Power meter for audio dummy load

[nikifena]

Hi everyone. I'm making a dummy load power meter for my lab. There are five 20ohm 200W resistors in parallel. I'm planning to put everything in a box with a cheap digital thermometer and some kind of a power meter.
I want to have a quick power indicator working well with frequencies up to 10khz. I prefer to use something cheap instead of making it.
There are a many AC panel meters from Aliexpres, but I'm not sure how they work with higher frequencies than 60Hz.

Any suggestions?

Best Regards.
Niki

[Dave]

Those resistors are supposed to be mounted flat onto a metal enclosure (preferably with a blob of thermal compound in between) if you want them to handle 200W continuously. Stacked up like this, they're going to heat up reasonably quickly.
Have you done any thermal calculations on these?

Personally, I'd put the binding posts below the LCD, because otherwise cables are going to be obstructing the view of the display.

Not aware of any existing modules that would suit your purpose.
The simplest solution I can think of is a RMS to DC converter (AD, for example) and then a microcontroller for interpreting the results and doing calibration.
Another solution would be to put an audio ADC in there and process the samples with something like a Cortex M4. This approach would be more involved and require some DSP knowledge.

[Audioguru]

Your continuous full power amplifier tests might destroy some amplifiers that work fine playing voices or music because they are not designed for continuous full power.

[mk_]

something like that is calle VU-Meter. An active rectifier, some comparators driving some LEDs and thats it.
look at the famos LM3916 or similar, stackable up to 60db

[nikifena]

Hi guys. Thanks for the replays.

I'm looking something available on the market. Easy to implement into this project. My goal is a display with up to 5% error. So something made in China will do the job. Most of the units are shown for 50-60Hz and I'm not sure are they true RMS and what's their maximum supported frequency. I prefer to read true RMS power to get precise readings when I load the amplifier into the distortion region. I tried to find something but without success...

[David Hess]

200 watts into 4 ohms is 28 volts.  At that level, a simple average reading diode-capacitor rectifier can be used to feed a DC voltmeter with good accuracy except at low levels.  A switch can be used to select between peak and average measurements.

[nikifena]

They are five resistors in parallel. A total 4ohm @ 1000W

I will use this load mostly with amplifiers in the range of 200W so no drama with heating. And there will be a digital thermometer to see what's the temperature of the resistors.

Dear David. I'll try your idea. Maybe using something simple is the best idea instead of finding a more precise power meter.

 

[David Hess]

The complication is that a simple diode based average or peak detector will return the voltage or current and not the power but the same objection applies to any AC voltmeter.  To make a power meter, more is required.

[nikifena]

^ I know that!
A simple AC voltmeter can't be used into the entire 1kW range due to the log function of the power to the constant resistance. I'm going to buy one AC power meter for mains frequency to see how it works.


Best Regards.
Niki

[David Hess]

It is actually a little easier with analog meters where the scale can be made with the proper response to indicate watts.

[Datman]

They are five resistors in parallel. A total 4ohm @ 1000W
I will use this load mostly with amplifiers in the range of 200W so no drama with heating.

200W power on 500W or 1000W or 10000W resistors make 200W of heat!

[Datman]

You could rectify the signal by a silicon diode, then calculate and show power by Arduino on a LCD.

[David Hess]

You could rectify the signal by a silicon diode, then calculate and show power by Arduino on a LCD.

It is not *that* difficult to do the needed square function in the analog domain to directly drive a meter.  A multiplier is one of the easier ways which does not require temperature compensation but it can also be done with a voltage-to-frequency converter meaning it should be possible with a 555 timer and some cleverness.

An Arduino could either directly sample the signal or sample the precision rectified version like a multimeter does and compute peak, average, and RMS.  The RMS value is equal to the standard deviation.

[The Electrician]

nikifena, you might find this interesting: http://www.angelfire.com/electronic/funwithtubes/Audio_Power_Meter.html