Can I measure AC vampire current this way?

(snip)

Yes, the watt hour meter on your house does a quite accurate job of totaling the actual energy you use from your power service. The old ones have a motor whose drove torque is produced by the instantaneous multiplication of voltage and current, while it is braked in proportion to its rotational speed. It may not handle microsecond pulses, but it does take phase shift and non sinusoidal waveforms into account.

More modern ones have microprocessors that total the product of high speed samples of voltage and current.

Neither will charge you for inductive or capacitive current, nor miss the energy a rectifier and capacitive filter draw from the service at the peaks of the voltage waveform.

You've missed several important points:

- Doing it manually can be dangerous.

- Doing it manually takes a lot more time and effort.

- Doing it manually has many traps

- The Kill-A-Watt gives you a LOT more useful info that is hard to get with just manual measurements.

How much energy have you wasted discussing this on this forum?

Geeze, just buy the Kill-A-Watt meter, they are under $20:

I believe some others have discussed that already.

Possibly not. Manual measurements may have a high degree of both absolute and relative error. Unless you know all the factors involved and understand exactly what's happening, you'll probably get caught out. For under $20 you can do it correctly and get a LOT more useful info safely.

Dave.

Bob Pease has a good article about how a watt meter works, and a working schematic.

OTOH, I concur with the folks who say a Kill-A-Watt is the way to go. Much more reliable, less dangerous, and quite fun.

Regarding how it works, all you need to do is to get a very low value shunt resistor, and then multiply the voltage across the resistor with the voltage across the device in series (in real time). That gives you instantaneous power. You then integrate that to get the result over time, which is of course energy. You need to scale the result to get it in kW hours. His meter given in the link above does it with an analog circuit, which is pretty cool. One can also do the math with a microcontroller, which is probably how the Kill-A-Watt does it.

Electricity is dangerous, of course, so if you do this, keep one hand in your back pocket at all times when you are near any live wires. The thing that kills is AC across your heart, which can induce fibrillation.

Regards, Bob Monsen

How do you account for phase errors, for instance when your DUT is a transformer?

petrus bitbyter

Thanks, Bob, but I'm unable to get to the site. I just get an unending series of ads. It says "Click here to go to electronic design, or wait 20 seconds." But no matter what I do, I just get the same ad and another countdown, and another "AD=1" added to the end of the address.

Nevermind. They want cookies.

I Googled:

kill-a-watt accuracy

and this Anandtech thread came up first:

which compared Kill-a-Watt readings for various things with readings on the same items from a Fluke 1735, which I gather is a VERY expensive watt meter, with not particularly good results. Included was:

Of course, even if it's wrong, it could still be useful if it's consistently wrong - if I could apply a fudge factor of some kind.

Anyway, I accept that the answer to my original question - can I get good measurements with a simple home-brew ciruit - is no, that doesn't work.

Download Opera and set it to discard ALL cookies at the end of a browser session.

Hmmm. Just tried the link and it counted *past* zero but eventually got to the article. Another MSIE-wired site?

I had session cookies enabled in IE7, but that wasn't enough. But when I added it to the allowed sites for hard drive cookies, it went right through. What was confusing was that usually a site will say at some point if cookies are required. This one didn't.

I don't want to discard all cookies. Some of them keep track of where I left off, and are useful. But I allow the specific sites I need, and just allow 1st party session cookies on everything else.

What I do in that case, and for sites where the cookie contains login credentials, is click to the cookie list (tools | advanced | cookies) and set the ones that I want to keep to be retained.

If you haven't tried Opera (or Firefox) give it (them) a whirl.