Wattvision (YC W09) Shows The Power Of Simple, Realtime Energy Monitoring | Hacker News Reader

Wattvision (YC W09) Shows The Power Of Simple, Realtime Energy Monitoring

techcrunch.com

67 pointsdbul87 comments

Honest question, is this solving a problem and who has this problem? I pay about $30/month for electricity and I already know it goes up to about $80/month during the winter months. $200 seems like a lot of money to pay for someone to tell me what I already know.

Wattvision tends to cut your power use, because you're probably wasting power in ways you don't know about. When you first install it, your initial reaction is generally "why are we using so much power?" Then you walk around turning things on and off and watching what happens to the graph. You usually discover some surprises in the process, and those change your behavior.

You make what you measure. Watch any number carefully enough and you'll tend to optimize it, and most households have significant room to.

I can understand the first usage being kind of interesting and enlightening, but I'm not sure if its $200 kind of interesting.

It seems like this would be of more interest for commercial applications.

How about a device that actually directly saves you power?

Well that's what my friend's startup focuses on currently piloted in South Africa. The startup built a device that reliably saves power. You don't need to pay any fee up front for the device, you "pay as you go" on amount of saving made on monthly basis. No catch. Average savings currently are around 10-15%, engineering team is pushing the numbers up as we speak. Business has gotten necessary funding and even the primary energy supplier has come on board to put an initiative together that would implement the solution to all street lights in the country in the near future. South Africa is currently going through energy crisis due to huge demand and lack of supply.

PS. My wife is one of the co-founders. :)

How does it work?

What is the average time to recoup the cost of the meter and start actually saving money?

If you believe the normal claim that you will save 10-30% and your average monthly power bill is > the cost of monitoring it will take you less than 1 year to pay itself off.

So if your bill is $150 a month, this system (or similar) should pay for itself within a year - IF you try to save power using it.

Saving energy doesn't only have to be about money.

I've always thought that light switches & sockets should have meters on them.

http://www.42u.com/images/power-tower-xl-ptxlv016.jpg

Lots of datacenter grade stuff does. It's really convenient.

Oops, I was about to upvote, but I hit the wrong arrow. And there's no undo. Sorry!

Until recently it was very difficult to get the data from the sensors to a useful place. Wireless networks and common always on internet connections have changed that.

What are the chances some sort of clever metering will be built in to new, higher end houses?

I wouldn't bet on it too soon. When I was involved in the home automation industry back in the mid/late 90's it seemed like that sort of thing was imminent.

Invensys and CutlerHammer both had power-monitoring devices, electric meters and so on (I still have some CH power monitoring sensors on the main power lines at my house with a system that I put together that tracks KwH, heating oil consumption and a few other metrics).

Anyway... I haven't really seen those products move an inch in the last dozen years (other than press releases).

While I do believe that the real-time feedback is helpful, I also more strongly believe that people who really want to cut their power usage will do so by simply turning more things off or using things less. For the most part, a monthly feedback cycle isn't really THAT long. In some cases the month-long feedback loop let's you measure a more cumulative effect in dollars instead of cents.

I also think that automation has been imminent for long enough to be suspect.

But people responding to feedback is not something to just dismiss. Monthly isn't the same as on demand which isn't the same as "out in the open." If sockets changed colour from blue to red as the burned electricity, consumption would go down.

Measuring power at the meter is nice, but I'm wanting to measure the power per circuit in the house. Given how small electronics have gotten, I've been wondering if it's possible to put a simple monitor in a circuit breaker. It would make it easier to narrow down the power hungry devices in the home.

I know it differs from the whole internet-connected/constant monitoring solution, but there is a device that you can use to do this manually called the Kill A Watt: http://www.p3international.com/products/special/P4400/P4400-...

Installed cost for these home automation solutions quickly tends to exceed $1000 making them impractical as a cost saver. The sensor costs are suddenly dropping as competition heats up.

Hopefully innovative, web enabled solutions from fast moving startups re-invigorate this sector. I'd sure love to.

Extremely likely. There are good chances that in 30 years most houses are as automated as our off-grid house on Great Barrier Island and come with power monitoring as standard.

http://blog.gridspy.co.nz/2009/08/the-barrier-house.html

We plan to provide the devices today.

Yes, extremely likely. Hey gridspy guys, do you want to see if we can find some common ground between wv and gridspy? :) ping info@wattvision.com

It seems to me that live monitoring is most useful to people who look at the power bill and wonder "where did that all go?" The biggest people asking this question would be large power users such as businesses.

Businesses spend more money on power and need more detail to really drill into their power usage. That is why Gridspy (my startup) is focusing on them first.

http://gridspy.co.nz/

Because of this, I think that Watt vision and Gridspy are in slightly different markets, perhaps even complimentary.

The consumers interested in our system typically live in big homes in the suburbs and run their AC or heating quite a bit. That's not to say we can't also help landlords and small business owners who are also interested in saving money on their energy expenses.

We spend $150-$200/mo. We're a small house in Chicago with 4 occupants. For any likely % of efficiency we'd gain by measuring electricity use in our house easily, it's trivial to work out the small number of months it would take for this to pay for itself.

There's also a big difference between knowing in the abstract that (say) water kettles are expensive, and knowing specifically what is costing us money in our house.

"you can't improve upon something until you can measure it" -- Albert Einstein

But what scale of measuring is necessary? Do I need to know what devices drain what power when it's simple to assume that electric heaters consume a vast amount of energy during winter and that air conditioners consume a vast amount of energy during the summer.

I believe the OP was asking is what is the benefit of this additional measuring and monitoring. What will it provide beyond the KW/h information I can get from my own meter?

But how much energy? Do you know how many $$ of your power bill are down to AC and how many are due to hot water? If you turn down your AC one degree, can you tell at the end of the month how much you saved?

It is hard to isolate particular elements of your house's usage without either a lot of legwork or a decent sensor solution. Knowing how you use your power lets you make smart compromises and invest in power saving measures such as better lights, solar hot water heating, insulation or smarter appliances.

It is especially important if there are many people living or working in the same building, making it hard for any one person to get an accurate picture in their minds of the power consumption.

I know exactly how much electricity my devices use, and I certainly don't need a meter to figure out how or what they use. I worked as an electrician, I can work out the power consumption of every device, so yes personally I know where virtually every penny of my money is going.

I also have an induction meter (IIRC accurate to ~10W, I've used ones accurate to ~1W) that cost ~$60, so I seriously don't see the value in a $200 system that can't do much more than me, a pencil and time.

I see the system as a bit dishonest when you can tell people to look at a plug and read the amperage draw and tell them to times by 110. It's especially dishonest when the majority of devices now tell the wattage due to EnergyStar requirements.

I agree that eyeballs + a calculator can go a long way. However, the amperage draw on the back of your power supply or device is typically worst-case (maximum) current, not average current. Similarly, EnergyStar numbers do a little better, but everyone's usage pattern is different. For instance, what is the wattage consumed by your computer? It is certainly not the summed TDP of all components in the box. It is also not the rated wattage of your power supply. For things like this that are not "dumb loads", you need a more sophisticated way of measuring power consumption.

Personally, I'm with you in preferring a different solution. I use a Kill-a-Watt (http://www.p3international.com/products/special/P4400/P4400-... ) to measure actual average power consumption over a few days for each major device in my home. Seems to work fairly well.

That's like a race-driver saying that ABS brakes are inefficient because he can brake better than the computer (true, but 90% of the rest of the population can't).

Or an amateur-chef scoffing at someone going to a steakhouse to get a good steak, because he knows how to make them good at home, much cheaper.

Or, say, a geek saying that graphical UIs are bad because he can work much faster on the command line and in vi.

I have two electrical heating devices, but both have a thermostat. I can read the label and do the math and tell you exactly what they cost to have running. But I can't tell you how often the thermostat turns them on and off, and I'm not going to spend a day with a stopwatch, looking at the light going on and off. I can't tell how the temperature outside might affect my heating usage. I can't tell how two persons showering affects the floor-heating in the bathroom, over one or none.

A logger like the one Wattvision offers could offer some input to that.

Many many people will pay $200 dollars for the time component in your example.

You cannot read your meter as you walk around the house switching appliances.

Many (if not most) home-owners are far less aware of the power usage of common appliances than we are. They probably don't even know what a kWh is.

I think it could be handy for businesses, universities, research labs, or anyone else that has a big power bill. Getting it itemized would be interesting.

That is exactly what we do. http://gridspy.co.nz/

And doubly so for businesses with high power needs, such as data-centers or even software houses.

But this doesn't seem to be their target market. The simplicity of installation, and the website, point to home users. If you're running a data center you probably already have more sophisticated systems in place.

Good feedback. Yeah we can do that, too. ;)

It would be more interesting to have a sensor on the electric meter, then a few other units that you can put between appliances and the wall outlets so you can measure specific devices and tweak usage based on that data.

I have electric heating, last winter my average bill was $400 however my first bill was $490 little changes like leaving some doors closed during the day, and others open had $50 differences. I definitely would pay for that data, the savings would pay for it.

Electric heat is like converting steak (high grade electric power) in to hamburger (low grade radiated heat). It is probably the least efficient way you could possibly choose to heat a house.

I'm one of the alpha users, and I've found it surprisingly interesting. When you install Wattvision, you can suddenly see all sorts of things you never knew about. For example, how much power electric kettles use. When I switch on our electric kettle, it doubles the power consumption of our house. Who knew? So now when I use it I only put in as much water as I need.

Yes, but that's not suprising, and not something you need Wattvision to tell you. An iron, a toaster, an electric kettle all consume a lot of power, but you only use these items for a short period of time. A smarter device would be more localized and attach itself and optimize the usage of individual systems like your boiler, house heating unit and HVAC.

It was surprising to me. It's one thing to know in theory that heating water uses a lot of power, and another to actually watch the power consumption of your house double when you flip on the kettle.

Okay, I can see Wattvision being a useful tool for educating the consumer. I like that they also allow you to look at other households power consumption data. Someday it could lead to a conversation like:

John: Hey neighbor!

Jill: Hey John. What's up with the super-hot weather these days? Good thing I installed that AC unit a few months ago.

John: Oh, you have AC? I do too, but I looked you up on Wattvision and your power consumption is way lower than mine.

Jill: Oh, that's because my unit has a variable speed motor. And it has power factor correction built into it so it draws much less current from the grid.

John: You're so cool Jill!

Like the Kill-A-Watt?

http://www.p3international.com/products/special/P4400/P4400-...

Kill-A-Watt is for a single outlet, which in certain cases is more useful and cheaper.

Wallvision will do it for your whole house.

has this team considered an expansion into device level power monitoring down the road? plugging a sensor into the jack itself?

that would be really compelling if it were possible to make the sensors small enough and inexpensive enough.

Something like ladyada's Tweet-a-Watt? (http://www.ladyada.net/make/tweetawatt/). Device level monitoring makes so much more sense! Especially when a utility provides variable pricing for the electricity. For example, an monitoring unit built into the dishwasher would be able to turn on the appliance when the electricity demand is low and hence the kWh cost is low. But yes, it's a good starting point for Wattvision. Their future releases will probably be smarter and richer in features.

Definitely looking in to that, it's a future iteration and a smart idea.

Sounds like something you could use ZigBee for.

Let me save you $200:

- turn stuff off when it is not in use

- get rid of your old fridge/freezer

- get rid of any crts that you still have

- take a good look at wall warts and decide which ones really need to be plugged in when they're not charging something

- use a laptop, not a desktop, and enable powersaving

- use CFLS instead of bulbs

- run your heater at 18 degrees celsius (whatever that is in Fahrenheit)

- do not use electric heat for anything other than your quick-boil kettle, so no electric range etc.

There, that will save you $200 and at least that much in electricity.

The above list is a compilation of the outcome of a years worth of experimenting in a house that was powered by solar cells, the great insight of all that is that it is 10x as cheap to save a Watt hour than to generate one.

Thank you.

And also, for the record, I am aware that FPL is testing a product that syncs in with home networks in some cities across Florida. They're charging something like $2-3 / month for the service, which in my mind is a much better deal, especially since their analytics are probably far better.

I'm looking for a link now.

Most smart meter based solutions give you data in 15 minute chunks at the end of each day. Real time data really is far more useful.

Good luck selling your wife / flatmate / employee on these tips without a nice visual aid. It gets much harder when there are multiple power users.

Great list by the way. Our offgrid house uses the same amount of power per day as two towel rails by applying this sort of approach. The main power hogs are pumping fresh water and sewerage.

I wasn't aware the 'towel rail' was a new measure of power consumption, would that be the bath-towel rail or the toilet one ?

Seriously though, how many KWh do you go through in a day ?

We managed to get by on 1600 W of solar panels in Northern Canada, which in some weeks was touch and go on battery depletion (you really don't want to get below 70% or so).

That's what kept us going in the dark of winter:

http://pics.ww.com/v/jacques/renewables/batteries/IM000398.J...

Yeah, the towel rail is for the people who don't understand kWh.

That is 3.5KWH per day, powered using a 640W solar array and a petrol generator when there are several days of particularly bad weather. The battery bank stores 1200AH at 24V, and like you we try to keep it at 70% charge or above.

Gridspy was born out of monitoring power usage at the house both while there and while back in Auckland (90km away).

Read more here (should probably add more numbers to this blog entry) http://blog.gridspy.co.nz/2009/08/the-barrier-house.html

I saw a CNN story last week about Google's free energy monitoring software. Then at the end of the story was a huge let down: you need special hardware to get it running.

I checked Wattvision right after that and there were no updates, so today I'm super happy to see that the hardware is available. If it proves to work well, hopefully we'll see a story on CNN (i.e. to the public at large) in the next few months.

Google PowerMeter works with TED (http://www.theenergydetective.com/ted-5000-overview.html)

With the wealth of data that they collect, this would be a great place for an economist (like myself:)) to study usage patterns and possibly conduct experiments.

Of course, the set of people that buy this device are more "green" than the average consumer and also more likely to change habits. Nonetheless, looking at correlations of household-level usage with location specific variables like weather, sporting events,etc. could be really interesting. Moreover, conducting experiments where alerts are different for 50% of the users to test how responses differ would really illustrate how public policy should move forward. Think of the A/B testing web developers use, but with possibly wide-ranging consequences.

Great work!

I've been using a similar device for a year now - the Wattson Home Energy Meter (http://www.diykyoto.com). It's not as full featured as Wattvision and doesn't connect over WiFi, but the information it reveals is extremely interesting and enlightening.

It's easy to merely understand power usage (I always have done), but to see it revealed and presented like this is particularly fascinating. It's makes you stop and think. This is a good thing, for both environmental and educational reasons.

I can't wait to get my hands on Wattvision, and good luck to them!

People won't pay $200 to save a little money on their heating bill. People WILL pay $200 to have a facebook widget that says they were the #1 greenest home in their neighborhood/city. A little game theory applied to reducing energy consumption. Doesn't really have to be realtime either, if I could plot the energy usage of all my neighbors on a google map it would certainly shame people into submission.

Nobody really gives a damn about being green, green chic on the other hand... I can't wait to try it out.

If you're interested in non-realtime energy monitoring (and speak German), check out my skunkworks project from way back: www.stromverbrauch-online.de

You sign up, regularly put in your current meter readout and get an estimate of your monthly and yearly electricity bill. No hardware required. I've gotten a number of signups since and some people really get religious about putting in their data - some do it every other day!

Getting your energy consumption down can be an interesting sport, see the Tweet-A-Watt.

Very nice seamless implementation: device, connection, web-based, iphone. Full circle. Very professional. Great Name.

Thanks for the kind words christof!

This looks like a great product. Nobody seems to have mentioned the competitors, though, which include some pretty mainstream brands.

Black & Decker EM100B Energy Saver Series Power Monitor: http://www.amazon.com/Black-Decker-EM100B-Energy-Monitor/dp/...

Blue Line PowerCost Monitor: http://www.amazon.com/PowerCost-Monitor-BLI-2800/dp/B0028LS5...

I believe I saw the B&D model on an episode of "Ask This Old House". Not sure if any competitors offer a web or iPhone interface, or just the receiver devices, but in any case this is a useful form of feedback that will likely become a standard feature as the monitoring equipment gets cheaper.

After my parents got a prius, they noticed that their gas mileage in their other car went up as well, because they learned how to drive more efficiently. What they really needed was feedback.

This looks like it could have the same benefit.

But the name...

No analog meter support. I'd have bought.

Coming soon. We'll email you as soon as we have it ready!

Congrats Savraj! Out of curiosity, how are you powering the WiFi sensor?

wall wart. :) ( < 1 watt )

Very well executed. Hope it doesn't turn out to be an ice cream glove.

http://www.google.com/powermeter

hmmmm.....

- Not live (data up to a day old)

- Requires a smart meter or another online meter such as a TED5000

- Extremely simple (at present)

Actually, I'm looking at the in-house readout and it is quite live. (to the second) The gadget lags, but maybe 10 minutes, not a day.

I'm using the ted5k. I don't know what you mean by simple, it's really detailed, and you can even load pricing models into the device and get an idea of what everything costs and when.

edit: Also, it uses induction coils around the mains feeds. It's easy enough to install that I did it myself.

(and disclaimer: I work for google)

I meant that the Google dashboard seems to be quite simple in its list of features by comparison with the TED5000, Wattvision or my planned features for Gridspy.

Also, saying that Google is competing with Wattvision is not quite accurate since vendors like ourselves are hoping to integrate with Google's (your) solution. The only way in which it competes is in exposing smart meter data to the average user. That data requires a smart meter and is often expensive to collect (using GPRS for example) so is collected relatively infrequently.

Now in many of these areas I have to speculate because smart meters aren't yet being rolled out where I live. To use Powermeter here in NZ the user has to spend money for a solution such as Wattvision, TED5000 or Gridspy.

Finally, all this is moot if you want a system that can localise the power usage to particular circuits in your building such as Gridspy.

I hope I didnt' come off as competitive with Wattvision, the more people helping people save energy, the better we all are.

Amen to that. We all have our own unique strengths and weaknesses. More choice for the consumer. Looks like there is a huge market to save power + money + the environment all in one.

I couldn't find any info on how the thing reads the meter. Is it plugging into the meter via an open port, do these digital meters all have wireless transmitters themselves, or does it do something crazy like use a camera?

Almost all of the digital meters have an IR LED that blinks once for every kilowatthour that is consumed by the house. Their device (along with the one that Black and Decker sells) sits on top of the meter, counts the number of blinks every few seconds, then sends that over some sort of wireless to the internet.

I have yet to figure out how to intercept the Black and Decker's signal to duplicate wattvision's functionality. This is easy(ish) to do with a mini Arduino and a pair of XBee radios.

I think they have different sensors for digital ones. I think it is easier to make a sensor for analog one by checking the spinning wheel. It is pretty cool they can have a light device that reads the rpm of the spinning wheel and connects to internet via wifi.

the analog prototype i saw actually used a laser to read the spinning wheel. it was very cool and clearly saved the household money.

i bet the digital meters are easier to read.

For a competing solution, see Gridspy usage in Stephen's apartment here : http://your.gridspy.co.nz/prototype2/

I live in a condo. Can I make use of the Wattvision?

Unlikely at the moment, but if you can find your electricity meter it's possible.

Why do I have to click checkout to find out what the sensor price is? Even the pricing tab doesn't help...

It's $149 after the $50 TC coupon, will update, thanks for noting this.

I would buy this except $200 seems a bit steep.

Wow this is cool. I thought you were in the deadpool. Your pricing scheme is a bit confusing. It's very Microsoft-esq.

GO ON SAVRAJ MY MAN!!

Reply on news.ycombinator.com