Hacking a Kill-a-Watt for very low power measurements

[NiHaoMike]

I actually do embedded software testing at a computer company, but that has to do with the EC (Embedded Controller) on modern motherboards, not low level power electronics control or monitoring. The use of embedded software elsewhere is just one of my interests and also a common interest among many of my friends.

[ICanHas]

because....

[Daox]

Settle down guys and treat each other with respect. I don't want to have to lock threads or ban people. Thanks.

[jeff5may]

Originally Posted by ICanHas

It's a false cause. You just pulled out what you think it is out of your rear, but you didn't prove the cause.

I didn't ask if you sell software retail. I asked if you have vested interest, financially, status, position, contract award, etc from software or embedded firmware, such as being a provider to embedded craptem based solution who indirectly derive profit from developing/coding or manipulating the software part of " embedded system "

I take offense to your statement/challenge/whatever it is.

Who are you? What gives you any right to judge anyone or accuse any of us of anything? Pah, I can't be bothered... Get out of my office!

There are a large number of individuals here with a wide range of skills and a broad level of collective education. A great deal of us have each other to thank for our own progress. Much respect has been earned and awarded between users for their accomplishments and contributions, and a "can do" environment has been fostered throughout the forums. Many of us (myself included) would not have attempted or had measurable amounts of success in our endeavors or projects if not for the guidance of other users.

I will admit wholeheartedly that I am a skull and crossbones pirate in the sense of what you speak. I will take what you have worked so hard to make a fortune and image off of, and twist and warp it into something it was never designed to do. I will share this with the world to no end for free until you stop me. If the factories in China can do it too, more power to them. If another million people in the world can grab some of this pirated crapola and enjoy just a little of the "first world" lifestyle, I consider my effort a success.

With this in mind, I challenge you to quit bashing the status quo, because it is blatantly obvious in the world around us. Instead, I dare you to change your vindictive and paranoid attitude, and actually contribute something useful, here and now.

This thread was originally about morphing a $20 kilowatt meter into something that will read low-power-draw devices or appliances to increase the functionality and useful range of the device. Many of us ecorenovators have more than one of these power loggers laying around or in use, and the mod can be done quickly and easily without much effort. As with any inexpensive instrument, high precision is not assumed.

I understand your statement that the same purpose can be met with an analog device, but why? You come off as a ham radio enthusiast stuck in the sixties. The reason things were done that way back then is because technology had not advanced to the level where any other means was economical.

Please adapt your reasoning to the current decade. The cheap, programmable convergence device is now in everyone's lap, pocket or purse. The idea that these millions of devices are ticking time bombs designed to ruin everyone's lives is absurd. The world did not implode with the year 2000 bug, nor will it if all the cell towers stop working tomorrow. If your stuff quits working tomorrow, you will have the rest of your life to teach us all how to make a custom collection of discrete, bulky, single-purpose devices that will do some of the things the chinese toy did. Pardon me if I don't hold my breath in suspense.

[ICanHas]

I take offense to your statement/challenge/whatever it is.

It was in response to NHM's unfounded and offensive accusation that Microsoft only secured software in response to "open source".

I understand your statement that the same purpose can be met with an analog device, but why? You come off as a ham radio enthusiast stuck in the sixties. The reason things were done that way back then is because technology had not advanced to the level where any other means was economical.

My issue is the misinformation in NHM's statement that a "multimeter MUST have DSP" implying that digital processing is a requisite to measure power.

Power instrumentation using four quadrant analog power measurement chip is still very much in use and in production in current time. They're not obsolete like vacuum tube controls. Analog is actually much faster and respond in real time. Something digital struggles to do. Analog front-end is still a very effective technology today.

bulky, single-purpose devices that will do some of the things the chinese toy did. Pardon me if I don't hold my breath in suspense.

Discrete components are actually far more easily repurposed. The modern "embedded systems" type design is far less flexible and difficult to tweak outside once it leaves the developer's hands. It's like a $2.99 calculator. There are no intuitive adjustments. it is "hidden" in the software. If you replace a component, the calibration parameters need to be readjusted. The instrument is scrap unless you know how to get into the embedded systems microcontroller software.

Like it or not, the "one chip design" is the least customization friendly and you really don't know exactly what's going on inside the silicon chip. When software or calibration becomes corrupt, you can't adjust it or use parts for anything. All you can do is turn it into hazmat e-waste

[NiHaoMike]

I didn't mean Microsoft improved security in response to open source, but rather the reliability of the OS. As in Windows ME crashing all the time (while Windows 2000 was quite reliable), but the home versions of Windows XP and later were seemingly just as reliable as the business versions. The main reason was because Linux was becoming usable as a desktop OS and as it was able to give home users a very reliable OS, Microsoft had to make consumer Windows reliable in order to counter that advantage.

It is possible to do power measurement in analog, as those classic analog watt hour meters show. It is just that in a portable instrument, doing it in digital makes a lot more sense. Given that just about everyone wants a digital readout instead of an analog dial, there's still going to be a microcontroller inside. And the reason why digital meters update slowly is so that the value is readable. The better digital meters with a bargraph actually respond faster than an analog meter, as they don't have to deal with inertia.

Hacking embedded systems can easily be a lot of fun! As an example, take a look at the WRT54G and how many hacks have been done on it.

[ICanHas]

Additionally, with nonlinear loads, a multimeter must have DSP in order to measure properly.

Proof to contrary here.





Those familiar with the arts of electronics circuitry thingies can see from the above explanation that the front-end known as the watt converter is ENTIRELY electronic and ANALOG.

Display is digital just like a common multimeter as there are A/D converters for the display output, but it's not just theoretical discussion.

What you'll see with many China made instruments is that accuracy specs are vague and only holds true under very narrow limits. For example, Kill-A-watt is only meant to be accurate with pure sine wave input waveform and it does not specify how it performs with spiky loads where peak to RMS ratio is very high.

Dismissing analog instruments as archaic and obsolete is narrow minded when many digital instruments have a lot of limitations such as power delivered through a dimmer, or at the OUTPUT end of PWM speed controller.

[NiHaoMike]

That's because they didn't have affordable high speed processors (where "high speed" is on the order of 50MHz or so) back then. Nowadays, 48MHz of CPU goes for well under a dollar in quantity.

Here's a teardown of a somewhat newer digital power meter:

http://www.youtube.com/watch?v=-6qopB8bd5w

[ICanHas]

So, if it was possible to accurately measure power without " DSP " back then, I can not see why it is no longer possible.

Looks like they still make the meter using the same tried and true analog front-end technology. DSP version is even more expensive.

[NiHaoMike]

It's still possible to do it in analog, but it's just not cost competitive with digital nowadays. On a fixed budget, digital does way better than analog for this application. It also allows features like long duration averaging that are difficult to do in analog. In particular, when dealing with time constants longer than a few seconds (plus high accuracy requirements), leakage currents become a big problem when designing analog circuits.

Pay more attention to the teardown. There is a lot of analog front end, but the real magic happens in the digital domain. The voltage and current are individually digitized, then some code in the processor convert them into useful measurements.