heat/AC Down and Unplug Frig while doing this. Turn off all lights, TV's etc.

To find "piggy devices" you watch the Meter Real Time and Turn Breakers Off One by One.
Watch how much the Meter Drops. Anything over 5 Watts is worth looking at IMHO. Often it's something you forgot you even owned.

Start with the largest loads.

Once you ID a Circuit that's using a lot, Turn it Back On (and Ideally everything else OFF).
It could be several motion sensors for lights or something. All good.
If it's a bunch of outlets, start unplugging one by one and watch for large drops.

Like my stupid TIVO DVR, that we barely use these days, uses a whopping 40 watts !!
It's a process of elimination. Cross check yourself, it's easy to screw up and make wrong judgements.

Watch out for "Start up" loads. When you turn things on they might initially take more power, then settle down. Startup loads are not of a concern.

I also have a Kilo-Watt Meter. This can help find culprits. It can also measure devices over a long period of time.
Like a Refrigerator. They only way to measure a frig is over time.

I also use an Amp meter. This is just easier and quicker than turning off/on breakers to determine loads but it not required.

I'd recommend a whole house meter plus a kilo-watt.

Whole house meters vary from $80 to $300. It's really hard to justify $300, but the one below is one of the best ones.
You really want "charting" to get a good picture of what's going on.

My whole house meter is built into my Solar System and cost me $500 extra as part of the Solar install.

Whole house Meter
https://www.amazon.com/Sense-Energy-Monitor-Electricity-Usage/dp/B075K6PHJ9

Kilo-Watt meter
https://www.amazon.com/P3-P4400-Electricity-Usage-Monitor/dp/B00009MDBU

Amp Meter
https://www.amazon.com/gp/product/B001TCWL1E

 

[mswlogo]

Last month when we were away, I checked from my PW2 gateway reading for the house's consumption, which was 4.2kWh in a day, or 175W in average. The numbers should make sense as I also used Belkin Insight to measure some appliances' idle consumption.

View attachment 373997

Is 4.2kWh total usage for the house? Was Heat running?

We use around 11Kwh a day in winter which is half the typical house hold.
We might use 9 kWh when away (have not been away since I found the A/C heater pig)

That is without Tesla Charging.

You can see the popup value on the lowest valley, it is around 187.9 watts.
Peaks are Heat and Frig. Evening is TV and other appliances going.

BTW my integrated usage meter is not great for debugging/finding piggy devices because it only reports like once every 15 minutes.
It's great for confirming/monitoring things long term though.

 

[Kant.Ing]

Is 4.2kWh total usage for the house? Was Heat running?

I believe the heater wasn't triggered, and our Nest Thermostat was on ECO mode (45F) when we were away--the house in California hardly gets indoor below 45F.

For parental control reasons, we use a smart plug to control the power to our entertainment system including TV, speakers, receivers, and gaming consoles. The switch consumes about 0.7W and is usually off.

 

[mswlogo]

I believe the heater wasn't triggered, and our Nest Thermostat was on ECO mode (45F) when we were away--the house in California hardly gets indoor below 45F.

For parental control reasons, we use a smart plug to control the power to our entertainment system including TV, speakers, receivers, and gaming consoles. The switch consumes about 0.7W and is usually off.

That is incredibly low for a primary home, even while away. My TIVO is a pig but I have Smarthings shut everything down in entertainment system. Except TV stays on (1W). I have a lot of motion sensors, probably 12, Those add up. Heat is down as low as it can go when we are gone. A/C off in summer when gone. Frig is always there doing it's thing.

We have a 2nd home we are mostly away from. With very few gadgets, no TV's, no A/C. Heat on 40. I just checked, it's averaging 5 kWh a day in winter. But it's -10F there .

 

[Kant.Ing]

We do have 5 always connected surveillance cameras each takes no more than 2.5W, five Google Minis around the house each 1W, and all lights are LEDs (including those in the refrigerator!) and light switches are all wifi controlled and every smart switch should be around 0.7W. Internet infrastructure is about 20W. Other appliances including Toto bidets are negligible in their power saving mode. OTOH, I really want to analyze and break down how Tesla car consumes vampire drain... when I have some time.

 

[HankLloydRight]

If installing Sense with a powerwall (no solar), would you put the CTs and sense unit inside the non-essential loads box (which has the mains from the street and the A/C breakers), or would you install it in the essential loads box after the powerwall? We're only using PW for backup power, not TOU shifting.

There's lots of room in the A/C panel for Sense's 240v breaker, so I'm tempted to put it there. Not sure if I'm overlooking some major difference between the locations.

Our house averages about 62kWh per day in the winter @ $.1982/kWh. Our electric bill is killing us. In the summer, it's a lot more due to A/C. So I think Sense will pay for itself if I can figure out the pig circuits.

 

[mswlogo]

There is a good energy monitoring system from Sense.com that provides great details on where your energy is going. It provides real time information. As you move around your house and turn on and off devices it will tell you the energy used. It takes some time, but it will learn all your home devices and provide information on how often each one is on and the energy used. The sense units are not cheap, $300-$350 depending if you have a solar system. It you confortable with electrical system, it takes less than 1 hour to install. I have had mine for 2 months and love it.

That’s awesome it knows how to handle Solar. Most do not. It’s not required to have the Solar option even if you have Solar depending on how your Solar is tied into your panel.

But I see a lot of complaints that it cannot figure what’s what. Read the reviews. That’s because it’s nearly impossible to do, by measuring from one point. It can make some good guesses over time. But it will never figure it all out and never would have found the problems I’ve found.

If one thing is on 24/7 vs 20 things are on 24/7 there is no way for it to tell what’s what. It can probably sort out some big items like heat, Frig, dryer. But it’s not gonna find an A/C heater or a dead battery left on a charger.

But other than it trying to sort things out, which nobody else even (smartly) attempts. It looks like a great monitor. Even though very pricey.

 

[David_Cary]

Actually pretty easy to find specific culprits if it works like other whole house monitors.
I had eguage on my last house. My background was fairly high - like 500 watts.
The response was instantaneous on my gauge so I just sequentially shut off circuit breakers. Turns out my highly efficient a/c was chewing up a lot but I need it so nothing I could do. Also my Arc fault breakers were users also. The house was built the first year with arc fault codes - I am hoping they got better.
Now I am sitting here in a new house. My gauge is Solaredge and it updates every few hours. So completely useless to find phantom loads. But I do see the nadir on the graph at about 250 last night. That with some external lights on, ceiling fan in bedroom, noise maker in bedroom and a few night lights. While the external are LED - they are big. I think 60 watts total. So then my true background is less than 200. Not bad for 3800 sqft. No cable boxes. Just one automated dimmer so far.

To find my true background, I would have to shut everything down and wait 3 hours. Don't get Solaredge. My solar installer switched to it and said it was better. Haven't complained to them yet....

 

[mswlogo]

Actually pretty easy to find specific culprits if it works like other whole house monitors.
I had eguage on my last house. My background was fairly high - like 500 watts.
The response was instantaneous on my gauge so I just sequentially shut off circuit breakers. Turns out my highly efficient a/c was chewing up a lot but I need it so nothing I could do. Also my Arc fault breakers were users also. The house was built the first year with arc fault codes - I am hoping they got better.
Now I am sitting here in a new house. My gauge is Solaredge and it updates every few hours. So completely useless to find phantom loads. But I do see the nadir on the graph at about 250 last night. That with some external lights on, ceiling fan in bedroom, noise maker in bedroom and a few night lights. While the external are LED - they are big. I think 60 watts total. So then my true background is less than 200. Not bad for 3800 sqft. No cable boxes. Just one automated dimmer so far.

To find my true background, I would have to shut everything down and wait 3 hours. Don't get Solaredge. My solar installer switched to it and said it was better. Haven't complained to them yet....

You don’t understand what Sense claims it can do and doesn’t. It claims it can identify everything all by itself without you switching anything off and on. And tracks each item going forward. All by measuring load on your mains.

I love the integrated SolarEdge monitor. One of the Best things I bought. Extremely accurate. Great UI. Well built. But yeah, it’s not easy to use for hunting real time. And you don’t have to wait 3 hours. It updates every 15 minutes. But will feel like 3 hours Your installer gave you good advice to get it. I made my installer put it in.

I also have an engage, it does the job and is cheap. But the company is horrible to deal with if you had an issue and the UI on the phone is crap. They sent me the WIFI module with a MAC address already assigned to an account. They could not clear it for me. Basically a brick. I didn’t initially set up WIFI. So I was stuck and had to buy another WIFI module.

If I had nothing today. I’d probably buy the pricey Sense. Especially where it can handle Solar. But I’d not expect much accuracy in correctly ID items (it’s claim to fame). But that’s fine, nobody else does.

 

[HankLloydRight]

I like the engage system with the wifi setup. Yes, there are many flaws with the company and the UI, but it's a basic system that works in real time. I'm using two CTs now to monitor each 120v branch of the two mains coming into the house. So if I'm looking for a lossy circuit, I can at least narrow it down by half.

That said, I'm |-- --| close to getting a Sense unit. It's strange I can't find anyone selling them used. I'd probably use mine just to find and fix the biggest drains, and then sell it for $200 used and go back to my engage units.

 

[mswlogo]

I like the engage system with the wifi setup. Yes, there are many flaws with the company and the UI, but it's a basic system that works in real time. I'm using two CTs now to monitor each 120v branch of the two mains coming into the house. So if I'm looking for a lossy circuit, I can at least narrow it down by half.

That said, I'm |-- --| close to getting a Sense unit. It's strange I can't find anyone selling them used. I'd probably use mine just to find and fix the biggest drains, and then sell it for $200 used and go back to my engage units.

You can find exactly what’s using what by flipping off breakers and unplugging stuff with any real-time monitor or amp meter.

The engage doesn’t separate the two phases either. It adds them and gives you one number. Last I looked. Maybe you can set it up that way.

BTW one trick you can do is leave one coil unclipped and clamp the other on any single 120v circuit to narrow down debugging one circuit. If it’s a 240v circuit use both coils. Like you could put the monitor just on your Tesla charging circuit. A Sense monitor could probably separate the EV charging out easily because it has a very unique load pattern on the mains.

Don’t buy the Sense expecting it doing the work of figuring out what is wasting watts. It might tag say maybe an old beer fridge that has no beer you might want to dump. You just can’t do a thorough job without manually flipping breakers, unplugging stuff etc.

 

[HankLloydRight]

I have two separate transmitters one for each branch. They’re not added together.

The loads I don’t know are periodic and short, so flipping breakers would take a long time waiting for a load not to pop up.

 

[HankLloydRight]

This is what I’m trying to diagnose... looks like several different drains.

 

[mswlogo]

View attachment 374206

I have two separate transmitters one for each branch. They’re not added together.

The loads I don’t know are periodic and short, so flipping breakers would take a long time waiting for a load not to pop up.

Use the real-time display to debug. I think it updates every 10 seconds or so. The charts update over much longer periods (e.g. 15 minutes). No chart will save data down to the real-time level (e.g. less than 10 seconds), it’s too much data to sift through and not very useful. I forget how course engage is but it’s worse than Solar Edge. The problem with Solar Edge is, is there is no real time read out at all. Only charts.

Real-time display on engage looks like this. I also have the physical Engage display which I know is around 10 seconds, which is what I’ve debugged with before. And you can set the update rate. But I’m pretty sure this updates almost as quickly.