W A R N I N G !


W A R N I N G !

This page is full of non-facts and bullsh!t, (just like the internet and especially forums and other blogs), please do not believe entirely without exercising your intellect. Any resemblance to real things in reality is purely coincidental. You are free to interpret/misinterpret the content however you like, most likely for entertainment, but in no case is the text written on this blog the absolute truth. The blog owner and Blogger are not responsible for any misunderstanding of ASCII characters as facts. *cough* As I was saying, you are free to interpret however you like. *cough*

Tuesday, October 5, 2010

The $20 wattmeter

$18 from eBay including shipping, $2 for the UK plug bought from SELFFIX.

$0 for the time and anxiety spent on settings this up.

You see, when I bought this on eBay, I told the seller I wanted a type G plug for use in Singapore. Not only did it come without, the sticker on the back of the unit says "110V".

Now who in his right mind would plug this in.

But there are hints - hints that say this thing might be usable with both voltages.

Too impatient for the seller's reply, I decided to uncover the secret.


Finally the pic.


Inside - the chip at the top corner beside the crystal is the energy metering IC ADE7755 - quick Googling of its datasheet shows that it has no issues with working with 230V. Also, the way it measures power by multiplying the current and voltage that it measures means that it has no problem with accuracy with different voltages compared to one that uses only current to calculate power. Why measuring both current and voltage works, if you don't get it, ask your physics/electronics teacher.

The yellow capacitor is rated for 250VAC. There is a single diode hidden behind for a half-wave-rectifier. Also hidden behind the capacitor(s) is a U2 in TO-92 package. 'U' stands for integrated circuit, don't ask me why. This should be the switch-mode power supply (transformerless?).

One thing good about SMPS, is that it can work with a large range of input voltages while ensuring up-to-spec output voltage. With unregulated, changing input voltage changes output voltage in the same ratio and fries things. With linear regulation, the regulators have to drop the extra voltage and fry. With switching power supplies, the PWM duty cycle is reduced so the higher input voltage still results in the same output voltage.

Now the issue is, I can't see the markings on U2. I can only hope that it can withstand 230V. If I were the designer, I would use U2 that can withstand 230V and use it for all models of my product, which is only differentiated with a sticker. And the power plug. Instead of having two versions and needing more inventory.

It is a guess. But a calculated guess.

So I wired up the thing and plugged it in. Nothing exploded.

So it is working.

Now time to have some fun with it.

How accurate this is - with a table lamp (bulb to be precise) rated at 7W, I got between 6.9 to 7.2W. So it's fairly accurate.

These are the power consumptions of my laptop (T4200, 14-inch LED-backlit screen) in different conditions:


Should be 17.8/17.9, the LCD was transiting between the two.


The screen consumes around 1.9W (~2W), bravo.

Note that the liquid crystal part of this screen is "always-on" - I can still see the screen and it changing with the backlight off, but just very barely, and only in the right conditions. The "switch off screen" (Fn + F6) of my Acer laptop only switches off the backlight.

Still, 2W for backlight is worth commenting. (I'm not sure whether it is worth commending.)




Battery removed - I removed the battery and do Linpack with screen on again. The battery has no effect compared to natural fluctuations in wattage.


I played a 1080p H.264 file that was lagging my CPU with 100% load all the time - it used to be just barely able to play with stuttering at the more difficult parts, I think it is due to virus.

Edit: Turns out I disabled ffmpeg-mt for the decoder, and for some reason CoreAVC was disabled, so it was using some unknown leftover codec to decode, which is obviously slower than ffmpeg-mt and CoreAVC.

Edit2: I found the unknown codec - DivX decoder. Yes it is still alive and long forgotten. And for good because their encoder sucks and their decoder is so slow.

ffmpeg-mt is actually not so much slower compared to CoreAVC. Not bad for a free codec.

With easier videos (those that can be played on an Atom, without hardware accelerated H.264 decoding), it was either 23.5W or 24.5W.


Worst case scenario? Probably Linpack, CPU not undervolted, with screen at full brightness.

I need to learn to document things as I move along.




The power that the laptop uses varies probably +-0.5W, which means there could be up to 1W difference between measurements of doing the same things.

Now some other measurements in text... I'm too lazy for more pics

Zhaolu D2.5A idling - 10.4W
Zhaolu D2.5A playing - 11.7W

What? 10.4W doing nothing? No wonder it is this hot.

Sony SRS-D4 w/MX5021 satellites -
Idling - 6.7W
Playing - 6.8W

which is pretty much the same - the readings were fluctuating between those two values anyway. Typical of most amplifiers - you'll only see higher power draw at way higher output levels.

The subwoofer and satellites use less power than the source, bravo.

The computer is the most power-wasting alright; this is a laptop which consumes very little compared to desktops.

And, why bother with overpriced "small" low-power SFF systems when you can have comparable power draw for much more awesome performance, a screen, and a battery so you can bring it out, and is cheaper for the cost of everything?

No comments: