Clamp Sensor Power Monitoring

[Bill Porter] has joined in the pursuit of an inexpensive yet effective way to monitor his power usage. He calls his project the Not So Tiny Power Meter, and shared both his successes and follies involved in seeing it through to implementation. There are problems; sizing issues with enclosures and his PCB, issues with noise when measuring low-current signals with the clamps, and even some wireless communications issues. But it looks like he’s got the system running despite these hurdles. Right now it streams data wirelessly so that he can display the current energy consumption, but he plans to add Google Power Meter functionality next.

We’ve seen commercial units using the same sensing principles that have been hacked to show data online.

45 thoughts on “Clamp Sensor Power Monitoring

  1. Thanks byohazrd,

    Yeah, I have a feeling I’m in for an onslaught of people telling me how dangerous it is to stick your hands inside a breaker panel.

    The display is temporary, I haven’t decided if I want to just build a larger display that shows everything (Volts, Watts, Frequency, Power Factor, etc) or have everything logged by a computer, or something online based.

    Maybe a receiving Arduino that tweets daily usage and power stats. Any thoughts guys?

  2. He could have saved himself a lot of time and placed a photodiode on his existing utility meter. It’s a standard, I believe- 1 blink per power unit (1kW/hr, I think?)

    No high voltage, no complex code. Could’ve done it off a parallel or serial port or using an Xbee’s existing IO functions.

  3. Brett,

    I could, but it’s a lot more complicated then you make it sound. My meter isn’t weather protected, and is located far away from any viable power sources(plugs). so now we’re talking about weather proofing, solar panels and battery chargers.

    And my way gets you way more then just KWs. Volts, Amps, frequency and power factor are all measured, in real time.

  4. It’s perfectly “real time”; the blinks happen as fast as you use power. Your utility meter is power-factor corrected.

    Your meter doesn’t need to be weather protected or near a power source. If you don’t like the idea of the Xbee, run a thin cable to the meter out your window. Fine gauge, 2 conductor wire would have gotten the job done.

    Volts, frequency? You can get that from any UPS, or by plugging in something into the wall in a project box.

    If it’s fair to scream blue-bloody-murder when someone uses an Arduino to talk to a garage door and an Xbee, it’s fair to point out that you completely ignored the simplest, easiest, cheapest, most elegant solution available to you.

  5. @Brett: You make some good points. Generally you are not so interested in voltage, current, or frequency as you are in the actual energy consumed, which if your meter has already done all the calculations is really easy to obtain that way.

    I think AC frequency is not interesting at the kind of precision this project provides since I believe it is extremely well regulated to 60 Hz in the U.S., and variations probably aren’t measurable without more precise instruments.

  6. A residential meter only reads Real Power. My meter reads much more. And it’s not real time. If I’m only consuming 1KW per hour, there’s only one blink per hour. For a whole hour, I have no idea what my real time power draw is until it blinks.

    In my case the cable would have to run down three stories, and travel another 800 feet to the meter. Not to mention be obvious to any safety inspectors walking around. Still think it’s practical?

    And combining data from multiple sources like UPS becomes complex. This one sensor gives it all, in a wireless signal I can pull down to multiple places like my wall mounted display and soon into a internet connection for logging. Your solution is not practical, cheap nor elegant.

  7. I am interested in voltage, it’s a good sign of trouble. I’ve lived in places with a poorly tied neutral connection where the phases were drifting, one would go as high as 150V RMS for brief periods.

    I’d also like to see PF, to see how much the power is actually costing the utility company to provide, vs. how much they are charging me for what they meter. The increasingly inductive residential load is somewhat of an issue.

    If I wanted just power at low precision by estimating voltage, I would have just tied the clamp to an Xbee. I wanted more.

  8. @ColinB; Just used my O-scope and voltmeter, my line voltage is coming in at 58Hz. I live in the Hampton Roads area of Virginia, fwiw.

    @Bill; I understand exactly. I have lived in plenty of complexes where it was inconvenient to even look at, let alone monitor. Thankfully, I own a house now, so upgrades / modifications are a snap now.

  9. Very neat indeed. Thanks for the writeup and sharing your designs!

    I’ve been looking for a new DIY project that involves some kind of wireless comm aspect so this would be very interesting. Useful, too!

  10. Bill, the blinking rate was such that I could turn on a light and see the effect about 10-15 seconds later on the wireless power meter my electric company sold to me for ~$30; it used a pair of AA batteries, attached to the meter, and watched for the blinks, then transmitted over RF to the base station, which showed live kw and accumulated power usage stats (total and daily, with a meter that compared the current day to the previous.) So yes, it IS “real time”. I threw out a (wrong) guess. It’s actually one thousand blinks per kilowatt-hour.

    http://blog.janbottorff.com/?p=109

    Second: the biggest consumer of electricity will be electric heat (if you have it), followed by your dryer (if you use an electric dryer), followed by your fridge, followed by lights and computer equipment, if it runs 24/7. Most of those are resistive loads, and thus have a PF of 1. If you computer gear was made in the last 10 years, its power supplies are PF-corrected. So that really leaves just your fridge.

    If you’re having power problems, call your utility company. They have power monitors they can plug in / connect.

    1. Brett,

      “real-time”, by your standards, means that Windows 1.0 running on a 100mHz 80286 computer is “real-time”. The term is actually meaningless without scale, and clearly your scale is different from Bill’s (and mine.) His desire to see the power factor is reasonable, and his desire for the data to be real-time is reasonable, etc etc and so on and so on. From some of your other statements, like what uses the most power in your house, (did you just forget water heater or do you enjoy dirty dishes and cold showers?) are obviously just semi-educated guesses, so as to who needs what Bill asks for, YOU DO. This about all the power that could be saved if every household had micro-monitoring of power usage… is it worthwhile to unplug your TV when not in use? Your computer? I wouldn’t worry so much about what draws 1200 watts for 5 minutes as what draws 12 watts nonstop. (20 or 30 such power suckers could point to potential reduction in power use.)

  11. That’s great Brett. You used a consumer product that gives you limited data. I wanted to know more about my electrical usage, so I built something to give me everything I wanted. And I get measurements 3 times a second, still way faster then the blinking. But hey, whatever works for you.

    Not to mention, like I said before, my meter is several hundred feet away with several floors and walls of concrete and steal. I doubt your product would have worked for me.

    And actually, I live in Florida. So my energy use is A/C (inductive), fridge (inductive), computers (inductive, with limited pf correcting, still not 1.00 though) and lighting, which i’m transitioning to CCFL (inductive). The rest is natural gas.

    I would not have know about my power problems had I not been monitoring my utility voltage.

  12. Woah! Brett, your solution is quick and simple. Perfect for not caring about intricate details. Bill, Your’s is far from quick or simple, but gets those intricate details. Let’s shake hands now…

    Bill, I do like the method/build. Great work.

  13. Wow. “Your solution isn’t what I would do, here’s my approach.”

    “Hey that’s neat, but my project is finished, so it’s what I’ve got, for these reasons.”

    “Okay that’s pretty reasonable, nice work.”

    is that so hard?

  14. what is everyones obsession with monitoring how much power they use?
    and don’t give me bs about trying to save the planet. there are easier ways to check things are turned off, like just turning them off. at least then you aren’t wasting electricity just to watch yourself waste it!
    oh maybe you think your electric company is cheating you out of money? well big deal they charge for the power that makes them most money what do you expect them to do? what are you going to go and write a letter to them because your home made power logger says you used a dollar less than they charged you?
    if i was them i would literally laugh in your face.
    seriously don’t get this at all.
    i could maybe understand the whole bored electrical engineer looking for a problem to solve with technical expertise if it was one or two projects but everyone seems to be doing it.

    I cannot believe there are two people arguing about methods of measuring energy usage in this comments section. i think you are all ill in the head to be quite frank.

  15. @pff,

    While I won’t deny being ill in the head, maybe I can shed some insight into my insanity. As an EE always did want to do this, but recently I had more of a reason.

    My girlfriend who lives with me. She had a few times now where she can’t imagine why our electric bill was so high. So now we have a readout of current usage. In fact, I’m changing the display soon to show $/hour or day of current usage rate, so she can see exactly how much money her little personal space heater is costing. (one example)

    I have had my own denials too. It’s just nice to see your money flow out of your pocket hour by hour, as opposed to all at once per month with not much to relate it too.

  16. Rusty check you scope it need calibration. There is no way you have 58hz on main, elecrrical grid synchronized to atomic clock and monitored, it critical for stations to be synchronized or they will literally explode

    1. I didn’t want to comment about this, but since you bring it up, I’ll second that. Nowadays (hey, that passed spell check!!!) we have a “national grid”. That means that it has to be in sync. It’s not simple to do that (the grid has capacitance and inductance and must obey the speed limit c) but it’s extraordinarily unlikely that you are off by that much. Note that there ARE some places where, for practical or legal reasons, the power will be converted from AC to DC and back to AC, but since my days working in that industry, that might even have been eliminated by now because it was a TITANIC pain in the arse and expensive and inefficient to do.

  17. @pff

    I have to say I agree. At the end of the month it shows how much you’ve used and how much you owe. If you want to use less use less. If you want to see which things are using more or less power so you can use them less or turn them off then why not just turn them all off and only use them when you need them. If your trying to waste less then use it less or buy a more efficent version of it.

  18. Not sure about the whole atomic clock thing, but yeah the US is made up of 3 separate electric grids. If one strays away from 60 Hz too far someone notices. Thought they do drift away from each-other. Any power sharing between them has to be done through Asynchronous Power Transfer; AC-DC-AC conversions or new transformers that actively ‘rotate’ to match frequency. So I don’t think they are sync’d to anything.

    I really did frequency because I could, I already did the same thing in code for Power Factor. Though I can think of places it would be nice to monitor. Any places ‘off the grid’ like islands off the main land, or towns that run there own power system. Also, if you run your own generator, for full use or backup only.

  19. @Bill,

    I was planning on setting up something as such for my home, power clamps distributed around so I get more in depth view of the consumption around the house and Id use the blinks from the main meter to confirm the measurement, All of this hooked to my home server for graphs and more fun. Id be interested in the progress of your project, thanks :)

  20. Having the ability to turn off comments on HAD would be great. I always get sucked into these waste of time reads of people just completely ripping apart someone that took the time to build an interesting project. Regardless how good or bad a project is, they will always contain useful information that can only add to the community.

    Cool build build Bill, I recently purchased these clamps and was planning on playing around with them. This info should be useful.

  21. Actually, years ago they used to reduce the frequency below 60 Hz, supposedly to save energy, and then periodically run it above 60 Hz to make all the clocks accurate again. These days, in North America, whenever the error exceeds 10 seconds for the east, 3 seconds for Texas, or 2 seconds for the west, a correction of ±0.02 Hz (0.033%) is applied. Time error corrections start and end either on the hour or on the half hour. Smaller power systems may not maintain frequency with the same degree of accuracy.
    http://en.wikipedia.org/wiki/Utility_frequency
    58 Hz is 2 orders of magnitude beyond the grid correction frequency deviation factor.

  22. Wow. 800 feet from your apartment to the meter? I don’t recall all the calculations (0.253 Om/100ft for 14 gauge), but you must be losing a dollar or two a month in transition along that feeder line. Perhaps you can ask to have the meter moved closer to get those 2 Ohms back?

  23. What’s up with the ‘Recent posts’ thing under the post? It shows:
    ‘Bill Meara – 2010-12-22 03:38:54 (12/22/2010)
    em – 2010-12-21 23:56:55 (12/21/2010)
    reza – 2010-12-21 22:54:57 (12/21/2010)
    Rob Wentworth – 2010-12-21 22:08:04 (12/21/2010)
    Frode – 2010-12-21 18:00:47 (12/21/2010)’
    ??

  24. @therian; As I said, I measured with two devices, both got the same reading. I don’t think they need to be calibrated, and I am a metrologist, so I am confident they are good.

    @Rob W; Thanks for the info. I do, in fact get my power from ‘alternative energy’ sources, as indicated by my provider. (I better, since I pay a premium for RE energy).

  25. Yes goddamn heaters and ac units sucking up all your energy. Personally I only care about monitoring the power to my computers and upses are good for that.

    Electric companies are big retards, if you would put something like this near to the meter they would report you for stealing electricity…

    Some areas in Europe they use wireless meters with GSM communication. Which means nobody from the power company goes out to read your meter in all months. They most likely never go out unless it has some problem or have to be replaced and then they might even call you first.
    Consider the possibilities getting free energy on those areas :D The power companies are know that the rich fuks living there wouldnt bother stealing it.

  26. @pff

    Some studies have suggested that having continuous feedback on energy usage can help people reduce.

    For example, some cars have a display that shows the current gas usage. With detailed feedback on how much energy one is using at the moment, people can learn to modify their behavior to increase efficiency. (This can be taken to extremes with hypermiling, but just generally being aware can encourage more efficient use.)

    Energy feedback devices may have similar effects in home. It isn’t always obvious what the big users are, or how significant a particular action is. Having a meter right there showing the CO emissions, $ used, etc. can be a reminder, incentive and education tool.

    I haven’t seen studies that look at the long term impact (are the reductions transient, do they continue if monitoring removed, etc.)

  27. One advantage to the above that hasn’t been mentioned yet – there are ways to save energy in the home that will have 0 impact on usage of creature comforts or any changes to appliances and computing devices. 2 power clamps will allow you to redistribute the loads on the 2 phases of the AC coming in, thus reducing the imbalance. (I am talking about the common 2 phase 60hz 120v common in North America.) Ideally, all power in the house will only travel on the 2 phases, with absolutely no current flow on the neutral. In reality, especially due to changing loads in the home, this isn’t going to happen, but you can move loads from one phase to the other to get it as close as you can to balanced.

  28. I think the idea is awesome. Im in the process of building one now – Want a good reason?
    I rent a 2 car garage on a residential property which i use as a shop. I have very limited access to the house. The house tenants have the hydro in their names. There is no separate meter for me.
    I have been sampling the consumption with an amp probe while waiting on my clamps to arrive, which works, but its hard to access the info that i have written on the pad when im at home or …
    The how and why of it do not really matter, I think. Its a great project to do if it interests you. if not then why not shove off and read some spam?

Leave a 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.