I've got one of those hooked up to a Loxjie A40 amp right now. Playing music at a level slightly louder than normal, it shows it is using 7.6 watts.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Fabrication of box containing AC power meter, AC volts meter and AC amps meter with AC outlet and power cord.
- Thread starter Doodski
- Start date
I've got one of those hooked up to a Loxjie A40 amp right now. Playing music at a level slightly louder than normal, it shows it is using 7.6 watts.
OP
- Thread Starter
- #22
Is the Amps metering fluctuating? if so what kind of delta?
Changing between .133 and .145. It stays on .133 for several seconds, changes to .145 for a couple of seconds and then back to .133.
OP
- Thread Starter
- #24
OK cool. What I am interested in with your providing the delta change is to see what the sample rate and resolution of a varying waveform is like when the meter gives us a readout. Interesting. Can you crank it a little to see how the delta responds then?
Cranked it to volume level 30 (way too loud for comfort). Amps changed to a low of .177 and a high of .189. Watts changed to 8.4.
OP
- Thread Starter
- #26
Hmmmz. I suspect the digital metering software and circuitry in the metering device is not fast enough/resolving enough for a musical source. So sine waves will be required for a steady state voltage and current.
OP
- Thread Starter
- #27
@Brian Hall thank you for taking those measurement from the device. Much appreciated.
AudiOhm
Senior Member
Purchased this on Amazon in 2021...
Ohms
Ohms
Of course it won't be... the power supply and its capacitors and stuff filter things both ways - It stops 50Hz of the mains from reaching your music, it also stops the fluctuations of our music from reading the mains. That's how your music continues playing when the AC is instantaneously at 0V.
In fact it is this filtering effect that makes the power factor an a$$ by making power flowing only during that 1ms capacitor recharging duration. Over 90% of the time your amp is not drawing power from the mains. (*Of course with APFC then PF becomes close to 1 and a lot of the discussion becomes simple)
OP
- Thread Starter
- #30
Yes, I was drooling on that device too and formulating a system to get it on a workbench with the metering gear. The issues with it is it needs a project box, a AC main output outlet, a extension cord for mains power and a bunch of labor to get everything mounted in a project box. These links in post #1 where also attractive till I was recommended to look to the Kill-a-Watt meter.>>>
OP
- Thread Starter
- #31
Yes, thanks for that techy info that I needed. The Watt meter also seems to update the display too slowly too.
restorer-john
Grand Contributor
It's only useful for power consumption on the supply side. For that purpose, it is excellent, accurate, cumulative and inexpensive.
OP
- Thread Starter
- #33
So basically for our needs one can run a 1 k Hz sine wave and set say an amps' power output to say 0.5W, 1W, 5W and maybe more to get repeatable reference numbers.
restorer-john
Grand Contributor
If you want to know how "efficient" your amplifier is, sure.
Set the amplifier to produce a specific output voltage (AC VRMS) across a known resistive load to produce a known power. Make that input continuous and then measure the input consumption in Watts/VA whatever. Do that for various (input levels)/outputs, ranging from idle through to rated/clipping power and drop all those numbers into a spreadsheet. Then plot consumed vs supplied power as an efficiency percentage.
In reality, you don't need many points. A bunch around idle to a few watts, then 2 or 3 up to and including clipping.
It's an exercise we all do at some stage, but it's not something you want to be doing every day.
There are smart power plugs that produce graphs with update rates of down to 1 second (but the cheap ones that update every 30s should be avoided).
We just need to leave that graph on, run the usual tests, look at the graph and we can roughly decipher what time was what test. Of course that would require the usual SINAD @ 5W to last for at least a few seconds to work, and I honestly have no idea how long that test is.
We just need to leave that graph on, run the usual tests, look at the graph and we can roughly decipher what time was what test. Of course that would require the usual SINAD @ 5W to last for at least a few seconds to work, and I honestly have no idea how long that test is.
OP
- Thread Starter
- #36
To have graphical plots is the 100% best goal for sure although is that a lot of work or is it pretty simple once the spreadsheet is made and then the variables are entered into the fields?
restorer-john
Grand Contributor
@amirm's FFT is 32k with 3 averages. So if he's sampling at 48kHz, it would be ~750mS per capture, so a total of around 2.25 seconds to get the numbers we see. If he's sampling faster (96/192kHz etc) the period is even less- like 200mS per capture and 600mS total. As far as I know, it's not a "one-shot" it's likely continuous up to and after the measurement, but you'd have to ask him. That's what I do, just continually monitor and either grab a screen or stop the aquisition.
I don't know what he's using as his FS for the 5W@4R plot, it's not shown, but with the FFT chopped off at 20kHz, it could be 44.1kHz or up.
At the end of the day, it's all about consistency. Then apples can be compared with one another.
Last edited:
Most Kill-a-Watt wallplugs also measures power factor... which is telling.
Would the use of such a 'consumer-grade' measuring device be considered ASR's... ummmmm... antithesis?
[vis-a-vis AP]
mhardy6647
Grand Contributor
- Joined
- Dec 12, 2019
- Messages
- 12,478
- Likes
- 27,423
Well, as I said elsewhere today (on roughly the same topic), NIST-traceable calibration would be nice...
OP
- Thread Starter
- #40
As soon as a calibration NIST certificate is completed and the label is signed and adhered to a device the price goes up considerably.Well, as I said elsewhere today (on roughly the same topic), NIST-traceable calibration would be nice...
Similar threads
