measuring output power

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Thats easy. Measure the true RMS voltage across what ever load you want to test it on and then calculate the square of that voltage, then divide by the load amount.
Example: 40 volts RMS AC measured across 8 ohms is 1600 volts. Divide that by 8 to get 200 Watts RMS.

If anyone does it another way I'd like to know about it. I've always used the above to calculate it as I have several very accurate RMS AC voltmeters.

Mark
 
Nelson, I have only a multimeter and a transistor tester, building this amplifier was my first electronic DIY project. By the way thank you so much for sharing your ideas! There aren't alot of people out in the world so generous with knowledge versus making money off that knowledge.

Mark, that sounds simple enough. So I can simply use a resistor for the load? What rating does the 8 ohm resistor have to have so I do not burn it up? Thanks Mike
 
MikeZ said:
So I can simply use a resistor for the load? What rating does the 8 ohm resistor have to have so I do not burn it up? Thanks Mike [/B]


Euh? :rolleyes:

How about a big fat resistor which can take the abuse of your amp! So for a Aleph 2, at least one which can take 100W for some while! I use two 4 ohm wire wound resistor (8 ohm in serie, 4 ohm a piece, 2 ohm parallel!) which have bought at a dump store for US$1 a piece. I have put 400W (to test my power supply) on to them. After 5 minutes they started to glow, but they did not break down!

Edwin
 

PRR

Member
Joined 2003
Paid Member
> So I can simply use a resistor for the load?

Everybody does. Loudspeakers are not really resistors, but the resistor-load power is a good indication of speaker-load power and a lot simpler.

What rating does the 8 ohm resistor have to have so I do not burn it up?

Bigger than the expected power of the amplifier. Small amps work OK with the 8Ω 20Watt resistor that Radio Shack used to sell. Big power amps need BIG resistors: use arrays of 20W to 100W resistors. Hmmmm... the Aleph 0 is rated 75 watts, but may have a "soft" distortion rise, and may deliver more power before it clips. And some were rated more. Better find over 100 watts worth of resistors that add up to 8Ω.

For testing above 1KHz, ideally you must use "non-inductive" resistors, though I've never seen much difference with the usual wire-wounds at mid audio frequency.

You bring up power until the amplifier distorts "too much". Since you have only a multimeter, you won't know when it is distorting. A scope will show gross clipping or severe warping. A THD meter is necessary for low-THD readings. But you can get the "gross undistorted (several-percent) power" reading by wiring a speaker in parallel with a 0.1Ω resistor in series with your load resistor. You can hear the test-tone in the speaker. When the sound quality changes, gets rougher, back-off until it cleans-up, then measure the output voltage.
 
Lightbulb resistor will change too much because of temperature changes.

If you don't have big power resistors, just use some 10 or 20 smaller metal film or carbon resistors in parallel. In the past I just use e.g. 10 half Watt resistors in parallel, with the wires twisted together, soldered some leads to it and dropped the whole thing in a mug with water. It is amazing how many Watts you can dissipate with these small water cooled resistors, even 100W for some minutes. At a certain moment the water starts to boil, then get yourself a free refill with cold water. And it has low inductance too. ;)

Steven
 
Mike-

One good source is at ApexJr.com. Steve there has some 50W, 8 ohm resistors. Put two in series and parallel that with another similar pair and you have something good for 200W. That's $8 worth.

Put it on some cheap heatsinks and you've got a new piece of test equipment for your laboratory.
 
Something not clear to me.

With a tester and a couple of
resistors I'll be able to measure
the output power.
But how can I notice the clipping
without an oscilloscope?

Under clipping the voltage reading
increases if I increase the volume.
Only when I have a square wave
it stops to increase but, at that point
I probably blow something.

I think one can use the audio
board of the PC and some free
software to see the wave shape.
 
Soundcard oscilloscope

Hi fscarpa58,

PRR already mentioned that you can quite easily hear when a sine wave starts clipping. If you use a pure sine wave as input signal, you can detect less than 1% distortion easily, suddenly the character of the sound changes from a whistle to a buzz; that's the onset of clipping.

But, you are right, there are a lot of (free) oscilloscope programs that you can use with your PC soundcard, including tone generators and even spectrum analyzers. Via Google I found e.g. http://www.sumuller.de/audiotester/maine.htm, which will probably do all that you need. :cool:

Steven
 

PRR

Member
Joined 2003
Paid Member
> Can I use a lightbulb for the resistor?

I did it. Once.

Two 6V6 amplifier. Book says a pair of 6V6 make 10-15 watts. The amp had a 16Ω tap. I found that a 120V 60W lamp is about 16Ω on the ohm-meter. I expected 12 to 16 volts (9 to 16 watts).

I wind it up. Voltage was around 30V, indicating 50-60 watts IF that was really a 16Ω load. And the lamp glows! Dim red glow, but visible in room light.

A moment of thought tells you that, when running normally, white-hot, the resistance of a 120V 60W lamp must be about 240Ω, instead of the 16Ω measured cold. Most metals increase resistance when hot, and lamps get very hot.

I checked the AC current and it was around 0.5A when the voltage was 32V. So my "16Ω resistor" was really more like 60-70Ω. The underloaded 6V6 would make a lot of voltage (they were somewhat overloaded when working in the rated load) so the voltage would rise to 30+V before gross clipping. The actual power was 30V*0.5A= 15 watts.

Since the resistance of a lamp changes rapidly with voltage, they are nearly useless. You might get-by with a 120V lamp on a sub-Watt amplifier, but don't trust it.

Get real resistors. Don't run them too hot. If they boil spit, they are about as hot as they can be and still be pretty accurate.

As Steven says, a 10-pack or 100-pack of 2W resistors makes a really nice load.

I have done water-cooled. Don't try a 130W amplifier in a coffee mug. Don't expect the resistor to last long. Some cheap 10W resistors won't take much of this at all. I have an industrial 44W resistor: the lugs kept being eaten-away by electrolysis and I had to re-clean and re-solder leads, but it did last for decades and sometimes over 200W. Even a quart of water is not really enough at those power levels. Don't use metal or plastic jugs unless you can support the resistor well. Don't spill water in the amplifier. Generally it would be wiser to get a bunch of good 50W-100W power resistors on ceramic insulators on a board over the bench.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.