Power amps: continuous output power measurements

[restorer-john]

I finally got my 300W test resistors. I can wire this thing for 8, 6, 4 and 3 ohms load, although I mostly will test 3 and 4 ohms I guess.

And, you can use it for heating your room in winter.

 

[Deleted member 49933]

I was thinking of a cheap way to measure my AVR power output, at minimum impedance, with speakers connected and if it goes into clipping at the listening level I use with movies, but without hearing protection from high SPL.

If I connect my AVR through a kill-a-watt meter and look at the idle power draw and then I play a 1 or 2 Hz full scale sine wave (with bass management disabled) and raise the volume slowly in 3dB steps, that should give me a rough idea, no?

 

[antcollinet]

I was thinking of a cheap way to measure my AVR power output, at minimum impedance, with speakers connected and if it goes into clipping at the listening level I use with movies, but without hearing protection from high SPL.

If I connect my AVR through a kill-a-watt meter and look at the idle power draw and then I play a 1 or 2 Hz full scale sine wave (with bass management disabled) and raise the volume slowly in 3dB steps, that should give me a rough idea, no?

Not even close.

 

[DonH56]

I was thinking of a cheap way to measure my AVR power output, at minimum impedance, with speakers connected and if it goes into clipping at the listening level I use with movies, but without hearing protection from high SPL.

If I connect my AVR through a kill-a-watt meter and look at the idle power draw and then I play a 1 or 2 Hz full scale sine wave (with bass management disabled) and raise the volume slowly in 3dB steps, that should give me a rough idea, no?

Some off-the-cuff thoughts:

1. Unless you know the efficiency of the AVR's amplifiers, voltage rails, and such, you will not be able to do the conversion from wall inlet power to output power. Even harder if it is a class G or H design with changing supply rails.
2. Such a low frequency is likely to fry your woofers as you increase power. At may not be the actual minimum impedance point; at port tune, the reverse (port) wave may actually "stop" the woofer, resulting in the low-impedance point.
3. Single-tone testing is very hard on amplifiers and speakers; usually it is best to do short (<1 s) bursts with plenty of cool-down time between (10-60 seconds).

Assuming you do not have an oscilloscope, a simple RMS digital multimeter (DMM) may provide a better reading of average power, though ignores the phase component of reactive power. A peak-hold RMS meter with a few hundred Hz bandwidth probably isn't too expensive. Hook it up to measure voltage and play a 1 kHz (or whatever) tone for a second, then read the answer. Convert to watts using an impedance plot for your speaker or using a 1 kHz tone and the nominal impedance (which is usually measured at 1 kHz): P = V^2/R (where R is the magnitude of the impedance).

 

[Deleted member 49933]

Convert to watts using an impedance plot for your speaker or using a 1 kHz tone and the nominal impedance (which is usually measured at 1 kHz): P = V^2/R (where R is the magnitude of the impedance).

I already measured max unclipped Vrms of my AVR with a multimeter, but I don't know if that output does not drop when speakers are connected.

 

[DonH56]

I already measured max unclipped Vrms of my AVR with a multimeter, but I don't know if that output does not drop when speakers are connected.

Well, it will drop, even if only a tiny amount but could be more. If you know the impedance of your speakers, buy a high-wattage load resistor of the right value, and measure using that to see how much it drops. Then you can use that number with the sensitivity of your speakers and distance from them to estimate your maximum SPL.

 

[Jukka]

This is a wrap for now. I've got :

• some interesting results
• one destroyed Hypex NC502MP
• burn scars on my fingers from touching the dummy load, that reached in excess of 150 degrees Celsius at the surface ......

I'm sorry I'm late, but I would have suggested some sort of active cooling to the dummy load. Like an floor standing fan and/or you could even dip one end of the cooling plate in water. With that much heat even heat radiation should be considered. The resistance of a resistor usually climbs when its temperature rises, and with such high temperature that might affect the results.

For comparison my water kettle, that I use to make myself tee, is rated for 1.5 kW. That boils 0.5 litres water in about 5 minutes. My electric cooking stove is something like 0.5-1 kW per plate.

In fact when building amps from DIY modules, it's on the builders responsibility to apply enough cooling. Active cooling on amp can sometimes be well justified.

 

[mhardy6647]

I'm sorry I'm late, but I would have suggested some sort of active cooling to the dummy load. Like an floor standing fan and/or you could even dip one end of the cooling plate in water. With that much heat even heat radiation should be considered. The resistance of a resistor usually climbs when its temperature rises, and with such high temperature that might affect the results.

As I have mentioned at some other point, somewhere or another on ASR -- an oil bath is the classic way to do it.
cf. the long-running Heathkit "Cantenna" kit (for amateur radio, but an AF version would be a pretty straightforward hack).

source: https://www.worldradiohistory.com/Archive-Catalogs/Consumer/Heathkit-1974-03.pdf

 

[DonH56]

As I have mentioned at some other point, somewhere or another on ASR -- an oil bath is the classic way to do it.
cf. the long-running Heathkit "Cantenna" kit (for amateur radio, but an AF version would be a pretty straightforward hack).
View attachment 388842

source: https://www.worldradiohistory.com/Archive-Catalogs/Consumer/Heathkit-1974-03.pdf

I had a power meter and Cantenna for years (had a D-104 mic instead of Heathkit), but sold them with my station when we moved and I left everything behind. Should have kept the Cantenna; I had made a second top for it with five-way binding posts to a parallel array of non-inductive power resistors to create an 8-ohm load. I have some of my old gear in boxes, somewhere. I kept thinking I'd renew my HAM license but never got around to it; didn't need another expensive hobby for which I had no time. Another regret is not turning in my 2nd class radiotelephone license; I passed the test but the 1st class exam was scheduled in a few weeks, so I held on to the paperwork so I could go 1st class. Never happened.

 

[mhardy6647]

I have always wanted a D-104. I do like the sound of those suckers*.

My father had a First Class FCC license, which he kept current for long after he actually stopped using it professionally.
He was very proud of it.

________________
* Suckers... Lollipop.

 

[DonH56]

I have always wanted a D-104. I do like the sound of those suckers*.

Oooooo, bad boy! No lollipop for you!

My father had a First Class FCC license, which he kept current for long after he actually stopped using it professionally.
He was very proud of it.

I would be too! Of course, I am still paying IEEE dues, and haven't really used the journals for several years. Hard to give it up. Might need it if I give some lectures at the local university, which is what I thought last year...

 

[Old_School_Brad]

So @StigErik do you have that .wav extraction uploaded somewhere? -For science, of course..

 

[Head_Unit]

I would have suggested some sort of active cooling to the dummy load. Like an floor standing fan and/or you could even dip one end of the cooling plate in water. With that much heat even heat radiation should be considered. The resistance of a resistor usually climbs when its temperature rises, and with such high temperature that might affect the results.

YES! Doing any longer duration power testing it is really incumbent on the tester to, at the test conclusion, quickly disconnect the load and measure the resistance to see if it has heated. This means testers really need a higher power rating than the test and lots of heatsinking and liquid nitrogen. Well, maybe a fan at least.

 

[StigErik]

So @StigErik do you have that .wav extraction uploaded somewhere? -For science, of course..

No. Sorry....