Tweet-a-Watt Kits

tweetawatt

[adafruit] wrote in to let us know that the Tweet-a-Watt is now available in kit form. While the plans have been available for a while, a kit is a welcomed addition. The kit doesn’t include the Kill-a-Watt monitor, but rather the XBee adapters and parts necessary to make it talk to your Arduino or computer. The kit is $90, while the Kill-a-Watt can be found for roughly $20.

41 thoughts on “Tweet-a-Watt Kits

  1. hey – it’s phil, i founded hackaday and later went on to become the senior editor at MAKE and i also co-designed this project.

    @ninja, this is the only “free” low-cost and public domain way to monitor and publish power usage at this time. you’ll see a few dozen companies that try and patent this and sell expensive versions. with this kit you can make your own. it’s also a great way to learn about xbee.

    i’m a little surprised by the snarks – it’s an open source hardware project, we released the project to the public domain, it hacks/mods a current device and then adds a lot of smarts on the “web 2.0” side for data (you can use it with google app engine, twitter, facebook, irc, whatever).

    i’d like to see what one of the snarky folks here could do better, cheaper and more open. please post up your projects…

  2. @phil / pt

    I agree with you 100%. A hack is a hack.

    This one is pretty slick, and doesn’t just apply to twitter.

    I believe that anything that shows up here with the word ‘twitter’ on it is shunned. I really do not understand why. This stuff can be ported on to so many different environments.

    Quit looking at ‘twitter’, and look at the hack itself, and what could be changed.

  3. @phil / pt

    Just ignore these guys. It is not worth
    the effort to reply to them.

    I’m a regular reader of hack-a-day, I’ve
    seen amazing hacks published in here. Just
    keep your good work and don’t let this
    project go down just because of this idiots.

  4. @pt

    the backlash isn’t about your hack – which is cool, let me make that clear – but it’s about Twitter. i posted about this when this originally hit hackaday: focusing on the twitter aspect raises the ire of people who don’t really like twitter (like me). it just sounds like hype-shooting marketingspeak that’s latching on to a current meme (open-source or not, everyone markets their own projects).

    ok, ok, so there’s some hype thrown in, what’s wrong with that? i think it dilutes the purpose of your technology to shackle it with something as limited as twitter. i won’t go over why i don’t like twitter, other than to say what’s relevant: it’s very narrow and very transient. the absolute most useful part of this project, aside from hacking the kill-a-watt, is that it outputs to a real, live spreadsheet. this project would be far better described as “Spreadsheet-a-Watt” or “Log-a-Watt” or “Anal-a-Watt” (!), not “Put-it-on-a-stack-of-poorly-organized-messages-a-Watt”.

  5. I am on the fence, I agree with the “twitter” comments.

    I like the idea of logging, and I like the idea of “wireless” transmission of power usage.

    I keep pushing the idea down because instead of logging to a database or spreadsheet “you can ‘tweet’ your power usage!!”

    As far as “cheaper” and “more open source” goes, you can apparently use an iron ballast and a few turns of wire to record your amperage usage. Combine that with an X10-like wire transmitter and a $1.50 Atmega and you can have nearly the same functionality for $30 or so (less if you have junk to raid for parts), not $110. Not to mention that you can probably attach multiple ammeters, or just one on your house shunt. (in fact, if you could design one to safely clamp around your bus-bar you would only need one.) You could use a portable device to measure your devices’ draw and then when your usage increments by that amount, the device is active.

    (somebody please show me a $20 Kill-a-watt, all I can find is ebay [dodgy] and Newegg, $25+ after tax and shipping.)

  6. @emilio – you’re saying that you have issues with twitter and the project would have been better if we named it “spreadsheet-a-watt”…. ok, next revision we will consider renaming it.

    @nubie – on the fence? it’s pretty clear that you can use *any* way to display the data, twitter is one easy and free way to do it (and it’s a free SMS gateway).

    the data goes to google app engine, or anywhere else (just read the how-to pages).

    please design and publish the project you outlined – i’d love to see that and make something better and cheaper, that’s the point of this too – to get others to make things.

    lastly, you can find a kill-a-watt for $15 or less, i see them in NYC and a trip to google’s old froogle thing has the same results.

  7. Oh, did I mention I am too broke to buy the kill-a-watt at $25 as opposed to $20?

    I live in California, not NYC, and the local Home Depot stores have them listed for $40.

    I do understand that you can use it any way you see fit, if you can afford it in the first place.

    Perhaps a version that can communicate via wire or Infra-red to the PC? It is just the Xbee system that seems frightfully expensive.

    I am looking for the links to the project I was thinking of, they may be on my other PC, DOH!

  8. Just a quick confirmation on prices, “froogle” (or Google shopping to me) has exactly 3 links for the kill-a-watt under $15:

    The first is an ebay “$5” link, to an auction that ended at $13.39 with $5 shipping, so $18.39

    the second is iKitchen for $14.98, but of course they are out of stock (and who knows what shipping will cost?)

    The most legit under $15 link is this one for $14.99: http://www.google.com/product_url?q=http://www.qualcollmore4u.com/catalog/item/1746053/5470665.htm&fr=ANlN89aO3-Nfsozl7oAI7ZKFBTbyl1snsZjFLpT_a86QYEWgk1tmCrOjPxN0LpwchKd6gML1SqyoqlDTbRVPabT0OUg5waEjD-CWrBJSImla2q4iBGMruPde6zZ1GMRRSCF7RvsAmIV0AAAAAAAAAAA&gl=us&hl=en&sa=title

    Unfortunately they expect you to pay this much for shipping:

    UPS / USPS Standard $13.99
    UPS 3 Day Select $47.99
    UPS 2nd Day Air $59.99
    UPS Next Day Air $95.99

    Sorry, it appears that you can’t get one for less than $25

    As a person with little money I try to save where I can, and have found that armchair budgeting from ‘Froogle’ adds $8-20 more at each store. (When will google add real shipping costs?)

  9. @nubie – you can get it online and also at other non-big-box store for less than $15. in general the price of things are also cheaper in most cities in CA as opposed to nyc.

    you could easily use a wire – everything we did is documented, it’s up to anyone and everyone to use the information for better and cheaper ways to make a project like this if they want.

    on a related note, the xbee is cheap compared to other wireless solutions – just my opinion.

  10. Hmmm ok PT..

    add a pic and a rs232 port to your PC. cheaper, easier, and does not clutter up the internet with useless information. I’d rather have this info on my Pc at home than on twitter.

    $90.00 is a really high price for what the Xbee stuff does. Honestly it’s why it is not used in much because it is really really overpriced. Use the low grade 433mhz modules that you can get from sparkfun for $8.99 each and do the exact same thing with just a bit more programming.

    It’s a neat idea, dont get me wrong, but It’s like seeing a “remote control light! uses 2 iphones!” using overkill parts for the sake of using them is silly. The foundation of hacking is to use the cheap stuff, not the highest priced RF link I can find.

  11. @fartface – sure there are many ways to do this – in fact, please make & publish your suggestion – it might be better… as far as ” and does not clutter up the internet with useless information” – really? you think power data from a modded power meter is worse than 99% of what’s on twitter (and the internet)? this is real data that actually is useful – we lowered our power bill and other power services are tapping in to our data for their own use and study.

    $90 is cheap compared to trying to do this with wifi – keep in mind, $90 = 2 xbee modules, 2 adapter kits, 1 ftdi cable and parts. if you were to buy all this on your own it’s well over $100 and if you were to try and re-create it using other tech it might not work and it would be a lot more expensive.

    we used an off the shelf $15 power meter and the cheapest/best way to do wireless – again, do a better a cheaper version and publish it, i’d love to check it out and make it :)

  12. Please again point me to a source for $15 pricing.

    What is a “non-big-box” store?

    I just find it wasteful that the main parts of the “hack” involve costly complicated parts that are all-together used as:

    1. An ammeter
    2. An input cable.

    over $100 is way too much to spend for what you receive in return (in my opinion).

    The necessity for wireless communication is rare, no matter the cost of the Xbee.

    In these ways I find the device inelegant (waste of hardware resources, high cost), and to add insult to the injury it is promoted with “tweeting”.

    I am looking for a better solution, namely a source of clamp ammeters similar to that used in this instructable: http://www.instructables.com/id/S5ZQL8VFPWWWR7N/

    It should be possible to make your own with a few winds of wire if you want to go that route.

    I would assume that an Atmega8 should be able to read the resulting voltage signal and then communicate the result over Serial or USB (Using this firmware http://www.obdev.at/products/avrusb/index.html)

    Cost goes way down, reliance on proprietary modules goes down as well.

  13. I like that there’s such a convo going on about this.

    my question would be, how much power to you use keeping a pc running to log all this. not everyone would normally keep one running and it promotes doing so. of course you *could* use a low watt laptop or netbook, but lets be honest, alot of people might wanna run this on an old p3 based system sitting in a closet burning 300watts.

    it seems more gimmick. now if it had an xport or wiznet 5100 built into the design, different story, less power usage, and you could skip the required xbee altogether, or still use it and transmit to an xbee + ethernet adaptor instead of the computer.

    I question whether this solves a problem or adds to product consumption for gimmicks sake. much like the botanical kit…

    props for making, don’t take me wrong, hopefully you see my point, but from the defensive tone here I think I know the response I’ll get…

  14. not to add to the flame wars, but while the XBees are great, they do seem like the biggest waste of money in the Tweet-a-Watt design. A complete wireless XBee link costs $40, while this complete link (which is slower, but high-speed data shouldn’t be necessary for the design) costs only $4 + $5 = $9 total:
    http://www.sparkfun.com/commerce/product_info.php?products_id=8946
    and
    http://www.sparkfun.com/commerce/product_info.php?products_id=8949

    Here’s an example design they’re used in (nicely documented):
    http://www.elxproject.com/news.php?readmore=36

  15. @fyrebug http://www.flickr.com/photos/ladyada/3318289854/ the router was $45

    @fartface sure, 433 mhz modules are cheap. dont forget you’ll also need: a microcontroller with A/D AND a microcontroller programmer PLUS all the code to deal with A/D, sleeping and retransmits, packetization, error correction, addressing, collision detection, etc. if you’re psyched to write that code go for it! but…im not. its not easy and not worth the time. the xbee modules Just Work with no programming or tweaking.
    then once you’ve spent that time you can reuse the code i’ve posted & comment out the single line of code (at the very end of the project) that sends a twitter

    @nubie if you live in new york, J&R http://www.jr.com/p3-international/pe/P3_P4400/ is right downtown and has them for…$20!

  16. There are a few suggestions here for “just a wire coil” around the bus-bar/feed. Which works pretty ok if your entire house is purely resistive, but is wildly inaccurate if you have a non-unity power factor.

    Given the killawatt represents the “it can kill you” and “burn your house down” part of the process, I think spending the extra money may be warranted. Especially given it may not be more than an equivalent clamp-on ammeter. killawatt was $8.50 from the library (they used to rent them out) while a used 75A clamp-on ammeter that wasn’t routinely overwhelmed by RF noise was $17.

    Granted you can build a functional current-shunt meter or hall-effect meter for less.

  17. correct me if I am wrong, but doesn’t the unit only measure power usage of whats plugged into it anyway? I mean, wow.. you managed to log your desk-lamp power usage to the internet.
    The real use of these devices is to go around your house with one and collect data on EVERYTHING. Then come back, sit with all of it on paper and figure out where you can cut your bill. (wow, that 40 watt bulb in the halway is really sucking the current down, maby I can change that for a 20?) constantly monitoring only one appliance is sort of like missing the point. Knowing that your washer is a power hog doesn’t change the fact that you still need to wash your clothes.
    as pointed out by several people, some way to watch your entire house power usage would be nice… ohh.. wait.. the electric company installed one of those fancy meters with the spinning wheel on it right there on the side of your house.

    So, here is my published hack:
    parts needed – pencil, paper (ruled), eyeglasses (Depends on your eyesight), house, gumption (that attribute that determines your willingness to get off your ass for 5 minutes a month, rather than relying on a fancy plug in module to wirelessly deliver all its data to your lazyboy while you watch the big game)
    optional – ruler (to draw straight lines)
    Step 1 – using pencil (and optional ruler) divide paper into 4 vertical columns. Mark the top of each column with the following headings: date, previous reading, this reading, monthly usage
    Step 2 – go to your meter on the side of your house. Apply eyeglasses (if necessary). Read meter. Enter this information in the “this reading” column. Also not today’s date in the date column.
    Step 3 – collect underpants, make profit, wait one month.
    Step 4 – (30 days later) go to meter. Re-apply eyeglasses (if necessary). Read meter. Note this information in “this reading” column. Also note the date.
    Step 5 – carry the data from the previous month’s reading in “this reading” down to today’s entry under the “last reading” column
    Step 6 0 subtract last reading from this reading. Enter the result under “monthly usage”
    Hack complete
    Variation – use PC and spreadsheet software in place of paper. You can also slightly automate the process.
    Tip: publish your data on twitter to track your progress and allow others to learn how much you depend on the air conditioner.

    I suppose if you wanted to make one of these devices really useful, wire it into your air conditioner, or behind your fridge, or your washer/dryer combo.

    at any rate, logging it to a spreadsheet sounds useful. But could the designers spell out for me why the heck you’d want to put it on twitter? I dislike all the twitterhype. Ive never bothered to look at twitter, so I don’t know what all the buzz is about. please… justify the reason for twitter. I just don’t get it.

  18. @nubie – there’s one for $14.50 – just google around. a big box store is like a home depot, best buy, etc. they have huge mark ups, don’t shop there for things like kill-a-watts.

    please please please make a better cheaper one, it’s fun to research around and try and figure a way to make things, the hard work is ahead – make one :)

    @fyrebug – we do not use a PC to “keep this running” – please read the page(s) – we use a hacked ASUS router that is already on and doing things. again, you can use other things – we wanted low-cost wireless. if you look around there are dozens of “funded” companies doing wireless power metering and they’ll all using xbee too it seems. i see your point, but i don’t think you read the site or the details of the project. most folks just react to the headline.

    @dano – you can use anything, the xbee is just one way of doing this project… and again, if you can do one better and cheaper and publish, i’d love to see it… just linking to lower-cost products doesn’t count :)

    @error404 – you can use ethernet (please read the project page).

  19. @mre “correct me if i am wrong, but doesn’t the unit only measure power usage of whats plugged into it anyway? i mean, wow.. you managed to log your desk-lamp power usage to the internet.”

    we have our entire apartment on this, so it’s more than just a desk lamp. a house would be another type of project – and you could use what we did to tap the breaker too and publish…

    “at any rate, logging it to a spreadsheet sounds useful. but could the designers spell out for me why the heck you’d want to put it on twitter? i dislike all the twitterhype. ive never bothered to look at twitter, so i don’t know what all the buzz is about. please… justify the reason for twitter. i just don’t get it.”

    it does log to a spreadsheet, including google app engine (please read the site).

    i can’t help your twitter issues, it’s a cheap and free sms gateway – you can comment the one line of code out if you want :)

  20. if someone can point me to a good resource for do-it-yourself ammeter, I’ll have a go at adapting a propeller to do this with a built in ethernet cable like the ybox. Word on the street is prop prices are gonna drop, we’ll see. t’would be cool to see this combined with a relay so you could turn the plug off and on over the net.

  21. @ Ladyada, Much respect, I love your site.

    Unfortunately for me I live in California and the JR website offers them for $26.95 shipped to me.

    At that point I might as well go with newegg (last I checked it was $17.99, it is now back to 19.99 + tax and shipping.)

    I like your use of the router, can the router be connected directly to the Kill-a-watt to bring the price down?

  22. Argh, I don’t really mean to fill up the comments here, but . . .

    I found at least one of the pages with the DIY clamp on current transformers

    http://www.edcheung.com/automa/power.htm

    I really wish I could find all the info that I saw before, it may be in some older bookmark backups.

    It would be great if there was an option that was entirely DIY that only provided the functionality to interface with a computer or micro-controller.

    (could it have been at eco-modder forums? Or a link from Dansdata? Or even here? Nuts, wish I knew, maybe a slashdot comment had the link I clicked.)

  23. @ladyada that’s great! you’re right I didn’t fully read that, and having had people do that on my projects I know it can be annoying, but now understandt it from both ends.
    much, much better with a router. though I’m still no fan of the xbee. just a personal preference unrelated and unneeded. seems like an overcost large hammer to drive in a tiny finsihing nail.

  24. For house bus bars there *should be no ‘shunt’ involved!* I would *highly* recommend some prior research before you start comparing a ‘shunt’ to a ‘current transformer’. frankly, you stand to **LEARN** more by designing a clamp-on current transformer and actually understand where your power usage is coming from (rather than buying one)

  25. @medix,

    Whoops, that was an error in vocabulary, I did indeed mean a current transformer or transducer.

    (I am aware that a shunt must be installed in-line, and thus is much harder to install and is potentially more dangerous)

    @rasz , that is very cool for $10

    (if you already have an xbee and a kill-a-watt laying around I guess the t%$*t a watt is a neat hack, but looking from a cost-performance ratio it leaves a little to be desired. It should top the product range and at least one cheaper version should be considered.)

  26. Once again I think people are missing the point here. Yes this may be expensive for some people. And if ~$110 is too much for you, having it is NOT going to save you any money probably. On other hand, for some people this $110 investment is returned within a first month on energy saved. And like others pointed out, you can just use an ammeter and have that data sent. The only flaw with the this meter is that its only use-full to monitor a single circuit. It’s more use-full to audit your per-device consumption but to get a better understanding or view of your energy consumption you have to monitor your total consumption. For most people the energy wasted is not caused by a single appliance, though outdated/rusted water heater will add a nice chunck to your monthly bill (as I learned). For most people it will be all those phantom devices that lay around plugged in but unused.

  27. If you placed it at the junctionbox you could not only see your use but you could watch your house for any unwanted guest since you’d see peaks if they switched on a light or something.
    Assuming one of these could handle the load of many devices at once.
    You’d have to make some baselines for when the fridge motor is on and off and such periodic peaks devices of course.

    Just musing.

  28. @adafruit:

    Maybe I’m missing something, but the project just seems to be about pumping serial data over xBee. The only mention of a non-computer-dependent Ethernet implementation was hacking an Asus router and adding xBee to it. I mean do away with all of this and build a simple data logging client and Ethernet interface into the device itself to cut out the cost of xBee and the need to have a power-sucking computer online to log how much power you’re sucking.

  29. I think this is a brilliant concept – it’s an idea I’ve been sitting on for about 15 years. If people are simply conscious of what energy is going where (and it’s presented in the right way) then they’ll instinctively move towards efficiency.

    The ultimate (in my most humble of opinions) is a web-based interface where you can measure every single outlet… and have a clamp meter over the entire supply… AND control every outlet via the web.

    I think this would lead to economisation software, and possibly a social-type thing where (for example) one school can compare itself to another school and exchange tips/tricks into how to get better results.

    I think the killer-ap aspect of this will be remote control – the energy monitoring part of it is an adjunct to this – more value than the control itself, but it will come as an “extra”

    I think the price is too high at the moment though… for any currently commercially available unit anyway. Although the ROI angle is pretty clear, it’s still not low enough to be an impulse buy.

Leave a Reply to mem.namefixCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.