C-clamp Current Monitoring

[Hydronic] did some tests to make his own current sensor using a c-clamp wrapped with wire. He tried several different cores including an aluminum carabiner, the C-clamp, and what he calls a u-lock (removable chain link). There is some success here that could be improved with cleaner winding and by adjusting readings based on the length of wire used in the wrapping.

This did make us perk up a bit right off the bat. Reader [Mure], who tipped us off about this, suggested that this could be used to make your own Kill-a-Watt without including it in the circuit. We made the jump to house monitoring. We’d love to have a data tracker for our home circuits to curb wasteful energy use. Perhaps we’ll try to make our own sensors and produce a diy Cent-a-Meter.

24 thoughts on “C-clamp Current Monitoring

  1. Seems I remember a project where a coil of wire was used to light an LED if current was flowing. I think the idea was to be able to tell which breaker was tripped. I also think the wire was wrapped directly around the power lead (rather than around a toroid circling the lead). Does anyone else remember this?

  2. pretty cool- another use for what those of us in the hvac industry consider a “10-wrap”- such as used on the heat anticipator on older (read non-digital) to figure out amperage draw. 10 turns of the wire on your ammeter/amp-clamp and read amperage, then divide by ten- this was done back in the day as most amp-clamps ten to fifteen years ago didn’t read well at those lower amperages.

    i love seeing old (and possibly forgotten) methods and tricks-of-the-trade breath new life into newer projects.

  3. dI/dt isn’t a problem, since we are looking at AC. The derivative of sine is sine with inconsequential phase shift, after all.

    Seems to me the essence of improving this hack is finding a material with good magnetic permeability that is cheap and machinable. Imagine half-toroids with Lego-like bumps and depressions on the end faces. Wind some turns on each half; snap them together around the power conductor; connect coils in series; and off you go.

  4. One problem with using this as a Kill-A-Watt replacement is if you clamp over both conductors in a cord, the currents will cancel out. You need to split the cord so you’re only clamping over one conductor. If you have to unplug the load to plug in a split cord, you might as well just plug in a Kill-A-Watt.

  5. This device alone cannot replace a kill-a-watt because it does not measure voltage. Also, The eddy currents in that c-clamp probably create significant error. I would be especially concerned about phase shift. To measure power, one must make phase correct measurements of both current and voltage and multiply them in real time.

  6. I’m trying to make a web based power meter….
    Once I get it up and running, I’d like a place to host the data… any ideas. It would be nice to have a live graph as well as a access to historical data. thanks for all the comments. I like the LED idea as it should work with enough windings. the solid, ferrous magnetic core concentrates the induced current.

    Once the data is available manufacturers should be able to create algorithms to sell their energy savings devices and equipment; such as, what the payback on a new refrigerator would be based on estimated refrigerator run time….

  7. An important point about Current Transformers – they should *never* be operated with an open circuit secondary or very high voltages will be induced.

    Any stepdown tranny can be used as a CT by passing the load current through the low voltage side and loading the mains side with a suitable resistor. The current in what is now the secondary will equal the primary current divided by the turns (voltage) ratio. The load current should be no greater than the rating of the low voltage winding, which is a bit limiting.

    For example the 2155 (MM2002) LV multitap has some tappings only one volt apart giving a ratio of 240:1 so at the rated 1 amp the secondary current will be 1/240th or 4.17mA, and if loaded with two 120 ohm (5w wirewound) in series will produce one volt per amp.

    For higher powers a few turns of well-insulated wire can often be added around the outside of the winding through the spare window space (or the LV winding could be stripped off). To find the turns ratio, energise the tranny normally and accurately measure the voltage across the new winding. I have one of these in a box with plug and outlet so it can be simply placed in series with an appliance (fridge consumption tests).

    Since it is the *current* that is defined in the CT secondary you set the required voltage span with the load resistor value. This resistor must be *securely* attached to the secondary so there is no chance of it coming adrift in use (with resulting fireworks). It’s overkill but I use 5 watt wirewound.

  8. Instead of the aluminum or steel C clamp (or G clamp as it may be) you can get actual ferrous cores from an old TV. You may have to check several TVs before you can find one that is not glued in (make sure you are aware of the high voltage dangers that lurk inside a TV). These cores are made of two C shaped pieces that come together to form a square O shape. These will be superior to a steel clamp because there will be no losses due to eddy currents.

    If I may simplify what Roly said, the proper way to use these is to make the secondary a closed circuit and measure current. This way you are actually measuring the current of the circuit (divided by the number of turns of wire) and not dI/dt as you would if the circuit was left open and voltage measured. The way I did this was by connecting a 0.1 ohm resistor (a bunch of 1 ohmers in parallel), which almost makes it a closed circuit. Then run the measured voltage through an amplifier and use ohm’s law to find the current.

    In order to measure power you also need to be measuring voltage and compute the average of current*voltage over time, taking thousands of samples per second. Or, just use a specialized power measurement chip like the one mentioned in the instructibles article linked above. To correct an earlier statement, the current waveform is NOT sinusoidal. Electronic devices (switching power supplies) create significant current spikes.

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.